HOUT-20. TIME-DEPENDENT ANALYSIS OF SELECTIVE SEROTONIN REUPTAKE INHIBITOR TREATMENT ON OVERALL SURVIVAL OF PATIENTS WITH GLIOBLASTOMA

Abstract

Abstract Glioblastoma (GBM) is the most common and aggressive form of primary brain tumor in adults. We recently investigated the hypothesis that treating GBM patients with psychosocial modifiers would be associated with improved overall survival (OS). Our study retrospectively analyzed 497 patients with GBM treated at Northwestern Medicine with or without selective serotonin reuptake inhibitors (SSRI) between the years 2000 and 2018. Information from the Northwestern Medicine Enterprise Data Warehouse was analyzed for baseline covariates including sex, age at diagnosis, type of surgery, and Charlson Comorbidity Index Score. Approximately one-third of analyzed patients were prescribed SSRIs, with highly variable treatment times. Several statistical methods were used to perform adjusted analyses including: (i) an extended Cox Proportional Hazards Model with SSRI as a time-dependent variable; (ii) a Cox Model using inverse probability weights; and (iii) a Cox Proportional Hazards model with landmark analyses. The hazard ratios (95% CIs) for each statistical model analyzing the association between SSRI treatment and OS were (i) 1.26 (0.97–1.63), (ii) 1.06 (0.8–1.4), and (iii) ranged from 1.01 (0.74–1.38) to 1.26 (0.75–2.09). Our analysis found no significant association between the time of SSRI treatment and GBM patient OS. Future work will study additional considerations for psychosocial modifier treatment and their potential effect(s) on GBM patient OS including: (i) confounders due to the extent of cancer treatment; (ii) comorbidities not associated with tumor burden; (iii) absolute leukocyte counts; and (iv) length of treatment time required for enhancing immune-mediated anti-GBM mechanisms.