P214 No association found between metformin use and overall survival in adults with melanoma receiving immunotherapy: a global retrospective study of 3788 patients

Abstract

The research recommendations in the NICE melanoma guidelines highlighted the need for cohort studies evaluating whether coprescriptions such as metformin influence melanoma survival. Despite improved melanoma outcomes with immunotherapy, survival varies in routine care and patients often take concomitant medications with unknown implications. Our aim was to evaluate whether concomitant metformin use is associated with overall survival (OS) in adults with type 2 diabetes mellitus receiving immunotherapy for melanoma. This retrospective cohort study used data from the TriNetX Global Collaborative Network. Adults (≥ 18 years) with melanoma and type 2 diabetes mellitus treated with immunotherapy were included. Metformin exposure was any recorded metformin within 12 months before or after immunotherapy initiation; comparators had no metformin exposure during the same period. OS was measured from first immunotherapy administration to death. Kaplan–Meier survival curves were compared using log-rank tests and Cox proportional hazards models. Propensity score matching on age, sex, race and ethnicity was performed. Matched analyses were repeated for 1, 3 and 5 years after first administration of immunotherapy. In total 3788 patients from 8 countries were included (1294 concomitant metformin cohort and 2494 no metformin cohort). Propensity score matching yielded 1247 patients per cohort for analysis. In matched cohorts, OS (concomitant metformin cohort vs. no metformin cohort) was 79.1% vs. 80.1% at 1 year [hazard ratio (HR) 1.07, 95% confidence interval (CI) 0.90–1.29; log-rank P = 0.45], 64.3% vs. 61.6% at 3 years (HR 0.96, 95% CI 0.83–1.11; log-rank P = 0.58) and 56.2% vs. 54.2% at 5 years (HR 0.97, 95% CI 0.85–1.11; log-rank P = 0.64). There were no statistically significant differences in OS between the cohorts with concomitant metformin and no metformin at 1, 3 or 5 years after first initiation of immunotherapy. Limitations of the study include quantifying the exposure time and residual confounding. Further research is being conducted including matching on a glycated haemoglobin control and melanoma stage. This will help us to understand the broader impact of coprescriptions on melanoma survival outcomes.