Long‐term oncological outcomes from a prospective cohort of patients with localised prostate cancer
Avi S. Baskin, Karen E. Hoffman, David F. Penson, Li‐Ching Huang, Zhiguo Zhao, Bashir Al Hussein Al Awamlh, Daniel D. Joyce, Alicia K. Morgans, Jeffrey J. Tosoian, Christopher J.D. Wallis, Michael Goodman, Ann S. Hamilton, Xiao‐Cheng Wu, Lisa E. Paddock, Antoinette Stroup, Brock B. O'Neil, Tatsuki Koyama, Daniel A. BarocasObjectives
To report long‐term oncological outcomes for men with clinically localised prostate cancer (PCa) treated with contemporary modalities in a population‐based United States cohort.
Patients and Methods
The Comparative Effectiveness Analysis of Surgery and Radiation (CEASAR) study prospectively enrolled men with clinically localised PCa from 2011 to 2012. Patients were stratified into two groups: favourable prognosis (clinical T stage [cT]1–T2a/bN0M0, prostate‐specific antigen [PSA] level ≤ 20 ng/mL, Grade Group 1–2) and unfavourable prognosis (cT2cN0M0, PSA level 20–50 ng/mL, or Grade Group 3–5). Main outcomes were PCa‐specific mortality (PCSM), composite progression to advanced disease (metastasis, PCSM event, or systemic therapy), and overall survival (OS). Cox regression models adjusted for demographic and clinical covariates.
Results
Of the 2604 men included, 73% were White, 15% Black, and 7% Hispanic. All hazard ratios (HRs) were adjusted for demographic and clinical covariates using multivariable Cox and Fine‐Gray models. In the favourable group, compared with surgery, external beam radiotherapy (EBRT; HR 1.5, 95% confidence interval [CI] 1.1–2.1, P < 0.01), brachytherapy (HR 2.1, 95% CI 1.3–3.3, P < 0.01), and active surveillance (HR 1.5, 95% CI 1.1–2.1, P = 0.02) were associated with higher all‐cause mortality; the 10‐year cumulative incidence of PCSM was ≤1.5% for all five strategies, and progression did not differ. In the unfavourable group, EBRT was associated with worse OS (HR 2.8, 95% CI 1.9–4.3, P < 0.01), with no adjusted differences in PCSM (10‐year cumulative incidence 3.5% after surgery vs 8.8% after EBRT) or progression across treatments.
Conclusion
In this population‐based cohort treated with contemporary modalities, the 10‐year risk of PCa death was low across all treatment strategies, including active surveillance. OS differences favouring surgery likely reflect residual confounding from baseline health and treatment selection. These findings underscore the importance of patient values and comorbidities in shared decision‐making for localised PCa.