37.06 Next Generation Sequencing Mutation Signature and Response to Immunotherapy in Melanoma

K. Loo1,2, I. Soliman2, J. D’Souza2, M. Renzetti1,2, T. Li2, H. Wu2, B. Luo2, S. Movva2, M. Lango2, A. Olszanski2, J. Farma2  1Temple University,Surgical Oncology,Philadelpha, PA, USA 2Fox Chase Cancer Center,Surgical Oncology,Philadelphia, PA, USA

Introduction:  Considerable advances in melanoma therapy have been realized through immunotherapy. Unfortunately, there remains a subset of patients who fail to respond. There is a need to identify patients likely to exhibit sustained response to immunotherapy agents, versus those who would benefit from alternate therapies. It remains to be seen whether a patient’s tumor molecular gene profile could serve as a marker of response. The principle aim of this study was to utilize molecular profiling to determine the difference in gene mutation status between responders and non-responders prior to initial immunotherapy.

Methods:  Tissue samples from n=42 melanoma patients were collected. Pre-treatment tumors were profiled using a Next Generation Sequencing (NGS) panel of 50 targetable cancer-related gene mutations. Gene mutation status was analyzed. Response to immunotherapy was assessed using RECIST v1.1 criteria. Statistical analysis was conducted using Fisher’s exact test.

Results: Within the total cohort, patients received initial immunotherapy of anti-CTLA4 (n=7), anti-PD1 (n=22), ipilimumab/nivolumab combination (n=7), or clinical trial immunotherapy combinations (n=7), and were divided into responders (n=17) or non-responders (n=25). 62% were male, with a median age of 68. 65 total mutations were identified affecting 36 unique genes. No mutations were found in 10% of patients, while 52% had 1 mutation, 28% had 2 mutations, 5% had 3 mutations, and 5% had 4 or more mutations. The overall hot-spot mutation burden in the responder cohort was 1.80 versus 1.32 in the non-responder cohort. The most common mutation among responders was NRAS, detected in 70.5% of responders, vs 29.4% in non-responders. Additionally, the most common mutation among non-responders was a BRAF mutation, identified in 18.2% of responders vs. in 81.8% of non-responders.

Conclusion: Using our NGS platform, this analysis demonstrated potential associations between NRAS mutation and responders to immunotherapy, and correlations between BRAF mutation and non-responders. This unique finding should be evaluated in a larger population to confirm these results and may help predict response in the future.