Abstract 4003: Comprehensive genomic profiling of Merkel cell carcinoma samples reveals bimodal distribution of tumor mutational burden and two mutually exclusive candidate mechanisms of malignant transformation

Abstract

Abstract Merkel cell carcinoma (MCC) is a rare but aggressive skin cancer. Etiologic risk factors include long-term UV exposure, Merkel cell polyomavirus (MCV) infection, age, and immunosuppression. Large-scale investigations of MCC are hindered by its rarity and comprehensive understanding of this disease is currently lacking. In a study of 275 patients, representing the largest sequenced cohort of MCC samples to date, we sought to characterize the genomic landscape of MCC as well as identify whether causality could be attributed to MCV infection, UV exposure, or a synergistic effect between the two. Comprehensive genomic profiling of 395 cancer-related genes, including evaluation of tumor mutational burden (TMB) and mutational signatures, was performed on MCC tumors from 275 patients. The TMB of MCC exhibited a stark bimodal distribution between TMB high (≥20 mutations/mb) and TMB low (<6 mutations/mb) specimens with only 22 specimens having an intermediate TMB. Overall, 101/275 (37%) of MCC samples were TMB high (median 53.1 mutations/mb) with the majority carrying short variant mutations in TP53, RB1, or NOTCH1. Analysis of mutational signatures found that 94/101 (93%) of TMB high samples exhibited signs of UV damage. No signatures indicative of APOBEC activity or of intrinsic DNA repair defects such as microsatellite instability were observed in the TMB high group. In contrast, 152/275 (55%) of MCC specimens were TMB low (median 1.2 mutations/mb) with 39% of these samples having no known or likely genomic driver. We hypothesized that the primary driver in the TMB low group was MCV integration and examined MCV infection through de novo assembly of non-human sequencing reads in all specimens of the MCC cohort. We found 99/152 (65%) of TMB low specimens to be positive for MCV integration as opposed to 0/101 (0%) of TMB high specimens, which was a statistically significant difference (P = 2.20e-31, Fisher's exact test). Overall, 3 of the 22 TMB intermediate specimens were MCV-positive. In a large cohort of 275 patient samples, we have described two genomically distinct subsets of MCC: a TMB high group defined by UV damage with mutations in well-known tumor suppressors and a TMB low group defined by MCV integration. We propose that MCV integration is the dominant driver in the TMB low cohort. Indeed, most MCV sequences that we observed appear to be derived from the large or small T antigens, two established viral oncogenes. These results are suggestive of two distinct etiologies of MCC that appear to be mutually exclusive, namely UV exposure vs. MCV integration. These etiologies may have significant impact on targeted and immuno-therapy options for MCC patients. Citation Format: Meagan Montesion, Ethan S. Sokol, Todd C. Knepper, Andrew S. Brohl, Garrett M. Frampton, Phil J. Stephens, James A. DeCaprio, Kenneth Y. Tsai, Lee A. Albacker. Comprehensive genomic profiling of Merkel cell carcinoma samples reveals bimodal distribution of tumor mutational burden and two mutually exclusive candidate mechanisms of malignant transformation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4003.