Oral Presentation Australasian Melanoma Conference 2026

Polygenic risk for idiopathic Inflammatory Bowel Disease predicts severe colitis following Immune Checkpoint Blockade in advanced melanoma (#12)

Hadis Ghajari 1 2 , Nurudeen Adegoke 3 4 5 , Stuart MacGregor 1 , Huanwei Wang 1 2 , Dale R Nyholt 2 , Ines Pires da Silva 3 4 5 , Georgina V Long 3 4 5 6 7 , Alexander M Menzies 3 4 6 7 , Serigne N Lo 3 4 5 , Mathias Seviiri 1 2 8 , Matthew H Law 1 2 8
  1. QIMR Berghofer Medical Research Institute, Brisbane, QLD, Australia
  2. School of Biomedical Sciences and Centre for Genomics and Personalised Health, Faculty of Health, Queensland University of Technology, Brisbane, Queensland, Australia
  3. Melanoma Institute Australia, The University of Sydney, Sydney, NSW, Australia
  4. Faculty of Medicine and Health, University of Sydney, NSW 2006, Australia, Sydney, NSW, Australia
  5. Charles Perkins Centre, The University of Sydney, NSW 2006, Australia, Sydney, NSW, Australia
  6. Mater Hospital, North Sydney, NSW 2060, Australia, Sydney, NSW, Australia
  7. Royal North Shore Hospital, St Leonards, NSW 2065, Australia, Sydney, NSW, Australia
  8. University of Queensland, School of Biomedical Sciences, Faculty of Health, Medicine and Behavioural Sciences, Brisbane, Australia, Brisbane, QLD, Australia

Introduction:
Immune checkpoint inhibitors (ICIs) improve survival in advanced melanoma but cause immune-related adverse events (irAEs). Colitis is among the most common and clinically significant irAEs, with the potential to cause hospitalization and death. Overlapping histopathological features between ICI–induced colitis and inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn’s disease (CD), suggest a shared biological basis and predisposition. This raises the possibility that genetic susceptibility to IBD could predict colitis risk. Defining such determinants could improve risk stratification, inform treatment decisions, and provide insight into disease mechanisms.

Methods:
We conducted genome-wide association studies (GWAS) of UC and CD using 90% of the UK Biobank (UKBB) cohort. We then performed meta-analyses integrating UKBB data with summary statistics from the Million Veteran Program, FinnGen, and IBD Genetics Consortium to generate discovery sets for the calculation of polygenic risk scores (PRS). The discovery GWAS comprised 17,263 UC cases and 1,336,274 controls, and 19,573 CD cases and 1,233,508 controls. PRSs were validated in the remaining 10% of the UKBB cohort and tested for association with ICI–induced colitis in an independent cohort of melanoma patients (106 with colitis, 1,305 without) genotyped across two from the Melanoma Institute Australia (MIA 1 and 2).

Results:
In the UKBB, the PRSs incorporating age and sex demonstrated strong predictive performance for both UC and CD, with ORs per SD of 2.29 (95% CI 2.05-2.55) and 2.15 (1.87-2.46), respectively. Areas under the receiver operating characteristic curve were 0.736 (95% CI 0.710-0.763) for UC and 0.707 (95% CI 0.671-0.744) for CD. Fixed-effect meta-analysis of MIA 1 and 2 showed significant associations between UC- and CD-derived PRSs and ICI-induced severe colitis (UC PRS: OR per SD increase = 1.30, 95% CI 1.07–1.58, P = 0.006; CD PRS: OR per SD increase = 1.24, 95% CI 1.02–1.51, P = 0.03).

Conclusion:
Genetic susceptibility to IBD, particularly UC, is associated with increased risk of ICI–induced colitis, supporting shared genetic mechanisms underlying IBD and irAEs. Identification of individuals at elevated genetic risk may enable more precise patient stratification, inform immunotherapy decisions, and support personalized prevention strategies to reduce the incidence and severity of colitis.