Germline biomarkers predict toxicity to anti-PD1/PDL1 checkpoint therapy

Area of Science:

• Genomics and Precision Medicine
• Oncology
• Immunotherapy

Background:

• Immune-related adverse effects (irAEs) are common and can be severe with immune checkpoint inhibitors like anti-PD1/PDL1 therapies.
• Identifying patients at risk for irAEs is crucial for personalized cancer treatment and managing long-lasting toxicities.

Purpose of the Study:

• To investigate if germline microRNA pathway functional variants can predict irAEs in patients receiving cancer therapies.
• To develop and validate a biomarker panel for predicting toxicity across different cancer types.

Main Methods:

• Evaluated microRNA pathway variants for association with grade 2 or higher toxicity using four machine learning classifiers.
• Trained and validated the biomarker panel on cohorts of patients with melanoma and other cancer types.
• Assessed the association between predicted toxicity probability and treatment response in the melanoma cohort.

Main Results:

• A biomarker panel accurately predicted toxicity (80% in melanoma, 77.6% in pan-cancer validation).
• The most significant toxicity biomarker, RAC1, indicated a >9-fold increased risk.
• Predicted toxicity was not associated with response to anti-PD1/PDL1 therapy in the melanoma cohort.

Conclusions:

• A germline microRNA-based signature effectively predicts irAEs to anti-PD1/PDL1 therapy across various cancer types.
• This biomarker represents a significant advancement in personalizing checkpoint inhibitor therapy.
• Further research can help tailor cancer treatments and mitigate adverse effects.