Real-world evidence for factors associated with development of castration-resistant prostate cancer (CRPC).
Abstract
92 Background: Androgen deprivation therapy (ADT) is foundational for advanced prostate cancer (PCa); luteinizing hormone-releasing hormone (LHRH) agonists and gonadotropin-releasing hormone (GnRH) antagonists are widely used. ~20% of PCa patients on LHRH analogs develop castration-resistant prostate cancer (CRPC) within 3 years. We evaluated factors associated with CRPC onset and time to CRPC onset for agonists vs. antagonist using real-world dataset. Methods: Data were collected from an EMR database purchased from Decision Resources Group. Analysis set included PCa patients who received ≥1 ADT 1991-2020 and excluded those who received both LHRH agonist and a GnRH antagonist and/or developed CRPC on the same day/before ADT initiation. CRPC onset date was defined as the earlier of 2 options: CRPC diagnosis or first CRPC therapy (standard of care treatments e.g., androgen pathway inhibitors, immunotherapy, chemotherapy, radiotherapy). Associations between CRPC and each available factor were evaluated using MV Cox regression. Time to CRPC onset was assessed for agonists vs. antagonist. Results: 41,379 patients were analyzed. Factors associated with MV HR >1.4, p<0.001 were oncology setting, metastatic disease, absence of diabetes, and no statin use (Table 1). Antagonist use had an adjusted HR of 1.4 (95% CI 1.06-1.85, p=0.02). Median time to CRPC onset after ADT initiation is 10.9 months for antagonist (n=238) vs. 17.3 months for agonists (n=5,231). Conclusions: Treatment-related (oncology setting; antagonist use) and comorbidity factors (older age, no diabetes, no statin use) were associated with CRPC onset. Oncology setting and antagonist use may reflect practice patterns in higher risk individuals. The association between diabetes and CRPC risk is under further evaluation e.g., whether antecedent diabetes therapies confer protection. Hazard ratio (95% CI) and P-value of CRPC for top 10 significant factors 1 in multivariable analysis. Multivariable (n=9,554) 2,3 Factors HR (95% CI) P-value Oncology vs Urology (Setting) 3.04 (2.56-3.59) <0.001 W vs W/o Metastasis (Baseline) 2.20 (1.58-3.05) <0.001 W/o vs W Diabetes 4 1.45 (1.22-1.73) <0.001 W/o vs W Statin 4 1.43 (1.25-1.63) <0.001 Antagonist vs Agonist 1.40 (1.06-1.85) 0.02 W/o vs W Hypertension 4 (HTN) 1.17 (1.01-1.36) 0.041 Decreasing ADT Exposure per Year 1.09 (1.05-1.13) <0.001 Increasing Age per Year (Older vs Younger) 1.02 (1.02-1.03) <0.001 1 Significant for multivariable analysis. 2 Excluded patients with CRPC on the same day/before ADT Start. 3 Excluded patients with both agonist and antagonist use; without BMI, PSA, and age data; non-white/black race; with unknown urology/oncology or with both urology/oncology setting. 4 Has taken statins/hypertension/diabetes medication or diagnosed with hypertension/diabetes disease.
Article Details
Journal Info
Journal of Clinical Oncology
Lippincott Williams & Wilkins
Authors (5)
Charles J. Ryan
Memorial Sloan Kettering Cancer Center, New York, NY
Kristine Peregrino Lacuna
Memorial Sloan Kettering Cancer Center, New York, NY
Lena Yoo
Tolmar, Inc., Buffalo Grove, IL
Zuzanna Szmukier
Tolmar Pharmaceuticals Canada Inc., Waterdown, ON, Canada
Deborah M Boldt-Houle
Tolmar, Inc., Buffalo Grove, IL