P076 A multicentre evaluation of surveillance outcomes and phenotypic trends in carriers of CDKN2A pathogenic variants

Abstract

Pathogenic variants (PVs) in CDKN2A confer a markedly increased lifetime risk of cutaneous melanoma. Despite this, there is currently no UK consensus regarding optimal dermatological surveillance for individuals with CDKN2A PVs. This study was designed to describe the clinical phenotype of carriers of CDKN2A PVs at two UK tertiary centres and to evaluate surveillance outcomes associated with current practices. A retrospective review of medical records was undertaken for individuals with confirmed pathogenic CDKN2A variants. Fifty-six individuals with CDKN2A PVs were identified. The median follow-up was 9 years (range 2–20). Surveillance incorporated total-body photography and serial dermoscopic imaging with 6-monthly review. Overall, 79% of individuals had a history of melanoma or melanoma in situ (MIS). The median number of melanoma or MIS diagnoses per individual was 2 (range 0–11); 75% were diagnosed at stage 0 or IA. Clinical phenotype was heterogeneous. Total naevus counts varied widely (< 15, 29%; 15–49, 36%; 50–100, 20%; > 100, 9%), as did atypical naevus counts (0, 27%; 1–5, 22%; 6–10, 22%; 11–20, 16%; > 20, 7%). Incremental increase in naevus count categories (< 15, > 15, > 50, > 100) was associated with higher odds of melanoma [odds ratio (OR) 3.0, 95% confidence interval (CI) 1.15–11.4, P = 0.02], adjusting for age, sex and skin type. Age was independently associated with melanoma risk (OR 1.10 per year, 95% CI 1.04–1.19, P < 0.001). The presence of five or more atypical naevi was associated with a trend towards higher odds of melanoma (OR 5.9, 95% CI 0.96–66.9, P = 0.056). In conclusion, carriers of CDKN2A PVs demonstrate a high melanoma burden with frequent multiple primaries. Surveillance was associated with predominantly early-stage melanoma detection. Marked phenotypic heterogeneity was observed. Naevus count, atypical naevi and age may stratify melanoma risk and may therefore inform surveillance frequency. We highlight the need for standardized surveillance practice for this high-risk cohort.