A retrospective analysis of cardiovascular outcomes in patients with prostate cancer treated with upfront gonadotropin-releasing agonist versus antagonists utilizing real-world data.

e17101 Background: Prostate cancer is commonly treated with androgen deprivation therapy (ADT). This is achieved using gonadotropin-releasing hormone (GnRH) agonists such as leuprolide or GnRH antagonists such as degarelix and relugolix. While GnRH agonists have been widely used for decades, antagonists offer more rapid testosterone suppression and avoid the initial testosterone surge associated with agonists. Although there is data to suggest that GnRH antagonists may have better cardiovascular safety compared to agonists, the available evidence is conflicting. Therefore, leveraging real-world data may provide important insights into outcomes across broader patient populations. Methods: We utilized the TriNetX global federated health research network which included 115 different healthcare organizations. By utilizing ICD-10 and ICD oncology codes, patients with prostate cancer were identified and separated into two cohorts based on time to initiation of either leuprolide (GnRH agonist arm), or degarelix or relugolix (GnRH antagonist arm) within 1 year of the index diagnosis event. Propensity score matching (PSM) was conducted based on age, race and stage of prostate cancer (Stage I to IV). Primary outcome measure of interest was major adverse cardiovascular events (MACE), a composite of myocardial infarction, cerebrovascular infarction and all-cause mortality. Cox-proportional hazards model for major adverse cardiovascular events (MACE) was performed with covariates including age, race, heart failure, personal history of malignant neoplasm, cerebral infarction, dementia, chronic pulmonary disease, liver disease, diabetes and chronic kidney disease. Results: 108,064 patients with prostate cancer treated with GnRH agonists and 12,242 such patients who received GnRH antagonists were identified. 12,242 patients were matched 1:1 based on PSM per cohort. Baseline characteristics including age and race were well balanced in the matched cohorts. MACE occurred in 2,554 (22.56%) of patients in the GnRH agonist arm versus 1,618 (15.02%) of patients in the GnRH antagonist arm HR 1.649, 95% CI 1.539-1.766, p < 0.0001). All-cause mortality was similarly higher in patients in the GnRH agonist arm (2,330 patients or 19%) versus the GnRH antagonist arm (1,589 patients or 13%) (HR 1.58, 95% CI 1.472-1.691, p < 0.0001). Cox-proportional hazards model showed statistically significant increase in MACE associated with treatment with GnRH agonists compared to GnRH antagonists (HR 1.16, 95% CI 1.11-1.21, p < 0.0001). Conclusions: Treatment with GnRH agonists was associated with significantly higher rates of MACE and all-cause mortality compared with GnRH antagonists among patients with prostate cancer. Prospective randomized studies are needed to inform individualized treatment decisions.