Original Article

Preoperative considerations and benefits of neoadjuvant chemotherapy: insights from a 12-year review of the Hong Kong Breast Cancer Registry

Hong Kong Med J 2023 Jun;29(3):198–207 | Epub 6 Apr 2023

https://doi.org/10.12809/hkmj219333

ORIGINAL ARTICLE CME

Preoperative considerations and benefits of neoadjuvant chemotherapy: insights from a 12-year review of the Hong Kong Breast Cancer Registry

Yolanda HY Chan, MB, BS, FHKAM (Surgery)¹; Carol CH Kwok, MB, ChB, FHKAM (Radiology)²; Desiree MS Tse, MPH, BA³; HM Lee, MPhil, BSc³; PY Tam, MMedSc, BSc³; Polly SY Cheung, MB, BS, FHKAM (Surgery)³

¹ Department of Surgery, Kwong Wah Hospital, Hong Kong SAR, China

² Department of Oncology, Princess Margaret Hospital, Hong Kong SAR, China

³ Breast Cancer Research Centre, Hong Kong Breast Cancer Foundation, Hong Kong SAR, China

Corresponding author: Dr Polly SY Cheung (pollycheung@hkbcf.org)

Full paper in PDF

Abstract

Introduction: Neoadjuvant chemotherapy (NAC) was initially used for locally advanced or inoperable breast cancers. Its extension to early disease has facilitated breast-conserving surgery (BCS). This study investigated the use of NAC in patients registered with the Hong Kong Breast Cancer Registry (HKBCR); it also assessed NAC effectiveness according to rates of pathological complete response (pCR) and BCS.

Methods: Records were retrieved from the HKBCR regarding 13 435 women who had been diagnosed with invasive breast cancer during the period of 2006 to 2017, including 1084 patients who received NAC.

Results: The proportion of patients treated with NAC nearly doubled from 5.6% in 2006-2011 to 10.3% in 2012-2017. The increase was most pronounced among patients with stage II or III disease. In terms of biological subtype, substantial increases in the receipt of NAC were evident among patients with triple-negative and human epidermal growth factor receptor 2 (HER2)–positive (non-luminal) tumours. The best rates of pCR were observed in patients with HER2-positive (non-luminal) [46.0%] tumours, followed by patients with luminal B (HER2-positive) [29.4%] and triple-negative (29.3%) tumours. After NAC, the rate of BCS was 53.9% in patients with clinical stage IIA disease, compared with 38.2% in patients with pathological stage IIA disease who did not receive NAC.

Conclusion: The use of NAC in Hong Kong increased from 2006 to 2017. The findings regarding rates of pCR and BCS indicate that NAC is an effective treatment; it should be considered in patients with stage ≥II disease, as well as patients with HER2-positive (non-luminal) or triple-negative breast cancers.

New knowledge added by this study

The use of neoadjuvant chemotherapy (NAC) in Hong Kong increased from 2006 to 2017.
Higher pathological complete response rates were detected in patients with human epidermal growth factor receptor 2–positive (non-luminal) and triple-negative tumours.
After treatment with NAC, greater proportions of patients with clinical stage IIA or IIB disease underwent breast-conserving surgery.

Implications for clinical practice or policy

Alterations in breast cancer biomarkers after NAC suggest that reassessments of residual tumour would provide useful guidance regarding further adjuvant therapy.
Under the care of a multidisciplinary team, patients with early breast cancer who have an appropriate indication should consider receiving NAC before surgery.

[Abstract in Chinese]

Introduction

Neoadjuvant chemotherapy (NAC)—chemotherapy delivered before definitive breast cancer surgery—was first described in the late 1970s as treatment for locally advanced (often inoperable) breast cancers; it was intended to reduce tumour size and facilitate surgery.1 Subsequently, the use of NAC has been extended to early operable breast cancers.2 3 4 5 This approach offers the advantages of down-staging the disease, potentially reducing the extent of surgery, and allowing breast-conserving surgery (BCS); in the current era of individualised treatment, it supports evaluations of therapeutic efficacy.2 6

There is evidence that NAC is equivalent to adjuvant chemotherapy in terms of preventing breast cancer recurrence.6 It demonstrated equal effectiveness in terms of disease-free survival and overall survival in the National Surgical Adjuvant Breast and Bowel Project B-18 trial.7 Furthermore, a recent meta-analysis by the Early Breast Cancer Trialists’ Collaborative Group showed no significant differences between NAC and adjuvant chemotherapy for distant recurrence, breast cancer mortality, or death from any cause.8

Here, we hypothesised that the use of NAC would change over time among patients with breast cancer in Hong Kong, considering its increasing acceptance as a treatment approach. Thus, the objectives of this study were to investigate the use of NAC over time in patients registered with the Hong Kong Breast Cancer Registry (HKBCR), and to assess the effectiveness of NAC among patients with breast cancer in Hong Kong according to rates of pathological complete response (pCR) and BCS. This study also evaluated alterations in breast cancer biomarkers, including oestrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and Ki-67 proliferation index.

Methods

Records were retrieved from the HKBCR regarding Hong Kong Chinese female patients who were diagnosed with invasive breast cancer in the period of 2006 to 2017. Patients were excluded for the following reasons: stage 0 or stage IV disease, missing or unknown information regarding surgery, and concurrent neoadjuvant endocrine treatment or NAC received outside Hong Kong (which may involve different clinical considerations).

Breast cancer was categorised into four biological subtypes based on clinicopathological criteria, in accordance with recommendations by the St Gallen 2013 Consensus Guideline.9 A cut-off of <14% reportedly has the strongest correlation with the gene-expression definition of the luminal A-like subtype; a cut-off of ≥14% is generally regarded as the threshold for a high Ki-67 proliferation index. Histological grade 3 was used as a surrogate indicator of the luminal B-like subtype if Ki-67 information was unavailable.10 Pathological complete response was defined as no histological evidence of malignancies (ypT0) or the presence of only in-situ residuals in breast tissue (ypTis) and complete disappearance of lymph node metastasis (ypN0) after surgery.11 The same definitions have been adopted by the MD Anderson Cancer Center,12 as well as the Austrian Breast & Colorectal Cancer Study Group.13

Ethics approval for this study has been obtained from six relevant approving bodies. Written informed consent for data collection was obtained during patient recruitment into the HKBCR, who were from 20 hospitals and 37 clinics (online supplementary Appendix). Patient demographics, pre-chemotherapy and post-chemotherapy disease staging, tumour characteristics, and prescribed chemotherapeutic agents were evaluated. The effectiveness of neoadjuvant chemotherapy was assessed in terms of the rates of pCR and BCS. Baseline tumour characteristics were analysed, including size, nodal stage, histological grade, Ki-67 level, hormone receptor status, and HER2 status.

Descriptive statistics were used to summarise demographic and clinical characteristics of patients. Continuous variables are shown as mean, standard deviation, and range; categorical variables are reported as frequency and percentage. Means were compared between groups using independent samples t tests. The Pearson Chi squared test was used to evaluate differences in pCR according to biological subtype and surgical approach. Data were analysed using SPSS (Windows version 22.0; IBM Corp, Armonk [NY], United States). All P values were derived from two-sided statistical tests, and P values <0.05 were considered statistically significant.

Results

Patient selection

In total, 13 990 patients with invasive breast cancer were initially screened for inclusion. After the exclusion of 555 patients, 13 435 patients (13 625 breast cancer cases) were included in this study (Fig 1). The NAC group comprised 1084 patients (1097 breast cancer cases) and the non-NAC group comprised 12 351 patients (12 528 breast cancer cases).

Figure 1. Flowchart of patient selection. The NAC group comprised 1084 patients (1097 breast cancer cases) and the non-NAC group comprised 12 351 patients (12 528 breast cancer cases)

Characteristics of patients who received neoadjuvant chemotherapy

In the NAC group, the median age was 49.7 years (interquartile range, 43.5-56.7; range, 21.9-81.6), and half of the patients (53.8%) were premenopausal. The median invasive clinical tumour size was 4.0 cm (range, 0.55-20.0). The patients’ clinical characteristics (eg, age, biological subtype, clinical tumour stage, nodal stage, and cancer stage) are shown in Table 1.

Table 1. Clinical characteristics of non-neoadjuvant chemotherapy and neoadjuvant chemotherapy cases in each cohort

Among the 13 625 breast cancer cases, 13.6% of affected patients aged <40 years were treated with NAC, compared with 8.0% and 1.9% of affected patients aged 40-69 years and ≥70 years, respectively (Table 1). The administration of NAC was positively associated with cancer stage at diagnosis: the proportion increased from 0.3% in patients with stage I disease to 26.9% among patients with stage III disease (Table 1). Furthermore, greater proportions of patients with luminal B (HER2-positive), HER2-positive (non-luminal), or triple-negative subtypes of breast cancer received NAC.

Use of neoadjuvant chemotherapy in two temporal cohorts

For the assessment of changes in NAC adoption, the 13 435 patients were divided into two groups according to the year of diagnosis: periods of 2006-2011 and 2012-2017. The proportion of patients treated with NAC nearly doubled from 5.6% in 2006-2011 to 10.3% in 2012-2017 (Table 1).

Further analysis indicated that the use of NAC was significantly increased in patients with stages II and III breast cancers, but not in patients with stage I breast cancer. It was most pronounced among patients with stages IIB (7.8% in 2006-2011 vs. 13.3% in 2012-2017) and III (20.7% vs. 32.6%) disease. An increase in the use of NAC was also observed in patients with all biological subtypes of breast cancer. In particular, substantial increases were observed among patients with triple-negative (6.4% vs. 14.3%), HER2-positive (non-luminal) [8.9% vs. 13.9%], and luminal B (HER2-positive) [8.0% vs. 18.9%] tumours (Table 1).

Regimens of neoadjuvant chemotherapy

Among the 1084 patients who received NAC, 353 were diagnosed with HER2-positive (non-luminal) cancer. Anti-HER2 agents were added to chemotherapy in 73.7% of these patients, and the proportions increased from 57.6% in 2006-2011 to 82.5% in 2012-2017; taxane-carboplatin-trastuzumab was the most frequently used regimen. In contrast, for patients with HER2-negative tumours or unknown HER2 status, NAC regimens most commonly consisted of anthracyclines (doxorubicin or epirubicin), administered in combination or sequentially with taxanes (paclitaxel or docetaxel).

Responses to neoadjuvant chemotherapy

Rates of pathological complete response

Two hundred and twenty-one (20.1%) of 1097 breast cancer cases treated with NAC achieved pCR in the breast and axillary lymph nodes. Subsequent analysis according to biological subtype revealed that outcomes were optimal in patients with HER2-positive (ER-negative and PR-negative) tumours, among which nearly half (46.0%) achieved pCR. Pathological complete response rates in luminal B (HER2-positive) and triple-negative subtypes were 29.4% and 29.3%, respectively; these were significantly higher than the rates in other hormone-positive subtypes (all P<0.05; Fig 2).

Figure 2. Proportions of breast cancer cases (n=1097) achieving pathological complete response according to biological subtype, among 1084 patients who received neoadjuvant chemotherapy

Factors significantly associated with pCR included ER/PR negativity and HER2 positivity. Within the HER2-positive population, pCR was more common for hormone receptor–negative tumours than for hormone receptor–positive tumours; it was also more common in patients who received trastuzumab. Other factors (eg, age, menopausal status, clinical tumour and nodal stages, ER status, and Ki-67 proliferation index) did not appear to influence the achievement of pCR.

Rates of breast-conserving surgery

Figure 3 shows the proportions of patients treated with NAC who subsequently underwent different types of breast surgery, categorised according to clinical cancer stages. Patients with clinical stage IIA disease were most likely to switch from mastectomy to BCS after NAC; 53.9% underwent BCS after NAC, compared with 38.2% of patients with stage IIA disease who did not receive NAC. The second highest proportion was observed among patients with clinical stage IIB disease, 38.3% of whom underwent BCS after NAC. Even among patients with clinical stage III disease, 14.1% underwent BCS after NAC. Significant differences in the rate of BCS were also observed between the NAC and non-NAC groups in patients with stages IIA (P=0.02) and IIB (P=0.031) disease.

Figure 3. Types of surgery in neoadjuvant chemotherapy (NAC) [n=1097] and non-NAC groups (n=12 528) according to cancer stage

Alterations in breast cancer biomarkers

Biomarkers were compared between diagnostic core biopsies and final surgical specimens. Excluding the 221 patients who achieved pCR after NAC, 844 breast specimens with residual tumours were evaluated after final surgery. Patients without data regarding biomarkers in either pre-chemotherapy or post-chemotherapy or both were excluded from this analysis. Alterations in ER, PR, and HER2 statuses after NAC are shown in Table 2. Most patients had no change in their ER status, but 7.6% switched from positive to negative or from negative to positive. With respect to PR status, a shift occurred in 17.4% of patients, and a shift in HER2 status was detected in 10.9% of patients. More than one-fifth (21.3%) of patients with residual tumours had a change in at least one receptor status after NAC. Ki-67 proliferation index was also evaluated; among the 297 cases assessed, 131 (44.1%) showed alterations after NAC.

Table 2. Changes in breast cancer biomarkers after neoadjuvant chemotherapy

Discussion

Use of neoadjuvant chemotherapy

During the early phase of the study period, a multidisciplinary approach was not widely used for breast cancer management; thus, most treatment decisions were based on the discretion of the attending surgeon or oncologist. Nevertheless, locally advanced diseases and hormonal receptor–negative tumours were generally the targets of NAC. Over time, NAC has been increasingly accepted, as shown in updates of various national and international guidelines (eg, National Comprehensive Cancer Network guidelines14 and European Society of Medical Oncology guidelines15). This inclination clearly contributed to the substantial increase in NAC use during the periods analysed in this study: from 5.6% in 2006-2011 to 10.3% in 2012-2017.

The increased use of NAC was mainly attributed to advancements in translational research, along with new evidence from clinical trials that have led to a better understanding of breast cancer biology and the establishment of tumour biology–based targeted treatments.16 After the expansion of its use in adjuvant therapy, trastuzumab was first registered for use as neoadjuvant therapy for breast cancer in 2006 under the Department of Health in Hong Kong. Its entry into the Hospital Authority Drug Formulary soon followed, and it was included in the safety net enlistment by 2009. This timeframe suggests that the drug has become accessible to a much broader spectrum of patients under the care of public sector hospitals in Hong Kong; it is also compatible with the considerable increase in use of trastuzumab over time. In our dataset, among patients with HER2-positive (non-luminal) tumours, the proportion of patients using anti-HER2 regimens in neoadjuvant therapy increased from 57.6% in 2006-2011 to 82.5% in 2012-2017.

Pathological complete response

Neoadjuvant trials allow rapid assessment of drug efficacy; they can accelerate the development and approval of treatments for early breast cancer. Pathological complete response has been proposed as a surrogate endpoint for predictions of long-term clinical benefit.17 Although it is difficult to compare outcomes among trials and individual series because of heterogeneity in terms of study design and patient populations, the results of some meta-analyses have suggested that the achievement of pCR after NAC is a predictor of overall survival, disease-free survival, and relapse-free survival.18

Our results are consistent with findings by von Minckwitz et al11 and the Collaborative Trials in Neoadjuvant Breast Cancer (CTNeoBC) meta-analysis,17 which concluded that frequency of pCR was low in patients with low-grade, hormone receptor–positive tumours, whereas it was much higher among patients with more aggressive subtypes (ie, triple-negative and HER2-positive [non-luminal] tumours). Overall, these data suggest that the underlying molecular subtypes influence the rates of pathological responses. Further improvements in the rate of pCR have been observed in cases of HER2-positive (non-luminal) tumours treated with dual anti-HER2 targeted agents, as well as cases of triple-negative breast cancer treated with platinum and immunotherapy. Moreover, trials have also been done or in progress to evaluate the need for additional chemotherapy in selected patients with residual disease after NAC; the results of those trials are expected to provide further insights regarding treatments for further improving survival outcomes in neoadjuvant setting.18 19 20

Standard prognostic indicators, such as tumour size at the time of surgical resection or the number of involved lymph nodes, are no longer applicable in the neoadjuvant setting; systemic therapy often down-stages the disease and may lead to eradication. There is increasing evidence that the tumour response to NAC can facilitate prognostic predictions. In the multidisciplinary management of breast cancer, the identification of prognostic variables for patients receiving NAC can help to determine whether additional therapy is warranted. Given the strong support for an association between prognosis and clinicopathological features in the neoadjuvant setting, clinicians may be able to avoid additional interventions after surgery (e.g., additional chemotherapy) in patients who are otherwise considered high risk at initial presentation since pCR has been achieved. This is because although HER2-positive and triple negative breast cancers carry poor prognosis, these tumours have higher pCR rates after NAC, and pCR in HER2-positive (non-luminal) and triple-negative tumours was associated with excellent prognosis.11 17 21

Breast-conserving surgery

Quality of life–focused research has shown that body image scores are significantly better among patients who undergo BCS than among patients who undergo mastectomy. Patients who undergo BCS are less worried about their appearance, have more freedom in their choice of clothing, feel less upset about changes in their bodies, and feel more accepted by their partners.22 These findings reinforce the benefits of NAC for breast cancer in terms of down-staging the disease, increasing resectability, and enhancing BCS eligibility among patients who would otherwise require mastectomy. Furthermore, a systematic review of NAC for operable breast cancer revealed that the mastectomy rate was lower among patients who received NAC than among patients who underwent surgery prior to adjuvant chemotherapy (relative risk=0.71; 95% confidence interval [CI]=0.67-0.75); the use of NAC did not hinder local control (hazard ratio=1.12; 95% CI=0.92-1.37).23 Long-term follow-up analyses also showed that preoperative chemotherapy increased rates of BCS without increasing the rates of locoregional recurrence.24 25 In a previous study in Hong Kong, univariate analysis revealed that patients who achieved pCR after NAC had a higher likelihood of successful BCS (P=0.028). Pre-chemotherapy disease staging (P=0.001) and tumour size (P=0.005) were also important factors that influenced successful conversion to BCS.5

However, a recent meta-analysis by the Early Breast Cancer Trialists’ Collaborative Group showed that, compared with adjuvant chemotherapy, NAC was associated with more frequent local recurrence; the 15-year rates of local recurrence were 21.4% for NAC and 15.9% for adjuvant chemotherapy (rate ratio=1.37; 95% CI=1.17-1.61; P=0.0001).8 Thus, continued follow-up of patients registered in the HKBCR and updates will provide important insights with respect to NAC on long-term outcomes.

Alterations in breast cancer biomarkers

Neoadjuvant chemotherapy can cause changes in ER, PR, and HER2 statuses, as well as the Ki-67 level, in patients with invasive breast cancer.26 27 A possible explanation for this phenomenon is that chemosensitive cancer cells are destroyed by chemotherapy, whereas chemoresistant cells survive; such a change could alter the receptor status. Furthermore, because ER, PR, and HER2 are highly interdependent, a change in one receptor could lead to changes in the other receptors.28 A systematic review showed that the rates of ER and/or PR discordance range from 2.5% to 51.7%; among patients who received NAC combined with trastuzumab, up to 43% exhibited a switch to HER2 negativity.29

Thus far, there are only limited data regarding the prognostic value of changes in biomarkers after NAC among patients with breast cancer.28 Several groups have reported that a switch from negative to positive status (for ER, PR, or HER2) is associated with better overall survival.30 31 Additionally, outcomes are better among patients with stable hormone receptor status profiles than among patients with altered profiles.32 Notably, Guarneri et al33 reported that patients with loss of HER2 overexpression tended to have a greater risk of relapse, compared with patients who remained HER2-positive; in contrast, a decrease in Ki-67 expression after NAC was reportedly associated with better outcomes.34

Because of the above observations, biomarkers and Ki-67 levels should be retested after NAC. Such retesting is particularly important for tumours that were ER/PR-negative and/or HER2-negative before treatment because a shift to a positive status would indicate a need for endocrine therapy and/or trastuzumab. The results of these changes may influence clinical decisions regarding subsequent treatment and help to identify patients with better outcomes after NAC.28 35

Limitations

This study had several limitations. First, it was a retrospective analysis and the earliest records in the database were incomplete; the missing information particularly affected breast cancer biomarkers, and Ki-67 was not routinely tested in Hong Kong public hospitals. Second, selection bias may have been present because the receipt of NAC was largely dependent on surgeon assessment and patient preference. In recent years, the potential for such bias has decreased because multidisciplinary management of breast cancer is gradually becoming the preferred approach. Considering the complexities of treatment planning, monitoring, and evaluation, decisions regarding preoperative systemic therapy require input from surgeons, oncologists, radiologists, and pathologists. Of note, the comparison of rates of surgery types between NAC and non-NAC groups can only be regarded as approximation, as assignment of patients into these two groups is not randomised; furthermore, clinical stages may differ from pathological stages, thus they may not be comparable.

Conclusion

Changes in the clinical management of breast cancer led to increased use of NAC in Hong Kong during the period of 2006 to 2017. Neoadjuvant chemotherapy was effective in tumour down-staging; one-fifth of patients subsequently achieved pCR in the breast and axillary lymph nodes. In particular, higher rates of pCR were detected in HER2-positive (non-luminal) and triple-negative subtypes. After NAC, greater proportions of patients with clinical stage IIA or IIB disease underwent BCS. Currently, post-NAC adjustments to treatment are based on whether pCR has been achieved. In the future, alterations in breast cancer biomarkers after NAC may provide useful guidance regarding further adjuvant therapy. The indications for NAC have expanded from the treatment of locally advanced breast cancers (to facilitate surgery) to the down-staging of early disease, thereby facilitating BCS. Under the care of a multidisciplinary team, patients with early breast cancer who have an appropriate indication should consider receiving NAC before surgery. Further studies are warranted to evaluate the benefits of individual NAC regimens.

Author contributions

Concept or design: PSY Cheung.
Acquisition of data: DMS Tse, HM Lee, PY Tam.
Analysis or interpretation of data: All authors.
Drafting of the manuscript: YHY Chan.
Critical revision of the manuscript for important intellectual content: PSY Cheung, CCH Kwok.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

All authors have disclosed no conflicts of interest.

Acknowledgement

The authors thank all patients who have joined the Hong Kong Breast Cancer Registry (HKBCR), as well as the research staff who have participated in data collection from 20 hospitals and 37 clinics throughout the territory (online supplementary Appendix). The authors also acknowledge the following steering committee members who provided guidance for the development of the HKBCR: Dr Sharon Wing-wai Chan (United Christian Hospital), Dr Wai-ka Hung (Pedder Clinic), Dr Lawrence Pui-ki Li (Alpha Oncology Centre), and Dr Chun-chung Yau (Hong Kong Sanatorium & Hospital).

Funding/support

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Ethics approval

Ethics approval for this study has been obtained from the following six approving bodies:
1. The Joint Chinese University of Hong Kong–New Territories East Cluster Clinical Research Ethics Committee, Hong Kong (Ref No.: CRE-2009.037)
2. Kowloon West Cluster Research Ethics Committee, Hospital Authority, Hong Kong (Ref No.: KW/EX/08-090)
3. Research Ethics Committee (Kowloon Central/ Kowloon East), Hospital Authority, Hong Kong (Ref No.: KC/KE-09-0013/ER-3)
4. Hong Kong East Cluster Research Ethics Committee, Hospital Authority, Hong Kong (Ref No.: HKEC-2010-004)
5. New Territories West Cluster Clinical & Research Ethics Committee, Hospital Authority, Hong Kong (Ref No.: (8) in NTWC/CREC/866/10)
6. Institutional Review Board of the University of Hong Kong/Hospital Authority Hong Kong West Cluster, Hong Kong (Ref No.: UW 09-378)

Written consent was also obtained from all patients in the study who were recruited from the participating hospitals and clinics.

References

1. Rubens RD, Sexton S, Tong D, Winter PJ, Knight RK, Hayward JL. Combined chemotherapy and radiotherapy for locally advanced breast cancer. Eur J Cancer (1965) 1980;16:351-6. Crossref

2. Thompson AM, Moulder-Thompson SL. Neoadjuvant treatment of breast cancer. Ann Oncol 2012;23 Suppl 10:x231-6. Crossref

3. Gampenrieder SP, Rinnerthaler G, Greil R. Neoadjuvant chemotherapy and targeted therapy in breast cancer: past, present, and future. J Oncol 2013;2013:732047. Crossref

4. Vugts G, Maaskant-Braat AJ, Nieuwenhuijzen GA, Roumen RM, Luiten EJ, Voogd AC. Patterns of care in the administration of neo-adjuvant chemotherapy for breast cancer. a population-based study. Breast J 2016;22:316-21. Crossref

5. Man VC, Cheung PS. Neoadjuvant chemotherapy increases rates of breast-conserving surgery in early operable breast cancer. Hong Kong Med J 2017;23:251-7. Crossref

6. Mauri D, Pavlidis N, Ioannidis JP. Neoadjuvant versus adjuvant systemic treatment in breast cancer: a metaanalysis. J Natl Cancer Inst 2005;97:188-94. Crossref

7. Rastogi P, Anderson SJ, Bear HD, et al. Preoperative chemotherapy: updates of National Surgical Adjuvant Breast and Bowel Project Protocols B-18 and B-27. J Clin Oncol 2008;26:778-85. Crossref

8. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Long-term outcomes for neoadjuvant versus adjuvant chemotherapy in early breast cancer: meta-analysis of individual patient data from ten randomised trials. Lancet Oncol 2018;19:27-39.

9. Goldhirsch A, Winer EP, Coates AS, et al. Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol 2013;24:2206-23. Crossref

10. Curigliano G, Burstein HJ, Winer EP, et al. De-escalating and escalating treatments for early-stage breast cancer: the St. Gallen International Expert Consensus Conference on the Primary Therapy of Early Breast Cancer 2017. Ann Oncol 2017;28:1700-12. Crossref

11. von Minckwitz G, Untch M, Blohmer JU, et al. Definition and impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol 2012;30:1796-804. Crossref

12. Green MC, Buzdar AU, Smith T, et al. Weekly paclitaxel improves pathologic complete remission in operable breast cancer when compared with paclitaxel once every 3 weeks. J Clin Oncol 2005;23:5983-92. Crossref

13. Steger GG, Greil R, Lang A, et al. Epirubicin and docetaxel with or without capecitabine as neoadjuvant treatment for early breast cancer: final results of a randomized phase III study (ABCSG-24). Ann Oncol 2014;25:366-71. Crossref

14. National Comprehensive Cancer Network. NCCN Guidelines. Breast Cancer. Available from: https://www.nccn.org/professionals/physician_gls/. Accessed 25 Jan 2021.

15. European Society for Medical Oncology. ESMO clinical practice guidelines: breast cancer. Early breast cancer. Available from: https://www.esmo.org/guidelines/breast-cancer. Accessed 25 Jan 2021.

16. Dent S, Oyan B, Honig A, Mano M, Howell S. HER2-targeted therapy in breast cancer: a systematic review of neoadjuvant trials. Cancer Treat Rev 2013;39:622-31. Crossref

17. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet 2014;384:164-72. Crossref

18. Masuda N, Lee SJ, Ohtani S, et al. Adjuvant capecitabine for breast cancer after preoperative chemotherapy. N Engl J Med 2017;376:2147-59. Crossref

19. von Minckwitz G, Huang CS, Mano MS, et al. Trastuzumab emtansine for residual invasive HER2-positive breast cancer. N Engl J Med 2019;380:617-28. Crossref

20. ClinicalTrials.gov, US National Library of Medicine. Platinum in treating patients with residual triple-negative breast cancer following neoadjuvant chemotherapy. Available from: https://clinicaltrials.gov/ct2/show/NCT02445391?term=NCT02445391&draw=2&rank=1. Accessed 17 Mar 2023.

21. Waljee JF, Newman LA. Neoadjuvant systemic therapy and the surgical management of breast cancer. Surg Clin North Am 2007;87:399-415, ix. Crossref

22. Ng ET, Russell ZA, Tran BX, et al. Comparing quality of life in breast cancer patients who underwent mastectomy versus breast-conserving surgery: a meta-analysis. Int J Environ Res Public Health 2019;16:4970. Crossref

23. Mieog JS, van der Hage JA, van de Velde CJ. Neoadjuvant chemotherapy for operable breast cancer. Br J Surg 2007;94:1189-200. Crossref

24. van der Hage JA, van de Velde CJ, Julien JP, Tubiana-Hulin M, Vandervelden C, Duchateau L. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001;19:4224-37. Crossref

25. Zhou X, Li Y. Local recurrence after breast-conserving surgery and mastectomy following neoadjuvant chemotherapy for locally advanced breast cancer—a meta-analysis. Breast Care (Basel) 2016;11:345-51. Crossref

26. Wu YT, Li X, Lu LJ, et al. Effect of neoadjuvant chemotherapy on the expression of hormone receptors and Ki-67 in Chinese breast cancer patients: a retrospective study of 525 patients. J Biomed Res 2017;32:191-7. Crossref

27. Shaaban AM, Provenzano E. Receptor status after neoadjuvant therapy of breast cancer: significance and implications. Pathobiology 2022;89:297-308. Crossref

28. Trifunovic J, Memisevic N, Nikolin B, Salma S, Dugandzija T, Vidovic V. Modulatory effect of neoadjuvant chemotherapy on the prognosis of patients with breast cancer. J BUON 2017;22:638-43.

29. van de Ven S, Smit VT, Dekker TJ, Nortier JW, Kroep JR. Discordances in ER, PR and HER2 receptors after neoadjuvant chemotherapy in breast cancer. Cancer Treat Rev 2011;37:422-30.Crossref

30. Tacca O, Penault-Llorca F, Abrial C, et al. Changes in and prognostic value of hormone receptor status in a series of operable breast cancer patients treated with neoadjuvant chemotherapy. Oncologist 2007;12:636-43. Crossref

31. Hirata T, Shimizu C, Yonemori K, et al. Change in the hormone receptor status following administration of neoadjuvant chemotherapy and its impact on the long-term outcome in patients with primary breast cancer. Br J Cancer 2009;101:1529-36. Crossref

32. Yang L, Zhong X, Pu T, Qiu Y, Ye F, Bu H. Clinical significance and prognostic value of receptor conversion in hormone receptor positive breast cancers after neoadjuvant chemotherapy. World J Surg Oncol 2018;16:51. Crossref

33. Guarneri V, Dieci MV, Barbieri E, et al. Loss of HER2 positivity and prognosis after neoadjuvant therapy in HER2-positive breast cancer patients. Ann Oncol 2013;24:2990-4. Crossref

34. Chen C, Zhang Y, Huang Z, Wu J, Huang W, Zhang G. Decrease in the Ki67 index during neoadjuvant chemotherapy predicts favorable relapse-free survival in patients with locally advanced breast cancer. Cancer Biol Med 2019;16:575-86. Crossref

35. Dede DS, Gumuskaya B, Guler G, Onat D, Altundag K, Ozisik Y. Evaluation of changes in biologic markers ER, PR, HER 2 and Ki-67 index in breast cancer with administration of neoadjuvant dose dense doxorubicin, cyclophosphamide followed by paclitaxel chemotherapy. J BUON 2013;18:366-71.

Public awareness of preventive measures against COVID-19: an infodemiology study

Hong Kong Med J 2023 Jun;29(3):214–23

https://doi.org/10.12809/hkmj219556

ORIGINAL ARTICLE

Public awareness of preventive measures against COVID-19: an infodemiology study

Alex Mok, MB, ChB¹; Oliver OY Mui, MB, ChB¹; KP Tang, MB, ChB¹; WY Lee, MB, ChB¹; CF Ng, MD, FRCSEd (Urol)¹; Sunny H Wong, MB, ChB, DPhil (Oxon)²; Martin CS Wong, MD, MB, ChB^3,4; Jeremy YC Teoh, MB, BS, FRCSEd (Urol)^1,5

¹ SH Ho Urology Centre, Department of Surgery, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China

² Department of Medicine and Therapeutics, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China

³ The Jockey Club School of Public Health and Primary Care, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China

⁴ Centre for Health Education and Health Promotion, Faculty of Medicine, The Chinese University of Hong Kong, Hong Kong SAR, China

⁵ Office of Global Engagement, The Chinese University of Hong Kong, Hong Kong SAR, China

Corresponding author: Prof Jeremy YC Teoh (jeremyteoh@surgery.cuhk.edu.hk)

Full paper in PDF

Abstract

Introduction: The coronavirus disease 2019 (COVID-19) pandemic has led to an increase in global awareness of relevant public health preventive measures. This awareness can be explored using online search trends from major search engines, such as Google Trends. We investigated the relationship between public awareness of preventative measures and progression of the COVID-19 pandemic.

Methods: Search data for five queries (‘mask’, ‘hand washing’, ‘social distancing’, ‘hand sanitizer’, and ‘disinfectant’) were extracted from Google Trends in the form of relative search volume (RSV). Global incidence data for COVID-19 were obtained from 1 January to 30 June 2020. These data were analysed and illustrated using a global temporal RSV trend diagram, a geographical RSV distribution chart, scatter plots comparing geographical RSV with average number of daily cases, and heat maps comparing temporal trends of RSV with average number of daily cases.

Results: Global temporal trends revealed multiple increases in RSV, associated with specific COVID-19–related news events. The geographical distribution showed top regions of interest for various preventive measures. For the queries ‘mask’, ‘hand washing’, ‘hand sanitizer’, and ‘disinfectant’, heat maps demonstrated patterns of early RSV peaks in regions with lower average number of daily cases, when the temporal element was incorporated into the analysis.

Conclusion: Early public awareness of multiple preventive measures was observed in regions with lower average number of daily cases. Our findings indicate optimal public health communication regarding masks, hand washing, hand sanitiser, and disinfectant in the general population during early stages of the COVID-19 pandemic. Early public awareness may facilitate future disease control efforts by public health authorities.

New knowledge added by this study

This study focused on the importance of early public awareness in controlling coronavirus disease 2019 (COVID-19); this effect was not extensively investigated in previous studies.
Early awareness trends among regions with lower average number of daily cases were illustrated using heat maps for the queries ‘mask’, ‘hand washing’, ‘hand sanitizer’, and ‘disinfectant’.
In contrast to prior infodemiology studies, this study used a global online approach and focused on specific preventive measures recommended by the World Health Organization.

Implications for clinical practice or policy

This study revealed correlations between regions with low average number of daily cases of COVID-19 and early public awareness of multiple preventive measures (ie, ‘mask’, ‘hand washing’, ‘hand sanitizer’, and ‘disinfectant’). Health policies should seek to promote these preventive measures among the general public, which could help to slow the spread of disease. Furthermore, early public awareness may help public health authorities to control future global public health crises.
This study also investigated the effects of authorities, public figures, and social media on public health awareness. Future healthcare policies should consider these factors to effectively promote correct public health information among the general public.

[Abstract in Chinese]

Introduction

Coronavirus disease 2019 (COVID-19) began in December 2019 in Wuhan, China, and became a public health crisis affecting millions of people worldwide.1 On 11 March 2020, the World Health Organization (WHO) declared that COVID-19 constituted a pandemic2; by 1 September 2020, the total number of confirmed COVID-19 cases had exceeded 26 million, with over 800 000 deaths.3

Accordingly, the WHO issued recommendations to the general public with the goal of reducing community transmission of COVID-19. These recommended preventive measures included the use of masks in specific situations as well as hand washing, social distancing, and various other disease prevention strategies.4 In the early and middle of 2020, there was no specific treatment to cure the aggressively spreading virus; thus, preventive measures and public awareness of such information had important roles in the formulation of public health policies.

Because of technological advancements in recent decades, the internet has become a convenient and effective channel for providing readily accessible and up-to-date public health information to members of the general public. In the era of big internet data, infodemiology—an emerging area of science that explores the distribution and determinants of information in electronic media—has been implemented in multiple areas of modern medicine.5 The analysis of large amounts of internet data enables researchers to identify trends in online search behaviour; this information can be used to analyse relationships among public health communication, public awareness, and the progression of disease outbreaks. Indeed, search trends from major search engines (eg, Google) have been extensively used in infodemiology and infoveillance studies focused on outbreak patterns and public awareness,6 particularly with respect to the Ebola,7 8 9 H1N1 influenza,10 11 and Zika12 13 14 viruses.

In the context of COVID-19, various infodemiology investigations have been conducted, ranging from the impacts of COVID-19 on domestic abuse15 and psychological distress,16 17 to its impacts on social media discourse.18 In particular, previous infodemiology studies used Google data to examine public awareness of COVID-19 in various countries.19 20 21 22 23 Analyses of other popular websites, such as Wikipedia, revealed an increase in public awareness of health-related topics during the COVID-19 pandemic.24 However, previous regional studies did not utilise extensive datasets with respect to time period, number of countries, and all five WHO-recommended preventive measures that were selected in this study. Additionally, previous studies did not explore how early public awareness of preventive measures is related to lower average number of daily cases in specific countries.

This study was conducted to explore relationships between early public awareness of preventative measures and the progression of the COVID-19 pandemic through the interpretation of Google searches regarding multiple public health preventive measures. The results are expected to provide guidance for future public health communication and policy decisions.

Methods

Overview of Google Trends and global incidence data

To explore the relationship between public awareness of specific preventive measures against COVID-19 and the progression of COVID-19 pandemic, search data were extracted from Google Trends (GT) and compared with global incidence data for COVID-19. The global incidence data, measured in number of cases, were retrieved from the COVID-19 Data Repository by the Centre for Systems Science and Engineering at Johns Hopkins University of the US.3 Next, the average number of daily cases in each country/region was calculated based on the total number of cases in that country/region between the date of the first locally reported case and 30 June 2020.3 Google Trends provides quantitative information regarding actual search requests on Google for specific terms, in the form of relative search volume (RSV). The RSV is the volume of a search query for a specified location and period of time, normalised both geographically and temporally. The data are expressed using a range from 0 to 100, depending on the ratio of searches for that topic to all searches for all topics on Google.25 26 When conducting an infodemiology study using GT data, accurate keyword, region, and period selections must be made according to the study aims.25 26

Keyword, region, and period selections

Based on WHO recommendations27 and top-ranked Google search queries related to COVID-19, we selected ‘mask’, ‘hand washing’, ‘social distancing’, ‘hand sanitizer’, and ‘disinfectant’ as keywords that represented public interest in preventive measures against COVID-19. For each keyword, data were retrieved from GT according to ‘search topic’ (where available), which allowed simultaneous analysis of five queries.28 In contrast to the ‘search term’ option, ‘search topic’ is a ‘group of terms that share the same concept in any language’.26 By analysing GT data in the form of search topics, we were able to accommodate differences in language, translation, terminology, and spelling of the same concept.

In terms of region selection, GT normalises data differently according to geographical level.26 In this study, we retrieved both global- and country-level data depending on the analysis; global-level data were used to analyse general trends in public interest and geographical distribution, whereas country-level data were used to analyse correlations. Global incidence data for 190 countries/regions worldwide are available from the aforementioned Johns Hopkins University database. To clarify the terminology used in this study, ‘geographical RSV’ data were normalised according to search volume in individual regions over a fixed period of time through analyses of ‘interest by region’ in worldwide searches. In contrast, ‘temporal RSV’ data were normalised according to daily search volume over time, either globally or regionally, through analyses of ‘interest over time’ in either worldwide or regional searches.

Furthermore, global incidence data and GT-based RSV data were collected for the period from 1 January to 30 June 2020. The period selected for GT data completely matched the study period, consistent with published guidance.26 As mentioned above, a primary goal of this study was the examination of global public awareness during early stages of the COVID-19 pandemic. To examine awareness before local outbreaks, a universal start date was selected, rather than the date of the first reported case in each country/region (used in the aforementioned calculation of average number of daily cases). According to the timeline of WHO’s response to COVID-19,29 the first event involving all three levels of the WHO (headquarters, regional, country) occurred on 1 January 2020; accordingly, this date was selected as the start date for this study. Because GT data are proportional to all searches for all topics over time, it is important to note that the GT data used in this study were last retrieved on 7 November 2020.26

Data analysis

To illustrate the global temporal RSV trends for each query throughout the study period, global RSV data for each search topic were extracted and plotted on line graphs, where RSV was proportional to worldwide temporal search volume. Moreover, for each individual query, the geographical distribution of RSV was analysed and summarised in a table listing the top 20 regions of interest.

We analysed correlations between geographical RSV trends for each query and average number of daily cases in each country/region, whereby RSV was normalised according to overall regional search volume throughout the study period. Correlations were presented using scatter plots, and Pearson correlation coefficients were calculated. To avoid pre-analytical errors, we used the default GT setting of excluding regions with low search volume.

The temporal element of RSV trends in each country/region is necessary to illustrate the importance of early awareness. Therefore, temporal RSV trends in each region were extracted separately for each query; the RSV for each region was normalised according to the search volume of individual days in that region. Temporal RSV trends were then plotted against the lists of regions (excluding regions with low search volume) on five individual heat maps. In each heat map, the y-axis depicts the region list sorted from highest to lowest average number of daily cases, whereas the x-axis represents the timeline from 1 January to 30 June 2020. A three-colour scale of green, yellow, and red was used to represent low, medium, and high RSV, respectively.

Results

Global temporal trends

Figure 1 shows the global temporal RSV trends of the five queries in this study, namely, ‘mask’, ‘hand washing’, ‘social distancing’, ‘hand sanitizer’, and ‘disinfectant’. ‘Mask’ was the query that consistently demonstrated the greatest global RSV throughout the study period; at its peak, it exceeded the peak of the second highest query, ‘hand sanitizer’, by more than threefold.

Figure 1. Global temporal relative search volume trends for the queries ‘mask’, ‘hand washing’, ‘social distancing’, ‘hand sanitizer’, and ‘disinfectant’ from 1 January to 30 June 2020

With respect to the query ‘mask’, the greatest peak occurred on 4 April 2020, and three other peaks were identified (31 January 2020, 26 February 2020, and 21 March 2020). In particular, the peak on 4 April 2020 (RSV=100) corresponded to the WHO’s announcement of 1 000 000 cases worldwide.30 The peak on 31 January 2020 (RSV=24) corresponded to the WHO Director-General’s Statement regarding the International Health Regulations Emergency Committee on 30 January 2020, in which COVID-19 was declared a ‘Public Health Emergency of International Concern’.31 Similarly, the peak on 26 February 2020 (RSV=33) corresponded to the WHO release of guidelines entitled ‘Rational use of personal protective equipment for coronavirus disease’,32 which detailed preventative measures such as hand hygiene (soap/alcohol sanitiser), use of masks, and social distancing.

The RSV peak for the query ‘hand sanitizer’ on 13 March 2020 (RSV=30) corresponded to the WHO’s press release declaring that COVID-19 was a pandemic, during a media briefing on 11 March 2020.2 This peak was also accompanied by an article in The New York Times describing a shortage of hand sanitiser.33

Another major peak, visible without extensive data analysis, was recorded for the query ‘disinfectant’ on 24 April 2020. Unlike the other peaks, which gradually increased, the query ‘disinfectant’ increased from an RSV of 1/100 on 23 April 2020 to 11/100 on the following day; this 11-fold increase is visible in Figure 1.

Geographical distribution

With respect to country-level interest in the query ‘mask’ (online supplementary Table), the highest RSV was observed in Hong Kong (100), followed by Singapore (87) and France (75). The highest country-level countrylevel RSV values for ‘hand washing’ were observed in Indonesia (100), Vietnam (100), and Hong Kong (88), while that for ‘social distancing’ were recorded in Canada (100), Indonesia (95), and Singapore (92). For ‘hand sanitizer’, the highest country-level RSV values were recorded in Hong Kong (100), Singapore (96), and Denmark (91). For ‘disinfectant’, the highest country-level RSV values were observed in the US (100), the Philippines (88), and Singapore (79). The full list of geographical distributions showing all countries/regions is included in the online supplementary Table.

Correlations between geographical relative search volume trends and average number of daily cases

Figure 2 shows the correlation between the average number of daily cases for each country/region and the LogRSV of each respective search query from 1 January to 30 June 2020. ‘Hand washing’ and ‘social distancing’ were the only queries with mild correlations, with Pearson correlations (r values) of -0.44 (‘hand washing’) and -0.38 (‘social distancing’). No strong correlations were observed for the other three terms ‘mask’ (r=0.03), ‘hand sanitizer’ (r=0.00), and ‘disinfectant’ (r=-0.06).

Figure 2. Correlation between country-specific geographical relative search volume (RSV) trend and average number of daily cases for the queries (a) ‘mask’, (b) ‘hand washing’, (c) ‘social distancing’, (d) ‘hand sanitizer’, and (e) ‘disinfectant’. Dots represent countries/regions

Correlations between temporal relative search volume trends in each region and average number of daily cases

The online supplementary Figure shows heat maps for the five search queries. In online supplementary Figure a, a divergence pattern was observed for the search query ‘mask’, which tended to display an earlier RSV peak in countries/regions with lower average number of daily cases and a later RSV peak in countries/regions with higher average number of daily cases. Among the countries/regions with an early RSV peak and low average number of daily cases, Hong Kong had an early RSV peak (100) on 29 January 2020 and an average number of daily case count of 7.75. Other notable examples include Taiwan (early RSV peak on 3 February 2020 and average number of daily case count of 2.79) and Vietnam (early RSV peak on 31 January 2020 and average number of daily case count of 2.23). In contrast, countries/regions with a late RSV peak and high average number of daily cases included the US (late RSV peak on 4 April 2020 and average number of daily case count of 16789.11), Brazil (late RSV peak on 3 April 2020 and average number of daily case count of 11590.02), and Russia (late RSV peak on 30 March 2020 and average number of daily case count of 4327.68).

Similarly, online supplementary Figures b, d, and e show heat maps for the search queries of ‘hand washing’, ‘hand sanitizer’, and ‘disinfectant’, respectively. Earlier increases in RSV tended to occur in countries/regions with lower average number of daily cases. However, the heat map of ‘social distancing’ did not display such a clear pattern; it showed a sudden global increase in late March 2020 (online supplementary Fig c).

Discussion

Principal findings

Overview

The rapid and aggressive infectivity of COVID-19 requires the general public to be vigilant about preventive measures. Although prevention is generally preferred over curative treatment, the effect of each preventive measure on COVID-19 transmission was unclear during early stages of the pandemic. For example, during early stages of the pandemic, there was controversy regarding the importance of wearing masks to prevent COVID-19 transmission via droplets.34 Indeed, the routine use of medical masks by normal healthy individuals had not been recommended by the WHO at the start of data collection.35 This controversy led to confusion regarding public health policies, as well as the stigmatisation of individuals who practised specific preventive measures. Thus, the present study retrospectively compared public awareness of the five aforementioned preventive measures with the progression of COVID-19; this analysis was intended to provide guidance regarding public health communication and policy decisions.

Early awareness in regions with low average number of daily cases

The heat maps (online supplementary Figs a-d) show a pattern of early awareness among countries/regions with lower average number of daily cases, according to analyses of the queries ‘mask’, ‘hand washing’, ‘hand sanitizer’, and ‘disinfectant’. These findings suggested that such queries were associated with the prevention of COVID-19 progression. Despite these positive findings, we did not find strong correlations between average number of daily cases in specific countries/regions and the overall geographical RSV trend throughout the study period (Fig 2). This negative result highlighted the importance of temporal element in the prevention of COVID-19 transmission, implying that increased public awareness in an earlier stage of the pandemic was superior to an increase in the overall volume of public awareness. Notably, a similar GT-based study of mask awareness conducted earlier in May 2020 demonstrated a significant correlation (Kendall rank correlation coefficient [τ] of -0.47) between mask awareness and average RSV data during a very early stage of the pandemic (21 January to 11 March 2020).36

Our positive findings regarding mask, hand washing, hand sanitiser, and disinfectant queries are consistent with the current understanding of COVID-19 transmission. The major routes of COVID-19 transmission include contact, droplets, and aerosols37; importantly, animate and inanimate surfaces participate in COVID-19 transmission. Face masks may slow the spread of COVID-19 by reducing aerosol and respiratory droplet transmission.36 Systematic reviews and meta-analyses have increasingly shown that mask usage in community or healthcare settings reduces COVID-19 transmission.38 39 40 41 42 43 In contrast, a Danish randomised controlled trial of mask usage in the general population suggested little to no evidence that facemask usage alone could prevent transmission of severe acute respiratory syndrome coronavirus 2, the virus causing COVID-19.44 Retrospective cohort studies and case-control studies have provided some evidence of the preventive effects of mask usage in communities such as Beijing and Thailand.45 46 Additional randomised controlled trials are needed to conclusively determine the benefits of mask usage in the general population.47 Importantly, the routine maintenance of good hand hygiene can reduce contact transmission. The use of an alcohol-based hand sanitiser can disrupt COVID-19 transmission via surface protein precipitation.37 Our findings regarding mask and hand washing queries were also consistent with previous regional infodemiology studies, including a Taiwanese GT-based study focused on masks and hand washing.22 In support of the regional results, the present study illustrated the importance of early awareness on a global scale.

Despite the lack of a clear pattern of early awareness concerning the search topic ‘social distancing’, a meta-analysis has confirmed that social distancing of ≥1 m reduces COVID-19 transmission.38 Therefore, the lack of positive findings regarding ‘social distancing’ in the present study does not necessarily indicate a lack of effectiveness. Instead, it suggests inadequate public awareness. Careful analysis of temporal RSV trends for all five queries (Fig 1) revealed that a lower overall volume of searches for ‘social distancing’. Although public awareness of the topics ‘mask’, ‘hand sanitizer’, and ‘disinfectant’ may spontaneously increase because of various other factors, such as a market shortage, social distancing during the COVID-19 pandemic was often implemented via governmental policy, rather than public awareness.48 49 This lack of public awareness was demonstrated by the decrease in confirmed COVID-19 cases in the US after government-imposed social distancing measures had been implemented.50 Despite their proven efficacies, specific preventive measures such as hand washing were often implemented via public health initiatives, rather than law enforcement.48 49 Future studies should seek to identify specific preventive measures beyond public awareness that can guide public health policy decisions regarding the COVID-19 pandemic.

Preventive measures against COVID-19 transmission are only effective if the majority of the general public acknowledge and practise them with the correct timing and knowledge. In addition to the determination of whether a preventive measure is effective, patterns of early public awareness should be explored to enhance the preventive effects of public health communication on COVID-19 transmission.

Effects of authorities, public figures, and social media on public awareness

As mentioned above, there were multiple instances of a sudden surge in public awareness. One of the most prominent patterns was the surge in ‘disinfectant’ queries on 24 April 2020. A substantial increase in global awareness of disinfectant occurred within a single day, leading to questions regarding the underlying cause and whether that cause can provide any insights concerning effective public health communication. Further investigation revealed a possible key event related to the timing and content of the surge in ‘disinfectant’ queries: a speech made by US President Donald Trump on 23 April 2020, in which he claimed that disinfectant ‘knocks it [severe acute respiratory syndrome coronavirus 2] out in a minute’ and suggested that scientists should conduct further research in this area.51 Although the scientific legitimacy of the contents of Trump’s speech was questionable, the speech itself had a substantial impact on public awareness, as demonstrated by the massive number of Google searches in such a short period of time.

The example above was not the only surge pattern evident in this study. Buried under the overwhelming search volumes of other queries, the RSV magnitude of ‘social distancing’ appears to be relatively negligible (Fig 1). However, closer inspection of the temporal RSV trend of ‘social distancing’ reveals an obvious surge from 10 March 2020 to 20 March 2020 (Fig 3). Over an interval of 10 days, the RSV of ‘social distancing’ increased from 3 to 86. Similar to Trump’s speech, a key event in early March was associated with the surge in ‘social distancing’ queries. A sentiment of ‘staying home’ was reportedly coined by Florian Reifschneider, a German engineer; it soon became a trend on social media and was heavily promoted by prominent celebrities.52 53 54 55 Although ‘staying home’ and ‘social distancing’ are distinct key terms, an approximately overlapping rise and fall pattern is evident upon comparison of both RSV trends (Fig 4). Notably, the RSV of ‘staying home’ overlapped with the RSV of ‘social distancing’, but the magnitude of the RSV of ‘staying home’ exceeded the magnitude of the RSV of ‘social distancing’ by more than 50%; this finding implies that the public response to social distancing may have been greater if the concept of social distancing had been promoted correctly.

Figure 3. Global temporal relative search volume trend for the query ‘social distancing’ from 1 January to 30 June 2020

Figure 4. Global temporal relative search volume trends for the queries ‘social distancing’ and ‘staying home’ from 1 January to 30 June 2020

The relationship between sudden surges in global RSV and key events suggests that the effect of global public awareness is secondary to promotion by authorities and public figures. The evolution of the internet and social media may offer new avenues for public health communication, particularly in times of crisis.

Limitations

There were a few limitations in this study. To begin with, GT data constitute an indirect representation of public awareness; these data do not indicate whether preventive measures were correctly implemented by the general public. Therefore, the analysis may have overestimated or underestimated correlations. Moreover, despite the use of search topics to explore GT data, the selected keywords may not accurately represent the concept of each preventive measure because of variations in language, translation, terminology, and spelling of the same concept. Furthermore, to facilitate comparison, this study exclusively analysed the queries in a single search platform (ie, Google). This limited focus may have led to sampling error based on access to Google, as well as regional search engine preferences. Internet accessibility also varies among regions; therefore, GT data may not accurately represent public awareness in regions with fewer internet users.

An important example is China (not including Hong Kong), which was regarded as a country with ‘low search volume’ for some queries, despite its 538 million netizens.56 There are multiple reasons for this bias. First, Google holds <20% of China’s online search market; Baidu is the most popular search engine.57 Future studies involving China should consider the use of Baidu, as in a previous internet query study specifically focused on China.57 However, a study by Kang et al56 revealed that Chinese GT data may be used as a valid complementary source of information for influenza surveillance in south China.

Second, this study did not consider potential confounders in the correlation analyses, including the stringency of public health measures, the containment capacities of the countries and regions included, and the degree to which those countries and regions are vulnerable to public health threats.58 59 60

Third, this study primarily focused on public awareness and progression of COVID-19 in the early stages of the pandemic; thus, factors identified during later stages of the pandemic were not evaluated.

Finally, research concerning preventive measures against COVID-19 is largely limited by the lack of randomised controlled trials. Considering the current scale of the pandemic, it is neither feasible nor ethical to conduct individual randomised controlled trials for each preventive measure in healthcare or non-healthcare settings. Therefore, infodemiology studies remain valuable in policy making for the foreseeable future.

Conclusion

Google Trends offers large-scale population data regarding public health events. The results of RSV trend analysis revealed an earlier RSV peak in countries/regions with lower average number of daily cases, suggesting that early public awareness can slow the spread of a pandemic. During future pandemics, global and local public health authorities could focus on early public awareness to facilitate effective disease control. Additionally, our findings illustrate the value of early public health communication regarding the use of masks, hand washing, hand sanitiser, and disinfection among the general public during the COVID-19 pandemic.

Author contributions

Concept or design: JYC Teoh, A Mok.
Acquisition of data: JYC Teoh, A Mok, OOY Mui.
Analysis or interpretation of data: JYC Teoh, A Mok, OOY Mui.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

As editors of the journal, CF Ng, MCS Wong and JYC Teoh were not involved in the peer review process. Other authors have disclosed no conflicts of interest.

Funding/support

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

1. Wang C, Horby PW, Hayden FG, Gao GF. A novel coronavirus outbreak of global health concern. Lancet 2020;395:470-3. Crossref

2. World Health Organization. WHO Director-General’s opening remarks at the media briefing on COVID-19-11—March 2020. Available from: https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---11-march-2020. Accessed 7 Sep 2020.

3. Johns Hopkins University and Medicine. Coronavirus resource center. Available from: https://coronavirus.jhu.edu/map.html. Accessed 7 Sep 2020.

4. World Health Organization. Coronavirus disease (COVID-19). Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/question-and-answers-hub/q-a-detail/q-a-coronaviruses. Accessed 7 Sep 2020.

5. Eysenbach G. Infodemiology and infoveillance: framework for an emerging set of public health informatics methods to analyze search, communication and publication behavior on the Internet. J Med Internet Res 2009;11:e11. Crossref

6. Chandrasekaran R, Mehta V, Valkunde T, Moustakas E. Topics, trends, and sentiments of tweets about the COVID-19 pandemic: temporal infoveillance study. J Med Internet Res 2020;22:e22624. Crossref

7. Alicino C, Bragazzi NL, Faccio V, et al. Assessing Ebola-related web search behaviour: insights and implications from an analytical study of Google Trends–based query volumes. Infect Dis Poverty 2015;4:54. Crossref

8. Odlum M, Yoon S. What can we learn about the Ebola outbreak from tweets? Am J Infect Control 2015;43:563-71. Crossref

9. van Lent LG, Sungur H, Kunneman FA, van de Velde B, Das E. Too far to care? Measuring public attention and fear for Ebola using Twitter. J Med Internet Res 2017;19:e193. Crossref

10. Dugas AF, Jalalpour M, Gel Y, et al. Influenza forecasting with Google Flu Trends. PLoS One 2013;8:e56176. Crossref

11. Chew C, Eysenbach G. Pandemics in the age of Twitter: content analysis of Tweets during the 2009 H1N1 outbreak. PLoS One 2010;5:e14118. Crossref

12. Pruss D, Fujinuma Y, Daughton AR, et al. Zika discourse in the Americas: a multilingual topic analysis of Twitter. PLoS One 2019;14:e0216922. Crossref

13. Daughton AR, Paul MJ. Identifying protective health behaviors on Twitter: observational study of travel advisories and Zika virus. J Med Internet Res 2019;21:e13090. Crossref

14. Stefanidis A, Vraga E, Lamprianidis G, et al. Zika in Twitter: temporal variations of locations, actors, and concepts. JMIR Public Health Surveill 2017;3:e22. Crossref

15. Xue J, Chen J, Chen C, Hu R, Zhu T. The hidden pandemic of family violence during COVID-19: unsupervised learning of tweets. J Med Internet Res 2020;22:e24361. Crossref

16. Chen X, Zhang SX, Jahanshahi AA, et al. Belief in a COVID-19 conspiracy theory as a predictor of mental health and well-being of health care workers in Ecuador: cross-sectional survey study. JMIR Public Health Surveill 2020;6:e20737. Crossref

17. Rana U, Singh R. Emotion analysis of Indians using google trends during COVID-19 pandemic. Diabetes Metab Syndr 2020;14:1849-50. Crossref

18. Chen E, Lerman K, Ferrara E. Tracking social media discourse about the COVID-19 pandemic: development of a public coronavirus Twitter data set. JMIR Public Health Surveill 2020;6:e19273. Crossref

19. Effenberger M, Kronbichler A, Shin JI, Mayer G, Tilg H, Perco P. Association of the COVID-19 pandemic with internet search volumes: a Google Trends^TM analysis. Int J Infect Dis 2020;95:192-7. Crossref

20. Hu D, Lou X, Xu Z, et al. More effective strategies are required to strengthen public awareness of COVID-19: evidence from Google Trends. J Glob Health 2020;10:011003. Crossref

21. Husain I, Briggs B, Lefebvre C, et al. Fluctuation of public interest in COVID-19 in the United States: retrospective analysis of Google Trends search data. JMIR Public Health Surveill 2020;6:e19969. Crossref

22. Husnayain A, Fuad A, Su EC. Applications of Google search trends for risk communication in infectious disease management: a case study of the COVID-19 outbreak in Taiwan. Int J Infect Dis 2020;95:221-3. Crossref

23. Rovetta A, Bhagavathula AS. COVID-19–related web search behaviors and infodemic attitudes in Italy: infodemiological study. JMIR Public Health Surveill 2020;6:e19374. Crossref

24. Chrzanowski J, Sołek J, Fendler W, Jemielniak D. Assessing public interest based on Wikipedia’s most visited medical articles during the SARS-CoV-2 outbreak: search trends analysis. J Med Internet Res 2021;23:26331. Crossref

25. Google. FAQ about Google Trends data. Available from: https://support.google.com/trends/answer/4365533?hl=en&dark=1. Accessed 7 Sep 2020.

26. Mavragani A, Ochoa G. Google Trends in infodemiology and infoveillance: methodology framework. JMIR Public Health Surveill 2019;5:e13439. Crossref

27. World Health Organization. Advice for the public: coronavirus disease (COVID-19). Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/advice-for-public. Accessed 7 Sep 2020.

28. Smallwood C, editor. The Complete Guide to Using Google in Libraries: Instruction, Administration, and Staff Productivity. Volume 1. Lanham, Maryland: Rowman & Littlefield; 2015.

29. World Health Organization. Listings of WHO’s response to COVID-19. 2020; Available from: https://www.who.int/news/item/29-06-2020-covidtimeline. Accessed 7 Sep 2020.

30. World Health Organization. Coronavirus disease 2019 (COVID-19) Situation Report–75. 2020. Available from: https://www.who.int/docs/default-source/coronaviruse/situation-reports/20200404-sitrep-75-covid-19.pdf?sfvrsn=99251b2b_4. Accessed 7 Sep 2020.

31. World Health Organization. WHO Director-General’s statement on IHR Emergency Committee on Novel Coronavirus (2019-nCoV). Available from: https://www.who.int/dg/speeches/detail/who-director-general-s-statement-on-ihr-emergency-committee-on-novel-coronavirus-(2019-ncov) Accessed 7 Sep 2020.

32. World Health Organization. Rational use of personal protective equipment for coronavirus disease 2019 (COVID-19) and considerations during severe shortages. 2020. Available from: https://www.who.int/publications/i/item/rational-use-of-personal-protective-equipment-for-coronavirus-disease-(covid-19)-and-considerations-during-severe-shortages. Accessed 7 Sep 2020.

33. Suthivarakom G. Coronavirus has caused a hand sanitizer shortage. What should you do? The New York Times. 11 March 2020. Available from: https://www.nytimes.com/2020/03/11/smarter-living/wirecutter/coronavirus-hand-sanitizer.html. Accessed 7 Sep 2020.

34. Feng S, Shen C, Xia N, Song W, Fan M, Cowling BJ. Rational use of face masks in the COVID-19 pandemic. Lancet Respir Med 2020;8:434-6. Crossref

35. World Health Organization. Coronavirus disease (COVID-19) advice for the public: when and how to use masks. Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/advice-for-public/when-and-how-to-use-masks. Accessed 7 Sep 2020.

36. Wong SH, Teoh JY, Leung CH, et al. COVID-19 and public interest in face mask use. Am J Respir Crit Care Med 2020;202:453-5. Crossref

37. Pradhan D, Biswasroy P, Kumar Naik P, Ghosh G, Rath G. A review of current interventions for COVID-19 prevention. Arch Med Res 2020;51:363-74. Crossref

38. Chu DK, Akl EA, Duda S, et al. Physical distancing, face masks, and eye protection to prevent person-to-person transmission of SARS-CoV-2 and COVID-19: a systematic review and meta-analysis. Lancet 2020;395:1973-87. Crossref

39. Howard J, Huang A, Li Z, et al. An evidence review of face masks against COVID-19. Proc Natl Acad Sci U S A 2021;118:e2014564118. Crossref

40. MacIntyre CR, Chughtai AA. A rapid systematic review of the efficacy of face masks and respirators against coronaviruses and other respiratory transmissible viruses for the community, healthcare workers and sick patients. Int J Nurs Stud 2020;108:103629. Crossref

41. Tabatabaeizadeh SA. Airborne transmission of COVID-19 and the role of face mask to prevent it: a systematic review and meta-analysis. Eur J Med Res 2021;26:1. Crossref

42. Fouladi Dehaghi B, Ghodrati-Torbati A, Teimori G, Ibrahimi Ghavamabadi L, Jamshidnezhad A. Face masks vs. COVID-19: a systematic review. Invest Educ Enferm 2020;38:e13. Crossref

43. Li Y, Liang M, Gao L, et al. Face masks to prevent transmission of COVID-19: a systematic review and meta-analysis. Am J Infect Control 2021;49:900-6. Crossref

44. Bundgaard H, Bundgaard JS, Raaschou-Pedersen DE, et al. Effectiveness of adding a mask recommendation to other public health measures to prevent SARS-CoV-2 infection in Danish mask wearers: a randomized controlled trial. Ann Intern Med 2021;174:335-43. Crossref

45. Wang Y, Tian H, Zhang L, et al. Reduction of secondary transmission of SARS-CoV-2 in households by face mask use, disinfection and social distancing: a cohort study in Beijing, China. BMJ Glob Health 2020;5:e002794. Crossref

46. Doung-Ngern P, Suphanchaimat R, Panjangampatthana A, et al. Case-control study of use of personal protective measures and risk for SARS-CoV 2 infection, Thailand. Emerg Infect Dis 2020;26:2607-16. Crossref

47. World Health Organization. Mask use in the context of COVID-19: Interim guidance, 1 December 2020. Available from: https://apps.who.int/iris/handle/10665/337199. Accessed 7 Sep 2020.

48. International Monetary Fund. Policy responses to COVID-19. Available from: https://www.imf.org/en/Topics/imf-and-covid19/Policy-Responses-to-COVID-19. Accessed 7 Sep 2020.

49. Mervosh S, Lu D, Swales V. See which states and cities have told residents to stay at home. The New York Times. 20 April 2020. Available from: https://www.nytimes.com/interactive/2020/us/coronavirus-stay-at-home-order.html. Accessed 7 Sep 2020.

50. Courtemanche C, Garuccio J, Le A, Pinkston J, Yelowitz A. Strong social distancing measures in the United States reduced the COVID-19 growth rate. Health Aff (Millwood) 2020;39:1237-46. Crossref

51. Agence France-Presse. Experts rubbish Donald Trump’s suggestion to inject disinfectant to kill coronavirus. South China Morning Post. 24 April 2020. Available from: https://www.scmp.com/news/world/united-states-canada/article/3081503/experts-rubbish-trumps-suggestioninject? module=perpetual_scroll&pgtype=article&campai gn=3081503. Accessed 7 Sep 2020.

52. American Medical Association. AMA, AHA, ANA: #StayHome to confront COVID-19. Available from: https://www.ama-assn.org/press-center/press-releases/ama-aha-ana-stayhome-confront-covid-19. Accessed 7 Sep 2020.

53. Hudson A. Busy Philipps Joins Cameo to Record “Stay the Fuck Home” messages for coronavirus. Exclaim!. 17 March 2020. Available from: https://exclaim.ca/film/article/busy_philipps_joins_cameo_to_record_stay_the_fuck_home_messages_for_coronavirus. Accessed 7 Sep 2020.

54. #StayTheFuckHome. A movement to stop the COVID-19 pandemic. Available from: https://staythefuckhome.com/. Accessed 7 Sep 2020.

55. Ehrlich B. Taylor Swift urges fans to stay home amid COVID-19 outbreak. Rolling Stone. 15 March 2020. Available from: https://www.rollingstone.com/music/music-news/taylor-swift-urges-fans-to-stay-home-amid-covid-19-outbreak-967629/. Accessed 7 Sep 2020.

56. Kang M, Zhong H, He J, Rutherford S, Yang F. Using Google Trends for influenza surveillance in south China. PLoS One 2013;8:e55205. Crossref

57. Yuan, Q, Nsoesie EO, Lv B, Peng G, Chunara R, Brownstein JS. Monitoring influenza epidemics in China with search query from Baidu. PLoS One 2013;8:e64323. Crossref

58. World Health Organization. e-SPAR State Party Annual Report. 2020. Available from: https://extranet.who.int/e-spar. Accessed 7 Sep 2020.

59. Wong MC, Huang J, Teoh J, Wong SH. Evaluation on different non-pharmaceutical interventions during COVID-19 pandemic: an analysis of 139 countries. J Infect 2020;81:e70-1. Crossref

60. Wong MC, Teoh JY, Huang J, Wong SH. The potential impact of vulnerability and coping capacity on the pandemic control of COVID-19. J Infect 2020;81:816-46. Crossref

Time for change? Feasibility of introducing micromodules into medical student education: a randomised controlled trial

Hong Kong Med J 2023 Jun;29(3):208–13

https://doi.org/10.12809/hkmj219267

ORIGINAL ARTICLE

Time for change? Feasibility of introducing micromodules into medical student education: a randomised controlled trial

CF Ng, FRCSEd (Urol), FHKAM (Surgery); Kevin Lim, MB, ChB; CH Yee, FRCSEd (Urol), FHKAM (Surgery); Peter KF Chiu, FRCSEd (Urol), FHKAM (Surgery); Jeremy YC Teoh, FRCSEd (Urol), FHKAM (Surgery); Franco PT Lai, BN

Department of Surgery, SH Ho Urology Centre, The Chinese University of Hong Kong, Hong Kong SAR, China

Corresponding author: Prof CF Ng (ngcf@surgery.cuhk.edu.hk)

Full paper in PDF

Abstract

Introduction: Didactic lectures have been the foundation of learning for many medical students. However, in recent years, the flipped classroom model has become increasingly popular in medical education. This approach enhances pre-class learning, allowing the limited contact time between clinicians and medical students to be focused on practical issues. This study evaluated the effectiveness and non-inferiority of online micromodule teaching in terms of knowledge transfer concerning specific urology topics.

Methods: Medical students without prior exposure to the urology subspecialty were enrolled in the study, then randomised to a traditional didactic lecture group or an online micromodule group. Knowledge transfer was assessed by pre-intervention and post-intervention multiple-choice questions and objective structured clinical examinations that involved the acquisition of medical histories from real patients.

Results: In total, 45 medical students were enrolled (22 in the traditional didactic group and 23 in the online micromodule group). In terms of knowledge transfer (assessed by objective structured clinical examinations), the efficacy of online micromodules was comparable to traditional didactic lectures, although the difference was not statistically significant (P=0.823). There were no significant differences in terms of knowledge acquisition, retention, or clinical application between the two groups.

Conclusion: terms of acquiring, retaining, and applying foundational urological knowledge, online micromodules can help medical students to achieve outcomes comparable with the outcomes of didactic lectures. Online micromodules may be a viable alternative to traditional didactic lectures in urology education.

New knowledge added by this study

Compared with traditional didactic lectures, online micromodules have similar knowledge transfer efficacy in medical student education.
The flipped classroom model may help to allow the limited contact time between clinicians and medical students to focus on practical training and experience sharing.

Implications for clinical practice or policy

Large-group didactic lectures will likely decline in the future.
There is an urgent need to develop teaching methods appropriate for the modern era.
Micromodules may be regarded as a flipped classroom component that can facilitate learning and knowledge transfer.

[Abstract in Chinese]

Introduction

The coronavirus disease 2019 pandemic dramatically changed modern life. Traditional didactic lecture methods suddenly became impossible,1 and there was a need to maintain social distancing. A shift to online didactic lectures was the most common solution. However, there is evidence that information acquisition becomes inefficient beyond the first 10 to 15 minutes of a lecture.2 It may be even more difficult to concentrate in online lectures that lack interaction between the speaker and audience. Notably, videos longer than 10 minutes are less likely to be viewed.3 4 Short online video lectures (ie, micromodules), with or without interactive elements, offer an attractive alternative. Such micromodules can be incorporated into the flipped classroom (FC) model, which is a pedagogical paradigm shift that rearranges how time is spent in and out of the classroom.5

The FC model is becoming increasingly popular in medical education. It is attractive to the current generation of students who are accustomed to utilising digital media; on average, 70% of students prefer this learning model.6 Students can learn pre-class materials at their own pace; they can also enjoy more in-class active learning and interaction. Moreover, they can negotiate the FC platform at their preferred time and in their preferred place. Instead of passively delivering information in class, educators can devote valuable contact time to interactions with students, exploration of their needs, and discussions of more nuanced and challenging topics.6 The acquisition of foundational information becomes an active self-directed process, outside of the classroom.

Considering the continuous growth of medical literature, today’s medical students must acquire an expanded field of knowledge before graduation. A modern urology clerkship should alleviate the intense time pressures placed on students by helping them to effectively and efficiently develop diagnostic and procedural core competencies. Where possible, students should be allowed to learn by active participation, rather than listening and reading, during the limited available contact time. The FC model holds great promise in achieving this goal.7

The success of the FC model requires an efficacious online platform that facilitates self-directed learning; stringent evaluation of the online platform is necessary. However, methodologically rigorous qualitative and quantitative studies and evidence-based recommendations are scarce.8 Most published works quote practical wisdom, anecdotes, and principles of educational theory as the basis for their recommendations.9

This pilot study was conducted to compare our institution’s online micromodule platform with traditional didactic lectures in facilitating the acquisition of foundational urological knowledge by medical students.

Methods

This prospective, single-centre, single-blind randomised controlled trial, performed at a tertiary academic hospital, investigated whether online micromodules are non-inferior to traditional didactic lectures as an instructional medium; this trial is a component of a larger movement towards the FC approach in clinical training.

Urology curriculum

The urology clerkship is a surgical subspecialty in our faculty curriculum. All medical students have 1 week of clinical attachment in their final year of medical clerkship training (Year 6). The standard curriculum consists of lectures, bedside tutorials, and clinical shadowing. Traditionally, lectures are delivered to the whole class at the beginning of the academic year. Students then shadow our team in small groups on the wards, in clinics, and in the operating theatre. Teaching is opportunistic, based on symptoms, signs, investigations, diseases, and procedures encountered in the clinical setting. Formal knowledge assessment is conducted during end-of-year examinations in the form of written examinations (multiple-choice questions [MCQs] and short-answer questions), objective structured clinical examinations (OSCEs), and clinical short case examinations.

Study intervention

In this study, we selectively assessed knowledge transfer with regard to two urology topics: approaches to lower urinary tract symptoms (LUTS) and haematuria. First, a didactic lecture on the management of LUTS and haematuria, along with other topics, was recorded during its delivery in our routine lecture series for final-year students. Subsequently, two micromodules were prepared concerning the management of LUTS and haematuria; the micromodule content was similar to the didactic lecture content. The study participants continued with their scheduled urology training in Year 6; therefore, the study intervention was regarded as supplemental curriculum. Because the participants’ overall learning opportunities were not affected, we decided to obtain only verbal consent for inclusion in the study.

Randomisation, allocation concealment, and blinding of participants

Medical students in Years 4 to 6 with no exposure to the urology subspecialty rotation were voluntarily recruited for the study. Participants were randomly allocated to either traditional didactic lectures or online micromodules; rigorous proctored tests were administered in accordance with the schedule shown in the Figure. Permuted block randomisation was conducted using a computer program. Random allocation sequences were placed into identical sealed and numbered envelopes. Designated research staff members were responsible for allocating consecutively numbered envelopes to the participants.

Figure. Flow of pre-intervention assessment, randomisation, intervention, and post-intervention assessment

Students randomised to the traditional didactic lecture were grouped into a class, which watched the pre-recorded 45-minute didactic lecture in the lecture theatre (as if the students were attending a standard lecture). Students randomised to the online micromodule group viewed the micromodules on separate computers at their own pace. The total runtime of these micromodules was 10 minutes each, and students were expected to explore the content in its entirety within 45 minutes. The breadth and depth of topics covered in both interventions were identical to each other and similar to past lectures; the only difference was the delivery medium. The students could not be blinded; however, all outcome assessors (including content creators) were blinded to intervention allocation because the didactic lecture was not delivered live or in person.

Assessment

We used the Kirkpatrick’s four-level training evaluation model as the basis for evaluations of instructional effectiveness. In the context of online learning, Level 1 (reaction) refers to the student’s affective responses to training quality or relevance, usually measured by surveys; Level 2 (learning) refers to knowledge directly obtained from the online lecture, usually measured by knowledge tests such as MCQs and true-false questions; Level 3 (behaviour outcomes/transfer of learning) refers to improvements in the outcomes of tasks not directly taught in the instructional content, typically measured through practical or standardised examinations; and Level 4 (results) refers to the impact of training on organisational goals (ie, actual benefit to patients).

Prior to randomisation, a pre-intervention MCQ test was used to determine participant baseline knowledge. Immediately after randomisation and completion of training, participants repeated the MCQ test to determine the degree of knowledge acquisition (ie, Kirkpatrick Level 2). Their confidence in the subject matter was also measured using a 10-point scale (ie, Kirkpatrick Level 1).

After 3 weeks of teaching, each participant underwent individual assessments in outpatient clinics. The MCQ test was administered again to test knowledge retention. Then, an OSCE was administered to assess the participant’s approach to a real patient with either LUTS or haematuria. A nurse was present as a chaperone and third-party assessor, who gave a subjective assessment score, measured using a 10-point scale. The participant then presented the case to a urologist, who assessed the collected information using a structured marking scheme. Additionally, the urologist gave a subjective assessment score, similar to the nursing assessment. All student assessors were blinded to the allocated teaching approach. The scores from the nurses and urologists were used to assess student performance in the OSCE (ie, Kirkpatrick Level 3); they also were used to assess the overall effectiveness and safety of the micromodule teaching approach. Due to the study design, the impact of training on organisational goals (ie, Kirkpatrick Level 4) was not assessed.

Statistical analysis

Statistical analysis was performed using SPSS (Windows version 23.0; IBM Corp, Armonk [NY], United States). There was no crossover between treatment arms. Data were analysed using an intention-to-treat approach. Descriptive statistics (means, standard deviations, and ranges) were used for demographic data. Independent samples t tests or one-way multivariate analysis of variance were used for parametric continuous variables; the Mann–Whitney U test was used for non-parametric continuous variables; and the Chi squared test was used for categorical variables. P values <0.05 were considered statistically significant.

Results

Between 4 December 2017 and 22 January 2018, 45 medical students voluntarily enrolled in this study at our hospital; 22 students were randomised to the didactic lecture group and 23 students were randomised to the online micromodule group. Most participants (77% and 74%, respectively) were in their final year of medical education. There were no significant differences in demographic composition between the two groups (Table 1). The difference in pre-intervention MCQ scores also was not statistically significant (P=0.471), indicating that the participants had similar baseline knowledge (Table 2).

Table 1. Baseline participant characteristics

Table 2. Assessment result of the students during different phases of the study

In this study, the primary outcome was the difference in OSCE scores between the didactic lecture and online micromodule groups, as assessed by the urologists. This outcome corresponds to Level 3 of the Kirkpatrick model. Three-quarters of participants assessed real patients with LUTS; the remaining participants assessed patients with haematuria. There was no difference in OSCE score between the groups (13.09 ± 1.59 vs 12.98 ± 1.75, P=0.823) [Table 2].

The secondary outcome was the difference in knowledge acquisition and retention between interventions. Knowledge acquisition was defined as the difference between pre-intervention and post-intervention MCQ scores. Knowledge retention was defined as the difference between pre-intervention MCQ score and pre-OSCE MCQ score (taken 3 weeks after the intervention). Both of these outcomes correspond to Level 2 of the Kirkpatrick model. There were improvements in MCQ scores after teaching in both groups, although not statistically significant. However, there was no difference in the degree of improvement between the groups. Therefore, knowledge acquisition for the two groups were similar. For the assessment of knowledge retention, one-way multivariate analysis of variance with adjustment for pre-intervention MCQ scores revealed no statistically significant difference between post-intervention MCQ score and pre-OSCE MCQ score (Wilks’ Lambda=0.894, P=0.101, partial η²=0.106).

Finally, subjective assessment of confidence and competence was conducted; this assessment corresponds to Kirkpatrick Level 1. There was a significant improvement in post-intervention self-rated confidence, but there was no difference in the degree of improvement between the groups (Table 2). In terms of clinical performance (Kirkpatrick Level 3), there were no differences between the groups in terms of subjective assessment score by the urologists (7.89 ± 0.91 vs 7.70 ± 0.91, P=0.487) or nurses (8.05 ± 0.72 vs 8.04 ± 0.71, P=0.993).

Discussion

Our results show that both didactic lectures and online micromodules can help medical students achieve comparable outcomes in terms of acquiring, retaining, and applying foundational urological knowledge. Thus, online learning platforms may be viable substitutes for didactic lectures in the broader context of a move towards the FC approach.

In a systematic review of literature concerning the use of online lectures in undergraduate medical education,6 45 studies were identified; only 21 (47%) of those studies had clearly established control and intervention groups. Among the 21 studies, only six (29%) assessed learning using an OSCE or equivalent practical examinations; the remaining studies used MCQ tests. There was considerable heterogeneity in the manner by which online lectures were integrated into existing surgical curricula, which hindered meta-analysis. However, online lectures generally tended to be non-inferior to traditional lectures.

Online learning offers many benefits to educators and students. First, it ensures round-the-clock access to learning materials. Second, it allows students to revisit these materials throughout the curriculum. Third, online learning platforms can track and verify that students have accessed and completed specific materials. Fourth, electronic content can be updated in a convenient manner; distribution is instantaneous and universal. Fifth, students have autonomy over the sequence and pace of learning, as well as the allocation of time; these aspects allow them to tailor their learning experience to meet personal objectives. Sixth, although a higher initial investment may be required, online learning platforms can be reused, exchanged, and collaborated on; they offer new economies of scale.10 11 Finally, the coronavirus disease 2019 pandemic led to concerns about the spread of infection, such that online micromodules became an attractive option for medical student education that permitted social distancing. Notably, online micromodules represent easily accessible media that can be used for continuing medical education, and interactive teaching can be added to enhance learning experience.

An important limitation of online learning is that educators may utilise the scheduling freedom offered by online platforms to overburden students with learning materials; they may not consider the large amount of non-classroom time that may be allocated to other tasks. To avoid this phenomenon, we established ‘bite-sized’ modules (ie, micromodules) and ensured that all topics covered are highly relevant to future clinical practice. Such short modules also match the students’ attention spans.2 3 4 However, we acknowledge that educators may initially expend greater effort in the preparation of online modules.4

Although there were some improvements in MCQ scores after the lecture or micromodules, they were weaker than expected, potentially because the post-intervention MCQ test occurred immediately after the lecture and there was insufficient time for participants to process the lecture content. Another limitation of the study design was that there were no tutorials or in-class interactions after the lectures. Thus, the acquired knowledge may not have been consolidated, resulting in suboptimal knowledge retention. Nevertheless, this study was designed to demonstrate non-inferiority between pedagogical approaches. Educators should remember that online learning is one component of the overall FC model. An overhaul of the broader teaching mentality and existing curriculum is required to realise the paradigm shift offered by the FC model. Thus, simple conversion of existing lecture notes to an electronic format will not effectively facilitate learning. There is a need for full utilisation of software/technologies to prepare multimedia/truly interactive micromodules; this approach is more likely to enhance student learning experiences. It is also challenging to develop effective methods for assessment of student competencies. Educators should support and collaborate with clinicians in this regard, thereby complementing each other’s efforts.4 12 13 14 15

In addition to video lectures, online platforms can be used to deliver diverse educational content, including interactive multimedia learning modules, discussion forums, polling, and virtual patients. We deliberately excluded these materials for the duration of this study because they represent distinct instructional configurations in terms of content and interactivity. The combination of interactive elements and lecture into a single intervention group would have confounded and invalidated the results.6 8 Thus, the video lectures solely consisted of slide decks, narration, and video. More studies are needed to determine how to best incorporate these teaching approaches into the instructional design of future curricula.11

The present study focused on the transfer of clinical knowledge and management of common urological symptoms via micromodules. Future research should examine whether online lectures can also effectively transfer practical procedural skills. Because of time constraints and the curriculum system, exposure during the clerkship period is extremely limited. Therefore, the current instructional approach for physical examination and basic clinical procedures (eg, insertion of urethral and central venous catheters) is often informal, opportunistic, and unstructured. Further studies may clarify the role of online education in procedural training.

Conclusion

Online micromodules were non-inferior to a traditional didactic lecture in terms of knowledge transfer focused on urology topics. Further enhancement of the interactive elements of the instructional medium will improve learning experience. Micromodule utilisation can be optimised during the development of the FC model of teaching. In times such as the recent pandemic era, where social distancing must be maintained throughout the educational process, there is an urgent need for curriculum reform that maximises the use of technology to enhance medical student learning.

Author contributions

Concept or design: CF Ng.
Acquisition of data: CH Yee, PKF Chiu, JYC Teoh, FPT Lai.
Analysis or interpretation of data: K Lim.
Drafting of the manuscript: K Lim, CF Ng.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

As editors of the journal, CF Ng and JYC Teoh were not involved in the peer review process. Other authors have disclosed no conflicts of interest.

Funding/support

This research received support from the Micro-Module Courseware Development Grant Scheme of The Chinese University of Hong Kong, Hong Kong (Ref No.: 3210817).

Ethics approval

This research aimed to improve instruction through the use of educational tests administered to the participants. All participants provided informed consent without the collection of personal or sensitive data.

References

1. Al Samaraee A. The impact of the COVID-19 pandemic on medical education. Br J Hosp Med (Lond) 2020;81:1-4. Crossref

2. Mehanna WN. e-Pedagogy: the pedagogies of e-learning. Res in Learn Technol 2004;12:279-93. Crossref

3. Stuart J, Rutherford RJ. Medical student concentration during lectures. Lancet 1978;2:514-6. Crossref

4. Koo CL, Demps EL, Farris C, Bowman JD, Panahi L, Boyle P. Impact of flipped classroom design on student performance and perceptions in a pharmacotherapy course. Am J Pharm Educ 2016;80:33. Crossref

5. Rotellar C, Cain J. Research, perspectives, and recommendations on implementing the flipped classroom. Am J Pharm Educ 2016;80:34. Crossref

6. Cook DA, McDonald FS. E-learning: is there anything special about the “E”? Perspect Biol Med 2008;51:5-21. Crossref

7. Johnson L, Becker S, Estrada V, Freeman A. NMC horizon report: 2014 higher education edition. Austin (Texas): The New Media Consortium. Available from: https://www.learntechlib.org/p/130341/. Accessed 10 Mar 2020.

8. Sinclair PM, Kable A, Levett-Jones T, Booth D. The effectiveness of Internet-based e-learning on clinician behaviour and patient outcomes: a systematic review. Int J Nurs Stud 2016;57:70-81. Crossref

9. Cook DA. The research we still are not doing: an agenda for the study of computer-based learning. Acad Med 2005;80:541-8. Crossref

10. Tang B, Coret A, Qureshi A, Barron H, Ayala AP, Law M. Online lectures in undergraduate medical education: scoping review. JMIR Med Educ 2018;4:e11. Crossref

11. Cook DA, Levinson AJ, Garside S, Dupras DM, Erwin PJ, Montori VM. Internet-based learning in the health professions: a meta-analysis. JAMA 2008;300:1181-96. Crossref

12. AAMC Institute for Improving Medical Education. Effective use of educational technology in medical education. Colloquium on educational technology: recommendations and guidelines for medical educators. 2007. Available from: https://store.aamc.org/downloadable/download/sample/sample_id/111/. Accessed 10 Mar 2020.

13. Sezer B, Abay E. Looking at the impact of the flipped classroom model in medical education. Scand J Educ Res 2019;63:853-68. Crossref

14. Hurtubise L, Hall E, Sheridan L, Han H. The flipped classroom in medical education: engaging students to build competency. J Med Educ Curric Dev 2015;2:35-43. Crossref

15. Hew KF, Lo CK. Flipped classroom improves student learning in health professions education: a meta-analysis. BMC Med Educ 2018;18:38. Crossref

Effect of location on out-of-hospital cardiac arrests involving older adults in Hong Kong: secondary analysis of a territory-wide cohort

Hong Kong Med J 2023 Apr;29(2):142–9 | Epub 29 Mar 2023

https://doi.org/10.12809/hkmj209140

ORIGINAL ARTICLE

Effect of location on out-of-hospital cardiac arrests involving older adults in Hong Kong: secondary analysis of a territory-wide cohort

Ronald TM Wong, MB, BS, FHKAM (Emergency Medicine)

Department of Emergency Medicine, The University of Hong Kong, Hong Kong SAR, China

Corresponding author: Dr Ronald TM Wong (drwongtmr@gmail.com)

Full paper in PDF

Abstract

Introduction: Most out-of-hospital cardiac arrests in Hong Kong involve older adults. The likelihood of survival varies among locations. This study investigated patient and bystander characteristics, as well as the timing of interventions, that affect the prevalences of shockable rhythm and survival outcomes among cardiac arrests involving older adults in homes, on streets, and in other public places.

Methods: This secondary analysis of a territory-wide historical cohort used data collected by the Fire Services Department of Hong Kong from 1 August 2012 to 31 July 2013.

Results: Bystander cardiopulmonary resuscitation was primarily performed by relatives in homes but not in non-residential locations. The intervals in terms of receipt of emergency medical services (EMS) call, initiation of bystander cardiopulmonary resuscitation, and receipt of defibrillation were longer for cardiac arrests that occurred in homes. The median interval for EMS to reach patients was 3 minutes longer in homes than on streets (P<0.001). Forty-seven percent of patients who developed cardiac arrest on streets had a shockable rhythm within the first 5 minutes after receipt of EMS call. Defibrillation within 15 minutes after receipt of EMS call was an independent predictor of 30-day survival (odds ratio=4.07; P=0.02). Fifty percent of patients who received defibrillation within 5 minutes in non-residential locations survived.

Conclusion: There were significant location-related differences in patient and bystander characteristics, interventions, and outcomes among cardiac arrests involving older adults. A large proportion of patients had a shockable rhythm in the early period after cardiac arrest. Good survival outcomes in out-of-hospital cardiac arrests involving older adults can be achieved through early bystander defibrillation and intervention.

New knowledge added by this study

Among out-of-hospital cardiac arrests involving older adults that occurred at different locations, there were significant differences in patient and bystander characteristics, as well as prehospital interventions, which influenced survival outcomes.
Many older adults who experienced cardiac arrest in non-residential locations had a shockable rhythm in the early period after receipt of emergency medical services (EMS) call, and early defibrillation was associated with favourable survival outcomes.
Low rates of shockable rhythm and significant delays in bystander and EMS processes were observed within homes.

Implications for clinical practice or policy

Additional measures are needed to overcome bystander inertia.
Interventions to mitigate the adverse factors related to cardiac arrests occurring in older adult households, such as volunteer dispatch via mobile applications, should be considered.

[Abstract in Chinese]

Introduction

The proportion of older adults in Hong Kong is expected to increase from 18% in 2019 to 26% by the year 2029.1 Overcrowding is a serious problem, such that population densities of 57 530 people/km² are present in ageing districts.2 3 Most residents of Hong Kong live in high-rise apartments that require elevators for access, but most elevators cannot accommodate an ambulance stretcher with a patient in a supine position.4 More than 50% of out-of-hospital cardiac arrest (OHCA) events occur in private homes, a location that is associated with poor survival outcomes.5 The proportion of domestic households consisting solely of people aged ≥65 years has increased by approximately 24% between 2011 and 2016, from 8.4% to 10.4%.6 Considering these demographic changes, there is a need for improved overall understanding of the prehospital management of cardiac arrests that involve older adults in homes and other locations. This study investigated patient characteristics, types of bystanders involved, and prehospital interventions that were associated with differences in survival outcomes among cardiac arrests involving older adults in homes, compared with cardiac arrests on streets and in public areas excluding streets (PAES).

Methods

Study design and setting

This secondary analysis focused on a historical cohort from a previous study.5 The Emergency Ambulance Service of the Fire Services Department (FSD) provides most emergency medical services (EMS) in Hong Kong through a one-tiered system that serves the entire 1104 km² region. At the time of data collection, the population was around 7.1 million.7 Ambulance personnel are required to perform cardiopulmonary resuscitation (CPR) on and transfer most cases of OHCA to hospitals. A small number of patients with obvious post-mortem changes (eg, rigor mortis) may be directly transferred to the public mortuary; such patients were not included in this study. Fire Services Department ambulances will only transfer patients to emergency departments under the Hospital Authority. At the time of data collection, callers requesting for EMS for OHCA patients were not provided with post-dispatch instructions to perform CPR.

Participants

This secondary analysis included all patients with OHCA who were transferred to the Emergency Departments (EDs) by FSD ground ambulances from 1 August 2012 to 31 July 2013. Exclusion criteria were cardiac arrests caused by trauma, patients not transferred by ground ambulance, and patients directly transferred to the public mortuary. After patient selection from the primary dataset, the following additional exclusions were made: cardiac arrests that involved patients aged <65 years, occurred within residential care homes for the elderly, or occurred in the ambulance en route to hospital.

Data sources

Data regarding patient characteristics and prehospital management were prospectively collected by EMS personnel who were directly involved in prehospital care for patients who experienced OHCA. The collected data included patient age and sex, location of cardiac arrest, whether the cardiac arrest was witnessed and the identity of the witness, whether bystander CPR was performed and who performed it, whether defibrillation with an automated external defibrillator (AED) was performed, what electrocardiogram rhythm was first detected, the timings of prehospital events (recognition of cardiac arrest, receipt of EMS call, initiation of bystander CPR, initiation of first defibrillation, EMS arrival at patient’s side, initiation of CPR by EMS personnel, and arrival at the ED), and return of spontaneous circulation (ROSC) before ED arrival.

Electronic medical records at the relevant ED (Accident and Emergency Information System, Hong Kong Hospital Authority) were reviewed to determine the time of defibrillation and time of ROSC at the ED, as well as whether a patient survived until admission. A patient was assumed to have received no resuscitative intervention unless specific documentation was present in the ED record. Neurological status upon discharge and survival at 30 days after cardiac arrest were determined from a territory-wide electronic medical record database (Clinical Management System, Hong Kong Hospital Authority).

Variables

Streets were defined as paved thoroughfares for pedestrians, including sidewalks. Public areas excluding streets were other areas that were accessible by the public throughout the day; these included outdoors (eg, parks and markets) and indoor facilities (eg, eateries, places of recreation, and day care facilities for older adults). Bystanders were defined in accordance with the guidelines of the Utstein Resuscitation Registry Templates for Out-of-Hospital Cardiac Arrest.8 Fire Services Department first responders dispatched to the scene were classified as EMS personnel. Older adult care workers (OACWs) are individuals who care for residents in various private and public housing arrangements for older adults. Older adult care workers accompanying patients were not dispatched as part of the organised emergency rescue team; thus, they were classified as bystanders. Public access defibrillation (PAD) was defined as a defibrillation shock delivered from an AED when a bystander performed CPR. Shocks delivered when FSD first responders performed CPR were excluded.

Time intervals were rounded to the nearest minute. The decision interval was the interval between recognition of cardiac arrest and receipt of EMS call. Call-to-bystander CPR was the interval between receipt of EMS call and initiation of bystander CPR. Call-to-EMS arrival was the interval between receipt of EMS call and EMS arrival at the patient’s side. Time of first defibrillation was defined as the time of the earliest of the following three events: PAD, defibrillation by EMS, or defibrillation in the ED. Call-to-bystander CPR intervals were grouped as 0-2, 3-5, 6-8, 9-11, and 12-31 minutes, as well as no bystander CPR. Call-to-first defibrillation intervals were grouped as 0-5, 6-10, 11-15, 16-20, and 21-55 minutes, as well as no defibrillation (>55 minutes/not applicable).

Post-cardiac arrest neurological status was classified using the 5-point Glasgow-Pittsburgh Cerebral Performance Categories (CPC) scale. In the scale, CPC 1 represents patients with good cerebral performance; CPC 2 includes patients who can manage activities of daily living independently or participate in part-time work in a sheltered environment; CPC 3 to CPC 5 ranges from patients who are unable to live independently because of cerebral disability to patients who have experienced brain death. Patients with CPC 1 or CPC 2 were presumed to have a favourable neurological outcome.

Statistical methods

Patient characteristics, interventions, and outcomes were analysed using descriptive statistics. Pearson’s χ² test was used to compare categorical variables; Fisher’s exact test was used if >20% of expected counts were <5. The Kruskal–Wallis rank sum test was used to compare non-parametric time intervals. A P value of <0.05 was considered statistically significant. Predictors of 30-day survival were analysed using univariate and multivariate logistic regression; findings were reported as odds ratios (ORs) with 95% confidence intervals. Adjusted variables included age; sex; arrest location; person witnessing the arrest (relative, OACW or other bystanders, EMS personnel, or unwitnessed); person performing bystander CPR (no bystander CPR, OACW, relative, or other); PAD (yes or no); first monitored rhythm (asystole, pulseless electrical activity, ventricular fibrillation/ventricular tachycardia); and call-to-EMS arrival, call-to-bystander CPR, and call-to-first defibrillation intervals.

Statistical analysis was performed using R software, version 3.6.1 (R Foundation for Statistical Computing, Austria). The original study was approved by the Institutional Review Board of The University of Hong Kong/Hospital Authority Hong Kong West Cluster (Ref No.: UW 15-599). No new data were collected for secondary analysis. This manuscript was prepared in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) reporting guidelines.

Results

Participant selection and characteristics

Figure 1 describes patient selection from the primary dataset. The original cohort comprised 5154 patients with OHCA who were transferred to the ED by FSD ground ambulances. After the application of exclusion criteria, 2255 patients were included in the analysis. Table 1 compares the patient and bystander characteristics, interventions, and outcomes of OHCA occurring in homes, in PAES, and on streets. Patients who experienced cardiac arrest in homes were significantly older (approximately 5 years; P<0.001) than patients who experienced cardiac arrest on streets or in PAES. In all groups, there were more male patients; the sex disparity was the greatest in the streets group, followed by the PAES group.

Figure 1. Patient selection from primary dataset

Table 1. Patient and bystander characteristics, interventions, and survival outcomes of out-of-hospital cardiac arrests involving older adults in homes, on streets, and in other public areas

Furthermore, most cardiac arrests (66.4% among all patients; P<0.001; Table 1) were unwitnessed. Relatives were the most common type of bystander present in witnessed arrests, whereas there were significant differences in the involvement of OACWs, EMS personnel, and other individuals at the three locations. Compared with EMS personnel, there were more OACWs as bystanders in PAES and more bystanders, represented by the ‘other’ group, in PAES and on streets. There was a significant difference in the proportion of bystanders performing CPR among the three locations (P<0.001), as illustrated in Figure 2. The bystander CPR rate was the highest in PAES and the lowest in homes. Among the nine patients who received PAD, six received it when OACWs provided CPR in PAES.

Figure 2. Identity of bystanders performing cardiopulmonary resuscitation in homes, in public areas excluding streets (PAES), and on streets

Initial monitored rhythm

Notably, asystole was the most common initial monitored rhythm (81.2% among all patients; P<0.001; Table 1). However, cardiac arrests on streets and in PAES had significantly higher rates of shockable rhythm and PEA, compared with cardiac arrests in homes. The prevalences of shockable rhythm relative to the time from receipt of EMS call and the location of cardiac arrest are shown in Figure 3. The highest rates of shockable initial rhythm (SIR) were observed within the first 5 and 10 minutes after receipt of EMS call for cardiac arrests on streets, which were 47% (8/17) and 41% (17/42), respectively.

Figure 3. Location, time from receipt of emergency medical services (EMS) call, and proportion of shockable rhythm of cardiac arrests

Timing of interventions

Patients with cardiac arrest in homes had significantly longer intervals in terms of receipt of EMS call, initiation of bystander CPR, and receipt of defibrillation (all P<0.001; Table 1). The median interval for EMS to reach patients was 3 minutes longer in homes than on streets. The interval between recognition of cardiac arrest and receipt of EMS call was 0 minutes in 57.5% of patients (1297/2255).

Survival and neurological outcomes

Additionally, patients with cardiac arrest in homes had significantly lower rates of ROSC, 30-day survival, and favourable neurological outcomes (all P<0.001; Table 1).

Independent predictors of 30-day survival are shown in Table 2. Older age and longer call-to-EMS arrival interval both decreased the overall likelihood of survival (ORs of 0.92 and 0.87, respectively). Pulseless electrical activity and ventricular fibrillation/ventricular tachycardia increased the likelihood of survival compared with asystole (ORs of 6.4 and 15.6, respectively). Cardiac arrest witnessed by EMS personnel and defibrillation within 15 minutes after receipt of EMS call increased the overall likelihood of survival (ORs of 6.23 and 4.07, respectively).

Table 2. Independent predictors of 30-day survival of cardiac arrests

The relationship among the location, timing of defibrillation, and 30-day survival of cardiac arrest is shown in Figure 4. Overall, patients who received defibrillation within 5 minutes and at 6 to 10 minutes after receipt of EMS call had survival rates of 33% (2/6) and 17% (15/86), respectively. For patients who received defibrillation on streets/in PAES within 5 minutes and at 6 to 10 minutes after receipt of EMS call, the survival rates were 50% (2/4) and 22% (10/45), respectively. Two patients in the homes group received defibrillation within 5 minutes; the survival rate was 0% (0/2).

Figure 4. Relationship among location, timing of defibrillation, and survival of cardiac arrests. Only two patients in the home group received defibrillation within 5 minutes. Streets and public areas excluding streets are combined because of the small number of patients in some subgroups

Cardiac arrest at home was a predictor of survival in univariate analysis (OR=0.076, 95% confidence interval [CI]=0.038-0.15) but not in multivariable analysis (OR=0.65, 95% CI=0.22-1.90). The effect of location on survival was mediated by the first monitored rhythm, and the call-to-EMS arrival interval.

Discussion

This study investigated factors that affect the prevalences of shockable rhythm and survival outcomes among cardiac arrests involving older adults in Hong Kong. The patient characteristics, proportion of witnessed arrests, and rates of SIR and PAD for cardiac arrests involving older adults in homes were similar between the present study and a previous analysis in Japan.9 Unlike many western countries, EMS personnel in Hong Kong and Japan generally do not terminate resuscitation in the field; this similarity facilitates comparison of data between the two studies. A notable difference was that in Japanese homes, 45% of older patients received bystander CPR; this receipt of CPR was associated with rate of ROSC, 30-day survival, and favourable neurological outcomes that were threefold higher than the corresponding rates in Hong Kong.

Bystander cardiopulmonary resuscitation

The bystander CPR rate in Hong Kong homes was low (3.8%) [Table 1], and there was a substantial delay in its initiation. Although the type of relatives involved as bystanders was not recorded in the present study, considering the proportion of older adult households in Hong Kong,6 many of the relatives presumably were cohabiting older adults. Such individuals may not be able to follow telephone instructions to perform CPR because of physical limitations or emotional distress10; thus, the provision of post-dispatch instructions and enhancement of community-wide CPR training will not improve survival among these patients.11 Although high-rise apartments create barriers to EMS personnel, they also increase the likelihood that trained volunteers will be present in the vicinity, where they may be dispatched using mobile applications.12 13 14

In non-residential locations, most bystanders performing CPR were not relatives of the patients. Fear of legal consequences is reportedly a major cause for intervention inertia in this situation.15 A previous survey in Hong Kong, in which one-third of respondents had prior first aid training, revealed that nearly all respondents were willing to call for help but only one-fifth were willing to perform bystander CPR.16 These findings suggest that knowledge transfer is insufficient to overcome bystander inertia in Hong Kong. Training programmes should ensure that factors inhibiting intervention (eg, legal concerns, fear of disease transmission, and bystander effect) are addressed.17 18

Shockable initial rhythm

Previous studies in Hong Kong revealed low rates of SIR in patients with OHCA, ranging from 5% to 14%, along with dismal survival rates of 0.6% to 3%.1 19 20 These low rates imply that aggressive bystander interventions (eg, defibrillation for older adults) are futile. However, the findings of the present study indicate that older adults in non-residential locations have much higher SIR rates in the initial 10 minutes after receipt of EMS call; moreover, early defibrillation is an independent predictor of survival among such patients, and high survival rates can be achieved with early defibrillation.

The present study revealed a 2% per-minute decrease in the rate of SIR. This is similar to the findings in a large multinational study from northern Europe.21 Differences in SIR rates between residential and non-residential locations may be partly related to patient factors (eg, age and presence of co-morbidities); they could also be related to differences in the decision interval (ie, time elapsed between recognition of cardiac arrest [as reported by a bystander] and receipt of EMS call). A previous study in Hong Kong showed that efforts to seek advice from relatives often contributed to delayed receipt of EMS call.4 Longer decision intervals and consequential delays in EMS arrival lead to interactions with later parts of the shockable rhythm downslope and lower SIR rates. In practice, the recall of decision intervals by bystanders is unreliable. This is consistent with the decision interval of 0 minutes reported by most bystanders in the present study. Despite this confounding factor, the findings in this study indicate that bystanders should not hesitate to provide aggressive resuscitation and early defibrillation for older patients.

Public access defibrillation

Notably, very few patients received PAD in this study, and most instances of PAD administration were performed by OACWs in PAES. According to a nationwide study in Japan, 16.5% of patients received PAD during witnessed ventricular fibrillation cardiac arrest.22 Estimation of the AED coverage rate in Hong Kong using a horizontal level walking route distance model revealed that only 11% of patients with OHCA would have an AED within 100 m.23 Considering the large number of OHCA events occurring within high-rise buildings, the actual coverage rate is presumably lower. Furthermore, there is evidence that most people in Hong Kong do not know the location of the AED nearest to their home or workplace.16 Unless AEDs are easy to locate and readily accessible at all times, PAD rates will remain low.24

Barriers to rescue in high-rise buildings

In a previous study in Hong Kong, the proportions of patients with OHCA who could be accessed by elevator or stairs and by stairs alone were 74% and 14%, respectively.4 In the present study, the median interval for EMS to reach patients was 3 minutes longer in homes than on streets. This represents the ‘vertical response time’ component of the call-to-EMS arrival interval.25 In a previous study, survival was lower among patients who experienced cardiac arrest at higher levels within buildings.26 Nearly 70% of lifts in Hong Kong do not have sufficient area to accommodate the ambulance stretcher.4 Therefore, the vertical response time leads to a delay in EMS interventions and deterioration in CPR quality, both of which may contribute to the poor outcomes of cardiac arrests that occur in homes. The use of circulatory adjuncts to enhance cerebral perfusion during head-up position CPR within lifts should be considered.27

Limitations

Importantly, only patients transported to hospital by FSD ground ambulances were included in this study; a small number of patients with OHCA may have been transported to hospital by other means.

Furthermore, data regarding the timings of recognition of cardiac arrest, bystander CPR, and PAD obtained from bystanders may have been subject to response bias. The lack of blinding of emergency department personnel towards patient factors (eg, absence of shockable rhythm and prehospital defibrillation, longer time to ROSC, co-morbidities, and advanced age) may have led to selection bias regarding treatment decisions, including the termination of resuscitation, arrangement of intensive care unit resources, and coronary angiography; such bias has been reported to negatively influence the survival rate.28 Data regarding pre-arrest co-morbidity and functional status were not available, which may have resulted in a confounding effect on survival outcomes. Additionally, a small number of patients received defibrillation within 5 minutes. All of the factors listed here may have affected the accuracy of conclusions drawn from this subset.

This study was based on territory-wide data collected in 2012 to 2013. Thus, it may not reflect the current situation because of changes in patient demography, prevalence of shockable rhythm, and survival enhancement interventions introduced in the past several years. A large multinational study in northern Europe investigated the rate of SIR among OHCA events occurring in residential and public locations from 2006 to 2015. The rate of SIR in public locations remained stable during that period. A decrease was observed in residential locations between 2006 and 2010, but the proportion has remained stable since 2011.21 Therefore, despite these limitations, the findings of the present study add to the broader understanding of OHCA involving older adults.

Conclusion

This study revealed significant differences in the patient and bystander characteristics and prehospital interventions among cardiac arrests involving older adults that occurred in homes, on streets, and in other public locations. Many older adults who experienced cardiac arrest in non-residential locations had a shockable rhythm in the early period after receipt of EMS call. Early defibrillation, an independent predictor of survival, was associated with favourable survival outcomes in older adults. These findings suggest that bystanders should provide aggressive resuscitation, including early defibrillation. Additionally, low rates of shockable rhythm and significant delays in bystander and EMS processes were observed within homes. New interventions (eg, volunteer dispatch via mobile applications) are needed to overcome unfavourable factors that affect cardiac arrests occurring within older adult households. Finally, the overall bystander CPR rate was low, indicating that additional measures are needed to overcome bystander inertia. The insights from this study will help to improve survival outcomes in OHCAs involving older adults.

Author contributions

The author contributed to the concept or design, analysis or interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. The author had full access to the data, contributed to the study, approved the final version for publication, and takes responsibility for its accuracy and integrity.

Conflicts of interest

The author has no conflicts of interest to disclose.

Acknowledgement

The author thanks Dr Ling-pong Leung, Emergency Medicine Unit of The University of Hong Kong, for providing the original dataset and permitting its use for secondary analysis in the study. The author also thanks Mr Min Fan and Ms Lujie Chen, both from Emergency Medicine Unit of The University of Hong Kong, for their technical support in this study.

Funding/support

This research received no specific grant from any funding in the public, commercial, or not-for-profit sectors.

Ethics approval

This study is a secondary analysis of a historical cohort study which was approved by the Institutional Review Board of The University of Hong Kong/Hospital Authority Hong Kong West Cluster (Ref No.: UW 15-599). The requirement for informed patient consent was waived because of the retrospective study design. All patient data in the dataset were anonymous.

References

1. Census and Statistics Department, Hong Kong SAR Government. Hong Kong population projections 2020-2069. Available from: https://www.censtatd.gov.hk/hkstat/sub/sp190.jsp?productCode=B1120015. Accessed 7 Nov 2020.

2. Census and Statistics Department, Hong Kong SAR Government. The profile of Hong Kong population analysed by District Council district, 2018. Table 2. Proportion of land-based non-institutional population by District Council district and age group, 2017. Available from: https://www.censtatd.gov.hk/en/data/stat_report/ product/FA100096/att/B71807FB2018XXXXB0100.pdf. Accessed 24 Mar 2020.

3. Census and Statistics Department, Hong Kong SAR Government. 2016 Population By-census Office. Main tables (demographic). Population density by District Council district and year. 2017. Available from: https://www.bycensus2016.gov.hk/en/bc-mt.html?search=A202. Accessed 24 Mar 2020.

4. Leung LP, Wong TW, Tong HK, Lo CB, Kan PG. Out-of-hospital cardiac arrest in Hong Kong. Prehosp Emerg Care 2001;5:308-11. Crossref

5. Fan KL, Leung LP, Siu YC. Out-of-hospital cardiac arrest in Hong Kong: a territory-wide study. Hong Kong Med J 2017;23:48-53. Crossref

6. Census and Statistics Department, Hong Kong SAR Government. 2016 Population By-census. Domestic households in Hong Kong. Available from: https://www.bycensus2016.gov.hk/en/Snapshot-04.html. Accessed 24 Mar 2020.

7. Census and Statistics Department, Hong Kong SAR Government. 2011 Hong Kong Population Census. Census results. Available from: https://www.censtatd.gov.hk/en/scode170.html. Accessed 17 Mar 2023.

8. Perkins GD, Jacobs IG, Nadkarni VM, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: update of the Utstein Resuscitation Registry Templates for Out-of-Hospital Cardiac Arrest: a statement for healthcare professionals from a task force of the International Liaison Committee on Resuscitation (American Heart Association, European Resuscitation Council, Australian and New Zealand Council on Resuscitation, Heart and Stroke Foundation of Canada, InterAmerican Heart Foundation, Resuscitation Council of Southern Africa, Resuscitation Council of Asia); and the American Heart Association Emergency Cardiovascular Care Committee and the Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation. Circulation 2015;132:1286-300. Crossref

9. Okabayashi S, Matsuyama T, Kitamura T, et al. Outcomes of patients 65 years or older after out-of-hospital cardiac arrest based on location of cardiac arrest in Japan. JAMA Netw Open 2019;2:e191011. Crossref

10. Dami F, Carron PN, Praz L, Fuchs V, Yersin B. Why bystanders decline telephone cardiac resuscitation advice. Acad Emerg Med 2010;17:1012-5. Crossref

11. Kiyohara K, Nishiyama C, Matsuyama T, et al. Out-of-hospital cardiac arrest at home in Japan. Am J Cardiol 2019;123:1060-8. Crossref

12. Ringh M, Rosenqvist M, Hollenberg J, et al. Mobile-phone dispatch of laypersons for CPR in out-of-hospital cardiac arrest. N Engl J Med 2015;372:2316-25. Crossref

13. Smith CM, Wilson MH, Ghorbangholi A, et al. The use of trained volunteers in the response to out-of-hospital cardiac arrest—the GoodSAM experience. Resuscitation 2017;121:123-6. Crossref

14. Mao DR, Ong ME. High-rise residential resuscitation: scaling the challenge. CMAJ 2016;188:399-400. Crossref

15. Coons SJ, Guy MC. Performing bystander CPR for sudden cardiac arrest: behavioral intentions among the general adult population in Arizona. Resuscitation 2009;80:334-40. Crossref

16. Fan KL, Leung LP, Poon HT, Chiu HY, Liu HL, Tang WY. Public knowledge of how to use an automatic external defibrillator in out-of-hospital cardiac arrest in Hong Kong. Hong Kong Med J 2016;22:582-8. Crossref

17. Resuscitation Council UK. Cardiopulmonary resuscitation, automated defibrillators and the law. 2018. Available from: https://www.resus.org.uk/sites/default/files/2020-05/CPR%20AEDs%20and%20the%20law%20%285%29.pdf. Accessed 24 Mar 2020. Crossref

18. Sayre MR, Barnard LM, Counts CR, et al. Prevalence of COVID-19 in out-of-hospital cardiac arrest: implications for bystander cardiopulmonary resuscitation. Circulation 2020;142:507-9. Crossref

19. Wong TW, Yeung KC. Out-of-hospital cardiac arrest: two and a half years experience of an accident and emergency department in Hong Kong. J Accid Emerg Med 1995;12:34-9. Crossref

20. Lau CL, Lai JC, Hung CY, Kam CW. Outcome of out-of-hospital cardiac arrest in a regional hospital in Hong Kong. Hong Kong J Emerg Med 2005;12:224-7. Crossref

21. Oving I, de Graaf C, Karlsson L, et al. Occurrence of shockable rhythm in out-of-hospital cardiac arrest over time: a report from the COSTA group. Resuscitation 2020;151:67-74. Crossref

22. Kitamura T, Kiyohara K, Sakai T, et al. Public-access defibrillation and out-of-hospital cardiac arrest in Japan. N Engl J Med 2016;375:1649-59. Crossref

23. Fan M, Fan KL, Leung LP. Walking route–based calculation is recommended for optimizing deployment of publicly accessible defibrillators in urban cities. J Am Heart Assoc 2020;9:e014398 Crossref

24. Agerskov M, Nielsen AM, Hansen CM, et al. Public access defibrillation: great benefit and potential but infrequently used. Resuscitation 2015;96:53-8. Crossref

25. Silverman RA, Galea S, Blaney S, et al. The “vertical response time”: barriers to ambulance response in an urban area. Acad Emerg Med 2007;14:772-8. Crossref

26. Drennan IR, Strum RP, Byers A, et al. Out-of-hospital cardiac arrest in high-rise buildings: delays to patient care and effect on survival. CMAJ 2016;188:413-9. Crossref

27. Moore JC, Segal N, Debaty G, Lurie KG. The “do’s and don’ts” of head up CPR: lessons learned from the animal laboratory. Resuscitation 2018;129:e6-e7. Crossref

28. Winther-Jensen M, Kjaergaard J, Hassager C, et al. Resuscitation and post resuscitation care of the very old after out-of-hospital cardiac arrest is worthwhile. Int J Cardiol 2015;201:616-23. Crossref

The real-world impact of the COVID-19 pandemic on patients with cancer: a multidisciplinary cross-sectional survey

Hong Kong Med J 2023 Apr;29(2):132–41 | Epub 14 Apr 2023

https://doi.org/10.12809/hkmj219326

ORIGINAL ARTICLE

The real-world impact of the COVID-19 pandemic on patients with cancer: a multidisciplinary cross-sectional survey

Kelvin KH Bao, MB BChir, FRCR; Ka-man Cheung, MB, ChB, FRCR; James CH Chow, MB, ChB, FRCR; Carmen WL Leung, MB, ChB, FRCR; Kam-hung Wong, MB, ChB, FRCR

Department of Clinical Oncology, Queen Elizabeth Hospital, Hong Kong SAR, China

Corresponding author: Dr Kelvin KH Bao (bkh641@ha.org.hk)

Full paper in PDF

Abstract

Introduction: The coronavirus disease 2019 (COVID-19) pandemic has caused unprecedented disruptions to cancer care worldwide. We conducted a multidisciplinary survey of the real-world impact of the pandemic, as perceived by patients with cancer.

Methods: A total of 424 patients with cancer were surveyed using a 64-item questionnaire constructed by a multidisciplinary panel. The questionnaire examined patient perspectives regarding COVID-19–related effects (eg, social distancing measures) on cancer care delivery, resources, and healthcare-seeking behaviour, along with the physical and psychosocial aspects of patient well-being and pandemic-related psychological repercussions.

Results: Overall, 82.8% of respondents believed that patients with cancer are more susceptible to COVID-19; 65.6% expected that COVID-19 would delay anti-cancer drug development. Although only 30.9% of respondents felt that hospital attendance was safe, 73.1% expressed unaltered willingness to attend scheduled appointments; 70.3% of respondents preferred to receive chemotherapy as planned, and 46.5% were willing to accept changes in efficacy or side-effect profile to allow an outpatient regimen. A survey of oncologists revealed significant underestimation of patient motivation to avoid treatment interruptions. Most surveyed patients felt that there was an insufficient amount of information available concerning the impact of COVID-19 on cancer care, and most patients reported social distancing–related declines in physical, psychological, and dietary wellness. Sex, age, education level, socio-economic status, and psychological risk were significantly associated with patient perceptions and preferences.

Conclusion: This multidisciplinary survey concerning the effects of the COVID-19 pandemic revealed key patient care priorities and unmet needs. These findings should be considered when delivering cancer care during and after the pandemic.

New knowledge added by this study

Most patients with cancer (73.1%) reported that their willingness to attend scheduled oncology appointments was not affected by the pandemic. All surveyed oncologists underestimated patient motivation to avoid treatment interruptions.
Patient acceptance of telerehabilitation varied according to age and socio-economic status, whereas the negative impact of social distancing on patients with cancer were substantial and multidimensional.
Psychometric analyses can stratify patients with cancer into psychological risk groups, based on their distinct perceptions of the pandemic.

Implications for clinical practice or policy

These findings will help to increase awareness of the effects of the coronavirus disease 2019 pandemic on patients with cancer, revealing their priorities and unmet needs.
This work aligns the expectations of oncologists and patients with cancer with respect to modifications of cancer services during the pandemic.
These results will promote better resource allocation and earlier multidisciplinary interventions to reduce pandemic-related impact on at-risk populations.

[Abstract in Chinese]

Introduction

The coronavirus disease 2019 (COVID-19) pandemic poses an unprecedented threat to health systems worldwide. During the first year of the pandemic, there were more than 110 million confirmed COVID-19 cases globally and more than 2.6 million deaths.1 In terms of scale, the number of COVID-19 cases during that period was at least sixfold more than the global number of new cancer cases in 2018; the mortality during that period exceeded the combined mortalities of lung cancer and breast cancer in 2018.2 The many consequences of COVID-19 have included unprecedented disruptions to cancer care services,3 4 5 such as cancellations of outpatient appointments to delays in scheduled systemic treatments and radiotherapy; during periods of increased transmission, such disruptions have forced oncologists to make difficult decisions in attempts to balance patient protection and disease control. There have been similar impact to the delivery of oncology-related allied health services, including physical therapy and occupational therapy,6 dietetics,7 diagnostic imaging,8 and psychological services for patients with cancer9; there has been a particularly large shift in the use of telemedicine. Throughout the COVID-19 pandemic, and particularly during periods of increased transmission, good multidisciplinary coordination has been a crucial aspect of cancer care. By acquiring comprehensive knowledge regarding perceptions of the pandemic, changes in healthcare-seeking behaviour, impact on daily life, and the newly emerged unmet needs (physical, socio-economic, and psychological) of patients with cancer, multidisciplinary cancer caregivers can customise and adjust their services accordingly, thus enabling appropriate resource allocation. Considering these challenges, we designed and conducted a prospective study to comprehensively examine the real-world impact of COVID-19 on patients with cancer; we sought to identify actionable solutions from the perspective of experienced multidisciplinary cancer caregivers.

Methods

A prospective survey regarding the perspectives of patients with cancer on the impact of the COVID-19 pandemic was jointly developed by a multidisciplinary team at Queen Elizabeth Hospital, Hong Kong that consisted of clinical oncologists, clinical psychologists, physiotherapists, occupational therapists, and dieticians specialising in cancer care. A pilot survey was administered to 88 patients, followed by interviews to further assess patient understanding of the questions and to develop additional items via thematic analysis. The multidisciplinary team then refined the questionnaire. The final version consisted of 64 items with a combination of Likert scales and polar questions; it covered topics such as patient perceptions of cancer care resources, treatment delivery and quality, changes in healthcare-seeking behaviour, adequacy of available pandemic-related information, social distancing–related adverse impact, and psychological repercussions of the pandemic. We also invited patients who were newly diagnosed with cancer to complete an extended questionnaire which focused on psychometric measurements of post-traumatic stress disorder (PTSD) [the PTSD Checklist for DSM-5 (PCL-5)],10 anxiety and depression (the Emotion Thermometers tool),11 and intolerance of uncertainty (the Intolerance of Uncertainty Scale-12 [IUS-12]).12 Patients were then stratified into risk groups. High-risk individuals had scores of ≥5 on the abbreviated PCL-5 scale,13 ≥3 on the Emotion Thermometers for depression or anxiety,14 and ≥25 on the IUS-12 scale.11 Associations between patient perceptions and psychological risk were then explored. Full details of the patient questionnaire are shown in online supplementary Table 1.

Furthermore, we surveyed clinical oncologists in Hong Kong (practising in Queen Elizabeth Hospital, United Christian Hospital, and Buddhist Hospital) regarding their perceptions of the pandemic; the oncologists were also asked to predict the responses of patients with cancer in various domains of interest.

The patient survey was conducted between 12 and 22 May 2020 at Queen Elizabeth Hospital. Patients with cancer and survivors aged ≥18 years who attended their outpatient oncology appointments were invited to participate. Patients who could not read English or Chinese were excluded, and participation was voluntary. Detailed survey information was provided on the questionnaire cover sheet, and a patient’s decision to participate in the survey was regarded as informed consent. Hardcopies of the questionnaire were anonymously completed by participants on site, then collected by dedicated nursing staff.

Data analysis

Descriptive analysis was used to describe various impact of the pandemic on patients. Patient demographics, disease characteristics, treatment details, and socio-economic information were summarised. Qualitative data are presented as the percentage of respondents who selected a particular response. Chi squared tests were used to determine associations between responses and categorical patient factors. P values <0.05 were considered indicative of statistical significance. Analyses were performed using SPSS software (Windows version 25.0; IBM Corp, Armonk [NY], United States).

Results

Demographic characteristics

Between 12 and 22 May 2020, 650 patients with cancer were invited to participate in the survey; 424 responses were received, yielding a response rate of 65.2%. Demographic and clinical characteristics of the participants are presented in the Table. Most survey respondents were female (70.0%), and more than half were aged 46 to 75 years. Nearly half (46.0%) of the respondents were receiving active cancer treatment. The cancer stage was III or below in half of the respondents and 20.3% of respondents were at stage IV; 29.2% of respondents were uncertain of their staging. Almost half (43.9%) of the respondents had a monthly family income of <HK$16 000 (around US$2000), which is Hong Kong’s 2018 poverty line for a family of three.15

Table. Demographic and clinical characteristics of patients with cancer (n=424)

Impact of coronavirus disease 2019 on cancer resources

As shown in online supplementary Table 1, most respondents (82.8%) believed that patients with cancer are more susceptible to COVID-19, while more than half (52.1%) believed that cancer-related resources will be depleted and 59.3% were concerned that healthcare workforce shortages during the pandemic would harm their treatment. Overall, 65.6% of respondents were concerned that COVID-19 would lead to delays in anti-cancer drug development. These concerns were significantly associated with education level of patients, in which the more educated respondents demonstrated less concern (tertiary level 57.6% vs secondary 64.4% vs primary 71.4%, P=0.01) [Fig 1a].

Figure 1. (a) Level of patient concern about the impact of coronavirus disease 2019 (COVID-19) on anti-cancer drug development by education level (n=424). (b) Effect of COVID-19 pandemic on patient willingness to attend hospital appointments by patient demographics (n=424). (c) Perceived availability of information about the impact of COVID-19 on patients with cancer by education level (n=424)

Impact of coronavirus disease 2019 on healthcare-seeking behaviour

As shown in online supplementary Table 1, fewer than one-third of respondents (30.9%) felt that it was safe to attend hospital appointments during the pandemic (the proportion was greater among men than among women: 37.8% vs 27.8%). Furthermore, most respondents reported that their willingness to attend oncology clinic appointments (73.1%) or undergo clinical tests (80.2%) was unaffected. Age (P=0.021), sex (P=0.029), and education level (P=0.003) were factors significantly associated with patient willingness; respondents aged 18-45 years or >75 years, female, and more educated individuals were more hesitant to attend their scheduled appointments (Fig 1b).

Compared with the pre-pandemic period, most respondents (79.0%) stated that they were equally willing (65.1%) or more willing (13.9%) to seek medical attention now if they felt unwell. Overall, 62.2% of respondents were equally willing or more willing to be hospitalised if requested by their oncologists. Male respondents were more willing to be hospitalised, compared with female respondents (68.3% vs 59.2%); additionally, respondents receiving radical treatment were more willing to be hospitalised, compared with respondents receiving palliative treatment (71.3% vs 59.0%).

Effects of social distancing on medical consultation and cancer treatment

Nearly all respondents (98.4%) felt that it was acceptable for medical staff to maintain an increased physical distance from patients during consultations, and most (83.3%) felt that it did not negatively impact the quality of their clinic experience. During the pandemic, some clinically stable patients were exempt from the requirement for oncologist examination prior to medication refills. Overall, 59.9% of respondents felt that such an arrangement should continue beyond the pandemic period (male respondents vs female respondents: 69.3% vs 55.9%; respondents receiving radical treatment vs respondents receiving palliative treatment: 54.4% vs 69.2%).

Concerning the effects of the pandemic on plans for cancer treatment, most respondents stated that their decisions to receive chemotherapy (70.3%) or radiotherapy (67.9%) were unaffected. However, 46.5% of the respondents were willing to accept changes in treatment efficacy or side-effect profile to allow an outpatient regimen; this preference was particularly strong among respondents receiving palliative treatment (61.5%), compared with respondents receiving radical treatment (33.7%).

Acquisition and adequacy of pandemic-related information

As shown in online supplementary Table 1, half of the respondents (49.1%) spent an average of 10 to 30 minutes daily interacting with news and information sources focused on COVID-19. Their most common news sources were television (41.7%), the internet (28.0%), and newspapers (12.9%). Only 3.5% of respondents received pandemic-related news or information from their hospitals. Young respondents (<45 years) were significantly more likely to receive news primarily from the internet, compared with older respondents (>75 years) (37.6% vs 13.4%; P=0.001); such a difference was also present between respondents with different education levels (tertiary education vs primary education: 35.5% vs 17.4%; P=0.001) and between respondents with different levels of monthly family income (>HK$40 000 vs <HK$16 000: 32% vs 25%; P=0.006).

Concerning the adequacy of information received regarding COVID-19 and its impact on patients with cancer, more than half of the respondents (54.1%) felt that it was inadequate; this sentiment was more prevalent among respondents with a higher education level, compared with those who were less educated (tertiary vs primary level: 60.2% vs 40.3%; P=0.017) [Fig 1c].

Effects of social distancing on allied health professional services

During the pandemic, social distancing became the new daily norm. Nearly half of the respondents (49.5%) reported exercising less, whereas 10.4% reported exercising more. In general, 55.4% of respondents noticed an overall deterioration in their physical well-being (Fig 2a); about one-third of respondents (32.1%) reporting reduced walking tolerance, and 25.9% of respondents noticed some reduction in limb power (online supplementary Table 1).

Figure 2. (a) Patient perception that being homebound during lockdown results in physical and functional deterioration by age-group (n=424). (b) Preferred method of physiotherapy delivery during the coronavirus disease 2019 (COVID-19) pandemic by patient demographics (n=424). (c) Level of patient concern about the waiting time at outpatient department and associated risk of infection, compared with the pre–COVID-19 pandemic period (n=424). (d) Perceived availability of information about the impact of COVID-19 on patients with cancer (n=424)

With respect to patient preferences regarding physiotherapy delivery during the pandemic, 42.9% of young respondents (18-45 years) preferred online sessions, whereas 50.0% of older respondents (>75 years) preferred home visits by a therapist [Fig 2b]. Education level (P=0.029) and income (P=0.034) were significantly associated with patient preference. Respondents with a higher education level (tertiary vs primary level: 33.3% vs 16.3%) and respondents with a higher income (monthly income >HK$40 000 vs <HK$16 000: 50.3% vs 26.0%) preferred online sessions, rather than in-person sessions.

During the period of social distancing, 64.5% of older respondents (>75 years) felt that their lives had become monotonous and lonely; significantly fewer (39.3%) younger respondents (<45 years) expressed this sentiment. Most respondents (58.0%) agreed that their home care support had improved because family members spent more time together; this sentiment was more prevalent among older respondents (>75 years) [70.9%].

As shown in online supplementary Table 1, the pandemic caused dietary habit alterations in 38.9% of respondents. Approximately one-fifth of respondents reported reduced appetite (22.2%) and increased consumption of junk food (processed or ready-made meals) (19.3%). Significantly more respondents in the low-income subgroup reported reduced appetite, compared with respondents in the high-income subgroup (30.6% vs 8.2%, P=0.001). Notably, the use of face masks led to a reduction in meal frequency among 30.7% of respondents; this reduction was more prevalent among respondents with lower income, compared with respondents who had higher income (32.6% vs 23.2%).

Psychological impact of coronavirus disease 2019

Overall, 41.0% and 23.1% of respondents had recently experienced anxiety and/or depressed mood. In total, 103 consecutive newly diagnosed patients responded to the extended psychometric questionnaire. The results revealed greater levels of concern regarding the impact of COVID-19 on cancer care manpower and the risk of infection during outpatient clinic waiting time in patients with higher risks of PTSD (P=0.011 and P=0.015, respectively), anxiety (P=0.013 and P=0.034, respectively), depression (P=0.017 and P=0.043, respectively), and uncertainty intolerance (P=0.004 and P=0.044, respectively) [Fig 2c]. A high IUS-12 score (uncertainty intolerance) was associated with the presence of greater concern regarding the effects of the pandemic on cancer research and drug development (P=0.03). As shown in online supplementary Table 2, respondents with a high risk of anxiety were less likely to agree with the ‘no visiting’ policy of hospitals (P=0.013). More respondents with high risks of anxiety (P=0.024) and depression (P=0.044) felt that there was an insufficient amount of information available in the media regarding the impact of COVID-19 on patients with cancer (Fig 2d). Moreover, respondents with a high risk of PTSD demonstrated significantly greater concern when asked about their fear of being infected by their caregiver or family members, compared with respondents who had a low risk of PTSD (P=0.005). Detailed results of the psychometric questionnaire are shown in online supplementary Table 2.

Comparison of oncologist and patient perspectives

We invited 30 practising clinical oncologists to predict patient healthcare-seeking behaviour during the pandemic. All 21 responding oncologists predicted significant reductions in patient willingness to attend appointments and patient willingness to be hospitalised, but most patients reported no change in either type of willingness (73.3% and 54.7%, respectively). A greater proportion of oncologists (50.0%) than patients (16.7%) reported a negative impact on their clinic experience because of doctor-patient distancing measures (Fig 3a). Furthermore, when asked about their confidence in identifying the cause of a new fever (COVID-19–related vs other causes), most oncologists reported little (73.3%) or no confidence (13.3%), whereas almost half of the patients (47.4%) reported that they were quite or very confident in their ability to identify the cause of a new fever (Fig 3b).

Figure 3. (a) Effect of physical distancing by medical staff during consultations on perception of clinic experience of patients (n=424) and oncologists (n=30) during coronavirus disease 2019. (b) Level of confidence among patients (n=424) and oncologists (n=30) in identifying the cause of a fever in the coming week

Discussion

This study investigated the perceptions of patients with cancer regarding the real-world impact of COVID-19 (during the early days of the pandemic) through the perspective of a multidisciplinary team that included clinical oncologists, clinical psychologists, physiotherapists, dieticians, and occupational therapists. Using a comprehensive set of questions, we identified key concerns, unmet needs, perceptions, and expectations of patients with cancer at different stages in their cancer care journeys. Cancer care16 and pandemic management17 are both resource-intensive endeavours. Because the COVID-19 pandemic has become the focus of healthcare worldwide, it is understandable that patients with cancer are concerned about pandemic-related negative impact on cancer care resources. Our results suggest that patients with cancer remained committed to attending scheduled appointments, despite the perceived risk of COVID-19 during the early days of the pandemic. This sustained clinical demand—along with general acceptance among patients regarding COVID-19 adaptive measures (staff-patient distancing), streamlined services (prescription-only clinics), and outcome trade-offs (efficacy and side-effect profile)—allowed our oncology services to continue with minimal disruptions despite the reduced availability of healthcare resources. Nevertheless, only 30.9% of surveyed patients felt that it was safe to attend the hospital; this observation highlights the need to ensure patients are informed about hospital safety measures for COVID-19 management, with details regarding rationale and efficacy. This study also revealed a discrepancy between male and female patients in terms of healthcare-seeking behaviour; moreover, patients receiving radical treatment demonstrated different perceptions and needs, compared with patients receiving palliative care. Oncology healthcare providers should consider the unique needs of various patient groups when implementing pandemic management strategies.

Between the initial outbreak and the time of this survey, the general public’s understanding of COVID-19 heavily relied on mainstream media coverage,18 which often did not focus on the needs of specific patient groups. International guidelines regarding cancer management during the pandemic began to emerge later in 2020,19 but they mainly targeted medical professionals. Accordingly, patients with cancer felt that the COVID-19–related information provided to patients was inadequate. This perception was particularly prevalent among patients with a higher education level, who tended to obtain news and information more frequently from multiple sources (eg, the internet and social media). Notably, this situation highlights the phenomenon of ‘the more you know, the more you realise you don’t know’, thereby emphasising the presence of an additional information barrier for underprivileged patient groups.20 The situation is further complicated by the presence of COVID-19–related misinformation, which has been widespread on social media since 2020.21 The findings in this study provide insights concerning the distinct pandemic-related information preferences and needs among patients according to age, education level, and income. Cancer services should focus on addressing these preferences and needs by providing patients with current COVID-19–related information from official sources, ensuring that the hospital remains a source of verified and practical pandemic-related information accessible to all patients.

The consequences of social distancing (eg, reduced exercise, poor diet, increased financial burden, and loss of social interactions) are detrimental to the physical and psychological well-being of patients with cancer,22 23 24 potentially reducing cancer treatment tolerance and compromising outcomes. Although the impact of pandemic-related lockdowns on dietary patterns of diabetic patients25 and older population26 have been studied, there are limited prospective data regarding the nutritional status of patients with cancer and their needs during the pandemic. This study has revealed some real-world patient needs, particularly among socio-economically disadvantaged patients; it also highlights the importance of individualised dietetic and occupational health assessments and early interventions (inpatient or outpatient) by dieticians and occupational therapists who specialise in cancer care. Dedicated self-help materials prepared by allied health professionals to address the adverse effects of social distancing may also serve as effective resources.

When the pandemic began, telerehabilitation emerged as a promising alternative method for patient–clinician interactions, with effective use in a physiotherapy context.27 28 However, our findings indicate that telerehabilitation may not be universally welcomed, particularly among older patients. Turolla et al29 described the challenges of implementing telerehabilitation; our findings highlight the need to carefully examine telemedicine accessibility and ‘telehealth literacy’30 among socio-economically underprivileged populations.31 32 When possible, conventional physiotherapy and rehabilitation should remain available, particularly for older adults, less-educated individuals, and low-income patients.33 Our findings offer a rationale for triaging appropriate patients towards telemedicine; they also highlight the need for improving telemedicine quality and access, as well as the importance of ensuring that alternatives are available.

This study demonstrated that psychometric analysis is a meaningful tool for identifying at-risk populations of patients with cancer during the pandemic. Patients in psychological high-risk groups clearly demonstrated distinct perceptions, expectations, and needs when simultaneously confronted with a cancer diagnosis and the COVID-19 pandemic. Without effective management, such patients could experience long-lasting psychiatric morbidities, as revealed during the severe acute respiratory syndrome epidemic in 2003.34 35 Wang et al36 emphasised the importance of mental healthcare attention and resources for patients with cancer during the COVID-19 pandemic. Along with routine cancer care, targeted psychotherapies and follow-up care for both the acute impact and long-term sequelae of COVID-19 are needed.

Oncologists and patients with cancer have different perceptions of cancer symptoms, treatment priorities, and psychosocial needs.37 38 39 We found that oncologists tended to underestimate patient motivation to avoid treatment interruptions, as well as patient risk acceptance, consistent with the observation by Catania et al40 that patients with cancer were more concerned about their cancers than about the pandemic. Moreover, compared with oncologists, a greater proportion of our surveyed patients expressed confidence in identifying COVID-19 symptoms. These results illustrate differences in priorities and perceptions of pandemic severity, along with the challenge of balancing disruptions to cancer treatment and maintaining COVID-19–related safety.

Proposed interventions to minimise the impact of coronavirus disease 2019 on cancer patients

The following are some proposed interventions to minimise the impact of COVID-19 on cancer patients:
1. Ensure that patients are informed about hospital safety measures for COVID-19 management, with details regarding rationale and efficacy.
2. Ensure that healthcare staff maintain appropriate physical distance from patients.
3. Operate prescription-only clinics and lengthen follow-up intervals for clinically stable patients.
4. Triage appropriate patients towards telemedicine; enhance general telehealth literacy by implementing user-friendly interfaces, step-by-step demonstrations, and support hotlines.
5. Establish a regularly updated COVID-19–related newsfeed that is customised for patients with cancer.
6. Work with dieticians, physiotherapists, and occupational therapists to create self-help pamphlets that can guide patients with cancer in coping with the effects of social distancing; facilitate the establishment of virtual support groups for patients with cancer.
7. Implement early allied health assessments and interventions for at-risk patients.
8. Ensure early psychological support, particularly for newly diagnosed patients.
9. Compassionately and flexibly enforce restrictive measures for newly diagnosed patients, individuals approaching the end of life, and selected at-risk patients.
10. Periodically review these measures as the pandemic progresses.

Study strengths and limitations

To our knowledge, this is one of the first studies to simultaneously explore perceptions of the real-world impact of the COVID-19 pandemic among patients with cancer and oncologists. Importantly, the efforts of the multidisciplinary team to construct the questions contributed to a multidimensional, holistic understanding of issues and unmet needs that affect patients with cancer at different stages of their cancer care journeys. Because of the in-person survey invitation and paper-and-pen methodology, our survey achieved a high response rate of 65%, ensuring that the results are representative of the surveyed population. However, sampling bias was present because survey respondents were patients who physically attended their clinic appointments; data were missing for around 10% of patients who declined to attend their clinic appointments. Furthermore, this survey was conducted within a short interval (2 weeks) towards the end of the first wave of the COVID-19 pandemic in Hong Kong, when there was a gradual easing of lockdown policies and personal protective equipment availability began to improve41; thus, this cross-sectional assessment may not adequately reflect the evolution of patient perceptions regarding the COVID-19 pandemic. Other key limitations of the study include its inclusion of patients from a single cancer centre, as well as the exclusion of patients who could not read Chinese or English and patients who underwent treatment in private clinics. There is a need to repeat the study at various time points throughout the pandemic; future analyses should focus on other affected countries and patient populations.

Conclusion

This multidisciplinary survey concerning the effects of the COVID-19 pandemic impact revealed key care priorities among patients with cancer, as well as their unmet needs; in particular, it highlighted the importance of distinct priorities and needs among socio-economically underprivileged groups and patients with diverse psychological phenotypes. Oncologists should be aware that their own perceptions of pandemic-related effects differ from their patients’ perceptions. These findings should be carefully considered as multidisciplinary teams modify their delivery of cancer care services during and after the pandemic.

Author contributions

Concept or design: KKH Bao, KM Cheung, JCH Chow.
Acquisition of data: All authors.
Analysis or interpretation of data: KKH Bao, KM Cheung, JCH Chow.
Drafting of the manuscript: KKH Bao, KM Cheung, JCH Chow.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

All authors have disclosed no conflicts of interest.

Acknowledgement

We thank all multidisciplinary team members and participating patients for their efforts and contributions.

Declaration

This research has been presented in part as poster presentations at the following conferences:
1. ESMO Congress 2020, virtual, 19-21 Sep 2020 (title: Cancer patients’ perspectives on the real-world impact of COVID-19 pandemic: a multidisciplinary survey)
2. ESMO Asia Congress 2020, virtual, 20-22 Nov 2020 (title: Psychometric interplay of the perception of the real-life impact of COVID-19 pandemic: a cross-sectional survey of patients with newly diagnosed malignancies)

Funding/support

This research was supported by the Hong Kong Hospital Authority Kowloon Central Cluster Research Grant 2020 (Ref No.: KCC/RC/G/2021-B01).

Ethics approval

The study was approved by the Kowloon Central/Kowloon East Cluster Clinical Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: KC/KE-20-0126/ER-1). All eligible respondents explicitly agreed to join the panel and provided informed consent to participate in the study.

References

1. World Health Organization. WHO Coronavirus Disease (COVID-19) Dashboard. Available from: https://covid19.who.int. Accessed 7 Nov 2020.

2. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424. Crossref

3. Jazieh AR, Akbulut H, Curigliano G, et al. Impact of the COVID-19 pandemic on cancer care: a global collaborative study. JCO Glob Oncol 2020;6:1428-38. Crossref

4. Chen-See S. Disruption of cancer care in Canada during COVID-19. Lancet Oncol 2020;21:e374. Crossref

5. Sud A, Torr B, Jones ME, et al. Effect of delays in the 2-week-wait cancer referral pathway during the COVID-19 pandemic on cancer survival in the UK: a modelling study. Lancet Oncol 2020;21:1035-44. Crossref

6. Prvu Bettger J, Thoumi A, Marquevich V, et al. COVID-19: maintaining essential rehabilitation services across the care continuum. BMJ Glob Health 2020;5:e002670. Crossref

7. Thibault R, Coëffier M, Joly F, Bohé J, Schneider SM, Déchelotte P. How the Covid-19 epidemic is challenging our practice in clinical nutrition—feedback from the field. Eur J Clin Nutr 2021;75:407-16. Crossref

8. Maringe C, Spicer J, Morris M, et al. The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study. Lancet Oncol 2020;21:1023-34. Crossref

9. Liu S, Yang L, Zhang C, et al. Online mental health services in China during the COVID-19 outbreak. Lancet Psychiatry 2020;7:e17-8. Crossref

10. Price M, Szafranski DD, van Stolk-Cooke K, Gros DF. Investigation of abbreviated 4 and 8 item versions of the PTSD Checklist 5. Psychiatry Res 2016;239:124-30. Crossref

11. Beck KR, Tan SM, Lum SS, Lim LE, Krishna LK. Validation of the emotion thermometers and hospital anxiety and depression scales in Singapore: screening cancer patients for distress, anxiety and depression. Asia Pac J Clin Oncol 2016;12:e241-9. Crossref

12. Carleton RN, Norton MA, Asmundson GJ. Fearing the unknown: a short version of the Intolerance of Uncertainty Scale. J Anxiety Disord 2007;21:105-17. Crossref

13. Geier TJ, Hunt JC, Hanson JL, et al. Validation of abbreviated four- and eight-item versions of the PTSD checklist for DSM-5 in a traumatically injured sample. J Trauma Stress 2020;33:218-26. Crossref

14. Schubart JR, Mitchell AJ, Dietrich L, Gusani NJ. Accuracy of the Emotion Thermometers (ET) screening tool in patients undergoing surgery for upper gastrointestinal malignancies. J Psychosoc Oncol 2015;33:1-14. Crossref

15. Office of the Government Economist Financial Secretary’s Office. Census and Statistics Department. Hong Kong SAR Government. Hong Kong Poverty Situation Report 2018. Available from: https://www.povertyrelief.gov.hk/eng/pdf/Hong_Kong_Poverty_Situation_Report_2018(2019.12.13).pdf. Accessed 17 Mar 2023.

16. Mariotto AB, Yabroff KR, Shao Y, Feuer EJ, Brown ML. Projections of the cost of cancer care in the United States: 2010-2020. J Natl Cancer Inst 2011;103:117-28. Crossref

17. Tan-Torres Edejer T, Hanssen O, Mirelman A, et al. Projected health-care resource needs for an effective response to COVID-19 in 73 low-income and middle-income countries: a modelling study. Lancet Glob Health 2020;8:e1372-9. Crossref

18. Casero-Ripollés A. Impact of Covid-19 on the media system. Communicative and democratic consequences of news consumption during the outbreak. El profesional de la información 2020;29:e290223. Crossref

19. Curigliano G, Banerjee S, Cervantes A, et al. Managing cancer patients during the COVID-19 pandemic: an ESMO multidisciplinary expert consensus. Ann Oncol 2020;31:1320-35. Crossref

20. Matsuyama RK, Wilson-Genderson M, Kuhn L, Moghanaki D, Vachhani H, Paasche-Orlow M. Education level, not health literacy, associated with information needs for patients with cancer. Patient Educ Couns 2011;85:e229-36. Crossref

21. Pennycook G, McPhetres J, Zhang Y, Lu JG, Rand DG. Fighting COVID-19 misinformation on social media: experimental evidence for a scalable accuracy-nudge intervention. Psychol Sci 2020;31:770-80. Crossref

22. Di Corrado D, Magnano P, Muzii B, et al. Effects of social distancing on psychological state and physical activity routines during the COVID-19 pandemic. Sport Sci Health 2020;16:619-24. Crossref

23. Ammar A, Brach M, Trabelsi K, et al. Effects of COVID-19 home confinement on eating behaviour and physical activity: results of the ECLB-COVID19 International Online Survey. Nutrients 2020;12:1583. Crossref

24. Galea S, Merchant RM, Lurie N. The mental health consequences of COVID-19 and physical distancing: the need for prevention and early intervention. JAMA Intern Med 2020;180:817-8. Crossref

25. Ruiz-Roso MB, Knott-Torcal C, Matilla-Escalante DC, et al. COVID-19 lockdown and changes of the dietary pattern and physical activity habits in a cohort of patients with type 2 diabetes mellitus. Nutrients 2020;12:2327. Crossref

26. Ceolin G, Moreira JD, Mendes BC, Schroeder J, Di Pietro PF, Rieger DK. Nutritional challenges in older adults during the COVID-19 pandemic [in Spanish]. Rev Nutrição 2020;33:e200174.Crossref

27. Smith AC, Thomas E, Snoswell CL, et al. Telehealth for global emergencies: implications for coronavirus disease 2019 (COVID-19). J Telemed Telecare 2020;26:309-13. Crossref

28. Scherrenberg M, Wilhelm M, Hansen D, et al. The future is now: a call for action for cardiac telerehabilitation in the COVID-19 pandemic from the secondary prevention and rehabilitation section of the European Association of Preventive Cardiology. Eur J Prev Cardiol 2021;28:524-40. Crossref

29. Turolla A, Rossettini G, Viceconti A, Palese A, Geri T. Musculoskeletal physical therapy during the COVID-19 pandemic: is telerehabilitation the answer? Phys Ther 2020;100:1260-4. Crossref

30. Norman CD, Skinner HA. eHealth literacy: essential skills for consumer health in a networked world. J Med Internet Res 2006;8:e9. Crossref

31. Jasemian Y. Elderly comfort and compliance to modern telemedicine system at home. 1st International ICST Workshop on Connectivity, Mobility and Patients’ Comfort. 2008. Available from: https://eudl.eu/doi/10.4108/icst.pervasivehealth2008.2516. Accessed 5 May 2020. Crossref

32. Bujnowska-Fedak MM, Grata-Borkowska U. Use of telemedicine-based care for the aging and elderly: promises and pitfalls. Smart Homecare Technol Telehealth 2015;3:91-105. Crossref

33. Chesser A, Burke A, Reyes J, Rohrberg T. Navigating the digital divide: a systematic review of eHealth literacy in underserved populations in the United States. Inform Health Soc Care 2016;41:1-19. Crossref

34. Chua SE, Cheung V, McAlonan GM, et al. Stress and psychological impact on SARS patients during the outbreak. Can J Psychiatry 2004;49:385-90. Crossref

35. Mak IW, Chu CM, Pan PC, Yiu MG, Chan VL. Long-term psychiatric morbidities among SARS survivors. Gen Hosp Psychiatry 2009;31:318-26.Crossref

36. Wang Y, Duan Z, Ma Z, et al. Epidemiology of mental health problems among patients with cancer during COVID-19 pandemic. Transl Psychiatry 2020;10:263. Crossref

37. Newell S, Sanson-Fisher RW, Girgis A, Bonaventura A. How well do medical oncologists’ perceptions reflect their patients’ reported physical and psychosocial problems? Data from a survey of five oncologists. Cancer 1998;83:1640-51. Crossref

38. Sakai H, Umeda M, Okuyama H, Nakamura S. Differences in perception of breast cancer treatment between patients, physicians, and nurses and unmet information needs in Japan. Support Care Cancer 2020;28:2331-8. Crossref

39. Söllner W, DeVries A, Steixner E, et al. How successful are oncologists in identifying patient distress, perceived social support, and need for psychosocial counselling? Br J Cancer 2001;84:179-85. Crossref

40. Catania C, Spitaleri G, Del Signore E, et al. Fears and perception of the impact of COVID-19 on patients with lung cancer: a mono-institutional survey. Front Oncol 2020;10:584612. Crossref

41. Coronavirus.gov. Hong Kong SAR Government. Together, We Fight the Virus. Available from: https://www.coronavirus.gov.hk/eng/index.html. Accessed 7 Nov 2020.

Implementation of ovarian tissue cryopreservation in Hong Kong

Hong Kong Med J 2023 Apr;29(2):121–31 | Epub 24 Feb 2023

https://doi.org/10.12809/hkmj2210220

ORIGINAL ARTICLE

Implementation of ovarian tissue cryopreservation in Hong Kong

Jacqueline PW Chung, MB, ChB, FHKAM (Obstetrics and Gynaecology)^1,2; David YL Chan, BSc (Ulster), DPhil (Oxon)^1,2; Y Song, BSc (Peking University), PhD (CUHK)³; Elaine YL Ng, BSc (CUHK), MPhil (CUHK)¹; Tracy SM Law, MB, ChB, FHKAM (Obstetrics and Gynaecology)¹; Karen Ng, MB, ChB, FHKAM (Obstetrics and Gynaecology)¹; Maran BW Leung, BSc (CUHK), PhD (CUHK)³; S Wang, MB, BS, MSc¹; HM Wan, BEng (Jinan University), MSc (Jinan University)¹; Joshua JX Li, MB, ChB, FHKAM (Pathology)⁵; CC Wang, MB, BS, PhD (Surgical Sciences in Obstetrics and Gynaecology)^1,2,6

¹ Assisted Reproductive Technology Unit, Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Hong Kong

² Fertility Preservation Research Centre, Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Hong Kong

³ Department of Obstetrics and Gynaecology, The Chinese University of Hong Kong, Hong Kong

⁴ Department of Obstetrics and Gynaecology, Union Hospital, Hong Kong

⁵ Department of Anatomical and Cellular Pathology, The Chinese University of Hong Kong, Hong Kong

⁶ Li Ka Shing Institute of Health Science, School of Biomedical Sciences; and Chinese University of Hong Kong–Sichuan University Joint Laboratory in Reproductive Medicine, The Chinese University of Hong Kong, Hong Kong

Corresponding author: Prof Jacqueline PW Chung (jacquelinechung@cuhk.edu.hk)

Full paper in PDF

Abstract

Introduction: Worldwide, >130 babies have been born from ovarian tissue cryopreservation (OTC) and ovarian tissue transplantation (OTT). Ovarian tissue cryopreservation can improve quality of life among young female cancer survivors. Here, we assessed the feasibility of OTC and subsequent OTT in Hong Kong via xenografts in nude mice.

Methods: This pilot study was conducted in a university-affiliated tertiary hospital. Fifty-two ovarian tissues were collected from 12 patients aged 29 to 41 years during ovarian surgery, then engrafted into 34 nude mice. The efficacies of slow freezing and vitrification were directly compared. In Phase I, non-ovariectomised nude mice underwent ovarian tissue engraftment. In Phase II, ovariectomised nude mice underwent ovarian tissue engraftment, followed by gonadotrophin administration to promote folliculogenesis. Ovarian tissue viability was assessed by gross anatomical, histological, and immunohistochemical examinations before and after OTC. Follicular density and morphological integrity were also assessed.

Results: After OTC and OTT, grafted ovarian tissues remained viable in nude mice. Primordial follicles were observed in thawed and grafted ovarian tissues, indicating that the cryopreservation and transplantation protocols were both effective. The results were unaffected by gonadotrophin stimulation.

Conclusion: This study demonstrated the feasibility of OTC in Hong Kong as well as primordial follicle viability after OTC and OTT in nude mice. Ovarian tissue cryopreservation is ideal for patients who cannot undergo the ovarian stimulation necessary for oocyte or embryo freezing as well as prepubertal girls (all ineligible for oocyte freezing). Our findings support the clinical implementation of OTC and subsequent OTT in Hong Kong.

New knowledge added by this study

This study assessed the viability of ovarian tissue cryopreservation and subsequent ovarian tissue transplantation in Hong Kong via xenografts in nude mice.
Grafted ovarian tissues remained viable after transplantation, regardless of protocol (slow freezing or vitrification).

Implications for clinical practice or policy

Ovarian tissue cryopreservation is ideal for patients who cannot undergo the ovarian stimulation necessary for oocyte or embryo freezing, as well as prepubertal girls (all ineligible for oocyte freezing).
These findings support the clinical implementation of ovarian tissue cryopreservation and subsequent ovarian tissue transplantation in Hong Kong.
Further studies are needed to clarify optimal cryopreservation and engraftment protocols.

[Abstract in Chinese]

Introduction

A diagnosis of cancer is disheartening news for every patient in terms of both disease and treatment. Anticancer treatments for common cancers (eg, chemotherapy and radiotherapy) are gonadotoxic and detrimental to future fertility.1 Advances in medical treatment have improved the 5-year survival rates of some cancers to >80% in children and adolescents.2 However, most surviving patients experience illness-related infertility.3 Infertility is also a concern for patients with severe endometriosis, patients with poor ovarian reserve, patients with benign medical conditions requiring chemotherapy, and transgender individuals undergoing gender-affirming surgery.4 Fortunately, advancements in fertility preservation (FP) technologies offer these patients the opportunity to have biological offspring in the future. In Hong Kong, only 45.6% of clinicians,5 22.2% of medical students,6 and 21.7% of the general public7 are familiar with FP. Therefore, FP technologies require greater attention in Hong Kong. Both clinicians and the public should be aware where and how to seek help when patients are diagnosed with cancer and need to use FP technologies to preserve their fertility.

For patients who cannot undergo the ovarian stimulation necessary for oocyte or embryo freezing as well as prepubertal girls (all ineligible for oocyte freezing), ovarian tissue cryopreservation (OTC) and subsequent orthotopic or heterotopic ovarian tissue transplantation (OTT) are ideal options for FP after recovery.8 Many European countries have provided OTC for patients with various medical reasons.4 Although OTC and OTT are widely available in Belgium, Denmark, Spain, France,9 Japan, Singapore,10 the United States, India, Australia, the Philippines, Korea,11 and some parts of China,12 these FP technologies remain unavailable in Hong Kong. Thus far, >130 babies worldwide have been born via OTC and OTT,13 and the American Society for Reproductive Medicine removed the ‘experimental’ designation for these technologies in 2019.14 Ovarian tissue cryopreservation can be performed via slow freezing or vitrification. Slow freezing has been the standard treatment for ovarian tissues,15 but there is increasing evidence to support the use of vitrification.16 17 Nevertheless, controversies remain concerning subsequent oocyte viability and the preservation of morphological integrity after ovarian tissues have been processed using these two cryopreservation techniques.18

The development of OTC, which can improve quality of life among young female cancer survivors,19 is urgently needed in Hong Kong. Here, we performed a pilot study to assess the feasibility of OTC and subsequent OTT in Hong Kong via xenografts in nude mice. In this study, we collected ovarian tissues, established both slow freezing and vitrification protocols, and evaluated tissue viability and follicle preservation after OTC and OTT in a nude mouse xenograft model. This mouse model provided important insights that will support the clinical implementation of OTC and OTT in Hong Kong. Our primary outcome was ovarian tissue viability after slow freezing or vitrification, as determined by histological analysis and immunohistochemistry. Our secondary outcomes were follicular density and the morphological integrity of grafted ovarian tissues.

Methods

This study was conducted between July 2019 and December 2021 at the Prince of Wales Hospital, a university-affiliated tertiary hospital in Hong Kong. All participants received a detailed explanation of the study, then provided written consent for inclusion. All researchers involved in the animal experiments were licensed by the Department of Health of the Hong Kong SAR Government.

Ovarian tissue collection

Ovarian tissues were collected from women or transgender individuals who underwent laparoscopic or open, unilateral or bilateral, ovarian cystectomy or salpingo-oophorectomy as treatment for benign ovarian cysts or tumours. During the operation, each patient underwent removal of a small section of ovarian tissue or the whole ovary; each specimen of donated ovarian tissue was retrieved from a routine surgical specimen or directly removed during surgery. To prevent thermal injury during tissue removal, cold scissors were used and diathermy was avoided. The amount of donated tissue varied among patients, depending on their age and clinical condition. For example, larger volumes of ovarian tissue were often collected from patients undergoing oophorectomy. Each patient was assigned a unique identification number linked to an encrypted file containing the patient’s data and demographic information; during analyses of tissue from each patient, the pathologist and research staff who conducted histology and immunohistochemistry analyses were blinded to the contents of the encrypted files.

Ovarian tissue cryopreservation

Tissues were transported to the laboratory in a standardised culture medium at 4°C and processed within 30 minutes after collection. After removing the medullary region, the ovarian tissue was frozen in accordance with a controlled-rate slow freezing machine protocol (Ovarian Tissue Cryopreservation Scientific Roundup; Planer, UK)20 or a vitrification manual (Ova Cryo Kit Type M, VT301S; Kitazato Corporation, Japan).21 The cortical region was cut into small fragments with a thickness of approximately 1 mm. Some collected ovarian tissues were very small and could not be sectioned for parallel slow freezing and vitrification fresh tissue controls; these small tissues were either frozen using the standard slow freezing method or vitrification.22 Larger ovarian tissues (typically collected from transgender individuals during oophorectomy) were cut into smaller fragments prior to slow freezing and vitrification, or prior to use as fresh tissue controls. Before the xenograft procedure, fragments were removed from fresh tissue, thawed slow-frozen tissue, and thawed vitrified tissue; these fragments were subsequently compared with grafted tissues to identify any differences related to engraftment.

A subset of fresh ovarian tissue fragments was fixed and subjected to histological analysis. When a large amount of ovarian tissue was available from a single patient, we compared cryopreservation methods using tissue from that patient; we also compared the cryopreserved tissue with fresh tissue.

Slow freezing

Slow freezing was performed in accordance with a validated protocol.23 24 25 Collected ovarian cortices were equilibrated at 4°C on a tilting shaker for 30 minutes in freezing solution (1.5 mol/L ethylene glycol and 0.1 mol/L sucrose in G-MOPS PLUS; Vitrolife, Sweden). After equilibration, the ovarian tissue pieces were placed into 1.8-mL cryogenic vials that had been pre-filled with 1 mL of freezing solution (two tissue pieces per vial). The cryogenic vials were then placed into an automated, computer-controlled freezing system (Kryo-360; Planer, UK).20 The slow freezing protocol was performed in accordance with the method described by Dolmans et al26 and the Planer Ovarian Tissue Cryopreservation Scientific Roundup.20

Vitrification

For vitrification of ovarian cortices, the Ovarian Tissue Vitrification Kit (Ova Cryo Kit Type M, VT301S; Kitazato Corporation, Japan)21 was used. The collected ovarian cortices were cut into 1 × 1 × 1 cm³ cubes using a surgical knife and a square measuring device provided in the kit. Vitrification was then performed in accordance with the kit manufacturer’s protocol, and the ovarian tissues were stored in liquid nitrogen.

Thawing of ovarian tissue for transplantation

Slow-frozen tissues were removed from liquid nitrogen and exposed to room temperature air for 5 seconds, then placed in 37°C water for 2 minutes. Subsequently, they were transferred to thawing solution 1 (0.75 mol/L ethylene glycol and 0.25 mol/L sucrose in G-MOPS PLUS) for 10 minutes, then to thawing solution 2 (0.25 mol/L sucrose in G-MOPS PLUS) for 10 minutes, and finally to a handling medium (G-MOPS) for 10 minutes. Vitrified ovarian tissue fragments were thawed using Ova Thawing Kit Type M (V302S; Kitazato Corporation, Japan), in accordance with the manufacturer’s instructions.21

Ovarian tissue transplantation into nude mice

The nude mouse xenograft model is ideal for assessment of OTC and OTT outcomes. Ovarian xenografts in immunodeficient nude mice can be used to test follicular viability and development. This approach can reveal whether freezing and thawing cause damage to ovarian tissue; it can also demonstrate the ability of cryopreserved tissue to support the development of large antral follicles.27 Considering the higher rate of immune leakiness in severe combined immunodeficient mice,28 we used BALB/c athymic nude mice to validate our OTC and OTT protocols before clinical implementation. Thirty-four female BALB/c athymic nude mice (age, 4-6 weeks; Laboratory Animal Services Centre, The Chinese University of Hong Kong) were used for this study. To prevent fighting between engrafted mice, only three mice were housed in individually ventilated cages at 28°C under controlled sterile conditions, with a 12-hour light/dark cycle and free access to an autoclaved pelleted diet and water. Mice were anaesthetised by intraperitoneal injection of ketamine (75 mg/kg)/xylazine (10 mg/kg) (AlfaMedic Limited, Hong Kong; manufactured in Holland). Ovarian tissues collected from patients were grafted onto nude mice. During ovarian tissue engraftment, the cortical surface was carefully oriented outward and tightly attached to the subcutaneous tissue or abdominal wall. There were two phases in our study, as described in the following sections (Fig 1).

Figure 1. Flowchart depicting Phase I and Phase II studies. In Phase I, tissues were collected from four patients and engrafted into nine mice. In Phase II, tissues were collected from six patients and engrafted into 25 mice

Phase I: Analysis of ovarian tissue xenograft viability in non-ovariectomised nude mice

To maintain endogenous hormone secretion and avoid the risk of ovariectomy-related death, mice in this phase were not subjected to ovariectomy. One fresh, slow-frozen, or vitrified tissue of approximately 4 × 6 × 1 mm³ was engrafted into the subcutaneous site on the neck of nine nude mice.29 The mice were then sacrificed by intraperitoneal injection of overdose of the anaesthetic. One mouse engrafted with vitrified tissue, two engrafted with slow-frozen tissues and one engrafted with fresh tissue were sacrificed after 2 weeks. Two mice engrafted with vitrified tissues, one engrafted with slow-frozen tissue and two engrafted with fresh tissues were sacrificed after 5 weeks.

Phase II: Analysis of ovarian tissue xenograft viability, folliculogenesis, and ovulation in ovariectomised nude mice

To promote graft survival and growth, mice in this phase were subjected to ovariectomy. Fresh, slow-frozen, or vitrified tissues of approximately 4 × 6 × 1 mm³ were either engrafted into the subcutaneous site on the neck of ovariectomised nude mice, or used for intraperitoneal engraftment in the left abdomen of those mice.29 Mice in this phase were divided into a saline group and a treatment group after 2 or 6 weeks of engraftment. The presence of gonadotrophins can optimise graft establishment and stimulate follicle growth.30 To promote folliculogenesis, mice in the treatment group underwent intraperitoneal injection (in the right abdomen) of 1 IU (100 μL) of follitropin alfa (GONAL-f; Merck Serono, Geneva, Switzerland) every other day for 5 to 8 weeks after 2 or 6 weeks’ engraftment.30 During the same period, mice in the saline group underwent intraperitoneal injection (in the right abdomen) of an equal volume of physiological saline every other day. Thirty-six hours before the mice were sacrificed, both groups of mice received a single dose of 10 international units of human chorionic gonadotrophin (Sigma-Aldrich, St Louis [MO], US) by injection to promote ovulation.

Grafted ovarian tissue viability

All grafted tissues were fixed in buffered formalin and embedded in paraffin wax, then sectioned and stained for analysis.

Histological analysis

Microscopic observations up to 400 times the original magnification (Leica DMIRB; Leica Microsystem, Wetzlar, Germany) of fresh and thawed ovarian tissues were performed after the tissues had been stained with haematoxylin and eosin (H&E). All follicles from the entire grafted tissue specimen on every slide were counted; section thickness and the presence/absence of a nucleolus were also considered.

Immunohistochemical assessment of stromal tissue viability

Stromal tissue viability was determined by assessing the morphologies of stromal cells on H&E-stained sections. Viability was defined as the presence of spindle cells with consistent cellularity; an intact nuclear membrane; the absence of pyknotic figures, apoptosis, or necrosis; and the absence of fibrosis or calcification. Viability was confirmed by immunohistochemical analyses using antibodies to cluster of differentiation 10 (CD10) and oestrogen receptors. Anti-CD10 antibody (clone NCL-CD10-270; Novocastra, Newcastle upon Tyne, UK) was used at a dilution of 1:50 with an incubation time of 30 minutes at a sustained temperature of 37°C. Anti–oestrogen receptor antibody (RM-9101; Thermo Fisher Scientific, Fremont [CA], US) was used at a dilution of 1:150 with an incubation time of 32 minutes at a sustained temperature of 37°C. Antigen retrieval was performed using ethylenediaminetetraacetic acid and microwave. Antibody detection was performed using Roche Diagnostics OptiView DAB IHC Detection Kit (Thermo Fisher Scientific, Waltham [MA], US). Immunohistochemical staining (original magnification × 400) was semi-quantitative and based on signal intensity (absent, weak, moderate, and strong). The presence of at least weak staining intensity in ovarian stromal tissue was regarded as a positive result.

Follicular density and quality after freezing, thawing, and transplantation

All follicles from the entire grafted tissue specimen on every H&E-stained slide were counted on multiple levels within thick sections (≥20 μm). The digital images were annotated on QuPath,31 obtaining the two-dimensional area of the slide and number of follicles. Follicular density was calculated by established methods described previously.32 33 Ovarian follicles were classified as primordial, primary, or secondary follicles according to morphological assessment of H&E-stained sections.33 Evaluation of grafted follicle quality was based on basement membrane integrity, cellular density, presence or absence of pyknotic bodies, and oocyte integrity. Only morphologically normal (ie, viable) follicles were counted. The results of gross anatomical examinations were confirmed by histological assessments. Gross tissue integrity was defined as the presence of a distinct vascularised tissue fragment that exhibited firmness and perfusion. Microscopic findings indicating viability were the presence of an intact nuclear membrane and the absence of necrosis, apoptosis, and pyknotic nuclei.

Results

In total, 52 ovarian tissues were collected from 12 patients aged 29 to 41 years. Ovarian tissues from different patients were cut into several pieces according to tissue size. These tissues were treated by vitrification or slow freezing, then engrafted into 34 mice as shown in Figure 1. Tables 1 and 2 only show data from patients with follicles to facilitate readability. Although there were nine tissues from four patients (Patients 1, 6, 7, and 10) in Phase I, Table 1 only shows data from the two patients with follicles (Patients 1 and 7). In Phase II, there were 25 tissues from six patients (Patients 1, 5, 8, 9, 11, and 12), but Table 2 only shows data from the four patients with follicles. In total, 18 control tissues were collected from 11 patients (Patients 1, 2, 4, 5, 6, 7, 8, 10, 11, 12, and 13). One patient (Patient 5) provided sufficient ovarian tissue for a comparison of cryopreservation methods using tissue from a single patient. Additionally, we compared fresh and slow-frozen tissues from Patient 5, and compared fresh and vitrified tissues from Patients 1, 5, and 12.

Table 1. Follicular density of viable grafted tissues in Phase I analysis: two of four patients with primordial follicles

Table 2. Follicular density of viable grafted tissues in Phase II analysis: five of six patients with primordial follicles

Graft recovery rate and macroscopic assessment

In Phase I, all xenografts were successfully retrieved from the experimental mice. Macroscopic observations of fresh and thawed tissues did not show substantial differences between cryopreservation methods in terms of tissue integrity or morphology (Fig 2a). Microscopic findings showed the presence of viable nuclei and the absence of necrosis, apoptosis, and pyknotic nuclei.

Figure 2. Macroscopic observations of fresh, thawed, and grafted tissues. (a) Fresh tissue (left) and thawed slow-frozen tissue (right). (b) Grafted slow-frozen ovarian tissues at subcutaneous sites on the neck in three BALB/c athymic nude mice. (c) Grafted fresh tissues (left and middle) and vitrified ovarian tissue (right) at intraperitoneal sites in three BALB/c athymic nude mice. Note that angiogenesis was observed around each xenograft

In Phase II, all xenografts were successfully retrieved from the experimental mice, with the exception of two calcified tissues. Most subcutaneous sites contained soft tissue fragments that were completely encased in membranes; the graft–murine tissue interface was vascularised (Fig 2b). Intraperitoneal sites contained soft tissue fragments with small vessels visible on the graft surface; the fragments were attached to surrounding tissue, and some grafts were encased in abdominal adipose tissue (Fig 2c).

Analysis of stromal tissue morphology

Immunohistochemical staining showed that all retrieved grafts had maintained viability, with the exception of two calcified tissues (Fig 3). Haematoxylin and eosin staining, CD10 staining, and oestrogen receptor staining showed no between-group differences (fresh vs slow-frozen, fresh vs vitrified, and slow-frozen vs vitrified). Moreover, stromal tissue viability did not differ between the treatment and saline groups in Phase II.

Figure 3. Haematoxylin and eosin (H&E) and immunohistochemical staining of ovarian stromal tissues in xenografts retrieved from nude mice (400 ×). (a) H&E staining; (b) cluster of differentiation 10 staining; (c) oestrogen receptor staining

Follicular histology and density

Retrieved grafts were embedded with paraffin and sectioned at a thickness of 4 or 30 μm. Microscopy analysis revealed primordial, primary, and secondary follicles (Fig 4). Tables 1 and 2 show the follicular densities of retrieved grafts from Phases I and II, respectively. Primordial follicles were observed in fresh and cryopreserved grafts from the same patient (Patient 5), regardless of cryopreservation method (slow freezing or vitrification) or gonadotrophin injection status.

Figure 4. Microscopic observation of different stages of follicles in the grafted tissues on nude mice (400 ×). (a) Haematoxylin and eosin (H&E) staining of primordial follicles in 4-μm sections; (b) H&E staining of primordial (left), primary (middle), and secondary (right) follicles in 30-μm sections

Discussion

Summary of main findings

Our pilot study demonstrated the feasibility of OTC with subsequent OTT in Hong Kong via xenografts of fresh and cryopreserved ovarian tissues in nude mice. Most grafted ovarian tissues remained viable after engraftment, as demonstrated by CD10 and oestrogen receptor staining results in stromal tissue, along with the presence of viable nuclei and the absence of necrosis, apoptosis, and pyknotic nuclei. Regardless of cryopreservation method, primordial follicles were observed in thawed ovarian tissues after engraftment; thus, both cryopreservation methods are feasible and effective. There were no differences in folliculogenesis after gonadotrophin injection. Overall, these findings validate our protocol for surgical collection of ovarian tissue, cryopreservation via slow freezing or vitrification, and subsequent tissue engraftment into mice; this protocol successfully generated primordial follicles in the xenografts. To our knowledge, this type of protocol was not previously validated in Hong Kong.

The benefits of OTC and OTT are not limited to gynaecology patients; they are also useful for patients in other specialties4 (eg, medicine, oncology, and paediatrics), including adolescents,34 as well as women who cannot undergo ovarian stimulation. To our knowledge, this is the first study to demonstrate the feasibility of OTC and OTT in Hong Kong; our findings support the clinical implementation of these technologies at medical centres in Hong Kong.

Current condition and success rate of ovarian tissue cryopreservation

Ovarian tissue cryopreservation has become an accepted FP technology in many fertility centres since the removal of its experimental designation.11 14 Notably, OTC allows the preservation of thousands of primordial follicles in a single procedure; compared with mature oocytes, preserved primordial follicles are more resistant to cryodamage.35 This technology is also appropriate for patients who cannot undergo ovulation stimulation because they require urgent chemotherapy or must avoid the enhancement of a hormone-sensitive malignancy36; it is also the only available FP technology for prepubertal girls.36 Furthermore, OTC allows natural conception; several spontaneous pregnancies have been reported after successful orthotopic autotransplantation.9 37 In some instances, both fertility and gonadal function are restored.34 According to a meta-analysis,38 endocrine function was restored in 63.9% of patients; the combined rate of pregnancies and live births was 28.4%. Dolmans et al9 also reported that 26% of women became pregnant and gave birth to one or two infants after the transplantation of frozen-thawed ovarian tissue; the live birth rate was 30.6%.

Barriers to clinical implementation of ovarian tissue cryopreservation

Despite its advantages, there are multiple barriers to the clinical implementation of OTC. Effective use of this technology involves two surgical procedures: the initial removal of ovarian tissue (prior to cryopreservation) and a future transplantation procedure, which may cause surgical and ethical problems (particularly in prepubertal patients).39 The technology also requires expertise that is not available in some parts of Asia. A Japanese group reported a live birth in 201540; another successful live birth was reported by a Chinese group in 2021,41 involving the cryopreserved ovarian tissue bank established by the Beijing Obstetrics and Gynecology Hospital.12 However, OTC is not widely available in Hong Kong. Reproductive health centres in Hong Kong may lack sufficient surgical expertise and/or an optimal cryopreservation environment.11 Thus, there is a need to reduce the obstacles to clinical implementation of OTC. From our experience, in terms of laboratory requirements, the protocol, equipment, and consumables can be incorporated into most assisted reproductive technology units. However, practical education is needed regarding OTC, including tissue management (eg, tissue thinning during removal of the medulla) and specific aspects of cryopreservation. Proper records of success measures (eg, freeze-thaw outcomes and graft survival rate) are essential; these data should be carefully documented in laboratory records. Additional equipment is also needed for the clinical implementation of OTC as a routine service because the harvesting surgery may be performed on an urgent basis that differs from the routine assisted reproductive technology laboratory programme.11 While planning for this study, we found that there have been inconsistencies in terms of selection criteria, cryopreservation methods, laboratory management of harvested tissue, and the transplantation technique itself. Although we found no differences in the morphological integrity of ovarian tissue after cryopreservation via slow freezing or vitrification, further studies with larger numbers of patients are needed to confirm the feasibility of follicular stimulation in vivo.

Current status of ovarian tissue transplantation

Human OTT remains unavailable in Hong Kong. Notably, our analysis of tissue engraftment was conducted in a mouse model. After the clinical implementation of OTC in Hong Kong, OTT involving autotransplantation could be established as a routine service. Ovarian autotransplantation is performed when a patient has fully recovered from disease, but this approach may carry a small risk of reintroducing malignant cells in patients with cancer.42 The results of some studies have suggested that the risk of reintroducing malignant cells could be minimised by meticulous examination of representative biopsy samples via histology, immunohistochemistry, and molecular biology techniques.43 Moreover, optical coherence tomography can be used to assess malignant cells in thawed ovarian tissue before transplantation.44

Limitations

There were several limitations in this study. First, tissue engraftment was not conducted in humans because of ethical concerns; thus, we analysed tissue engraftment in a nude mouse xenograft model, which was the best available model. Second, some ovarian tissues were collected from transgender individuals who had undergone testosterone replacement therapy, which might have affected the hormonal milieu of the ovarian tissue.45 Third, we only retrieved small fragments of ovarian tissue (~1 cm) from random locations in the ovaries of included patients; this may have led to sampling error if the sampled cortical layers did not contain primordial follicles. Fourth, the small number of included patients hindered our ability to compare the effects of cryopreservation methods on tissue from a single patient. Moreover, the small sample size might have reduced the strength of the findings. Finally, there are no standardised protocols for freezing, gonadotrophin stimulation, or transplantation in nude mouse xenograft models. However, our study demonstrated the feasibility of OTC in our centre. Further randomised controlled trials are needed to confirm our findings.

Future trends

We plan to conduct a randomised controlled trial of the two cryopreservation methods used in this study to determine which is best for clinical implementation. From the experience of the Danish group Rosendahl et al24 on OTC, they suggested that before applying the technique to humans, each laboratory should thoroughly test and validate the OTC method. In the future, implantation of artificial ovaries or the engraftment of human ovarian tissue into mice may enable fertility restoration without the potential reintroduction of malignant cells. These approaches may be particularly useful in women with a high risk of blood-borne leukaemia or cancers with a high risk of ovarian metastasis, as well as women who cannot undergo autotransplantation.46

Conclusion

Our study demonstrated the feasibility and viability of OTC with subsequent OTT in Hong Kong via xenografts in nude mice. These findings support the clinical implementation of OTC and subsequent OTT in Hong Kong, particularly for prepubertal young girls and for women who cannot undergo the ovarian stimulation necessary for oocyte or embryo freezing. Further studies are needed to clarify optimal cryopreservation and engraftment protocols.

Author contributions

Concept or design: JPW Chung, DYL Chan.
Acquisition of data: All authors.
Analysis or interpretation of data: JPW Chung, DYL Chan, Y Song, MBW Leung, JJX Li, CC Wang.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: JPW Chung, DYL Chan, CC Wang.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

As an editor of the journal, JPW Chung was not involved in the peer review process of the article. All other authors have no conflicts of interest to disclose.

Funding/support

This research was supported by Basecare Medical Device Co., Ltd. and the Theme-based Research Scheme funded by the Research Grants Council of the Hong Kong SAR Government (Ref No.: T13-602/21-N).

Ethics approval

The research was approved by the Institutional Review Board of the Joint Chinese University of Hong Kong–New Territories East Cluster Clinical Research Ethics Committee (Ref No.: 2019.356) and overseen by an independent data and safety monitoring committee. The trial was registered with the World Health Organization Primary Registry–Chinese Clinical Trials Registry (Trial No.: ChiCTR2100041611). The experimental animal protocol was approved by The Chinese University of Hong Kong Animal Experimentation Ethics Committee (Ref No.: 19-214-MIS). All participants received a detailed explanation of the study and provided written consent for inclusion. All researchers involved in animal experiments were licensed by the Department of Health of the Hong Kong SAR Government.

References

1. Maltaris T, Seufert R, Fischl F, et al. The effect of cancer treatment on female fertility and strategies for preserving fertility. Eur J Obstet Gynecol Reprod Biol 2007;130:148-55. Crossref

2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69:7-34. Crossref

3. Brydøy M, Fosså SD, Dahl O, Bjøro T. Gonadal dysfunction and fertility problems in cancer survivors. Acta Oncol 2007;46:480-9. Crossref

4. ESHRE Guideline Group on Female Fertility Preservation, Anderson RA, Amant F, et al. ESHRE guideline: female fertility preservation. Hum Reprod Open 2020;2020:hoaa052. Crossref

5. Chung JP, Lao TT, Li TC. Evaluation of the awareness of, attitude to, and knowledge about fertility preservation in cancer patients among clinical practitioners in Hong Kong. Hong Kong Med J 2017;23:556-61. Crossref

6. Ng EY, Ip JK, Mak DR, Chan AY, Chung JP. Awareness of fertility preservation among Chinese medical students. Hong Kong Med J 2020;26:184-91. Crossref

7. Yeung SY, Ng EY, Lao TT, Li TC, Chung JP. Fertility preservation in Hong Kong Chinese society: awareness, knowledge and acceptance. BMC Womens Health 2020;20:86. Crossref

8. Dolmans MM, Donnez J. Fertility preservation in women for medical and social reasons: oocytes vs ovarian tissue. Best Pract Res Clin Obstet Gynaecol 2021;70:63-80. Crossref

9. Dolmans MM, von Wolff M, Poirot C, et al. Transplantation of cryopreserved ovarian tissue in a series of 285 women: a review of five leading European centers. Fertil Steril 2021;115:1102-15. Crossref

10. Harzif AK, Santawi VP, Maidarti M, Wiweko B. Investigation of each society for fertility preservation in Asia. Front Endocrinol (Lausanne) 2019;10:151. Crossref

11. Takae S, Lee JR, Mahajan N, et al. Fertility preservation for child and adolescent cancer patients in Asian countries. Front Endocrinol (Lausanne) 2019;10:655. Crossref

12. Jin F, Ruan X, Du J, et al. Analysis on the characteristics of the patients and effects of ovarian tissue cryopreservation in the first ovarian tissues cryopreservation bank in China [in Chinese]. J Capital Univ Med Sci 2021;42:521-5.

13. Donnez J, Dolmans MM. Fertility preservation in women. N Engl J Med 2017;377:1657-65. Crossref

14. Practice Committee of the American Society for Reproductive Medicine. Fertility preservation in patients undergoing gonadotoxic therapy or gonadectomy: a committee opinion. Fertil Steril 2019;112:1022-33. Crossref

15. Rivas Leonel EC, Lucci CM, Amorim CA. Cryopreservation of human ovarian tissue: a review. Transfus Med Hemother 2019;46:173-81. Crossref

16. Marques LS, Fossati AA, Rodrigues RB, et al. Slow freezing versus vitrification for the cryopreservation of zebrafish (Danio rerio) ovarian tissue. Sci Rep 2019;9:15353. Crossref

17. Glujovsky D, Riestra B, Sueldo C, et al. Vitrification versus slow freezing for women undergoing oocyte cryopreservation. Cochrane Database Syst Rev 2014;2014:CD010047. Crossref

18. Bianchi V, Macchiarelli G, Borini A, et al. Fine morphological assessment of quality of human mature oocytes after slow freezing or vitrification with a closed device: a comparative analysis. Reprod Biol Endocrinol 2014;12:110. Crossref

19. Turan V, Oktay K. Sexual and fertility adverse effects associated with chemotherapy treatment in women. Expert Opin Drug Saf 2014;13:775-83. Crossref

20. Planer Ovarian Tissue Cryopreservation Scientific Roundup. Ovarian Tissue Cryopreservation Scientific Roundup. 2019. Available from: https://mail.planer.com/__80258426005B6A5E.nsf/0/C4C23666734672DF802584B1002F3777/$File/Ai091V1-Ovarian-Tissue-Cryopreservation-Scientific-Round-Up-28th-October-2019.pdf?OpenElement. Accessed 1 Mar 2022. Crossref

21. Kitazato Corporation. Ova Cryo Kit—Ovarian Tissue Vitrification Kit. 2021. Available from: https://www.kitazato.co.jp/en/products/vitrification/ova-cryo-kit-type-m. Accessed 31 January 2023.

22. Huang L, Mo Y, Wang W, Li Y, Zhang Q, Yang D. Cryopreservation of human ovarian tissue by solid–surface vitrification. Eur J Obstet Gynecol Reprod Biol 2008;139:193-8. Crossref

23. Jensen AK, Kristensen SG, Macklon KT, et al. Outcomes of transplantations of cryopreserved ovarian tissue to 41 women in Denmark. Hum Reprod 2015;30:2838-45.Crossref

24. Rosendahl M, Greve T, Andersen CY. The safety of transplanting cryopreserved ovarian tissue in cancer patients: a review of the literature. J Assist Reprod Genet 2013;30:11-24. Crossref

25. Macklon KT, Jensen AK, Loft A, Ernst E, Andersen CY. Treatment history and outcome of 24 deliveries worldwide after autotransplantation of cryopreserved ovarian tissue, including two new Danish deliveries years after autotransplantation. J Assist Reprod Genet 2014;31:1557-64. Crossref

26. Dolmans MM, Luyckx V, Donnez J, Andersen CY, Greve T. Risk of transferring malignant cells with transplanted frozen–thawed ovarian tissue. Fertil Steril 2013;99:1514-22. Crossref

27. Nisolle M, Casanas-Roux F, Qu J, Motta P, Donnez J. Histologic and ultrastructural evaluation of fresh and frozen–thawed human ovarian xenografts in nude mice. Fertil Steril 2000;74:122-9. Crossref

28. Bosma GC, Fried M, Custer RP, Carroll A, Gibson DM, Bosma MJ. Evidence of functional lymphocytes in some (leaky) scid mice. J Exp Med 1988;167:1016-33. Crossref

29. Dath C, Van Eyck AS, Dolmans MM, et al. Xenotransplantation of human ovarian tissue to nude mice: comparison between four grafting sites. Hum Reprod 2010;25:1734-43. Crossref

30. Dittrich R, Lotz L, Fehm T, et al. Xenotransplantation of cryopreserved human ovarian tissue—a systematic review of MII oocyte maturation and discussion of it as a realistic option for restoring fertility after cancer treatment. Fertil Steril 2015;103:1557-65. Crossref

31. Bankhead P, Loughrey MB, Fernández JA, et al. QuPath: open source software for digital pathology image analysis. Sci Rep 2017;7:16878. Crossref

32. Schmidt KL, Byskov AG, Nyboe Andersen A, Müller J, Yding Andersen C. Density and distribution of primordial follicles in single pieces of cortex from 21 patients and in individual pieces of cortex from three entire human ovaries. Hum Reprod 2003;18:1158-64. Crossref

33. Gougeon A, Chainy GB. Morphometric studies of small follicles in ovaries of women at different ages. J Reprod Fertil 1987;81:433-42. Crossref

34. Anderson RA, Mitchell RT, Kelsey TW, Spears N, Telfer EE, Wallace WH. Cancer treatment and gonadal function: experimental and established strategies for fertility preservation in children and young adults. Lancet Diabetes Endocrinol 2015;3:556-67. Crossref

35. Anderson RA, Wallace WH, Baird DT. Ovarian cryopreservation for fertility preservation: indications and outcomes. Reproduction 2008;136:681-9. Crossref

36. Jeruss JS, Woodruff TK. Preservation of fertility in patients with cancer. N Engl J Med 2009;360:902-11. Crossref

37. Anderson RA, Wallace WH, Telfer EE. Ovarian tissue cryopreservation for fertility preservation: clinical and research perspectives. Hum Reprod Open 2017;2017:hox001. Crossref

38. Pacheco F, Oktay K. Current success and efficiency of autologous ovarian transplantation: a meta-analysis. Reprod Sci 2017;24:1111-20. Crossref

39. Wallace WH, Kelsey TW, Anderson RA. Fertility preservation in pre-pubertal girls with cancer: the role of ovarian tissue cryopreservation. Fertil Steril 2016;105:6-12. Crossref

40. Suzuki N, Yoshioka N, Takae S, et al. Successful fertility preservation following ovarian tissue vitrification in patients with primary ovarian insufficiency. Hum Reprod 2015;30:608-15. Crossref

41. Sun N, Li Z, Pang W, Wang L, Li W. Live birth after transplantation of cryopreserved ovarian tissue with two-year follow-up: report of the first Chinese case [in Chinese]. Chin J Reprod Contraception 2021;41:1026-30.

42. Peek R, Eijkenboom LL, Braat DD, Beerendonk CC. Complete purging of Ewing sarcoma metastases from human ovarian cortex tissue fragments by inhibiting the mTORC1 signaling pathway. J Clin Med 2021;10:4362. Crossref

43. Lotz L, Dittrich R, Hoffmann I, Beckmann MW. Ovarian tissue transplantation: experience from Germany and worldwide efficacy. Clin Med Insights Reprod Health 2019;13:1179558119867357. Crossref

44. Peters IT, Stegehuis PL, Peek R, et al. Noninvasive detection of metastases and follicle density in ovarian tissue using full-field optical coherence tomography. Clin Cancer Res 2016;22:5506-13. Crossref

45. Irwig MS. Testosterone therapy for transgender men. Lancet Diabetes Endocrinol 2017;5:301-11. Crossref

46. Telfer EE, Andersen CY. In vitro growth and maturation of primordial follicles and immature oocytes. Fertil Steril 2021;115:1116-25. Crossref

Efficacy, toxicities, and prognostic factors of stereotactic body radiotherapy for unresectable liver metastases

Hong Kong Med J 2023 Apr;29(2):105–11 | Epub 30 Mar 2023

https://doi.org/10.12809/hkmj208868

ORIGINAL ARTICLE CME

Efficacy, toxicities, and prognostic factors of stereotactic body radiotherapy for unresectable liver metastases

Calvin KK Choi, FHKCR, FHKAM (Radiology); Connie HM Ho, FHKCR, FHKAM (Radiology); Matthew YP Wong, MSc; Ronnie WK Leung, MSc; Frank CS Wong, FHKCR, FHKAM (Radiology); Stewart Y Tung, FHKCR, FHKAM (Radiology); Francis AS Lee, FHKCR, FHKAM (Radiology)

Department of Clinical Oncology, Tuen Mun Hospital, Hong Kong SAR, China

Corresponding author: Dr Calvin KK Choi (calvinkkchoi@hkbh.org.hk)

Full paper in PDF

Abstract

Introduction: This study aims to determine the outcomes of stereotactic body radiotherapy (SBRT) for liver metastases in patients not eligible for surgery.

Methods: This study included 31 consecutive patients with unresectable liver metastases who received SBRT between January 2012 and December 2017; 22 patients had primary colorectal cancer and nine patients had primary non-colorectal cancer. Treatments ranged from 24 Gy to 48 Gy in 3 to 6 fractions over 1 to 2 weeks. Survival, response rates, toxicities, clinical characteristics, and dosimetric parameters were evaluated. Multivariate analysis was performed to identify significant prognostic factors for survival.

Results: Among these 31 patients, 65% had received at least one prior regimen of systemic therapy for metastatic disease, whereas 29% had received chemotherapy for disease progression or immediately after SBRT. The median follow-up interval was 18.9 months; actuarial in-field local control rates at 1, 2, and 3 years after SBRT were 94%, 55%, and 42%, respectively. The median survival duration was 32.9 months; 1-year, 2-year, and 3-year actuarial survival rates were 89.6%, 57.1%, and 46.2%, respectively. The median time to progression was 10.9 months. Stereotactic body radiotherapy was well-tolerated, with grade 1 toxicities of fatigue (19%) and nausea (10%). Patients who received post-SBRT chemotherapy had significant longer overall survival (P=0.039 for all patients and P=0.001 for patients with primary colorectal cancer).

Conclusion: Stereotactic body radiotherapy can be safely administered to patients with unresectable liver metastases, and it may delay the need for chemotherapy. This treatment should be considered for selected patients with unresectable liver metastases.

New knowledge added by this study

Stereotactic body radiotherapy (SBRT) for unresectable liver metastases was effective and well-tolerated. It may delay the need for chemotherapy while prolonging progression-free survival.
The receipt of post-SBRT chemotherapy is a significant prognostic factor for survival.

Implications for clinical practice or policy

Stereotactic body radiotherapy can be regarded as an alternative to surgery for patients with liver metastases, particularly patients with unresectable tumours.
We recommend offering SBRT to patients with unresectable liver metastases if they have good performance status (ie, Eastern Cooperative Oncology Group 0-1), liver tumours ≤6 cm in diameter, three or fewer liver tumours, normal liver volume >700 cm3, adequate organ function, and adequate liver function (Child-Pugh class A).

[Abstract in Chinese]

Introduction

The liver is a common site of metastases, which most frequently originate from primary colorectal cancer via portal circulation. Surgical resection is the standard treatment for medically and technically operable liver metastases, particularly from primary colorectal cancer. However, most patients are not eligible for surgery because of co-morbidities or unfavourable tumour factors. Most patients receive systemic therapy as initial treatment for liver metastases, but such treatment rarely leads to permanent elimination of the metastases; some form of local ablative intervention is required. For patients with unresectable limited liver metastases, numerous local therapeutic approaches are available, such as radiofrequency ablation, transcatheter arterial chemoembolisation, cryotherapy, and high-intensity focal ultrasound. However, all of these approaches exhibit a degree of invasiveness and are currently limited by tumour size (usually <3 cm), distance from critical structures, and distance from critical vasculature.1

In the past, radiotherapy has had a limited role in the management of liver metastases because of concerns regarding radiation-induced liver disease.2 3 Because the liver is subject to the parallel architecture principles of radiobiology, the risk of radiation-induced liver disease is generally proportional to the mean dose of radiation delivered to normal liver tissue. Therefore, small hepatic lesions can be safely treated with high doses of radiation via stereotactic body radiotherapy (SBRT). Advances in tumour imaging, radiotherapy planning and delivery, and motion management have facilitated the delivery of highly precise and four-dimensional SBRT. This non-invasive method can be used to deliver ablative treatments on an outpatient basis, thereby decreasing morbidity and cost.4

Ablative techniques offer a minimally invasive treatment option for selected patients with oligometastatic liver disease.5 There is increasing evidence to support the use of SBRT.6 To our knowledge, there is limited published information regarding the role of SBRT in the treatment of unresectable liver metastases in Hong Kong. In this study, we investigated the efficacy, toxicities, and prognostic factors of SBRT in patients with unresectable liver metastases.

Methods

Patient eligibility

Data regarding consecutive patients with unresectable liver metastases who received SBRT between January 2012 and December 2017 were retrospectively retrieved from the treatment database of the Department of Clinical Oncology at Tuen Mun Hospital. All patients with liver metastases were evaluated in multidisciplinary team meetings involving radiation oncologists and hepatobiliary surgeons. Eligibility was determined using the following criteria: (1) histologically confirmed malignancy (hepatic lesion biopsy not required); (2) biphasic computed tomography (CT) scan or positron emission tomography–CT of the liver within 4 weeks of radiation planning demonstrating liver tumours ≤6 cm in diameter, presence of three or fewer liver tumours, and normal liver volume >700 cm³; (3) discussion of the case in a multidisciplinary team meeting that included an opinion regarding the lack of qualification for radiofrequency ablation, along with a determination of non-resectability by a qualified hepatic surgeon; (4) patient refusal of surgical treatment; (5) Eastern Cooperative Oncology Group performance status 0 or 1; (6) adequate organ function (absolute neutrophil count ≥1.5×10⁹/L; platelet count ≥75×10⁹/L; creatinine level ≤1.5×upper limit of normal), liver function test results (aspartate aminotransferase and alanine aminotransferase levels ≤1.5×normal level), and Child-Pugh score of ≤6 (class A); (7) controlled extrahepatic disease and life expectancy >6 months; (8) no chemotherapy concurrent with radiotherapy (previous chemotherapy was not an exclusion criterion); and (9) previous treatment with radiofrequency ablation was not an exclusion criterion if recurrence had been confirmed.

Radiotherapy treatment

During four-dimensional CT scans, patients were positioned supine on an evacuated foam bag (Klarity Medical, China) with both arms abducted. The extent of tumour motion during respiration was used to determine whether treatment would be administered with free breathing plus abdominal compression or active breathing control. The gross tumour volume (GTV) was determined using contrast CT and co-registered with positron emission tomography–CT. For patients who required optimal abdominal compression to mitigate organ motion, planning was conducted using the mid-ventilation–based planning target volume (PTV) approach, and the GTV was determined using intravenous contrast CT. The clinical target volume was 0 mm outside of the GTV within the liver (ie, equal to GTV); it included the position of the tumour in all phases of respiration. The PTV was defined by adding an isotropic margin of 3 to 5 mm from the clinical target volume or 7 to 10 mm in the cranial-caudal axis and 4 to 6 mm in the anterior-posterior and lateral axes. Pretreatment four-dimensional cone-beam CT was performed prior to each treatment for all patients to adjust for setup uncertainties. Tumour localisation was conducted using the diaphragm or whole liver as a surrogate for the tumour. A two-step four-dimensional registration approach was used to align the diaphragm/liver surrogate to its time-weighted mean position. The SBRT dose, ranging from 8 to 16 Gy × 3 fractions to 5 to 7.5 Gy × 6 fractions, was individualised according to the following normal tissue constraints: (1) maximum spinal cord dose <15 Gy; (2) ≥700 cm³ of liver should receive <15 Gy, and D5% <30 Gy; (3) maximum stomach point dose of 25 Gy; and (4) maximum duodenum point dose of 25 Gy.

Evaluation

Patients were evaluated weekly during SBRT, immediately after completion of treatment, at 6 weeks after treatment, every 3 months for the first 2 years, and every 4 months thereafter. Physical examinations and blood tests were performed at each follow-up visit. Triphasic CT of the liver was conducted at 3 months after SBRT and then every 6 months until disease progression. Tumour response was assessed using modified response evaluation criteria for solid tumours.

The primary endpoint of the study was local control; secondary endpoints were overall survival and toxicity. Local control was defined as the absence of progressive disease within the PTV. The appearance of new lesions outside of the PTV was regarded as intrahepatic out-field failure. Overall survival was calculated from the start of SBRT until the end of follow-up or death.

Toxicity was graded using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0. Toxicities were defined as adverse events that occurred <3 months after SBRT. Newly developed toxicities or toxicities that progressed to one grade above baseline were regarded as adverse events. Grade 5 liver failure related to SBRT was defined as death from liver failure in the presence of acute grade 3 liver toxicities during <6 months without intrahepatic progression.

Statistical analysis

Data were analysed using SPSS software (Windows version 23.0; IBM Corp, Armonk [NY], United States). Fisher’s exact test and independent t tests were used for univariate analysis of patient, disease, and treatment factors associated with liver toxicity. Binary logistic regression analysis was used for univariate analysis of dose-volumetric parameters associated with liver toxicity. Kaplan–Meier test was used for univariate analysis of overall survival, with a significance threshold of P<0.25; it was used for multivariate analysis of overall survival, with a significance threshold of P<0.05. Cox regression was used for further evaluation of variables which were significant in univariate analysis of overall survival.7 8

Results

Patients and treatment

During the study period, 31 consecutive patients with unresectable liver metastases underwent SBRT at our institution. Their characteristics are shown in Table 1. Colorectal cancer was the most common primary cancer. A total of 64.5% of patients received systemic treatment before SBRT; 71% of liver lesions were ≤ 30 mm. All patients received a fixed course of 3 or 6 fractions with total prescribed dose ranges of 24-48 Gy. The mean GTV was 26.9 cm³ (range, 1.5-137) and mean PTV was 91.8 cm³ (range, 21.7-269). The mean biological equivalent dose (BED10) to GTV was 79.8 Gy (range, 43.2-124.8). The median BED10 to GTV was 76.8 Gy. Surgical resection or radiofrequency ablation were performed in 32% of patients before SBRT. Targeted or non-targeted systemic chemotherapy was administered to 65% and 29% of patients before and after SBRT, respectively.

Table 1. Patient characteristics

Toxicities

Stereotactic body radiotherapy was well-tolerated. There were no grade 2-4 toxicities. Most patients were asymptomatic (grade 0) during radiotherapy; 19% of patients had grade 1 fatigue, 10% of patients had grade 1 nausea, and 3% of patients had skin reaction. No patients exhibited a change in Child-Pugh class after SBRT, and no significant prognostic factors for liver toxicities were identified.

Local control, survival, and prognostic factors

The median follow-up interval was 18.9 months. The 1-year, 2-year, and 3-year local control rates were 94% (29/31), 55% (17/31) and 42% (13/31), respectively. Only two patients (9% of all patients) with primary colorectal cancer had in-field recurrence at 1 year after SBRT. Sixteen patients in all treatment groups had out-field recurrence at 1 year after SBRT. The median time to progression was 10.9 months.

The median survival duration in all treatment groups was 32.9 months. The 1-year, 2-year, and 3-year survival rates were 89.6%, 57.1%, and 46.2%, respectively. The only significant prognostic factor for overall survival was receipt of post-SBRT chemotherapy for disease progression (P=0.039). Figures 1 and 2 show the survival curves and prognostic factors for all treatment groups. Previous local treatment, rat sarcoma virus status of colorectal cancer, number of liver metastases, extrahepatic metastases, BED to the liver, extrahepatic metastasis status, number of chemotherapy lines before or after SBRT, and carcinoembryonic antigen level after SBRT were not significant prognostic factors for overall survival. Table 2 summarises the factors that affected overall survival.

Figure 1. Overall survival of the whole group after stereotactic body radiotherapy (SBRT)

Figure 2. Overall survival of patients who received chemotherapy after stereotactic body radiotherapy (SBRT) for disease progression (PD) versus those who did not

Table 2. Prognostic factors affecting overall survival

The median survival duration in the colorectal cancer subgroup was 32.9 months. The only significant prognostic factor for overall survival was receipt of post-SBRT chemotherapy for disease progression (P=0.001). No other significant prognostic factors for overall survival were identified. Figures 3 and 4 show the survival curves and prognostic factors for the colorectal cancer subgroup.

Figure 3. Overall survival of colorectal cancer patients after stereotactic body radiotherapy (SBRT)

Figure 4. Overall survival of colorectal cancer patients who received chemotherapy after stereotactic body radiotherapy (SBRT) for disease progression (PD) versus those who did not

Discussion

Although surgical resection is the standard treatment for liver metastases, many patients are not eligible for such treatment. Multiple retrospective and prospective studies have demonstrated SBRT is a promising, safe, and non-invasive alternative to surgery for unresectable liver metastases.9 10 To our knowledge, there is limited published information regarding the use of SBRT to treat liver metastases in Hong Kong. In the present study, we retrospectively collected data regarding consecutive patients who received SBRT for unresectable liver metastases after multidisciplinary team evaluation; we assessed outcomes in terms of safety, local control, and survival. Among the 31 patients treated with SBRT, the 1-year and 2-year local control rates were 93% and 55%, respectively. The median survival duration was 32.9 months; the 1-year and 2-year survival rates were 89.6% and 57.1%, respectively. In the colorectal cancer subgroup, the 1-year and 2-year survival rates were 84.7% and 62.1%, respectively.

Multiple retrospective and prospective studies have been performed regarding SBRT for liver metastases from colorectal cancers (Table 3).11 12 13 14 In the present study, local control rates and survival rates were comparable with findings in previous reports. Notably, McPartlin et al11 conducted a prospective study using SBRT doses of 22-62 Gy in 6 fractions. The present study, with SBRT doses of 24-48 Gy in 3-6 fractions, demonstrated better 1-year local control (93% vs 50%) and 2-year survival (62.1% vs 26%) than the study by McPartlin et al.11

Table 3. Summary of literature regarding stereotactic body radiotherapy for liver metastases from colorectal cancers

Three other SBRT trials12 13 14 (45-75 Gy in 3 fractions) all demonstrated better local control rates than the findings in the present study (Table 3). These results indicate that a higher local control rate is associated with a higher radiation dose. Compared with the present study, Scorsetti et al12 and Joo et al14 showed higher 2-year survival rates (65% and 75%, respectively vs 62.1% in the present study), whereas Hoyer et al13 revealed a considerably lower 2-year survival rate (38%). These discrepant findings may be related to radiation dose—Scorsetti et al12 and Joo et al14 reported higher BED than that achieved by Hoyer et al13 and the present study. Among patients with primary colorectal tumours, the survival rate in the present study was comparable with rates in the previous studies.11 12 13 14 However, overall survival is dependent on many factors other than local control of irradiated liver metastases. Compared with earlier studies, overall survival is expected to be better in more recent studies because of stage migration, improvements in diagnostic techniques, and enhanced systemic treatment. Importantly, although the present study showed that post-SBRT chemotherapy was a prognostic factor for longer survival, selection bias may have been involved in the decision to administer chemotherapy to patients with better performance status.

In the present study, the incidence of toxicities was low, and there were no grade 2-4 toxicities. Among patients who received SBRT, only grade 1 toxicities were reported (fatigue, nausea, and skin reaction); these findings indicate that SBRT was well-tolerated.

Based on our results, we recommend that patients with unresectable liver metastases are evaluated in multidisciplinary team meetings; patients should be offered SBRT if they have good performance status (ie, Eastern Cooperative Oncology Group 0-1), liver tumours ≤6 cm in diameter, three or fewer liver tumours, normal liver volume >700 cm³, adequate organ function, and adequate liver function (Child-Pugh class A). Considering its minimal invasiveness and toxicity, as well as its potential for improving progression-free survival, SBRT should be regarded as an alternative to surgical resection of liver metastases to those patients who refuse surgical treatment.

There were some limitations in the present study. First, the BED to the tumour was low (median BED10 >100 Gy was administered to 35.5% of patients), and the mean GTV was high (26.9 cm³). The local control rate may have been influenced by the lower total radiation dose administered and larger tumour volume. Second, this was a retrospective study, and the sample size was small. Thus, a randomised controlled trial with a large number of patients is needed to determine whether SBRT can prolong overall survival in patients with liver metastases.

Conclusion

Stereotactic body radiotherapy can be safely administered to patients with unresectable liver metastases, and it may delay the need for chemotherapy. Considering its minimal invasiveness and toxicity, this treatment should be offered to selected patients with unresectable liver metastases; such an approach may improve progression-free survival. A phase III randomised study is needed to confirm these results.

Author contributions

All authors contributed to the concept or design of the study, acquisition of data, analysis or interpretation of data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgement

The authors thank Mr Jia-jie Huang from Quality and Safety Division of New Territories West Cluster, Hospital Authority, Hong Kong for his statistical analysis support.

Funding/support

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Ethics approval

Ethics approval document was issued by New Territories West Cluster Research Ethics Committee of Hospital Authority, Hong Kong (Ref No.: NTWC/REC/20035). Informed consent was obtained from patients for stereotactic body radiotherapy.

References

1. Aitken KL, Hawkins MA. Stereotactic body radiotherapy for liver metastases. Clin Oncol (R Coll Radiol) 2015;27:307-15. Crossref

2. Schefter TE, Kavanagh BD, Timmerman RD, Cardenes HR, Baron A, Gaspar LE. A phase I trial of stereotactic body radiation therapy (SBRT) for liver metastases. Int J Radiat Oncol Biol Phys 2005;62:1371-8. Crossref

3. Rusthoven KE, Kavanagh BD, Cardenes H, et al. Multi-institutional phase I/II trial of stereotactic body radiation therapy for liver metastases. J Clin Oncol 2009;27:1572-8. Crossref

4. Pan CC, Kavanagh BD, Dawson LA, et al. Radiation-associated liver injury. Int J Radiat Oncol Biol Phys 2010;76(3 Suppl):S94-100. Crossref

5. Aloia TA, Vauthey JN, Loyer EM, et al. Solitary colorectal liver metastasis: resection determines outcome. Arch Surg 2006;141:460-6. Crossref

6. Høyer M, Swaminath A, Bydder S, et al., Radiotherapy for liver metastases: a review of evidence. Int J Radiat Oncol Biol Phys 2012;82:1047-57. Crossref

7. Prentice RL, Zhao S. Regression models and multivariate life tables. J Am Stat Assoc 2021;116:1330-45. Crossref

8. Hashemi R, Commenges D. Correction of the p-value after multiple tests in a Cox proportional hazard model. Lifetime Data Anal 2002;8:335-48. Crossref

9. Rusthoven CG, Lauro CF, Kavanagh BD, Schefter TE. Stereotactic body radiation therapy (SBRT) for liver metastases: a clinical review. Semin Colon Rectal Surg 2014;25:48-52. Crossref

10. Kobiela J, Spychalski P, Marvaso G, et al. Ablative stereotactic radiotherapy for oligometastatic colorectal cancer: systematic review. Crit Rev Oncol Hematol 2018;129:91-101. Crossref

11. McPartlin A, Swaminath A, Wang R, et al. Long-term outcomes of phase 1 and 2 studies of SBRT for hepatic colorectal metastases. Int J Radiat Oncol Biol Phys 2017;99:388-95. Crossref

12. Scorsetti M, Comito T, Tozzi A, et al. Final results of a phase II trial for stereotactic body radiation therapy for patients with inoperable liver metastases from colorectal cancer. J Cancer Res Clin Oncol 2015;141:543-53. Crossref

13. Hoyer M, Roed H, Traberg Hansen A, et al. Phase II study on stereotactic body radiotherapy of colorectal metastases. Acta Oncol 2006;45:823-30. Crossref

14. Joo JH, Park JH, Kim JC, et al. Local control outcomes using stereotactic body radiation therapy for liver metastases from colorectal cancer. Int J Radiat Oncol Biol Phys 2017;99:876-83. Crossref

Asthenopia prevalence and vision impairment severity among students attending online classes in low-income areas of western China during the COVID-19 pandemic

Hong Kong Med J 2023 Apr;29(2):150–7

https://doi.org/10.12809/hkmj219864

ORIGINAL ARTICLE (HEALTHCARE IN MAINLAND CHINA)

Asthenopia prevalence and vision impairment severity among students attending online classes in low-income areas of western China during the COVID-19 pandemic

Y Ding, PhD¹; H Guan, PhD¹; K Du, PhD²; Y Zhang, PhD¹; Z Wang, MD¹; Y Shi, PhD¹

¹ Center for Experimental Economics for Education, Shaanxi Normal University, Xi’an, China

² College of Economics, Xi’an University of Finance and Economics, Xi’an, China

Corresponding author: Dr H Guan (hongyuguan0621@gmail.com)

Full paper in PDF

Abstract

Introduction: This study explored the impact of online learning during the coronavirus disease 2019 (COVID-19) pandemic on asthenopia and vision impairment in students, with the aim of establishing a theoretical basis for preventive approaches to vision health.

Methods: This balanced panel study enrolled students from western rural China. Participant information was collected before and during the COVID-19 pandemic via questionnaires administered at local vision care centres, along with clinical assessments of visual acuity. Paired t tests and fixed-effects models were used to analyse pandemic-related differences in visual status.

Results: In total, 128 students were included (mean age before pandemic, 11.82 ± 1.46 years). The mean total screen time was 3.22 ± 2.90 hours per day during the pandemic, whereas it was 1.97 ± 1.90 hours per day in the pre-pandemic period (P<0.001). Asthenopia prevalence was 55% (71/128) during the pandemic, and the mean visual acuity was 0.81 ± 0.30 logarithm of the minimum angle of resolution; these findings indicated increasing vision impairment, compared with the pre-pandemic period (both P<0.001). Notably, asthenopia prevalence increased by two- to three-fold, compared with the pre-pandemic period. An increase in screen time while learning was associated with an increase in asthenopia prevalence (P=0.034).

Conclusion: During the COVID-19 pandemic, students spent more time on online classes, leading to worse visual acuity and vision health. Students in this study reported a significant increase in screen time, which was associated with increasing asthenopia prevalence and worse vision impairment. Further research is needed regarding the link between online classes and vision problems.

New knowledge added by this study

Online learning has become increasingly popular during the coronavirus disease 2019 pandemic. Students reported a nearly twofold increase in screen time during the pandemic, compared with the pre-pandemic period.
Students reported greater asthenopia prevalence and demonstrated worse vision impairment during the pandemic, compared with the pre-pandemic period.
Screen time was associated with asthenopia prevalence but not with the progression of vision impairment.

Implications for clinical practice or policy

Policymakers should carefully consider the prevalence of asthenopia and progression of vision impairment among students who are increasingly using digital devices and enrolling in online classes.
Policies regarding vision care should be implemented in response to the increasing use of online learning approaches.

[Abstract in Chinese]

Introduction

The World Health Organization announced that the coronavirus disease 2019 (COVID-19) outbreak had become an international public health emergency on 30 January 2020; on 11 March 2020, it declared that the outbreak had become a pandemic.1 Governments and public health authorities worldwide implemented public health policies to reduce the risk of viral transmission, including strict physical distancing, severe travel restrictions, and the closure of many businesses and schools. On 25 January 2020, China’s Central Government announced a nationwide travel ban and quarantine policy2; it initiated nationwide school closures as an emergency measure to prevent the spread of COVID-19.3 Thus, >220 million school-aged children and adolescents were confined to their homes; online classes were offered and delivered via the internet.4

Vision problems are public health challenges; among school-aged children, these problems often involve asthenopia and vision impairment. Asthenopia is defined as a subjective sensation of visual fatigue, eye weakness, or eyestrain; it can manifest through various symptoms, including epiphora, ocular pruritis, diplopia, eye pain, and dry eye.5 Vision impairment is defined as visual acuity (VA) of 6/12 or worse in either eye6; it is often caused by uncorrected refractive errors, and its estimated prevalence is 43%.7 Although both asthenopia and vision impairment have negative effects on students, the effects of vision impairment are greater. A previous global analysis revealed that vision impairment was present in 12.8 million children aged 5 to 15 years, half of whom lived in China.8 Moreover, students with vision impairment have lower scores on various motor and cognitive tests.9 10

Excessive use of digital devices contributes to increases in asthenopia prevalence and vision impairment among school-aged children.4 11 12 13 14 15 The COVID-19 pandemic has led to increased use of digital device–supported online classes,16 17 18 which require extended exposure to those devices.19 20 Importantly, long durations of exposure to digital devices can contribute to many vision problems in children.14

Asthenopia and vision impairment related to the excessive use of digital devices during the COVID-19 pandemic have been investigated in developed countries and urban China.4 11 12 To our knowledge, no similar studies have been conducted in western rural China. Additionally, online classes are increasingly implemented in rural areas, and the use of digital devices is becoming more prevalent11; thus, there is a need for research that focus on vision health in students.

The primary purpose of this study was to assess screen time, asthenopia prevalence, and vision impairment progression during the COVID-19 pandemic among students in western rural China. To achieve this goal, we first conducted a general descriptive analysis of student characteristics and screen time trends before and during the pandemic. We then investigated the prevalence of asthenopia and progression of vision impairment. Finally, we explored factors influencing the prevalence of asthenopia and progression of vision impairment before and during the pandemic.

Methods

Setting

This study focused on areas that were broadly representative of rural western China because of limited resources. Thus, the study was conducted in Shaanxi and Ningxia regions in western China. In 2019, the per capita gross domestic product in Shaanxi Province was US$10 167; this is similar to that in Ningxia Autonomous Region (US$8236).21

Sample selection

Vision data were acquired from local vision care centres (VCs), which had been established by the Center for Experimental Economics in Education at Shaanxi Normal University, in cooperation with county-level organisations such as the local education ministries and hospitals.

Before the pandemic, VC screenings were performed in each county, except during summer and winter vacations. Staff conducted one to two screenings per week (covering 2 to 4 schools); they completed one round of screening in one town each month. In practice, approximately 1 year is needed to complete one round of vision screening for all eligible children in a particular county. The second round and subsequent rounds of vision screening were performed using a similar workflow. After the completion of vision screening, students who required further assessment were referred to the VC for full eye and refractive examinations. This study included students who had visited the VC 3 months before the beginning of the COVID-19 pandemic.

During the pandemic, VC staff could not attend schools to perform vision screenings. To maintain vision screening services for students, we telephoned all students who had visited the VC before the pandemic. Participants in this panel study were students who participated in data collection before and during the COVID-19 pandemic.

Data collection

We conducted two cycles of surveys in the VC. The first survey cycle was conducted from October to December 2019 (before the pandemic); the second survey cycle was conducted among a group of students who visited the VC for follow-up from July to December 2020 (during the pandemic), based on their enrolment in the study before the pandemic. The same information was collected during the two survey cycles. During the vision screening process, VC staff administered questionnaires to students for collection of the following information: sex (male=1), age, ethnicity (Han=1), residence (non-rural=1), only-child status (yes=1), parental education (parents with ≥12 years of education=1), and parental migration status (one or both out-migrated=1; defined as one or both parents worked away from home during the semester). Household assets were calculated by summing the values of 13 items owned by the family, in accordance with the China Rural Household Survey Yearbook.22

The survey also included the collection of information regarding screen time and asthenopia. Students completed a previously described, self-administered questionnaire concerning mean time spent throughout the day on near activities (including computer and smartphone use, television viewing, and studying/homework after school). Reports of time spent on near activities during different parts of the day were categorised as screen time while learning and screen time while playing. Information regarding asthenopia was collected via three questions focused on ocular discomfort: whether the student had experienced dry eyes (yes=1), eye pain and swelling (yes=1), and eye fatigue and watery eyes (yes=1). Asthenopia was defined as the presence of at least one of these three types of vision health problems (yes=1).23 Furthermore, information regarding VA was collected when students visited the VC. The optometrist in the VC conducted a VA test to measure the clarity of each student’s vision. All students completed VA tests without refractive correction; students with spectacles completed VA tests with their routine method of vision correction.

The questionnaire regarding asthenopia was developed and reviewed by a group of health experts from Shaanxi Normal University and Zhongshan Ophthalmic Center, a well-known ophthalmology institution in China. The included questions were constructed to ensure that they could be clearly understood by students aged 9 to 17 years with the aid of trained VC staff. These three questions can serve as good indicators of symptoms representing different degrees of asthenopia in students, and they have been used in previous research.23

Visual acuity assessment

Visual acuity was assessed using Early Treatment Diabetic Retinopathy Study tumbling-E charts (Precision Vision, La Salle [IL], United States). In an indoor area with sufficient light, VA was separately assessed for each eye without refraction at a distance of 4 m. Students were first examined using a 6/60 line; if they correctly identified the orientation of at least four of five optotypes, they were examined using a 6/30 line, followed by a 6/15 line and a 6/3 line. In this manner, the VA for an eye was defined as the lowest line on which four of five optotypes were correctly identified. If the participant could not read the top line at a distance of 4 m, they were tested at a distance of 1 m, and the VA result was divided by 4.

In this study, VA levels were calculated and compared using the logarithm of the minimum angle of resolution (logMAR) scale, which is a linear scale with regular increments that offers a reasonably intuitive interpretation of VA measurement.24 In this study, vision impairment was defined as logMAR ≥0.3 (ie, VA of 6/12 or worse) in either eye.

Statistical methods

This balanced panel study compared student data between two periods (before and during the COVID-19 pandemic). Mean screen time, asthenopia prevalence, and vision impairment progression were compared among students using t tests, after stratification according to various demographic and behavioural factors. Fixed-effects logistic and regression models were used to explore factors influencing the prevalence of asthenopia and progression of vision impairment before and during the pandemic. Fixed-effects models were adjusted for sex, age, ethnicity, rural or non-rural residence, only-child status, parental migration status, parental education level, household assets, screen time while learning, and screen time while playing. All analyses were performed using Stata Statistical Software, version 14.1 (StataCorp, College Station [TX], United States). All tests were two-sided, and P values <0.05 were considered statistically significant.

Results

This study included 128 students from western rural China (mean age before pandemic, 11.82 ± 1.46 years; mean age during pandemic, 12.32 ± 1.54 years; 80 girls [62.5%] and 48 boys [37.5%]). All participants had vision impairment and were attending online classes (Table 1).

Table 1. Screen time before and during the coronavirus disease 2019 pandemic, stratified according to student characteristics (n=128)

During the pandemic, screen time significantly increased because of enrolment in online classes. The mean total screen time during the pandemic was 3.22 hours per day, compared with 1.97 hours during the pre-pandemic period (P<0.001). The mean screen time while learning during the pandemic was 1.70 hours per day, compared with 0.90 hours during the pre-pandemic period (P<0.001); the mean screen time while playing during the pandemic was 1.52 hours per day, compared with 1.33 hours during the pre-pandemic period (P=0.019). Additionally, rural students had significantly greater screen time while learning during the pandemic, compared with the pre-pandemic period (P<0.001); there was no such difference among non-rural students (Table 1).

The prevalence of asthenopia and progression of vision impairment significantly differed between the pandemic and pre-pandemic periods. The prevalence of asthenopia during the pandemic was 55% (71/128), whereas it was 27% (35/128) during the pre-pandemic period (P<0.001). The mean logMAR VA was worse during the pandemic compared with the pre-pandemic period (0.81 vs 0.65; P<0.001). The prevalence of asthenopia was higher during the pandemic than during the pre-pandemic period, regardless of the characteristics used to stratify participants. The mean logMAR VA was worse during the pandemic than during the pre-pandemic period, although the difference being insignificant among participants with non-Han ethnicity and participants in the top quartile of household assets (Table 2).

Table 2. Asthenopia prevalence and visual acuity (in logarithm of the minimum angle of resolution [logMAR]) before and during the coronavirus disease 2019 pandemic, stratified according to student characteristics (n=128)

Fixed-effects logistic models for asthenopia revealed that screen time while learning was associated with asthenopia prevalence, and the probability of asthenopia increased by 24.6% for each 1-hour increase in screen time while learning (95% confidence interval [CI]=1.02-1.53; P=0.034). Additionally, older age (odds ratio [OR]=2.073, 95% CI=1.13-3.81, P=0.019), Han ethnicity (OR=2.405, 95% CI=1.22-4.74; P=0.011), and only-child status (OR=0.488, 95% CI=0.21-1.13; P=0.095) were factors associated with asthenopia; screen time while playing was not (Table 3).

Table 3. Fixed-effects logistic analysis of factors associated with asthenopia before and during the coronavirus disease 2019 pandemic (n=128)

Fixed-effects regression models showed that residence in a non-rural area (OR=-0.200, 95% CI=-0.355 to -0.046; P=0.011) and only-child status (OR=-0.099, 95% CI=-0.197 to 0.000; P=0.049) were factors associated with logMAR VA. The probability of worse logMAR VA increased by 0.200 in non-rural areas, compared with rural areas. However, screen time while learning and screen time while playing were not associated with vision impairment (Table 4).

Table 4. Fixed-effects regression analysis of factors associated with visual acuity (in logarithm of the minimum angle of resolution [logMAR]) before and during the coronavirus disease 2019 pandemic (n=128)

Discussion

The global spread of the COVID-19 pandemic has affected the education of >1.5 billion children and adolescents worldwide.25 The participants in our study were representative of this important population. They demonstrated declines in VA and vision health during the pandemic, in relation to the excessive use of digital devices; these findings were consistent with the results of previous studies.19 26

All students in our study were attending online classes during the pandemic. We observed an increase in the mean daily time spent on digital devices between the pre-pandemic and pandemic periods; these results are consistent with international findings that screen time was greater during the pandemic than before the pandemic.19 Notably, we found that total screen time and screen time while learning significantly changed among rural students but not among non-rural students; these results are also consistent with previous findings.19 This difference presumably occurred because, compared with rural students, non-rural students were more likely to use digital devices and online classes before the pandemic.

We observed a significant difference in asthenopia prevalence among students in low-income areas of western China before and during the pandemic; this finding supports the results of previous studies.26 27 Although the risk of asthenopia reportedly increases with screen time,28 there is no published literature concerning changes in asthenopia among students in relation to the COVID-19 pandemic. Similar to previous studies,14 we found that the prevalence of asthenopia was approximately twofold greater among students aged 13 to 17 years than among those aged 9 to 12 years. Furthermore, Moon et al26 reported that symptoms of dry eye diseases were more common among older children than among younger children. Older children spend more time using digital devices, leading to a higher prevalence of asthenopia.29

This study showed significant progression of vision impairment in relation to the pandemic; similarly, a study in eastern China revealed that students had worse vision during the pandemic, compared with their vision at pre-pandemic examinations.4 However, screen time has not been associated with vision impairment among students. Furthermore, evidence regarding the impact of digital devices use on vision impairment has been inconsistent,30 31 with computer screen time made students’ vision worse while television viewing had no effect. We speculate that the association will become clearer as school-aged children spend increasing amounts of time using these devices.

This study had three important limitations. First, the screen time data were retrospectively collected through a self-reporting mechanism, which may have led to recall bias. However, considering the resource and measurement limitations that researchers encountered during the pandemic, self-reported recall was regarded as the optimal method for collection of screen time data in the present study. Second, the selection of students with poor vision may lead to underestimation of screen time effects on the general population, and the results should be generalised with caution. Third, the study was not designed to accurately distinguish between vision impairment caused by intrinsic factors and vision impairment caused by pandemic-related eye strain.

Our findings provide new evidence regarding the effects of increased screen time on asthenopia and vision impairment among students in western rural China during the pandemic; they can also serve as a basis for future research. Although pandemic-related school closures are temporary, the increasing popularity of online classes may accelerate the overall acceptance of digital devices. The use of online learning approaches is associated with multiple vision problems, which merit attention in future studies.

Conclusion

The present study demonstrated that asthenopia and vision impairment among students in western rural China were also affected by the pandemic; these findings provide critical insights regarding the effects of the pandemic on vision health in rural students. Moreover, the findings highlight important issues related to childhood vision health during the pandemic; parents, teachers, and eye care providers should consider evidence-based measures to avoid asthenopia and vision impairment in children. The current pace of economic and technological development is leading to increased use of digital devices and online learning approaches, but vision problems in rural China have not received sufficient consideration. Thus, there is a critical need for greater efforts to monitor VA and vision health among students in this region.

Author contributions

Concept or design: All authors.
Acquisition of data: Y Ding, H Guan, K Du.
Analysis or interpretation of data: Y Ding, H Guan, K Du, Y Shi.
Drafting of the manuscript: Y Ding, Y Zhang, Z Wang.
Critical revision of the manuscript for important intellectual content: H Guan, Y Shi.

All authors contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

As an International Editorial Advisory Board member of the journal, Y Shi was not involved in the peer review process. Other authors have disclosed no conflicts of interest.

Acknowledgement

We thank Dr Wenting Liu, Dr Jiaqi Zhu, and staff from the Center for Experimental Economics in Education of Shaanxi Normal University, China for their valuable contributions.

Funding/support

H Guan received funding for this study from the National Natural Science Foundation of China (Grant No.: 7180310) and Soft Science Project of Shaanxi Province (Grant No.: 2023-CX-RKX-127). Y Ding received funding for this study from the Fundamental Research Funds for the Central Universities (Grant No.: 2020CSWY018). This study was supported by the 111 Project (Grant No.: B16031). The funders had no role in designing the study, collecting, analysing or interpreting the data, or in drafting this manuscript.

Ethics approval

This study protocol was approved by Sun Yat-sen University, China (Registration No.: 2013MEKY018) and all procedures followed the principles of the Declaration of Helsinki. Permission was obtained from the local boards of education in the study area, as well as the principals of all participating schools. All participating children provided oral assent before baseline data collection, and legal guardians provided written informed consent for their children to be enrolled in the study.

References

1. World Health Organization. WHO timeline—COVID-19. 2020. Available from: https://www.who.int/news-room/detail/27-04-2020-who-timeline---covid-19. Accessed 13 Sep 2021.

2. Li D, Liu Z, Liu Q, et al. Estimating the efficacy of quarantine and traffic blockage for the epidemic caused by 2019-nCoV (COVID-19): a simulation analysis. medRxiv [Preprint]. 25 Feb 2020. Available from: https://doi.org/10.1101/2020.02.14.20022913. Accessed 13 Sep 2021. Crossref

3. Wang G, Zhang Y, Zhao J, Zhang J, Jiang F. Mitigate the effects of home confinement on children during the COVID-19 outbreak. Lancet 2020;395:945-7. Crossref

4. Wang J, Li Y, Musch DC, et al. Progression of myopia in school-aged children after COVID-19 home confinement. JAMA Ophthalmol 2021;139:293-300. Crossref

5. Kowalska M, Zejda JE, Bugajska J, Braczkowska B, Brozek G, Malińska M. Eye symptoms in office employees working at computer stations [in Polish]. Med Pr 2011;62:1-8.

6. Cumberland PM, Peckham CS, Rahi JS. Inferring myopia over the lifecourse from uncorrected distance visual acuity in childhood. Br J Ophthalmol 2007;91:151-3. Crossref

7. Pascolini D, Mariotti SP. Global estimates of visual impairment: 2010. Br J Ophthalmol 2012;96:614-8. Crossref

8. Resnikoff S, Pascolini D, Mariotti SP, Pokharel GP. Global magnitude of visual impairment caused by uncorrected refractive errors in 2004. Bull World Health Organ 2008;86:63-70. Crossref

9. Jan C, Li SM, Kang MT, et al. Association of visual acuity with educational outcomes: a prospective cohort study. Br J Ophthalmol 2019;103:1666-71. Crossref

10. Roch-Levecq AC, Brody BL, Thomas RG, Brown SI. Ametropia, preschoolers’ cognitive abilities, and effects of spectacle correction. Arch Ophthalmol 2008;126:252-8. Crossref

11. Zhang Z, Xu G, Gao J, et al. Effects of e-learning environment use on visual function of elementary and middle school students: a two-year assessment—experience from China. Int J Environ Res Public Health 2020;17:1560. Crossref

12. Wong CW, Tsai A, Jonas JB, et al. Digital screen time during COVID-19 pandemic: risk for a further myopia boom? Am J Ophthalmol 2021;223:333-7. Crossref

13. Kim J, Hwang Y, Kang S, et al. Association between sexposure to smartphones and ocular health in adolescents. Ophthalmic Epidemiol 2016;23:269-76. Crossref

14. Mohan A, Sen P, Shah C, Jain E, Jain S. Prevalence and risk factor assessment of digital eye strain among children using online e-learning during the COVID-19 pandemic: digital eye strain among kids (DESK study-1). Indian J Ophthalmol 2021;69:140-4. Crossref

15. Guan H, Yu NN, Wang H, et al. Impact of various types of near work and time spent outdoors at different times of day on visual acuity and refractive error among Chinese school-going children. PLoS One 2019;14:e0215827.Crossref

16. Sultana A, Tasnim S, Hossain MM, Bhattacharya S, Purohit N. Digital screen time during the COVID-19 pandemic: a public health concern. Available from: https://f1000research.com/articles/10-81. Accessed 13 Sep 2021. Crossref

17. Nigg CR, Wunsch K, Nigg C, et al. Are physical activity, screen time, and mental health related during childhood, preadolescence, and adolescence? 11-year results from the German Montorik–Modul Longitudinal Study. Am J Epidemiol 2021;190:220-9. Crossref

18. Schmidt SC, Anedda B, Burchartz A, et al. Physical activity and screen time of children and adolescents before and during the COVID-19 lockdown in Germany: a natural experiment. Sci Rep 2020;10:21780. Crossref

19. Aguilar-Farias N, Toledo-Vargas M, Miranda-Marquez S, et al. Sociodemographic predictors of changes in physical activity, screen time, and sleep among toddlers and preschoolers in Chile during the COVID-19 pandemic. Int J Environ Res Public Health 2020;18:176. Crossref

20. Bates LC, Zieff G, Stanford K, et al. COVID-19 impact on behaviors across the 24-hour day in children and adolescents: physical activity, sedentary behavior, and sleep. Children (Basel) 2020;7:138. Crossref

21. National Bureau of Statistics of China, PRC Government. China Statistical Yearbook 2020. Available from: http://www.stats.gov.cn/tjsj/ndsj/2020/indexch.htm. Accessed 14 Sep 2021.

22. National Bureau of Statistics of China, PRC Government. China Statistical Yearbook 2013. Beijing, China: China State Statistical Press; 2013.

23. Seguí Mdel M, Cabrero García J, Crespo A, Verdú J, Ronda E. A reliable and valid questionnaire was developed to measure computer vision syndrome at the workplace. J Clin Epidemiol 2015;68:662-73. Crossref

24. Yi H, Zhang L, Ma X, et al. Poor vision among China’s rural primary school students: prevalence, correlates and consequences. China Econ Rev 2015;33:247-62. Crossref

25. United Nations International Children’s Emergency Fund. Don’t let children be the hidden victims of COVID-19 pandemic. Available from: https://www.unicef.org/press-releases/dont-let-children-be-hidden-victims-covid-19-pandemic. Accessed 6 Oct 2020.

26. Moon JH, Kim KW, Moon NJ. Smartphone use is a risk factor for pediatric dry eye disease according to region and age: a case control study. BMC Ophthalmol 2016;16:188. Crossref

27. Moon JH, Lee MY, Moon NJ. Association between video display terminal use and dry eye disease in school children. J Pediatr Ophthalmol Strabismus 2014;51:87-92. Crossref

28. Rechichi C, De Mojà G, Aragona P. Video game vision syndrome: a new clinical picture in children? J Pediatr Ophthalmol Strabismus 2017;54:346-55.Crossref

29. Mowatt L, Gordon C, Santosh AB, Jones T. Computer vision syndrome and ergonomic practices among undergraduate university students. Int J Clin Pract 2018;72:e13035. Crossref

30. Terasaki H, Yamashita T, Yoshihara N, Kii Y, Sakamoto T. Association of lifestyle and body structure to ocular axial length in Japanese elementary school children. BMC Ophthalmol 2017;17:123. Crossref

31. Fernández-Montero A, Olmo-Jimenez JM, Olmo N, et al. The impact of computer use in myopia progression: a cohort study in Spain. Prev Med 2015;71:67-71. Crossref

Artificial intelligence for detection of intracranial haemorrhage on head computed tomography scans: diagnostic accuracy in Hong Kong

Hong Kong Med J 2023 Apr;29(2):112–20

https://doi.org/10.12809/hkmj209053

ORIGINAL ARTICLE CME

Artificial intelligence for detection of intracranial haemorrhage on head computed tomography scans: diagnostic accuracy in Hong Kong

Jill M Abrigo, MD, FRCR; Ka-long Ko, MPhil; Qianyun Chen, MSc; Billy MH Lai, MB, BS, FHKAM (Radiology); Tom CY Cheung, MB ChB, FHKAM (Radiology); Winnie CW Chu, MB ChB, FHKAM (Radiology); Simon CH Yu, MB, BS, FHKAM (Radiology)

Department of Imaging and Interventional Radiology, Prince of Wales Hospital, The Chinese University of Hong Kong, Hong Kong SAR, China

Corresponding author: Dr Jill M Abrigo (jillabrigo@cuhk.edu.hk)

Full paper in PDF

Abstract

Introduction: The use of artificial intelligence (AI) to identify acute intracranial haemorrhage (ICH) on computed tomography (CT) scans may facilitate initial imaging interpretation in the accident and emergency department. However, AI model construction requires a large amount of annotated data for training, and validation with real-world data has been limited. We developed an algorithm using an open-access dataset of CT slices, then assessed its utility in clinical practice by validating its performance on CT scans from our institution.

Methods: Using a publicly available international dataset of >750 000 expert-labelled CT slices, we developed an AI model which determines ICH probability for each CT scan and nominates five potential ICH-positive CT slices for review. We validated the model using retrospective data from 1372 non-contrast head CT scans (84 [6.1%] with ICH) collected at our institution.

Results: The model achieved an area under the curve of 0.842 (95% confidence interval=0.791-0.894; P<0.001) for scan-based detection of ICH. A pre-specified probability threshold of ≥50% for the presence of ICH yielded 78.6% accuracy, 73% sensitivity, 79% specificity, 18.6% positive predictive value, and 97.8% negative predictive value. There were 62 true-positive scans and 22 false-negative scans, which could be reduced to six false-negative scans by manual review of model-nominated CT slices.

Conclusions: Our model exhibited good accuracy in the CT scan–based detection of ICH, considering the low prevalence of ICH in Hong Kong. Model refinement to allow direct localisation of ICH will facilitate the use of AI solutions in clinical practice.

New knowledge added by this study

A deep learning–based artificial intelligence model trained on an international dataset of computed tomography (CT) slices exhibited good accuracy in the detection of intracranial haemorrhage (ICH) on CT scans in Hong Kong.
Considering the 6% prevalence of ICH in our institution, and using a pre-specified probability threshold of ≥50%, the model detected 74% of ICH-positive scans; this outcome improved to 93% via manual review of model-nominated images.

Implications for clinical practice or policy

Considering the expected clinical applications, model refinement is needed to improve diagnostic performance prior to additional tests in a clinical setting.
Our model may facilitate assessment of CT scans by physicians with different degrees of experience in ICH detection, an important aspect of real-world clinical practice.

[Abstract in Chinese]

Introduction

Head computed tomography (CT) scans constitute the main imaging investigation during the evaluation of trauma and stroke; they are also important in the initial work-up of headache and other non-specific neurological complaints. In Prince of Wales Hospital of Hong Kong alone, >25 000 head CT scans were performed in 2019 during the clinical management of patients who presented to the Accident and Emergency Department. Computed tomography scans are composed of multiple cross-sectional images (ie, slices), which may be challenging to interpret. Typically, these scans are initially reviewed by frontline physicians prior to assessment by radiologists, and delays during the review process can be substantial. Thus, the timely recognition of an acute finding, such as intracranial haemorrhage (ICH), is limited by the competence and availability of frontline physicians.

The presence and location or type of ICH impacts the next clinical step, which can be further imaging investigations, medical management, or surgical intervention.1 Furthermore, a confirmation of ICH absence can also be useful in clinical management. For example, it can facilitate safe discharge from the hospital when appropriate; in patients with acute stroke, the absence of ICH is an important exclusion criterion that influences treatment selection.2

The use of artificial intelligence (AI) for ICH detection is a topic with global relevance considering its diagnostic impact and ability to optimise workflow, both of which have high practical value.3 4 In the accident and emergency department, AI can facilitate ICH detection in head CT scans during times when a radiologist is unavailable. Although there have been multiple reports of deep learning methods with high accuracy in the detection of ICH, the models in those reports were developed using in-house labelled training datasets and validated using a limited number of cases.3 5 6 7 8 Recently, the Radiological Society of North America (RSNA) publicly released >25 000 multi-centre head CT scans with slices that have been labelled with or without ICH by experts.9 Here, we developed a model using this RSNA dataset, then validated its performance on CT scans from our institution to determine its potential for clinical application in Hong Kong.

Methods

Ethical considerations

This study was approved by the Joint Chinese University of Hong Kong—New Territories East Cluster Clinical Research Ethics Committee (Ref No.: 2020.061). The model was developed from a publicly available dataset and validated on retrospectively acquired data from our institution. The requirement for patient consent was waived by the Committee given the retrospective design of the study and anonymisation of all CT scans prior to use.

The results of this diagnostic accuracy study are reported in accordance with the Standards for Reporting of Diagnostic Accuracy Studies guidelines.10

Public dataset: model development and internal validation

We acquired 25 312 head CT scans from four institutions in North and South America available in the RNSA open dataset,11 and were split into slices (each slice ≥5 mm thick), which were then randomly shuffled and annotated by 60 volunteer experts from the American Society of Neuroradiology. Each CT slice was labelled to indicate the presence and type of ICH. When present, ICH was classified according to its location, namely, intraparenchymal haemorrhage (IPH), subarachnoid haemorrhage (SAH), subdural haemorrhage (SDH), epidural haemorrhage (EDH), and intraventricular haemorrhage (IVH). The RSNA dataset comprised 752 807 CT slices, which were divided into a training subset (85%) and test subset (15%) for internal validation. Each subset consisted of approximately 86% negative ICH slices and 14% positive ICH slices, along with the following proportions of ICH subtypes: 4.8% IPH, 4.7%-4.8% SAH, 6.3% SDH, 0.4% EDH, and 3.4%-3.5% IVH.

The convolutional neural network (CNN) VGG (named after the Visual Geometry Group from the University of Oxford, United Kingdom) is an effective end-to-end algorithm for image detection.12 In this study, we adopted the VGG architecture with a customised output layer and loss function optimised for multi-label classification. To adjust for the low prevalence of ICH in the training set, each subtype’s logit outputs z_i were concatenated as independent channels after a sigmoid output layer:

The performance of the CNN model was evaluated by binary cross-entropy loss and Sørensen–Dice loss13:

The loss functions were linearly combined with weighted values to produce the multi-label classification loss:

Where w_i denotes the class prevalence weight, and α and β denote respective loss mix ratios. For simplicity, w_i=1∕(n-1) for all subtype classes and w_i=1 for ‘ANY’ was treated as an independent ICH class.

The model was trained with software written in our laboratory using the end-to-end open-source machine learning platform TensorFlow on an Nvidia Titan Xp graphics processing unit.

During internal validation (ie, slice-level performance for the detection of any type of ICH), the model achieved an area under the curve (AUC) of 0.912 (95% confidence interval [CI]=0.909-0.915) with sensitivity and specificity of 85% and 80%, respectively. Additionally, for the detection of specific types of ICH, the following AUC (95% CI) and sensitivity/specificity values were obtained: 0.860 (0.853-0.867) and 77%/88% for IPH, 0.835 (0.829-0.842) and 75%/82% for SAH, 0.850 (0.845-0.855) and 74%/83% for SDH, 0.813 (0.790-0.836) and 72%/80% for EDH, and 0.870 (0.861-0.879) and 79%/89% for IVH.

Prince of Wales Hospital dataset: external validation

The consecutive head CT scans of patients aged ≥18 years who underwent initial brain CT scans in the Accident and Emergency Department of Prince of Wales Hospital from 1 to 31 July 2019 were included, thereby simulating the point prevalence of ICH.

Head CT scans were acquired on a 64-slice CT scanner. Data analyses were conducted using reformatted 5-mm-thick slices, which can be accessed and viewed by physicians at all hospital workstations. DICOM (Digital Imaging and Communications in Medicine) images were de-identified prior to data analyses. The large volume of data was explored through the identification of relevant CT data using an automated program which selected scans with specific DICOM tags. Computed tomography scans performed for follow-up purposes or after recent intracranial surgery, as well as scans without radiologist reports, were excluded from the analysis.

We reviewed the corresponding radiology reports to determine the presence and type of ICH (IPH, SAH, SDH, EDH, or IVH). The CT scans were assessed by radiologists or senior radiology trainees; the corresponding reports were regarded as scan-level ground truth labels for analysis, consistent with their use as clinical reference standards in Hong Kong. Considering its rarity, EDH was grouped with and labelled as SDH, which has a similar appearance on CT. For scans with false-negative results, we performed post-hoc labelling of model-nominated CT slices. All scans were assessed prior to model construction; thus, the scan reports were established without knowledge of the AI results. Furthermore, all scans comprised the external validation dataset and constituted ‘unseen data’ for the model.

Statistical analysis

The diagnostic accuracies of the model for the detection of any type of ICH and each type of ICH were determined by calculation of the AUC with 95% CI, using DeLong et al’s method.14 To construct the confusion matrix during external validation, CT scans were classified as ICH-positive using a pre-specified probability threshold of ≥50%8; the corresponding sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were calculated. Additional probability thresholds were established to achieve 90% sensitivity and 90% specificity for the presence of any type of ICH. Statistical analysis was performed using R software (version 4.0.2; R Foundation for Statistical Computing, Vienna, Austria), and the threshold for statistical significance was set at P<0.05.

Results

Model output

Figure 1 shows an example of the model output. The model report includes an overall probability for the presence of ICH (labelled ‘A’ in Fig 1). Additionally, the model selects five representative CT image slices which are likely to contain ICH (one such slice is labelled ‘B’ in Fig 1), along with the probability of each ICH type in each representative slice (labelled ‘C’ in Fig 1). All scans were successfully analysed by the model.

Figure 1. Sample model output, highlighting three types of information provided by the model. A: intracranial haemorrhage (ICH) probability; B: model-nominated image with possible ICH and corresponding slice number in the computed tomography (CT) scan; C: probability of each ICH type for the corresponding CT slice

Prince of Wales Hospital data and model validation

The Standards for Reporting of Diagnostic Accuracy Studies diagram and corresponding confusion matrix are shown in Figure 2. In total, 1372 head CT scans (84 [6.1%] with ICH) were included in the analysis. The distribution of ICH types is summarised in the Table.

Figure 2. Standards for the Reporting of Diagnostic Accuracy flowchart for external validation of the model using computed tomography (CT) scans from Prince of Wales Hospital. The confusion matrix is shown below the flowchart

Table. Distribution of computed tomography (CT) scans without and with intracranial haemorrhage in the Prince of Wales Hospital dataset (n=1372)

Diagnostic performance of scan-based detection for any type of intracranial haemorrhage

The model achieved an AUC of 0.842 (95% CI=0.791-0.894; P<0.001) for the identification of any type of ICH. Using a probability threshold of ≥50% for the presence of ICH, the accuracy, sensitivity, specificity, PPV, and NPV were 78.6%, 73%, 79%, 18.6%, and 97.8%, respectively. In total, 62 scans were true positive, 22 were false negative, 1017 were true negative, and 271 were false positive (Fig 2).

Among the 62 true-positive scans, the model output in two cases did not contain ICH-positive CT slices: 6-mm IPH in the pons (n=1) and trace SAH in a patient with multiple metastatic tumours (n=1). Figure 3 shows selected cases of model-nominated CT slices with subtle ICH.

Figure 3. Representative computed tomography slices from model outputs for selected true-positive scans showing small or subtle intracranial haemorrhage. Arrowheads have been added to indicate intracranial haemorrhage. (a) Haemorrhage within a cystic tumour; (b, d, and e) subarachnoid haemorrhage; (c) intraparenchymal haemorrhage; (f) subdural haemorrhage

Among the 22 false-negative scans, 19 had one type of ICH (6 IPH, 7 SAH, 5 SDH, and 1 IVH), two had two types of ICH (1 IPH+SAH and 1 SAH+SDH), and one had three types of ICH (IPH+SAH+IVH). In 16 scans, the model selected at least one ICH-positive CT slice which allowed correct reclassification (Fig 4). The remaining six scans with undetected ICH (Fig 5) comprised small midbrain IPH (n=1), trace SAH (n=3), and thin SDH/EDH (n=2). One of the three cases of undetected trace SAH was visualised on thin CT slices but not on thick CT slices.

Figure 4. Representative computed tomography slices from model outputs for false-negative scans showing intracranial haemorrhage. Arrowheads have been added to indicate intracranial haemorrhage. (a) Intraventricular haemorrhage; (b, c, f, g, h, and i) subarachnoid haemorrhage; (d, e, l, m, n, o, and p) intraparenchymal haemorrhage; (j and k) subdural haemorrhage

Figure 5. False-negative computed tomography scans with undetected intracranial haemorrhage. Arrowheads have been added to indicate intracranial haemorrhage. (a-e) Representative images of intracranial haemorrhage in thick computed tomography slices [(a) intraparenchymal haemorrhage; (b and d) subdural haemorrhage; (c and e) subarachnoid haemorrhage]. (f) Trace subarachnoid haemorrhage that was visible in reformatted coronal thin computed tomography slices but not thick computed tomography slices

A probability threshold of 20.4% yielded a sensitivity of 90% (40% specificity, 9% PPV, and 98.3% NPV), whereas a threshold of 65.7% yielded a specificity of 90% (64% sensitivity, 30% PPV, and 97.4% NPV), for the detection of ICH.

Diagnostic performance of scan-based detection for each type of intracranial haemorrhage

At a probability threshold of ≥50%, the following AUC (95% CI) and corresponding sensitivity/specificity were obtained for each type of ICH: 0.930 (0.892-0.968) and 4%/100% for IPH, 0.766 (0.684-0.849) and 12%/96% for SAH, 0.865 (0.783-0.947) and 75%/90% for SDH/EDH, and 0.935 (0.852-1.000) and 85%/93% for IVH.

Discussion

In this study, we used a large international training dataset to construct a model for ICH detection, then conducted external validation using data from Hong Kong. To overcome the discrepancy between the training dataset (composed of CT slices) and the validation dataset (composed of CT scans), and considering our goal of clinical application, we designed a model that iteratively conducts assessments at the slice level to generate an overall probability at the scan level, then nominates the slices with the highest ICH probability for clinician evaluation. Furthermore, we performed validation using a point-prevalence approach to determine the diagnostic performance of the model in a real-world setting. Considering the 6% prevalence of ICH in our institution, and using a pre-specified probability threshold of ≥50%, the model detected 74% of ICH-positive scans; this outcome improved to 93% via manual review of model-nominated images.

Artificial intelligence for intracranial haemorrhage detection: research and reality

Multiple studies have successfully used AI for ICH detection via deep learning methods, typically involving variants of CNNs. For example, Arbabshirani et al5 (deep CNN, >37 000 training CT scans) reported an AUC of 0.846 on 342 CT scans; Chang et al4 (two-dimensional/three-dimensional CNN, 10 159 training CT scans) reported an AUC of 0.983 on 862 prospectively collected CT scans. Furthermore, Chilamkurthy et al3 (CNN, >290 000 training CT scans) reported an AUC of 0.94 on 491 CT scans; Lee et al7 (four deep CNNs, 904 training CT scans) reported an AUC of 0.96 on 214 CT scans. Finally, Ye et al8 (three-dimensional joint CNN-recurrent neural network, 2537 training CT scans) reported an AUC of 1.0 on 299 CT scans; Kuo et al6 (patch-based fully CNN, 4396 training CT scans) reported an AUC of 0.991 on 200 CT scans. Although these results demonstrate the high diagnostic performance that can be achieved using deep learning methods for ICH detection, the studies were conducted using in-house training datasets, which are laborious to produce and limit subsequent clinical applications. Moreover, the results may not be directly applicable to clinical practice, considering the limited number (generally <500) of CT scans during validation, as well as the effect of prevalence on sensitivity and specificity. Yune et al15 demonstrated this problem with a deep learning model that had an AUC of 0.993 on selected cases, which decreased to 0.834 when validated on CT scans collected over a 3-month period; notably, this is comparable with the AUC of our model. Thus, model performance in a real-world setting can reduce the risk of bias and serve as a better indicator of clinical relevance.16

Artificial intelligence for intracranial haemorrhage detection: our approach

The development of an AI model is the first step in a long process of clinical translation. In this study, we aimed to construct an algorithm that was reasonably comparable with radiologist performance, prior to further tests in a clinical setting. We recognise that our model is not an end-product; it constitutes an initial exploration of the potential for an international dataset–derived algorithm to be implemented in our institution. To avoid problems associated with the lack of an annotated dataset from Hong Kong, we utilised a dataset labelled by international experts, which is the most extensive open-access dataset currently available. However, the model achieved limited diagnostic accuracy, mainly because of type 1 error (ie, identification of false positives). The training dataset was composed of CT slices, whereas the model functioned at the CT scan level, iteratively assessing all slices to identify slices with highest ICH probability. If any slice identified in a single scan is considered positive, the model reports the CT scan as ‘ICH-positive’. Thus, any detection of false positives at the slice level will lead to amplification of the false-positive rate at the scan level. This strategy resulted in a low PPV (~19%) and a high NPV (~98%). To reduce the detection of false positives, we included a CT slice nomination feature in the model, which highlights CT slices with the highest probability of ICH. This facilitates manual review and reduces the black-box nature of the model.

Potential implications of artificial intelligence–detected intracranial haemorrhage in clinical practice

During validation, the model was tested using an ICH point–prevalence approach to elucidate the potential clinical implications of the classification outcomes. With respect to true positives, most ICH-positive scans were detected; most of these scans had large areas of ICH, which presumably could be easily identified by non-radiologists. However, in six cases, the model correctly nominated CT slices with small areas of ICH. In two cases, the nominated images did not have ICH, which could potentially have led to incorrect reclassification of the scan as a false positive.

Furthermore, there were many false positives. Such results may reduce physician confidence despite the correct interpretation of an ICH-negative scan; they may lead to overdiagnosis (with prolonged hospitalisation) or further investigations, such as a follow-up CT scan that involves additional radiation exposure.

With respect to false negatives, the model output includes a secondary mechanism of image review that allowed correct reclassification of 16 scans, increasing the rate of ICH detection from 74% to 93%. In five cases, ICH was conspicuous on the nominated images; in 11 cases, the nominated images displayed subtle ICH. In cases of subtle ICH, it is possible to overlook the trace amount of ICH on the nominated CT slice. The same problem may affect true-positive scans, which may be misclassified as false positives unless subtle ICH is recognised in the nominated image. Unfortunately, the model-generated probability of each type of ICH in each selected image did not facilitate the localisation of ICH.

Based on our primary clinical motivation to develop this model, we focused on CT scans with reformatted thick CT slices that can be viewed in all hospital workstations by non-radiologists. In practice, radiologists use dedicated imaging workstations to view sub-millimetre thin CT slices with greater sensitivity, which can display smaller or subtler pathologies. Thus, there is limited capacity for ICH detection in thick CT slices; this was highlighted in a case of trauma-related trace SAH, which was visible on thin CT slices but not thick CT slices. Subarachnoid haemorrhage is reportedly the most difficult type of ICH to interpret.17 In practice, a patient with a very small amount of isolated traumatic SAH would likely receive conservative treatment, and the pathology could reasonably await detection via radiologist assessment.

Limitations

This study had some limitations. First, diagnostic accuracy would have been more comprehensively assessed using a larger number of CT scans or a longer point prevalence; however, we limited the assessment to CT scans collected over a 1-month period, considering the preliminary stage of model development. Second, the CT scans were assessed by radiologists and senior radiology trainees who may have different degrees of experience in ICH detection17; importantly, this limitation reflects the real-world setting where model deployment is intended. Finally, the model was specifically trained for the detection of ICH; it was not trained for the detection of other clinically significant non-ICH findings (eg, non-haemorrhagic tumours, hydrocephalus, or mass effect). The detection of these other pathologies will require dedicated models with customised training datasets.

Conclusion

In this study, we used a CT slice–based dataset to develop an algorithm for CT scan–based ICH detection; we validated the model using our institutional data with a point-prevalence approach, yielding insights regarding its utility in real-world clinical practice. Although the model demonstrated good accuracy, its diagnostic performance is currently limited to the intended clinical application. However, our results support further development of the model to improve its accuracy and incorporate a mechanism that can facilitate visual confirmation of ICH location. These modifications would enhance the interpretability of the deep learning model and would be useful for further evaluation of clinical applications.

Author contributions

Concept or design: JM Abrigo, KL Ko, WCW Chu, SCH Yu.
Acquisition of data: JM Abrigo, Q Chen, WCW Chu, BMH Lai, TCY Cheung.
Analysis or interpretation of data: All authors.
Drafting of the manuscript: JM Abrigo, KL Ko, Q Chen.
Critical revision of the manuscript for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

All authors have disclosed no conflicts of interest.

Acknowledgement

We thank our department colleagues Mr Kevin Lo for anonymising and downloading Digital Imaging and Communications in Medicine data, and we thank Mr Kevin Leung for preparing figures for this manuscript.

Funding/support

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Ethics approval

This research was approved by the Joint Chinese University of Hong Kong—New Territories East Cluster Clinical Research Ethics Committee (Ref No.: 2020.061). The requirement for patient consent was waived by the Committee given the retrospective design of the study and anonymisation of all computed tomography scans prior to use.

References

1. Caceres JA, Goldstein JN. Intracranial hemorrhage. Emerg Med Clin North Am 2012;30:771-94. Crossref

2. Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the early management of patients with acute ischemic stroke: 2019 update to the 2018 guidelines for the early management of acute ischemic stroke: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2019;50:e344-418. Crossref

3. Chilamkurthy S, Ghosh R, Tanamala S, et al. Deep learning algorithms for detection of critical findings in head CT scans: a retrospective study. Lancet 2018;392:2388-96. Crossref

4. Chang PD, Kuoy E, Grinband J, et al. Hybrid 3D/2D convolutional neural network for hemorrhage evaluation on head CT. AJNR Am J Neuroradiol 2018;39:1609-16. Crossref

5. Arbabshirani MR, Fornwalt BK, Mongelluzzo GJ, et al. Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration. NPJ Digit Med 2018;1:9. Crossref

6. Kuo W, Häne C, Mukherjee P, Malik J, Yuh EL. Expert-level detection of acute intracranial hemorrhage on head computed tomography using deep learning. Proc Natl Acad Sci U S A 2019;116:22737-45. Crossref

7. Lee H, Yune S, Mansouri M, et al. An explainable deep-learning algorithm for the detection of acute intracranial haemorrhage from small datasets. Nat Biomed Eng 2019;3:173-82. Crossref

8. Ye H, Gao F, Yin Y, et al. Precise diagnosis of intracranial hemorrhage and subtypes using a three-dimensional joint convolutional and recurrent neural network. Eur Radiol 2019;29:6191-201. Crossref

9. Flanders AE, Prevedello LM, Shih G, et al. Construction of a machine learning dataset through collaboration: The RSNA 2019 Brain CT Hemorrhage Challenge. Radiol Artif Intell 2020;2:e190211. Crossref

10. Bossuyt PM, Reitsma JB, Bruns DE, et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. Radiology 2015;277:826-32. Crossref

11. Radiological Society of North America. RSNA intracranial hemorrhage detection: identify acute intracranial hemorrhage and its subtypes. Available from: https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/data. Accessed 28 Mar 2023. Crossref

12. Zhang X, Zou J, He K, Sun J. Accelerating very deep convolutional networks for classification and detection. IEEE Trans Pattern Anal Mach Intell 2016;38:1943-55. Crossref

13. Milletari F, Navab N, Ahmadi SA. V-Net: fully convolutional neural networks for volumetric medical image segmentation. 2016 Fourth International Conference on 3D Vision (3DV). USA (CA): Stanford; 2016: 565-71. Crossref

14. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 1988;44:837-45. Crossref

15. Yune S, Lee H, Pomerantz S, et al. Real-world performance of deep-learning–based automated detection system for intracranial hemorrhage. Radiological Society of North America (RSNA) 104th Scientific Assembly and Annual Meeting. McCormick Place, Chicago (IL); 2018.

16. Nagendran M, Chen Y, Lovejoy CA, et al. Artificial intelligence versus clinicians: systematic review of design, reporting standards, and claims of deep learning studies. BMJ 2020;368:m689. Crossref

17. Strub WM, Leach JL, Tomsick T, Vagal A. Overnight preliminary head CT interpretations provided by residents: locations of misidentified intracranial hemorrhage. AJNR Am J Neuroradiol 2007;28:1679-82. Crossref

Evaluation of contemporary olanzapine- and netupitant/palonosetron-containing antiemetic regimens for chemotherapy-induced nausea and vomiting

Hong Kong Med J 2023 Feb;29(1):49–56

https://doi.org/10.12809/hkmj209182

ORIGINAL ARTICLE

Evaluation of contemporary olanzapine- and netupitant/palonosetron-containing antiemetic regimens for chemotherapy-induced nausea and vomiting

Christopher CH Yip¹; L Li, MB, BS, FHKAM (Medicine)¹; Thomas KH Lau, MB, BS, FHKAM (Medicine)¹; Vicky TC Chan, MB, BS, FHKAM (Medicine)¹; Carol CH Kwok, MB, BS, FHKAM (Radiology)²; Joyce JS Suen, MB, BS, FHKAM (Radiology)²; Frankie KF Mo, PhD¹; Winnie Yeo, MB, BS, FHKAM (Medicine)¹

¹ Department of Clinical Oncology, Prince of Wales Hospital, Hong Kong

² Department of Clinical Oncology, Princess Margaret Hospital, Hong Kong

Corresponding author: Prof Winnie Yeo (winnieyeo@cuhk.edu.hk)

Full paper in PDF

Abstract

Introduction: This post-hoc analysis retrospectively assessed data from two recent studies of antiemetic regimens for chemotherapy-induced nausea and vomiting (CINV). The primary objective was to compare olanzapine-based versus netupitant/palonosetron (NEPA)-based regimens in terms of controlling CINV during cycle 1 of doxorubicin/cyclophosphamide (AC) chemotherapy; secondary objectives were to assess quality of life (QOL) and emesis outcomes over four cycles of AC.

Methods: This study included 120 Chinese patients with early-stage breast cancer who were receiving AC; 60 patients received the olanzapine-based antiemetic regimen, whereas 60 patients received the NEPA-based antiemetic regimen. The olanzapine-based regimen comprised aprepitant, ondansetron, dexamethasone, and olanzapine; the NEPA-based regimen comprised NEPA and dexamethasone. Patient outcomes were compared in terms of emesis control and QOL.

Results: During cycle 1 of AC, the olanzapine group exhibited a higher rate of ‘no use of rescue therapy’ in the acute phase (olanzapine vs NEPA: 96.7% vs 85.0%, P=0.0225). No parameters differed between groups in the delayed phase. The olanzapine group had significantly higher rates of ‘no use of rescue therapy’ (91.7% vs 76.7%, P=0.0244) and ‘no significant nausea’ (91.7% vs 78.3%, P=0.0408) in the overall phase. There were no differences in QOL between groups. Multiple cycle assessment revealed that the NEPA group had higher rates of total control in the acute phase (cycles 2 and 4) and the overall phase (cycles 3 and 4).

Conclusion: These results do not conclusively support the superiority of either regimen for patients with breast cancer who are receiving AC.

New knowledge added by this study

The olanzapine-based regimen (aprepitant, ondansetron, dexamethasone, and olanzapine) and the NEPA-based regimen (netupitant, palonosetron, and dexamethasone) demonstrated similar efficacies in terms of controlling chemotherapy-induced nausea and vomiting among patients with early-stage breast cancer.
Quality of life did not significantly differ between patients receiving the olanzapine-based regimen and patients receiving the NEPA-based regimen.

Implications for clinical practice or policy

The available data suggest that olanzapine-containing antiemetic regimens can be used without aprepitant, particularly when seeking to reduce medical expenses.
Antiemetic efficacy may potentially be enhanced if NEPA is administered in combination with dexamethasone and olanzapine as a four-drug antiemetic regimen.

[Abstract in Chinese]

Introduction

Patients with breast cancer receiving (neo)adjuvant treatment exhibit improved prognoses.1 However, chemotherapy regimens for breast cancer are associated with various degrees of chemotherapy-induced nausea and vomiting (CINV). The doxorubicin/cyclophosphamide (AC) regimen is one of the most frequently prescribed regimens for patients with breast cancer who are receiving (neo) adjuvant chemotherapy; AC is among the highly emetogenic chemotherapies with ≥90% risk of nausea and vomiting.

In situations where a neurokinin-1 receptor antagonist (NK1RA) is accessible, most current guidelines for AC(-like) chemotherapy recommend the use of a prophylactic triplet antiemetic regimen that consists of an NK1RA, a 5-hydroxytryptamine type-3 receptor antagonist (5HT3RA), and a corticosteroid, with or without olanzapine.2 3 4 In addition to earlier NK1RAs (eg, aprepitant, fosaprepitant, and rolapitant), netupitant/palonosetron (NEPA) [Akynzeo], which is a combination of an NK1RA (netupitant 100 mg) and a second-generation 5HT3RA (palonosetron 0.5 mg), has been available in the past decade. Although palonosetron constitutes a more potent 5HT3RA,5 it also has synergistic interactions with netupitant that include interference with 5HT3 receptor cross-talk and enhancement of the netupitant-mediated effect on NK1 receptor internalisation.6 7

In a recent systematic review and meta-analysis, Yokoe et al8 compared different antiemetic regimens to assess their control of CINV in patients receiving highly emetogenic chemotherapy regimens. The authors arbitrarily defined the ‘conventional’ regimen as a three-drug regimen that contained dexamethasone, a first- or second-generation secondgeneration 5HT3RA, and an earlier NK1RA compound (aprepitant, fosaprepitant, or rolapitant); they defined ‘new’ regimens as regimens that contained NEPA or olanzapine. The results indicated that, compared with conventional regimens, new regimens containing NEPA were more effective in terms of producing a complete response (ie, absence of vomiting and no use of rescue therapy). Additionally, Yokoe et al8 showed that olanzapine-containing regimens were most effective in terms of producing a complete response, particularly when olanzapine was added to a triplet regimen of an NK1RA, a 5HT3RA, and dexamethasone. These findings were supported by the results of a prospective randomised study published in 2020, which directly compared an olanzapine-containing four-drug regimen with a standard triplet antiemetic regimen (consisting of aprepitant, ondansetron, and dexamethasone) for the prevention of CINV in patients receiving AC chemotherapy.9

Here, we conducted a post-hoc analysis through retrospective assessment of individual patient data from two previously reported prospective antiemetic studies that involved Chinese patients with breast cancer.9 10 We hypothesised that a four-drug antiemetic regimen (consisting of an NK1RA, a 5HT3RA, dexamethasone, and olanzapine) would remain superior to a three-drug regimen (consisting of an NK1RA, a 5HT3RA, and dexamethasone) that included NEPA as a combination NK1RA and 5HT3RA agent. The primary objective was to compare the efficacies of olanzapine- and NEPA-containing antiemetic regimens in terms of controlling CINV during the first cycle of AC. The secondary objectives were: (1) to assess quality of life (QOL) outcomes in patients receiving these treatments during the first cycle of AC, and (2) to assess emesis control outcomes in patients receiving these treatments over multiple cycles of AC.

Methods

Patients

This study constituted a post-hoc analysis of data from two recently reported prospective studies. The first prospective study investigated emesis outcomes in patients with breast cancer who received a standard triplet antiemetic regimen (ie, aprepitant, ondansetron, and dexamethasone) with or without olanzapine9; after the first study, a second prospective study was conducted to assess the antiemetic efficacy of NEPA and dexamethasone.10 These studies were conducted with institutional ethics approval and were registered at ClinicalTrials.gov (NCT03386617 and NCT03079219, respectively). For the post-hoc analysis, data were extracted from the first study regarding patients who received an olanzapine plus aprepitant-containing four-drug antiemetic regimen; data were extracted from the second study regarding patients who received NEPA and dexamethasone. These patients were categorised into the ‘olanzapine’ and ‘NEPA’ groups, respectively.

Inclusion criteria were similar for the two studies. Specifically, patients were eligible if they were women of Chinese ethnicity, were aged >18 years, had early-stage breast cancer, and planned to receive a regimen of (neo)adjuvant AC. All study participants were required to read, understand, and complete study questionnaires and diaries in Chinese. Exclusion criteria included abnormal bone marrow, renal, or hepatic functions; receipt or planned receipt of radiation therapy to the abdomen or pelvis within 7 days prior to initial administration of study treatment; presence of grade 2 to 3 nausea, as defined by the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0,11 or vomiting within 24 hours prior to initial administration of the study treatment; presence of an active infection or any uncontrolled disease; a history of illicit drugs, including marijuana or alcohol abuse; mental incapacitation; and/or presence of a clinically significant emotional or psychiatric disorder. Written consent was provided by eligible patients prior to enrolment in the studies.

Study treatment

Patients in the olanzapine group received olanzapine 10 mg, aprepitant 125 mg, dexamethasone 12 mg, and ondansetron 8 mg before chemotherapy on day 1; they also received ondansetron 8 mg 8 hours after chemotherapy. Subsequently, they received aprepitant 80 mg daily on days 2-3 and olanzapine 10 mg daily on days 2-5.

Patients in the NEPA group received one capsule of NEPA (netupitant 300 mg/palonosetron 0.50 mg) with dexamethasone 12 mg before chemotherapy on day 1. Subsequently, they received dexamethasone 4 mg twice per day on days 2-3.

Study assessments

At the initiation of chemotherapy on day 1, individual patients were provided a diary to record the date and time of their symptoms of vomiting and nausea for 120 hours after the AC infusion; the use of any rescue medication was also recorded. On days 2-6, patients rated their symptoms of nausea for the previous 24 hours using a visual analogue scale (in which 0 mm implied no nausea, whereas 100 mm implied nausea that was ‘as bad as it could be’). Additionally, on day 1 (before infusion of AC) and day 6 (after completion of the diary), patients completed the Functional Living Index-Emesis (FLIE) questionnaire. A research nurse/assistant called individual patients on days 2-6 to remind them to take the study medications, complete the patient diary, and complete the FLIE questionnaire.

Assessment of efficacy and safety

Antiemetic efficacy was measured across three overlapping time periods. The ‘acute’ phase comprised 0 to 24 hours from the infusion of AC; the ‘delayed’ phase comprised 24 to 120 hours from the infusion of AC; the ‘overall’ phase comprised 0 to 120 hours from the infusion of AC.

Variables used to assess antiemetic efficacy were ‘complete response’, ‘no vomiting’, ‘no significant nausea’, ‘no nausea’, ‘no use of rescue therapy’, ‘complete protection’, and ‘total control’; definitions of these variables are provided in Table 1. The proportions of patients who exhibited these variables were recorded separately. Additionally, the ‘time to first vomiting’ in cycle 1 was determined using information recorded in patients’ diaries.

Table 1. Definitions of variables used to assess antiemetic efficacy

Quality of life was evaluated using the Chinese version of self-reported FLIE questionnaires from individual patients.12 The FLIE questionnaire consists of a nausea domain (9 items) and a vomiting domain (9 items). All scores were transformed to ensure that higher scores indicated worse impact on QOL.

Statistical analyses

A modified intention-to-treat approach was used for all efficacy analyses; specifically, analyses included patients who had received chemotherapy, had completed the study procedures from 0 to 120 hours in cycle 1 of AC, and had no major protocol violations.

To achieve the primary objective of this study, the efficacies of the two antiemetic regimens were based on the proportions (including 95% confidence intervals) of patients who achieved complete response during the acute, delayed, and overall phases after AC infusion in cycle 1. Other parameters compared in cycle 1 of AC were ‘time to first vomiting’, ‘no vomiting’, ‘no significant nausea’, no nausea’, ‘no use of rescue therapy’, ‘complete protection’, and ‘total control’.

To achieve the secondary objectives, QOL was compared between the two antiemetic regimens based on assessments of the nausea domain, vomiting domain, and total score (sum of nausea and vomiting domains) of the FLIE questionnaire during cycle 1 of AC. Emesis control over multiple cycles was compared between the two antiemetic regimens by assessing the proportions (including 95% confidence intervals) of patients who achieved ‘complete response’, ‘complete protection’, and ‘total control’ in the acute, delayed, and overall phases.

Comparisons between the two antiemetic regimens were made using the Wilcoxon rank-sum test for continuous data and Pearson’s Chi squared test for dichotomous data. Two-sided P values <0.05 were considered statistically significant. The SAS Software version 9.4 (SAS Institute, Cary [NC], United States) was used for analyses.

Results

Patient characteristics

Data from 120 patients were included in this study; 60 patients each were enrolled in the NEPA and olanzapine groups. Fifty-six patients (93.3%) in the olanzapine group completed all four cycles of AC, whereas 60 patients (100%) in the NEPA group completed all four cycles of AC.

Patient characteristics, including characteristics that could potentially affect CINV, are shown in Table 2. The olanzapine and NEPA groups had very similar patient characteristics, with median ages of 54.5 and 56 years, respectively. Nearly two-thirds of patients in each group had Stage II breast cancer (63.3% and 66.7%, respectively). The percentage of patients with a history of motion sickness was higher in the NEPA group (35%) than in the olanzapine group (16.7%). Furthermore, 30% of patients in the NEPA group and 20% of patients in the olanzapine group received AC as neoadjuvant treatment.

Table 2. Baseline characteristics of Chinese patients with early-stage breast cancer included in this analysis

Efficacy assessment

Antiemetic efficacies during cycle 1 of AC in the olanzapine and NEPA groups are shown in Table 3. Complete response rates in acute, delayed, and overall phases in cycle 1 did not differ between groups. In the acute phase, the olanzapine group exhibited a higher rate of ‘no use of rescue therapy’ (olanzapine vs NEPA: 96.7% vs 85.0%, P=0.0225). No parameters differed between groups in the delayed phase. In the overall phase, the olanzapine group exhibited significantly higher rates of ‘no use of rescue therapy’ (91.7% vs 76.7%, P=0.0244) and ‘no significant nausea’ (91.7% vs 78.3%, P=0.0408).

Table 3. Comparison of antiemetic efficacy during cycle 1 of doxorubicin/cyclophosphamide between olanzapine and netupitant/palonosetron groups

The median time to first vomiting was not reached in either group (P=0.3902). Quality of life results during cycle 1 of AC in the olanzapine and NEPA groups, determined using the FLIE questionnaire, are shown in the Figure. There were no significant differences in the nausea domain, vomiting domain, or total score of the FLIE questionnaire between the two groups.

Figure. Comparison of quality of life (assessed using Functional Living Index-Emesis questionnaire) throughout cycle 1 of doxorubicin/cyclophosphamide between olanzapine and netupitant/palonosetron groups

Antiemetic efficacies over multiple cycles of AC in the olanzapine and NEPA groups are shown in Table 4. In the acute phase, the NEPA group exhibited significantly higher rates of total control in cycle 2 (olanzapine vs NEPA: 59.6% vs 81.7%, P=0.0087) and cycle 4 (63.2% vs 86.7%, P=0.0032). No parameters differed between groups in the delayed phase. In the overall phase, the NEPA group exhibited significantly higher rates of total control in cycle 3 (55.4% vs 73.3%, P=0.0430) and cycle 4 (54.4% vs 75.0%, P=0.0195).

Table 4. Comparison of complete response, complete protection, and total control over multiple cycles between olanzapine and netupitant/palonosetron groups

Discussion

Chemotherapy-induced nausea and vomiting is a frustrating adverse effect for patients receiving anticancer treatment.13 The administration of optimal antiemetic prophylaxis can help to maintain QOL, while potentially improving patient compliance in terms of completing planned therapies. In current antiemetic prophylaxis guidelines, the European Society of Medical Oncology/Multinational Association of Supportive Care in Cancer, the American Society of Clinical Oncology, and the United States National Comprehensive Cancer Network offer several options regarding antiemetic regimens for patients receiving AC(-like) chemotherapy. These options mainly involve the combination of a 5HT3RA and corticosteroids, with or without an NK1RA and olanzapine.2 3 4 In particular, the incorporation of olanzapine, an antipsychotic drug with antagonistic effects on various receptors (eg, dopamine and serotonin receptors),14 is increasingly regarded as a component of antiemetic prophylaxis for patients receiving anticancer treatment.

In an attempt to identify the best antiemetic regimen, Yokoe et al8 conducted a meta-analysis of randomised trials that tested various antiemetic regimens. The results indicated that olanzapine-based regimens demonstrated the best efficacy. Specifically, olanzapine in combination with an NK1RA, a 5HT3RA, and dexamethasone exhibited the greatest efficacy; other olanzapine-containing regimens (consisting of a 5HT3RA and dexamethasone) were also superior to regimens that lacked olanzapine. Moreover, even in the presence of earlier NK1RAs (eg, aprepitant, fosaprepitant, or rolapitant), regimens lacking olanzapine remained inferior.

Similar to the findings with olanzapine, Yokoe et al8 reported that triplet antiemetics involving NEPA were superior to conventional NK1RAs (eg, aprepitant, fosaprepitant, or rolapitant). Furthermore, Zhang et al15 directly compared NEPA-based antiemetic regimens with aprepitant-based triplet regimens in a randomised study that involved 800 patients who underwent administration of a cisplatin-containing regimen. Their results revealed that patients receiving NEPA and dexamethasone exhibited similar control of CINV, compared with patients receiving aprepitant, granisetron, and dexamethasone; however, NEPA-treated patients had a significantly lower requirement for rescue therapy. Additionally, in a recent study focused on patients with breast cancer who were undergoing AC chemotherapy, patients who received NEPA and dexamethasone demonstrated significantly higher rates of complete response, complete protection, and total control with enhanced QOL, compared to historical controls who received aprepitant, ondansetron, and dexamethasone; these benefits persisted over multiple cycles of chemotherapy.10

To our knowledge, no study has directly compared olanzapine- and NEPA-containing regimens. Using an indirect comparison approach, the present study showed that the olanzapine-based regimen had higher rates of ‘no use of rescue therapy’ and ‘no significant nausea’ in cycle 1 of AC, compared to the NEPA-based regimen. In contrast, assessments in subsequent cycles revealed that the NEPA-based regimen led to higher rates of total control in the acute phase (cycles 2 and 4) and the overall phase (cycles 3 and 4). The lack of difference in QOL between the two groups of patients may be related to the difference in adverse-effect profiles of the antiemetics used. For instance, the continued use of dexamethasone on days 2-3 in the NEPA group may have affected QOL among those patients because of its effects on mood, insomnia, gastrointestinal symptoms, and metabolic profiles.16 Indeed, a recent meta-analysis showed that, among patients receiving AC or moderately emetogenic chemotherapy, 3 days of dexamethasone did not provide additional benefit compared to 1 day of the agent.17 However, olanzapine has been associated with sedation and somnolence.18 Thus, after the completion of a phase 2 trial in Japan that suggested olanzapine was more effective at 5 mg than at 10 mg,19 the same group of investigators conducted a phase 3 study in which they tested the addition of daily olanzapine 5 mg to an aprepitant-based three-drug regimen; the results showed that, even at a lower dose of olanzapine, the olanzapine-containing regimen remained more efficacious than the olanzapine-free regimen for patients receiving cisplatin.20 Other adverse effects have been reported. Our analysis of olanzapine in combination with aprepitant, ondansetron and dexamethasone revealed a significantly higher incidence of grade ≥2 neutropenia in the olanzapine arm than in the standard arm, although this altered incidence was not associated with a significant difference in neutropenic fever.9 A few cases of olanzapine-induced neutropenia have been reported21; additionally, a recent randomised antiemetic study showed that patients who received an olanzapine-containing regimen had a higher frequency of severe neutropenia (without an increased incidence of neutropenic fever).22 Although the underlying mechanism remains unknown, the results of the aforementioned Japanese study20 suggest that olanzapine 5 mg could reduce the incidence of neutropenia. In contrast, in our previous trial regarding a NEPA-based regimen, we found that patients in the NEPA arm had significantly lower incidences of grade ≥2 neutropenia and neutropenic fever, compared to historical controls who received an aprepitant-based regimen.9 10

This study had some potential limitations. First, dexamethasone was only used for 1 day in the olanzapine-based regimen, whereas it was administered for 3 days in the NEPA-based regimen; this difference may have influenced the findings. Second, the use of data from two separate studies may have affected the generalisability of the findings because of slight variations in patient characteristics; the lack of blinding in both studies also increased the potential for patient-related reporting biases. Nonetheless, the original studies were consecutively conducted during the period from 2017 to 2019; both the data from Chinese patients enrolled in a homogenous group with early-stage breast cancer who were receiving (neo)adjuvant AC chemotherapy and the present analysis were analysed based on individual patient data. These factors support the validity of our comparison approach.

Conclusion

In conclusion, the present findings do not conclusively support the superiority of either the olanzapine-based regimen or the NEPA-based regimen in terms of antiemetic efficacy or QOL among patients with breast cancer who are receiving AC. Our previous study demonstrated that aprepitant has a limited effect when used with a 5HT3RA and dexamethasone23; we also found that NEPA was superior to aprepitant.10 Overall, the available data suggest that olanzapine-containing antiemetic regimens can be used without aprepitant, particularly when seeking to reduce medical expenses. Moreover, the available data support the previous conclusion that, in parts of the world where socio-economic limitations restrict the availability of NK1RAs, the use of olanzapine combined with a 5HT3RA and dexamethasone may be an effective low-cost alternative antiemetic regimen.8 24 Antiemetic efficacy may be enhanced if NEPA is administered in combination with dexamethasone and olanzapine as a four-drug antiemetic regimen; however, the efficacy of an olanzapine plus NEPA regimen in terms of controlling CINV should be confirmed in a trial setting.

Author contributions

Concept or design: W Yeo.
Acquisition of data: FKF Mo, W Yeo.
Analysis or interpretation of data: W Yeo, CCH Yip, FKF Mo.
Drafting of the manuscript: W Yeo, CCH Yip.
Critical revision of the manuscript for important intellectual content: L Li, TKH Lau, VTC Chan, CCH Kwok, JJS Suen, FKF Mo.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

Conflicts of interest

W Yeo has been involved in the Chemotherapy-Induced Nausea and Vomiting (CINV) Network in Asia and has provided lectures on CINV at events organised by Mundipharma International Limited, which supported the design of the NEPA study10 analysed in this post-hoc analysis but had no role in the present comparative analysis, data collection and analysis, decision to publish, or preparation of the manuscript.

Acknowledgement

We thank Ms Dong KT Lai, Ms Elizabeth Pang, Ms Vivian Chan, and Ms Maggie Cheung of the Department of Clinical Oncology, The Chinese University of Hong Kong for their support in contributing to patient enrolment and study monitoring.

Declaration

Data from this study were presented at the European Society of Medical Oncology Asia Virtual Congress 2020 on 27 November 2020.

Funding/support

This study was supported by an education grant from Madam Diana Hon Fun Kong Donation for Cancer Research (Grant No.: CUHK Project Code 7104870). The Donation had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Ethics approval

The studies examined in this post-hoc analysis were approved by The Joint Chinese University of Hong Kong–New Territories East Cluster Institution Review Board of The Chinese University of Hong Kong and the Hong Kong Hospital Authority, and the Kowloon West Cluster Research Ethics Committee of the Hong Kong Hospital Authority (Ref No.: CREC 2016.013, CREC 2017.1609 and KW/FR-18-019[119-19]). All patient data in this study were anonymous and were based on the abovementioned reported studies. There was no additional work on retrieving patient records in this study.

References

1. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG); Peto R, Davies C, et al. Comparisons between different polychemotherapy regimens for early breast cancer: meta-analyses of long-term outcome among 100,000 women in 123 randomised trials. Lancet 2012;379:432-44. Crossref

2. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology, Antiemesis Version 1; 2019.

3. Herrstedt J, Roila F, Warr D, et al. 2016 updated MASCC/ESMO consensus recommendations: prevention of nausea and vomiting following high emetic risk chemotherapy. Support Care Cancer 2017;25:277-88. Crossref

4. Hesketh PJ, Kris MG, Basch E, et al. Antiemetics: American Society of Clinical Oncology Clinical Practice Guideline update. J Clin Oncol 2017;35:3240-61. Crossref

5. Popovic M, Warr DG, Deangelis C, et al. Efficacy and safety of palonosetron for the prophylaxis of chemotherapy-induced nausea and vomiting (CINV): a systematic review and meta-analysis of randomized controlled trials. Support Care Cancer 2014;22:1685-97. Crossref

6. Thomas AG, Stathis M, Rojas C, Slusher BS. Netupitant and palonosetron trigger NK1 receptor internalization in NG108-15 cells. Exp Brain Res 2014;232:2637-44. Crossref

7. Stathis M, Pietra C, Rojas C, Slusher BS. Inhibition of substance P-mediated responses in NG108-15 cells by netupitant and palonosetron exhibit synergistic effects. Eur J Pharmacol 2012;689:25-30. Crossref

8. Yokoe T, Hayashida T, Nagayama A, et al. Effectiveness of antiemetic regimens for highly emetogenic chemotherapy-induced nausea and vomiting: a systematic review and network meta-analysis. Oncologist 2019;24:e347-57. Crossref

9. Yeo W, Lau TK, Li L, et al. A randomized study of olanzapine-containing versus standard antiemetic regimens for the prevention of chemotherapy-induced nausea and vomiting in Chinese breast cancer patients. Breast 2020;50:30-8. Crossref

10. Yeo W, Lau TK, Kwok CC, et al. NEPA efficacy and tolerability during (neo)adjuvant breast cancer chemotherapy with cyclophosphamide and doxorubicin. BMJ Support Palliat Care 2022;12:e264-70.

11. National Cancer Institute. Cancer Therapy Evaluation Program. 2021. Available from: https://ctep.cancer.gov/protocoldevelopment/electronic_applications/ctc.htm#ctc_40. Accessed 3 Feb 2023.

12. Martin AR, Pearson JD, Cai B, Elmer M, Horgan K, Lindley C. Assessing the impact of chemotherapy-induced nausea and vomiting on patients’ daily lives: a modified version of the Functional Living Index-Emesis (FLIE) with 5-day recall. Support Care Cancer 2003;11:522-7. Crossref

13. Griffin AM, Butow PN, Coates AS, et al. On the receiving end. V: patient perceptions of the side effects of cancer chemotherapy in 1993. Ann Oncol 1996;7:189-95. Crossref

14. Bymaster FP, Nelson DL, DeLapp NW, et al. Antagonism by olanzapine of dopamine D1, serotonin2, muscarinic, histamine H1 and alpha 1-adrenergic receptors in vitro. Schizophr Res 1999;37:107-22. Crossref

15. Zhang L, Lu S, Feng J, et al. A randomized phase III study evaluating the efficacy of single-dose NEPA, a fixed antiemetic combination of netupitant and palonosetron, versus an aprepitant regimen for prevention of chemotherapy-induced nausea and vomiting (CINV) in patients receiving highly emetogenic chemotherapy (HEC). Ann Oncol 2018;29:452-8. Crossref

16. Roila F, Ruggeri B, Ballatori E, et al. Aprepitant versus metoclopramide, both combined with dexamethasone, for the prevention of cisplatin-induced delayed emesis: a randomized, double-blind study. Ann Oncol 2015;26:1248-53. Crossref

17. Okada Y, Oba K, Furukawa N, et al. One-day versus three-day dexamethasone in combination with palonosetron for the prevention of chemotherapy-induced nausea and vomiting: a systematic review and individual patient data-based meta-analysis. Oncologist 2019;24:1593-600. Crossref

18. Navari RM, Qin R, Ruddy KJ, et al. Olanzapine for the prevention of chemotherapy-induced nausea and vomiting. N Engl J Med 2016;375:134-42. Crossref

19. Abe M, Hirashima Y, Kasamatsu Y, et al. Efficacy and safety of olanzapine combined with aprepitant, palonosetron, and dexamethasone for preventing nausea and vomiting induced by cisplatin-based chemotherapy in gynecological cancer: KCOG-G1301 phase II trial. Support Care Cancer 2016;24:675-82. Crossref

20. Hashimoto H, Abe M, Tokuyama O, et al. Olanzapine 5 mg plus standard antiemetic therapy for the prevention of chemotherapy-induced nausea and vomiting (J-FORCE): a multicentre, randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol 2020;21:242-9. Crossref

21. Malhotra K, Vu P, Wang DH, Lai H, Faziola LR. Olanzapine-induced neutropenia. Ment Illn 2015;7:5871. Crossref

22. Gjafa E, Ng K, Grunewald T, et al. Neutropenic sepsis rates in patients receiving bleomycin, etoposide and cisplatin chemotherapy using olanzapine and reduced doses of dexamethasone compared to a standard antiemetic regimen. BJU Int 2021;127:205-11. Crossref

23. Yeo W, Mo FK, Suen JJ, et al. A randomized study of aprepitant, ondansetron and dexamethasone for chemotherapy-induced nausea and vomiting in Chinese breast cancer patients receiving moderately emetogenic chemotherapy. Breast Cancer Res Treat 2009;113:529-35. Crossref

24. Babu G, Saldanha SC, Kuntegowdanahalli Chinnagiriyappa L, et al. The efficacy, safety, and cost benefit of olanzapine versus aprepitant in highly emetogenic chemotherapy: a pilot study from South India. Chemother Res Pract 2016;2016:3439707. Crossref

Pages