EPID-20. The effect of medical comorbidities on survival in GBM: a database study using computed phenotype algorithm for rapid, reliable patient identification

Abstract INTRODUCTION Recent clinical trials have demonstrated some improved overall survival of glioblastoma (GBM) compared to historical trials, but this data is often criticized for being homogenous and not easily translated to real-world settings. The discrepancy between trial outcomes and clinical experience emphasizes the need for more investment in real-world evidence with population-based, non-exclusionary real-word data studies to complement clinical trial data and identify ongoing trends in patient characteristics and outcomes. This real-world data can also be used to identify more specific risk factors for poor outcomes. METHODS Using a computed phenotype, we identified 1,630 GBM patients from the OneFlorida+ CRN database from 2011-2021. The effect of medical comorbidities and the Charlston Comorbidity Index on overall survival was evaluated. RESULTS Patients with comorbidities of connective tissue/ rheumatic diseases, liver disease, or diabetes with complications had the lowest median survival (1.51, 4.11, 4.7 months, respectively) while those with seizures, hypertension, and dementia had the longest median survival (14.08, 14.84, 15.72 months, respectively). Older age at diagnosis (>65 years) (Hazard Ratio 2.52), paraplegia or hemiplegia (HR 1.47), diabetes without complications (HR 1.41), liver disease (HR 1.99), and hyperlipidemia (HR 0.69) were all found to be significant predictors of mortality on multivariate cox analysis. Age-adjusted Charlston Comorbidity Index and Area Deprivation Index were not significantly associated with mortality risk. CONCLUSION An estimated 20% of GBM patients participate in clinical trials, limiting the generalizability of results. Thus, it is imperative to continue to work towards obtaining high-volume real-world data to improve our understanding of this devastating condition. Here, using this real-world cohort, we identified several comorbidities that confer worse or better survival for patients with GBM. This data can be used to improve patient stratification and counseling.