Meta-Analysis and Systematic Review of the Genomics of Mucosal Melanoma

Abstract

Mucosal melanoma is a rare subtype of melanoma. To date, there has been no comprehensive systematic collation and statistical analysis of the aberrations and aggregated frequency of driver events across multiple studies. Published studies using whole genome, whole exome, targeted gene panel, or individual gene sequencing were identified. Datasets from these studies were collated to summarize mutations, structural variants, and regions of copy-number alteration. Studies using next-generation sequencing were divided into the “main” cohort (n = 173; fresh-frozen samples), “validation” cohort (n = 48; formalin-fixed, paraffin-embedded samples) and a second “validation” cohort comprised 104 tumors sequenced using a targeted panel. Studies assessing mutations in BRAF, KIT, and NRAS were summarized to assess hotspot mutations. Statistical analysis of the main cohort variant data revealed KIT, NF1, BRAF, NRAS, SF3B1, and SPRED1 as significantly mutated genes. ATRX and SF3B1 mutations occurred more commonly in lower anatomy melanomas and CTNNB1 in the upper anatomy. NF1, PTEN, CDKN2A, SPRED1, ATM, CHEK2, and ARID1B were commonly affected by chromosomal copy loss, while TERT, KIT, BRAF, YAP1, CDK4, CCND1, GAB2, MDM2, SKP2, and MITF were commonly amplified. Further notable genomic alterations occurring at lower frequencies indicated commonality of signaling networks in tumorigenesis, including MAPK, PI3K, Notch, Wnt/β-catenin, cell cycle, DNA repair, and telomere maintenance pathways. This analysis identified genomic aberrations that provide some insight to the way in which specific pathways may be disrupted. Implications: Our analysis has shown that mucosal melanomas have a diverse range of genomic alterations in several biological pathways.

Mucosal Melanoma: Pathological Evolution, Pathway Dependency and Targeted Therapy

“…It has been reported that silencing of PTEN cooperates with activated AKT to promote metastasis of melanoma ( 102 , 103 , 106 , 107 ), but MM does not show a weaker performance in metastasis than CM in the clinic possibly because there are other gene alterations in the PI3K pathway promoting invasiveness of tumor ( 108 ). In a cohort of 91 MM patients, 18% of cases show TSC1 loss-of-function mutations which plays a suppressive role in cellular proliferation initiated by mTOR ( 109 ) and a similar result also showed in a recently published meta-analysis review of MM ( 105 ). Due to the limited sample size, the alteration levels for TSC need to be further verified.…”

“…Compared with 12% of PTEN loss in CM where deletions and mutations both account for the changes ( 17 , 79 ), PTEN is deeply deleted in merely 6.0% of MM cases and has hardly any mutations ( 16 ) Furthermore, there is rare co-occurrence of deleted PTEN and amplified KIT that possibly implies that loss of PTEN or gain of KIT are redundant for the activation of the PI3K pathway in MM. This is also supported by the fact that besides KIT and PTEN , mutations in PI3K and AKT homologous of PI3K-AKT-mTOR are scarce (4.8% PIK3CA , 3.8% PIK3CG , and 4.8% AKT3 ) ( 105 ). It has been reported that silencing of PTEN cooperates with activated AKT to promote metastasis of melanoma ( 102 , 103 , 106 , 107 ), but MM does not show a weaker performance in metastasis than CM in the clinic possibly because there are other gene alterations in the PI3K pathway promoting invasiveness of tumor ( 108 ).…”

Association of Melanoma-Risk Variants with Primary Melanoma Tumor Prognostic Characteristics and Melanoma-Specific Survival in the GEM Study

“…A limitation could be insufficient power to detect associations of SNPs with lower minor allele frequencies (e.g., SLC45A2 rs16891982, MAF 0.017). Another limitation is that our study only included participants with cutaneous melanoma, not mucosal [ 76 , 77 ] or uveal melanomas [ 78 , 79 , 80 ], which seemingly have different genetic landscapes.…”

CDK4/6 Inhibitors in Melanoma: A Comprehensive Review

“…The prevalence of the amplification of CCND1 is controversial [28,29] and the frequency differs among melanoma types, with an established higher detection rate in acral melanoma [30][31][32]. Gene copy numbers and amplification levels are frequently associated with the expression of cyclin D, but, in contrast, the overexpression of the cyclin D does not necessarily follow gene expression alterations, suggesting the influx of other pathways [29].…”