Ancestry and somatic profile indicate acral melanoma origin and prognosis
Article metadata
Article Date: 18 February 2026
Article URL: https://www.nature.com/articles/s41586-025-09967-z
Article image: Figure 1 (full size)
Summary
This Nature study analysed 123 acral melanoma samples from 92 Mexican patients (predominantly Amerindian ancestry) using genotyping, whole-exome sequencing and RNA profiling. Key findings: a positive correlation between European ancestry and somatic BRAF-activating mutations; distinct transcriptomic clusters with prognostic value; many tumours are ‘quadruple wild-type’ for classic MAPK drivers; copy-number gains (for example, TERT, CRKL, CCND1) and frequent CDKN2A deletions; and mutational signatures dominated by clock-like signatures and SBS40a. The authors show that BRAF-mutant acral tumours have a transcriptional programme closer to non-acral cutaneous melanoma, suggesting differences in cell of origin or tumour classification. Transcriptomic clustering identified three groups with different recurrence and survival outcomes, with a proliferative/pigmentation cluster having the worst prognosis. The study highlights ancestry-driven somatic differences and the value of including underrepresented Latin American cohorts in cancer genomics.
Key Points
- Large Mexican cohort (123 samples, 92 patients) — 90% of genotyped samples had predominantly Amerindian ancestry (median ~81%).
- European genetic ancestry is positively correlated with somatic BRAF-activating mutations (mostly V600E).
- Classic melanoma drivers (BRAF, NRAS, KIT, NF1) are present in ~40% of samples; most tumours are triple/quadruple wild type.
- BRAF-mutated acral tumours show a transcriptomic profile closer to non-acral cutaneous melanoma (not explained simply by downstream BRAF signalling).
- Three RNA expression clusters were defined: an epidermal/immune-like cluster with better prognosis, a proliferative/pigmentation cluster with worse recurrence and survival, and an oxidative phosphorylation cluster with intermediate outcomes.
- Copy-number landscape shows more amplifications than deletions; recurrent amplifications include TERT, CRKL, GAB2, CCND1; CDKN2A deletions are frequent.
- Mutational signatures are dominated by SBS1, SBS5 and SBS40a; acral melanoma has lower SNV burden but higher structural variant burden vs UV-driven melanomas.
- Tumour driver mutations associate with higher recurrence; mutated tumours had higher recurrence and a tendency to worse overall survival.
- Study underscores the importance of ancestry and diverse cohorts for accurate genomic characterisation and for informing patient selection in clinical trials.
Content summary
The researchers performed stringent variant calling on FFPE-derived whole-exome data and RNA sequencing. They combined ancestry estimation from genotyping with somatic mutation, copy-number, mutational-signature and transcriptomic analyses. Logistic regression controlling for age, sex and total TMB revealed a statistically significant increase in odds of BRAF mutations with higher European ancestry. Copy-number analysis (GISTIC2) identified recurrent amplifications (TERT, CRKL, GAB2, CCND1) and frequent deletions (CDKN2A, ATM, TP53). Transcriptome consensus clustering (n=44 high-quality primaries) produced three biologically distinct groups linked to clinical features: an epidermal/immune cluster with smaller Breslow thickness and better outcomes, a mitotic/proliferative cluster with higher mitotic index and worse prognosis, and an oxidative phosphorylation cluster with intermediate risk. BRAF-mutant acral tumours scored more like cutaneous melanomas using a derived acral:cutaneous gene signature; this was replicated in an independent dataset. The authors discuss the possibility that some BRAF-mutant tumours classified clinically as acral may arise from melanocyte subtypes more similar to non-acral skin, with implications for classification and trial design. Limitations include FFPE sample artefacts, exome (not whole-genome) scope and a recruitment window spanning many years that may bias survival estimates.
Context and relevance
This work addresses two pressing issues: (1) underrepresentation of Latin American / Amerindian ancestry in cancer genomics and (2) the understudied biology of acral melanoma, which dominates melanoma cases in many non-European populations. The ancestry–somatic mutation link (European ancestry ↔ BRAF) is part of a growing literature showing germline background shapes tumour genomics. Clinically, the finding that BRAF-mutant acral tumours are transcriptionally cutaneous-like suggests molecular stratification could refine diagnosis and therapeutic decisions (for example, eligibility for BRAF/MEK-targeted approaches). The RNA clusters with prognostic signal also offer potential biomarkers for risk stratification and trial enrichment. For translational researchers, oncologists and genomic medicine teams, these results argue for including diverse cohorts when deriving biomarkers and designing trials so treatments benefit all populations equitably.
Why should I read this?
Because it actually matters — not just another dataset. If you care about why some acral melanomas behave differently or who benefits from BRAF-targeted therapies, this paper links ancestry to mutation patterns and shows measurable prognostic transcriptomic groups. Short version: ancestry influences tumour drivers, some acral BRAF+ tumours look more like ordinary skin melanomas at the RNA level, and expression clusters predict recurrence. If you’re in melanoma research, clinical trials or genomic medicine, this saves you the time of digging through raw data — the high-level takeaways are right here.
Author style
Punchy: This is a must-see for researchers and clinicians focussed on melanoma genomics. The study is one of the largest acral melanoma cohorts from a Latin American centre and delivers actionable insights — ancestry-linked BRAF prevalence, a cutaneous-like transcriptional subset of acral melanomas and expression clusters tied to prognosis. Read the methods if you’ll use these signatures or plan trials — the sample handling and stringent calling matter for reproducibility.