Survival Outcomes of a Large Cohort of Acral Melanoma Patients Treated at a South African Referral Hospital.

Abstract

Acral melanoma (AM) is the most common type of melanoma arising in people with skin of color. AM is often diagnosed late and associated with poor outcomes. Melanoma outcomes are also impacted by socioeconomic status. Although uncommon, AM is a public health concern in South Africa because of its epidemiology and association with health access issues, which predispose to late diagnosis. South African patients are managed based on staging systems and treatment guidelines developed for other populations. This cohort study aimed to determine the survival outcomes in a cohort of South African AM patients and how these were associated with demographic, clinical, pathological, and management data. The study included patients diagnosed at a referral hospital between 1 January 2010 and 31 December 2021. Analysis occurred in 2022 and 2023. The main outcomes of interest were survival time in months (overall and progression-free, OS and PFS). Survival outcomes were analyzed using Kaplan-Meier survival curves. Survival probabilities were compared between subgroups using log-rank tests. Univariate and multivariable analyses were performed using the Cox proportional hazards models to assess factors associated with survival. Ninety-one patients were included in the analysis. After a median follow-up of 28 months (range: 0-151 months), 48 patients (52.7%) had died. The 3- and 5-year survival rates for the entire cohort were 64.8% and 56.0% respectively. Notably, the OS and PFS were not affected by the population group (p value = 0.628, not significant). The examination of OS and PFS by the clinical stage group demonstrated proportional hazard. Although SLNB comprised a small group, the results appear to be prognostically valid, specifically for OS. The results indicate that the AJCC eighth edition staging is broadly applicable to AM in this population; however, a rigorous comparison was not possible. SLNB appears to be prognostically valid. No difference in survival outcomes by population group was shown.