Abstract
BACKGROUND: Aspirin and non-aspirin non-steroidal anti-inflammatory drugs (NA-NSAID) have been associated with improved survival in individuals with epithelial ovarian cancer (EOC); however, findings to date are inconsistent. METHODS: We conducted a registry-based cohort study evaluating survival following an incident invasive EOC diagnosis including individuals diagnosed between 2004-2018 (n = 4325; n = 2206 deaths; n = 1973 EOC deaths). Evaluated exposures were low-dose aspirin and NA-NSAIDs. Two primary post-diagnosis exposure windows were evaluated: fixed post-diagnostic baseline exposure ≤ 305 days after diagnosis (use, non-use) and updated "time-varying" exposure (never, past, current use; cumulative defined daily dose (DDD)). Pre-diagnostic exposure (use, non-use) was further evaluated. Multivariable Cox-proportional hazard models were used to estimate hazard ratios (HRs) and 95% confidence intervals [95% CIs]. The primary outcome was cause-specific survival. Restricted mean survival time (RMST) in exposure groups was estimated at 5 years following start of follow-up. RESULTS: Baseline post-diagnosis aspirin use was not associated with survival following an EOC diagnosis (e.g., use vs. no use: aspirin, HR = 1.02 [95% CI = 0.84-1.24]). Inverse associations were observed between current aspirin use post-diagnosis and survival in the time-varying exposure models (HR 0.68 [0.57-0.81]), and with higher post-diagnosis cumulative DDD of aspirin. Findings for NA-NSAIDs were less consistent. No associations were observed for pre-diagnostic use. Results for overall survival were similar to those for cause-specific survival. Compared to never use, post-diagnosis low-dose aspirin use was associated with a longer RMST (e.g., ever vs. never use, difference in RMST = 2.67 months). CONCLUSIONS: This study provides further evidence of a potential beneficial effect of post-diagnosis low-dose aspirin use for ovarian cancer survival.