Hong Kong Med J 2024;30(Suppl 3):S34-8
ImmuneMirror for evaluation of the genomic and transcriptomic features of resistance to immunotherapy for gastrointestinal tract cancer: abridged secondary publication
W Dai1, ML Lung1, KO Lam1, CL Chiang1, H Zheng2
1 Department of Clinical Oncology, The University of Hong Kong, Hong Kong SAR, China
2 Department of Medicine, BMIR-ITI Institute, Stanford University, United States
 
 
  1. We developed an integrative ImmuneMirror pipeline to evaluate tumour mutation burden, microsatellite instability status, human leukocyte antigen type, predicted neoantigen load, top-ranked neoantigens with T cell immunogenicity, and expression of innate anti-PD-1 resistance signatures.
  2. We established a web server incorporating a machine learning model for neoantigen prediction and prioritisation.
  3. In gastrointestinal tract cancers (including colorectal cancer, oesophageal squamous cell carcinoma, and hepatocellular carcinoma), an elevated neoantigen load was associated with good clinical outcomes in patients with oesophageal squamous cell carcinoma and poor clinical outcomes in patients with hepatocellular carcinoma.
  4. A subset of microsatellite instability–high colorectal cancers with lower neoantigen load was shown to exhibit an advanced T stage.
  5. The neopeptide YMCNSSCMGV derived from the TP53 hotpot mutation G245V restricted by HLA-A02 was identified in a patient with oesophageal squamous cell carcinoma. Experimental validation revealed high binding affinity between HLA-A02 and TP53G245V (YMCNSSCMGV).