Whole-genome landscape of mucosal melanoma reveals diverse drivers and therapeutic targets

Abstract

Knowledge of key drivers and therapeutic targets in mucosal melanoma is limited due to the paucity of comprehensive mutation data on this rare tumor type. To better understand the genomic landscape of mucosal melanoma, here we describe whole genome sequencing analysis of 67 tumors and validation of driver gene mutations by exome sequencing of 45 tumors. Tumors have a low point mutation burden and high numbers of structural variants, including recurrent structural rearrangements targeting TERT, CDK4 and MDM2. Significantly mutated genes are NRAS, BRAF, NF1, KIT, SF3B1, TP53, SPRED1, ATRX, HLA-A and CHD8. SF3B1 mutations occur more commonly in female genital and anorectal melanomas and CTNNB1 mutations implicate a role for WNT signaling defects in the genesis of some mucosal melanomas. TERT aberrations and ATRX mutations are associated with alterations in telomere length. Mutation profiles of the majority of mucosal melanomas suggest potential susceptibility to CDK4/6 and/or MEK inhibitors.

Fig. 1

Mutational signatures. a Mutation burden (top) and proportion of mutational signature (second from top) per sample. The tumor purity, country of origin, site, and region for each sample is shown beneath the plot. b Seven mutational signatures were identified in the WGS cohort. For each signature, the mutational type probability for each substitution in a trinucleotide context is shown (total 96 contexts). UVR ultraviolet radiation. c Proportion of each signature in tumors from the upper: nasal, oral eye and lower: anorectal and genitourinary body sites. Source data for Fig. 1c are provided as a Source Data file

Fig. 2

Structural variant complexity. a Five rearrangement signatures (RS) were identified. Rearrangements were classified into 32 categories based on the rearrangement size, type (Del = deletion, Dup = duplication, Inv = inversion, T = translocation) and whether breakpoints are clustered (left) or non-clustered (right). b Two groups of tumors following clustering with ConsensusClusterPlus. Plots from top to bottom are: number and type of SVs; proportion of each rearrangement signature (light blue = lower, dark blue = higher); evidence of localized complexity per chromosome, per sample (light = lower, dark = higher); number of kataegis loci per chromosome, per sample; total number of kataegis loci per sample; tumor ploidy; tumor purity; sample origin, sample ancestry, tumor body site, and body region

Fig. 3

Significantly mutated genes affected by SNVs and indels. a Number of coding mutations and oncoplot of mutations in 10 significantly mutated genes in the WGS cohort (n = 67). If a sample has multiple SNV/indel in a gene, the SNV/indel with the most severe predicted consequence is shown. b Positions of BRAF, NRAS, SF3B1, SPRED1, KIT, and NF1 somatic mutations in the protein. c Number of coding mutations and oncoplot of mutations in eight significantly mutated genes in the WES replication cohort (n = 45). If a sample has multiple SNV/indel in a gene, the SNV/indel with the most severe predicted consequence is shown. There were no mutations in CHD8 or HLA-A in the replication cohort

Fig. 4

Recurrent genes affected by copy number and rearrangement breakpoints. a Number and type of SV events. b Percent of the genome affected by copy number deletions (CN0), loss (CN1), copy neutral LOH, amplification (CN ≥ 6). c Rearrangement breakpoints in genes identified as recurrently affected by SNVs or CNVs are previously identified melanoma driver genes or are COSMIC cancer genes with rearrangements breakpoints that occur in greater than four samples. d Copy number amplifications per sample. e Copy number loss and deletions per sample

Fig. 5

Associations between somatic mutations and relative telomere length. Plot of relative telomere length (log2 telomere ratio) per sample is shown at the top and below are plots showing SNV/indels, SV breakpoints, and copy number amplifications (CN ≥ 6, magenta), loss (CN1, light turquoise), deletion (CN0, dark turquoise), and copy neutral LOH (cnLOH) in telomere-associated genes or melanoma-associated genes

Fig. 6

Driver summary and actionable mutations. a Number of mutations (SNV, indel, CNV, SV) in mucosal melanoma driver genes (defined as SMGs and known oncogenic activating mutations or LoF mutations in tumor suppressor genes in other cancer types) identified in this study. b Mutations per sample in driver genes. c Samples that have mutations (SNVs, indels, copy number amplification or homozygous deletion or SV fusion gene) that are predicted to be responsive to inhibitors by Cancer Genome Interpreter. Each actionable mutation is colored by evidence level: case report, early trials, late trials, NCNN (National Comprehensive Cancer Network) guidelines, FDA (Food and Drug administration) guidelines. d Samples that have mutations (SNVs, indels, copy number amplification or homozygous deletion or SV fusion gene) that are predicted to be resistant or non-responsive to inhibitors by Cancer Genome Interpreter. Each actionable mutation is colored by evidence level: case report, early trials, late trials, NCNN guidelines, FDA guidelines