Utilization and survival outcomes between definitive versus conservative treatments among elderly men with high-risk prostate cancer.

Abstract

204 Background: Prostate cancer (PCa) is the second leading cause of cancer death in men > 80 years old. However, studies have shown that older men are less likely to undergo curative treatment for localized PCa, possibly due to competing comorbidities or inability to accurately estimate life expectancy. Herein, we investigate utilization trends and survival outcomes amongst guideline-supported treatment options in elderly men with high-risk PCa in the United States. Methods: Men ≥ 80 years diagnosed with high-risk PCa (cT3-4 or Gleason 8-10 or PSA > 20) between 2004-2016 were analyzed from the National Cancer Database. Those missing risk-stratification or treatment data were excluded. Eligible patients were grouped based on their primary treatment modality: no treatment (observation), androgen deprivation therapy (ADT) alone, radiation therapy (RT) alone, RT + ADT, or radical prostatectomy (RP). Cochran-Armitage was used to evaluate treatment trends over time, and multivariable logistic regression was used to identify sociodemographic predictors of treatment. Overall survival (OS) between treatments was evaluated using Kaplan-Meier, log-rank, and multivariable Cox proportional hazards. Results: With a median follow up of 42 months, 19,920 men were eligible for analysis. The most utilized treatment modality was RT+ADT (37.2%), followed by ADT alone (29.4%), observation (23.9%), RT alone (7.8%), and RP (1.7%). There was a significant increase in use of RT+ADT and RT alone (p < 0.001) and decrease in use of ADT alone and observation ( p< 0.001); no change was seen in RP use. There was no OS difference between observation versus ADT alone (aHR 1.04, 95% CI 0.99-1.09, p = 0.11). Definitive local treatment was associated with improved OS compared to ADT alone (RT+ADT: aHR 0.48, 95% CI 0.46-0.50, p < 0.0001; RT alone: aHR 0.54, 95% CI 0.50-0.59, p < 0.0001; RP: aHR 0.50, 95% CI 0.42-0.59, p < 0.0001). Black men and uninsured status were independently associated with lower likelihood of undergoing definitive treatment (i.e. RT+ADT, RT, or RP). Conclusions: For men ≥ 80 years old with high-risk PCa in this large US registry, definitive local therapy using RT +/- ADT or RP was associated with a 50% reduction in overall mortality compared to observation or ADT alone. Less than half of men in this time period underwent a definitive treatment, and Black and uninsured men remained at particularly high risk of undertreatment.