Association of medication (Med) and antibiotic (Abx) use with response and survival in advanced melanoma (MEL) receiving PD-1 inhibitors

9572 Background: Retrospective studies suggest that various med could dichotomous effects in regards to immunotherapy. These include adverse (antibiotics) and positive (aspirin, beta-blockers) influences. To evaluate potential additive or detrimental effects of various med in patients (pts) receiving PD-1 immunotherapy, we performed a retrospective evaluation of med intake in 172 pts with stage IV cutaneous MEL focusing on aspirin (asp), antacid (ant), antibiotic (abx), bisphosphonate (bisp), metformin (met) and statin (stat) intake. Methods: Pts with stage IV cutaneous MEL who received anti PD-1 therapy at the University of Pittsburgh between 2014-2018 were included in this analysis. PD-1 blockade was continued until progression or intolerable toxicity. Tumor assessment was performed at baseline and every 12 weeks and response classified per RECIST v1.1. Clinical and demographic data were obtained. Med intake was documented based on chart review in all pts. Intake was confirmed by analyzing at least one other note from a non-oncological provider. Descriptive statistics were created for all covariates. Kaplan Meier and Cox proportional hazard regression were performed to assess how categorical variables related to response (ORR), overall survival (OS) and progression free survival (PFS) measured in months (mths). Results: 172 pts with advanced MEL were evaluated. Asp, ant, abx, bisp, met and stat use was documented in 62, 82, 29, 4, 15 and 57 pts respectively. ORR was not significantly related to intake of asp, ant, bisp, met and stat use; although ORR was lower in pts who received abx (p=0.0328). There was no significant difference in PFS and OS in pts who received asp, ant, bisp, met and stat. In patients who received abx compared to those who did not, median PFS (16.6 mths vs. 19.8 mths) and median OS (23.8 mths vs. 35.4) were both lower. Abx use did not interact with other meds. Conclusions: In this retrospective series of advanced MEL pts treated with PD-1 blockade, abx use was adversely associated with response to PD-1 blockade. Abx use was also associated with poorer PFS and OS. Conversely, neither a positive nor negative association with ORR, PFS and/or OS was seen with asp, ant, bisp, met and stat use. These results validate prior studies suggesting that abx use is associated with worse outcomes in pts receiving PD-1 blockade possibly by mediating intestinal dysbiosis.