BRAF mutation status in primary and metastatic melanomas in two regions with differing potential ultraviolet radiation exposure

Abstract

Background Melanoma is seen as a heterogeneous molecular entity, with solar ultraviolet radiation (UVR) and BRAF mutation status being important determinants. Aim To study primary and metastatic melanomas from two UVR-distinct regions to elucidate correlations between prognostic predictors, UVR and BRAF mutation status. Methods Extended BRAF testing for 9 mutations was obtained for 95 primary melanomas [Lebanon (LB) n = 55, Pakistan (PK) n = 40)] and 65 metastatic melanomas (LB n = 36, PK n = 29). Collected data included patient age and sex, melanoma size and anatomical location, prognostic parameters and solar elastosis grade for primary melanomas. For metastatic melanomas, site of metastasis, magnitude of necrosis and degree of pigmentation were assessed. Cumulative 21-year averages of potential UVR exposure for Lebanon (110 kJ/m2/year) and Pakistan (128 kJ/m2/year) were derived from the National Center for Atmospheric Research databases. Results BRAF mutation status was obtained for 146/160 cases (91.3%). Overall mutation rate was 24/88 (27.3%) in primary and 25/58 (43.1%) in metastatic melanoma. V600E was the predominant mutation in 21/24 (87.5%) of primary and 23/25 (92%) of metastatic melanomas. A 60% discordant mutation rate was identified; of three patients, two lost the mutation in the metastasis and one gained it. The relative incidence of BRAF mutation with potential UVR exposure showed a similar trend in primary (low vs. high UVR: 32.1% vs. 20.0%) and metastatic (57.1% vs. 21.7%) melanomas (P < 0.05). Predictors of BRAF mutations were trunk location and epithelioid and mixed cytology for primary and subcutaneous metastasis, low UVR exposure and absence of pigmentation for metastatic melanomas (P < 0.05). BRAF-positive status in primary melanomas was predicted by multivariate binary logistic regression with reasonable accuracy (C-statistic = 0.67, 95% CI 0.530-0.81 with one independent predictor, namely, epithelioid cytology (OR = 5.11, 95% CI 1.38-8.88, P = 0.01). In metastatic melanomas, high UVR (OR = 0.21, 95% CI 0.06-0.07; P < 0.01) was an independent negative predictor of BRAF mutation. Conclusions We have documented the rate of different BRAF mutation types in a Lebanese and Pakistani cohort, and assessed correlations with prognostic markers and potential UVR exposure. © 2014 British Association of Dermatologists.