Prospective tumour mutation burden and neoantigen profiling predicts immunotherapy response in metastatic melanoma

Tumour mutation burden (TMB) is a promising biomarker in predicting immunotherapy response, yet its reproducibility across target panels needs to be established. This study assessed the reproducibility of TMB estimates in melanoma using TruSight Oncology 500 across two laboratories and compared these results with the FoundationOne CDx and QIAseq TMB IO panels. High concordances in TMB estimation, mutation calls, and BRAF and N/K/HRAS hotspot variants were observed between platforms. In a cohort of 198 pre-treatment biopsies from patients treated with immune checkpoint inhibitors, high TMB (≥10 mut/Mb) was associated with significantly improved response and progression-free survival (PFS), while somatic mutations in PTPRD and a germline variant in PIK3CA (I391M) were associated with favourable outcomes independent of TMB. Neoantigen profiling of 135 samples demonstrated that neoantigen load, particularly neoepitopes with strong-binding to class II MHC, had a superior predictive value over TMB for PFS. Mutations in NF1 and ROS1 that produce a neoantigen were also linked to improved outcomes. These results support the reproducibility of TMB estimation and highlight the added value of neoantigen profiling in predicting immunotherapy benefits in melanoma.