Competing Risk of Specific Mortality in Prostate Cancer Patients in Brazil: A Retrospective Cohort Study

Abstract

Background Prostate cancer is among the most common malignancies worldwide and a leading cause of premature mortality in men. Although survival rates have improved globally, disparities persist. Here, we aim to analyze the risk factors associated with prostate cancer‐specific mortality among patients in a southeastern Brazilian state. Methods We conducted a retrospective cohort study using data from 10,556 patients diagnosed with prostate cancer between 2000 and 2016. Data were extracted from hospital‐based cancer registries linked to the state Mortality Information System. Patients were followed up for a minimum of 5 years and classified as alive, deceased from prostate cancer, or deceased from other causes. Subdistribution hazard ratios (SHRs) were estimated using Fine–Gray competing‐risks models. Results By the end of 2021, 6388 patients were alive, 1936 had died from prostate cancer, and 2232 had died from other causes. Older age increased prostate cancer‐specific mortality (SHR per 10 year increment = 1.098; 95% CI: 1.024–1.176), while distant metastasis was the strongest clinical predictor (SHR = 5.315; 95% CI: 4.676–6.041). Higher educational attainment remained statistically associated with lower prostate cancer‐specific mortality in the multivariable competing‐risks model (SHR = 0.767; 95% CI: 0.629–0.935), although the protective effect was attenuated after adjustment. Surgery (SHR = 0.382; 95% CI: 0.309–0.471) and radiotherapy (SHR = 0.477; 95% CI: 0.396–0.575) were also associated with lower cancer‐specific mortality, while hormone therapy remained associated with higher mortality, reflecting treatment selection among patients with a more advanced disease rather than a causal treatment effect. Conclusion Age, education, metastatic disease, and treatment modalities were significantly associated with prostate cancer‐specific mortality. These findings reinforce the importance for equitable access to early diagnosis pathways and curative treatment options and highlight the value of competing‐risks methods for accurately estimating prostate cancer outcomes in population‐based settings.