Disorders of sex development (DSD) are defined as congenital conditions in which development of chromosomal, gonadal, or anatomical sex is atypical.1 Traditionally, diagnosis in these patients relies on extensive endocrine investigation. With advances in the understanding of the genes involved in sexual determination and differentiation,2 molecular diagnosis is playing an increasingly important role and may even overtake the role of hormonal assessment as the first-line test, with the latter being reserved for assessment of disease severity rather than diagnosis.3

One of the most common causes of 46,XY DSD in the western population is androgen insensitivity syndrome (AIS).4 Whether the same is true in our local population remains unknown. We performed a prospective multicentre study to explore the possible aetiological basis of 46,XY DSD in the Hong Kong Chinese population.

Patients who were referred to a paediatric endocrinologist for the first time or were followed up in their clinic at five public hospitals in Hong Kong between July 2009 and June 2011 were recruited for the study. Inclusion criteria were 46,XY ethnic Chinese patients who presented with incompletely virilised, ambiguous, or completely female external genitalia. Criteria that suggested DSD at birth were overt genital ambiguity, apparent female genitalia with an enlarged clitoris, posterior labial fusion, or an inguinal/labial mass, apparent male genitalia with bilateral undescended testes, micropenis, isolated perineal hypospadias, or mild hypospadias with undescended testes, and discordance between genital appearance and prenatal karyotype.1 Micropenis is defined as stretched penile length of <2.5 cm based on the published norm for Chinese.1 Written informed consent was obtained from the patients and/or parents and the study was approved by the local ethics committee. None of the patients/parents refused to participate in the study although seven refused genetic testing (Table 1).

Blood was taken from patients for electrolyte and baseline endocrine assessment and included measurement of cortisol, 17-hydroxyprogesterone (17-OHP), dehydroepiandrosterone sulfate, testosterone (T), androstenedione (A4), dihydrotestosterone (DHT), anti-Müllerian hormone (AMH), and gonadotropins. Human chorionic gonadotropin (hCG) stimulation test was performed to test for testicular Leydig cell function. The short synacthen test was also performed when indicated.

Cortisol, dehydroepiandrosterone sulfate, and gonadotropins were measured by electro-chemiluminescence immunoassay (Modular Analytics E170; Roche, Mannheim, Germany); T was measured by a competitive immunoenzymatic assay (ACCESS 2; Beckman Coulter, Brea [CA], US); 17-OHP was measured by liquid chromatography–tandem mass spectrometry using an in-house method; AMH was measured by an enzyme-linked immunosorbent assay (AMH Gen II ELISA, A73818; Beckman Coulter, Brea [CA], US); DHT was measured by radioimmunoassay (DSL9600i; Beckman Coulter, Prague, Czech Republic); A4 was measured by solid-phase competitive chemiluminescent enzyme-labelled immunoassay (L2KAO2, Immulite 2000; Siemens, Tarrytown [NY], US). Male reference intervals were considered the most appropriate for data interpretation in this study.

Spot urine from patients under 3 months of age and 24-hour urine from those at or older than 3 months of age were processed for steroid profiling as described previously.5

DNA was extracted from peripheral whole blood using a QIAamp DNA blood kit (Qiagen, Hilden, Germany). Polymerase chain reaction and direct DNA sequencing were performed on targeted genes when suggested by the clinical and hormonal findings. Otherwise all patients had their AR (androgen receptor) and NR5A1 (steroidogenic factor 1) genes sequenced. Those patients with negative genetic findings were subjected to multiplex ligation-dependent probe amplification (MLPA) analysis (P185 Intersex probemix; P074 Androgen Receptor probemix and P334 Gonadal probemix; MRC-Holland) to test for gross deletion or gene duplication. The results were analysed by Coffalyser.Net. Family genetic studies were performed when mutation(s) were identified in the index patients.

The functional effect of novel missense mutations detected was tested by online in-silico analysis software SIFT, PolyPhen2, and Align GVGD.

Overall, 64 patients (53 male, 11 phenotypic female), including 14 new patients, with 46,XY DSD were recruited into the study. The clinical and hormonal findings of individual patients are listed in Table 1. A genetic diagnosis was made in 10 patients prior to the study. Other major structural abnormalities were evident in eight (Table 2). Their age at presentation ranged from birth to 17 years. Five (8%) were born prematurely (24-35 weeks) and nine (14%) with low birth weight (0.59-2.32 kg). All had non-consanguineous parents. A family history of sexual ambiguity was present in six. Overall, 61 (95%) presented with ambiguous external genitalia including 15 with isolated micropenis, eight with isolated severe hypospadias, and one with discordance between the prenatal karyotype and the postnatal phenotype. Three presented after birth, one each with inguinal hernia, delayed puberty, and primary amenorrhoea.

Regarding the hormonal findings, Figure 1 shows the baseline and post-hCG–stimulated T and DHT levels in patients with mutations detected in the AR gene and those without, where the results overlapped between the two groups. Eight patients (patients 13, 19, 21, 30, 31, 40, 49, and 56) underwent short synacthen test and with the exception of patient 19, all had an adequate cortisol response (>550 nmol/L). Patient 27 had a relatively low T/A4 ratio before and after hCG stimulation but sequencing revealed no mutation in his HSD17B3 (17β-hydroxysteroid dehydrogenase III) gene. All other patients had unremarkable T and A4 levels, as well as T/A4 ratio.

Eleven patients had characteristically low 5α- to 5β-reduced steroid metabolite ratios in their urine, compatible with the diagnosis of 5α-reductase 2 deficiency (5ARD). This was also confirmed by mutational analysis of the SRD5A2 (steroid 5α-reductase 2) gene. All other patients had unremarkable urinary steroid metabolite pattern.

Overall, 22 (39%) patients had a confirmed genetic diagnosis (Table 3). The most common diagnoses in our cohort were 5ARD (n=11) and AIS (n=7). Other genetic diagnoses included cholesterol side-chain cleavage enzyme deficiency (n=1), Frasier syndrome (n=1), NR5A1-related sex reversal (n=1), and persistent Müllerian duct syndrome (PMDS; n=1). The clinical and laboratory findings of patients 19 and 20 have been reported previously.6 7 Patients 12 and 15 had de-novo mutations in the AR gene and were in mosaic pattern. Patient 21 had a novel missense variant p.Ala260Val detected in his NR5A1 gene. His AMH level was not low, contrary to some of the previously reported cases.8 There was also a clinically significant rise in T level after hCG stimulation. Short synacthen test demonstrated an adequate cortisol response (baseline: 720 nmol/L; post–adrenocorticotropin hormone: 822 nmol/L). His father also carried the same heterozygous mutation although he denied any symptoms of DSD. This novel genetic variant was not detected in 100 normal Chinese subjects (control). Patient 22 had bilateral undescended testes. He underwent orchidopexy at the age of 1 year during which the presence of Müllerian duct structures was suspected. Further workup including pelvic ultrasound revealed Müllerian duct structures and extremely low AMH level. The diagnosis of PMDS was confirmed by the presence of three heterozygous novel missense variants in the AMH gene (Tables 3 and 4).

Six novel genetic variants were identified in the AMH, AR, and NR5A1 genes (Fig 2). At least two of the three in-silico analysis programmes predicted the variants to be pathogenic (Table 4). Multiple sequence alignment showed that the amino acids of concern were highly conserved across different animal species. All these findings support the pathogenic nature of these variants accounting for the patients’ phenotypes.

Eleven patients were reared as girls because of severe under-virilisation at birth, including three with 5ARD, three with AIS, and one with Frasier syndrome. The underlying genetic causes in the remaining four patients were undetermined. The longest follow-up period was 27 years. None of them has requested change of gender to date. Five patients (patients 2, 4, 7, 12, and 15) exhibited ‘tom-boy-like’ behaviour during childhood and required counselling by a clinical psychologist while two males (patients 17 and 47) requested exogenous T to augment penile growth after puberty. Patient 20 developed germinoma in her dysgenetic gonad with no recurrence after surgery.

46,XY DSD is a heterogeneous condition caused by a wide spectrum of disorders. Making an accurate diagnosis is difficult but important for emergency medical treatment as some DSDs are associated with life-threatening Addisonian crisis. In addition, the diagnosis is essential so that relevant information and counselling can be provided to parents and clinical management can be formulated, bearing in mind the best interests of the child. Initial workup includes a detailed antenatal and postnatal history, physical examination, karyotyping, and hormonal assays. This will guide further workup such as imaging and genetic analysis. Nonetheless, there are often limitations to hormonal studies as illustrated in the present series. The non-distinct pattern of T and DHT at baseline and following hCG stimulation in AR mutation–positive and –negative patients suggest the need to reconsider our laboratory diagnostic algorithm for AIS.

Androgen insensitivity syndrome is reported to be the most common cause of 46,XY DSD in a few ethnic groups,9 10 11 while 5ARD, which is believed to be rare, was also a major aetiology in our cohort. It is important to differentiate between 5ARD and AIS as soon as possible so that patients with 5ARD can be raised as boys whenever practical.12 The penile growth of patients with 5ARD can be promoted by topical DHT treatment and spontaneous virilisation may occur during puberty. Most of these patients who are reared as girls during childhood identify themselves as male and change their gender as an adult, although we have not received any such request from our cohort. Exposure to androgen during the antenatal, postnatal, and pubertal period may masculinise the brain and influence gender identity.13 It was found that 5ARD is easy to diagnose by its characteristic urinary steroid excretion pattern and its high mutational detection rate in the SRD5A2 gene.14 Of the 11 patients with 5ARD, eight harboured the missense mutation p.Arg227Gln in their SRD5A2 gene, a useful fact to enable screening for this mutation before proceeding to sequencing of the whole gene. Unfortunately, patients have previously been too easily labelled with AIS when laboratory diagnostic services were less advanced. This is illustrated by patient 7 who was labelled as AIS until her urine steroids were analysed and revealed classic features of 5ARD.15 We recommend that 5ARD is excluded in all 46,XY DSD patients before other differential diagnoses are considered. Moreover, since the baseline and post-hCG–stimulated T and DHT results are unreliable when diagnosing AIS, genetic study of the AR gene should also be performed as a first-line investigation.

HSD17B3 deficiency has been reported to be the most common cause of T biosynthetic defect leading to 46,XY DSD in some populations, with an estimated incidence of 1:147 000 in the Netherlands and as high as 1:200 to 1:300 in Arabians due to their high consanguinity rate.16 17 Nonetheless, no patient in our cohort was diagnosed with this condition based on the hormonal pattern. Ethnic differences in disease spectrum may be one of the reasons for this observation. Another possible explanation is the lack of reliable diagnostic cutoff for the pre- and post-stimulated T/A4 ratios. George et al18 have summarised the cutoffs used by various researchers, with the pre-stimulated cutoff range set at 0.006 to 1.64, and the post-stimulated level set at 0.09 to 3.4 for newborn to teenage groups. The difficulties in setting up reliable diagnostic cutoffs for the T/A4 ratio are similar to the T/DHT ratios and have been discussed in our previous study.14 Furthermore, HSD17B3 deficiency gives no characteristic findings on urinary steroid profiling.5 19 Molecular analysis of the HSD17B3 gene may have offered a means to diagnose this condition but unfortunately, due to budget constraints, we were unable to perform mutational analysis of this gene in all our patients, although a normal MLPA result in our patients made gross deletion in this gene unlikely.

The two novel mutations p.Asp266Asn and p.Thr576Pro in the AR gene lie within the N-terminal domain of the androgen receptor that is involved in transcription regulation and DNA binding, respectively. Missense mutations around these two codons have been reported in patients with AIS according to the Androgen Receptor Gene Mutations Database, April 2013.20 Multiple sequence alignment shows that both amino acids are highly conserved among different species, suggesting that aspartic acid at codon 266 and threonine at codon 576 are critical for proper receptor function. Similarly, the alanine at codon 260 of the NR5A1 gene is located in helix 3 of the ligand-binding domain of the nuclear receptor,21 and is also a highly conserved region. Mutation in this region has been reported to result in 46,XY DSD.8 Replacing alanine at this position by valine is therefore expected to be deleterious to the protein function. Phenotypic variability in NR5A1 gene mutation within a kindred has been reported and this may explain why patient 21 had ambiguous external genitalia to such an extent that he required the attention of a paediatric specialist, even though his father was fertile, and denied any symptoms of DSD or need for medical attention.22 For the AMH gene, the 3’ end of exon 5 is one of the mutational hotspots in patients with PMDS.23 Exon 5 encodes the bioactive C-terminal domain. The three mutations detected in patient 22 are all located at highly conserved regions. Although in-vitro functional characterisation for the mutant proteins was not performed, the undetectable serum AMH level in this patient was compatible with the mutations being pathogenic, possibly due to abnormal protein folding and increased instability, as reported previously in mutations located in this region.24

Gonadal malignancy was rare in our series, probably because gonadectomy was performed early in life when the decision of female sex assignment was made. Although this helps to avoid further virilisation and to establish gender identity, the timing of corrective surgery and gonadectomy remain controversial. Patient advocacy groups have suggested delaying any surgery for cosmetic reasons until the patient is mature enough to give informed consent25 but such practice has not been validated in our Chinese patients. Whether cultural factors have any impact on gender assignment remains uncertain in our community.

Prematurity or low birth weight was not uncommon in our series. This made diagnosis of DSD in our patients even more difficult because ethnic-specific and gestational age– or weight-adjusted anthropometric measurement of the external genitalia was not available. Assessment of the genital anatomy relies solely on the experience of the paediatric specialist and is obviously far from ideal. A conjoint effort by local paediatricians is needed to set up these normative data.

Less than half of our patients had a confirmed diagnosis in the present study. With the increasing availability of next-generation sequencing technology, and with its established role in molecular diagnostic services, including DSD,3 26 it is hoped that sooner rather than later, most patients will have a confirmed genetic diagnosis. Nonetheless, we speculate that some patients have a non-genetic aetiology since environmental factors may alter the phenotypic expression. Several animal and human studies have shown that antenatal exposure to pesticides and plasticisers may lead to fetal genital malformation.27

Altogether there was an average of 11 250 male live births every year in the five public hospitals that participated in this study. Since 11 newborns with 46,XY DSD were born in these five hospitals and were recruited during our study period, this gives an estimated incidence of 46,XY DSD of 1:2045 male births requiring the input of paediatric endocrinologists. This figure may underestimate the true incidence of this group of diseases as some patients present late and others may have subtle defects that go unnoticed by our specialists. If the actual number of patients with chromosomal and 46,XX DSD in our population is considered, the actual incidence of DSD can be expected to be much higher.

There are a few limitations in this study. First, the number of patients was relatively small. This may have resulted in bias in our observation and the data do not represent the prevalence of disease in our population. Second, in-vitro study was not performed on the novel genetic variants for functional characterisation, although we believe that all the available evidence indicates the pathogenic nature of these variants. Third, due to budget constraints, we were unable to sequence all genes related to 46,XY DSD.

Our findings indicate that 5ARD and AIS are possibly the major causes of 46,XY DSD in the Hong Kong Chinese population. Molecular analyses of the SRD5A2 and AR genes were demonstrated to be more reliable than hormonal testing. Since the missense mutation p.Arg227Gln was a recurrent hotspot mutation in 5ARD in our local patients, all patients should be screened for this mutation.

We thank Mr YC Ho, Ms YF Wong, and Ms YP Iu for their technical assistance. The study was supported by the Queen Elizabeth Hospital Research Grant 2009 QEH/RC/G/0910-A04/R0901 and Kowloon Central Cluster Research Grant 2012 KCC/RC/G/1213-B01.

1. Lee PA, Houk CP, Ahmed SF, Hughes IA; International Consensus Conference on Intersex organized by the Lawson Wilkins Pediatric Endocrine Society and the European Society for Paediatric Endocrinology. Consensus statement on management of intersex disorders. International Consensus Conference on Intersex. Pediatrics 2006;118.e488-500. Crossref

2. Ono M, Harley VR. Disorders of sex development: new genes, new concepts. Nat Rev Endocrinol 2013;9:79-91. Crossref

3. Arboleda VA, Lee H, Sánchez FJ, et al. Targeted massively parallel sequencing provides comprehensive genetic diagnosis for patients with disorders of sex development. Clin Genet 2013;83:35-43. Crossref

4. Jääskeläinen J. Molecular biology of androgen insensitivity. Mol Cell Endocrinol 2012;35:4-12. Crossref

5. Chan AO, Shek CC. Urinary steroid profiling in the diagnosis of congenital adrenal hyperplasia and disorders of sex development: experience of a urinary steroid referral centre in Hong Kong. Clin Biochem 2013;46:327-34. Crossref

6. Parajes S, Chan AO, But WM, et al. Delayed diagnosis of adrenal insufficiency in a patient with severe penoscrotal hypospadias due to two novel P450 side-chain cleavage enzyme (CYP11A1) mutations (p.R360W; p.R405X). Eur J Endocrinol 2012;167:881-5. Crossref

7. Chan WK, To KF, But WM, Lee KW. Frasier syndrome: a rare cause of delayed puberty. Hong Kong Med J 2006;12:225-7.

8. Allali S, Muller JB, Brauner R, et al. Mutation analysis of NR5A1 encoding steroidogenic factor 1 in 77 patients with 46,XY disorders of sex development (DSD) including hypospadias. PLoS One 2011;6:e24117. Crossref

9. Bangsbøll S, Qvist I, Lebech PE, Lewinsky M. Testicular feminization syndrome and associated gonadal tumors in Denmark. Acta Obstet Gynecol Scand 1992;71:63-6. Crossref

10. Boehmer AL, Brinkmann O, Brüggenwirth H, et al. Genotype versus phenotype in families with androgen insensitivity syndrome. J Clin Endocrinol Metab 2001;86:4151-60. Crossref

11. Abdullah MA, Saeed U, Abass A, et al. Disorders of sex development among Sudanese children: 5-year experience of a pediatric endocrinology clinic. J Pediatr Endocrinol Metab 2012;25:1065-72. Crossref

12. Mieszczak J, Houk CP, Lee PA. Assignment of the sex of rearing in the neonate with a disorder of sex development. Curr Opin Pediatr 2009;21:541-7. Crossref

13. Imperato-McGinley J, Peterson RE, Gautier T, Sturla E. Androgens and the evolution of male-gender identity among male pseudohermaphrodites with 5alpha-reductase deficiency. N Engl J Med 1979;300:1233-7. Crossref

14. Chan AO, But BW, Lee CY, et al. Diagnosis of 5α-reductase 2 deficiency: is measurement of dihydrotestosterone essential? Clin Chem 2013;59:798-806. Crossref

15. Chan AO, But BW, Lau GT, et al. Diagnosis of 5α-reductase 2 deficiency: a local experience. Hong Kong Med J 2009;15:130-5.

16. Boehmer AL, Brinkmann AO, Sandkuijl LA, et al. 17β-hydroxysteroid dehydrogenase-3 deficiency: diagnosis, phenotypic variability, population genetics, and worldwide distribution of ancient and de novo mutations. J Clin Endocrinol Metab 1999;84:4713-21. Crossref

17. Rosler A. 17 Beta-hydroxysteroid dehydrogenase 3 deficiency in the Mediterranean population. Pediatr Endocrinol Rev 2006;3 Suppl 3:455-61.

18. George MM, New MI, Ten S, Sultan C, Bhangoo A. The clinical and molecular heterogeneity of 17βHSD-3 enzyme deficiency. Horm Res Paediatr 2010;74:229-40. Crossref

19. Lee YS, Kirk JM, Stanhope RG, et al. Phenotypic variability in 17β-hydroxysteroid dehydrogenase-3 deficiency and diagnostic pitfalls. Clin Endocrinol 2007;67:20-8. Crossref

20. Androgen Receptor Gene Mutations Database. Available from: http://androgendb.mcgill.ca/. Accessed Aug 2013.

21. El-Khairi R, Martinez-Aguayo A, Ferraz-de-Souza B, Lin L, Achermann JC. Role of DAX-1 (NR0B1) and steroidogenic factor-1 (NR5A1) in human adrenal function. Endocr Dev 2011;20:38-46.

22. Ciaccio M, Costanzo M, Guercio G, et al. Preserved fertility in a patient with a 46,XY disorder of sex development due to a new heterozygous mutation in the NR5A1/SF-1 gene: evidence of 46,XY and 46,XX gonadal dysgenesis phenotype variability in multiple members of an affected kindred. Horm Res Paediatr 2012;78:119-26. Crossref

23. Josso N, Belville C, di Clemente N, Picard JY. AMH and AMH receptor defects in persistent Müllerian duct syndrome. Hum Reprod Update 2005;11:351-6. Crossref

24. Belville C, Van Vlijmen H, Ehrenfels C, et al. Mutations of the anti-Müllerian hormone gene in patients with persistent Müllerian duct syndrome: biosynthesis, secretion, and processing of the abnormal proteins and analysis using a three-dimensional model. Mol Endocrinol 2004;18:708-21. Crossref

25. Consortium on the Management of Disorders of Sex Development: Clinical Guidelines for the Management of Disorders of Sex Development in Childhood. California, US: Intersex Society of North America; 2006. Available from: http://www.dsdguidelines.org/files/clinical.pdf. Accessed Aug 2015.

26. Hersmus R, Stoop H, Turbitt E, et al. SRY mutation analysis by next generation (deep) sequencing in a cohort of chromosomal Disorders of Sex Development (DSD) patients with a mosaic karyotype. BMC Med Genet 2012;13:108. Crossref

27. Kalfa N, Philibert PH, Baskin LS, Sultan C. Hypospadias: interactions between environment and genetics. Mol Cell Endocrinol 2011;335:89-95. Crossref