Risk assessment for prostate cancer metastasis and mortality at the time of diagnosis

Abstract

Background: Although many tools for the assessment of prostate cancer risk have been published, most are designed to predict only biochemical recurrence, usually after a single specified treatment. We assessed the accuracy of the Cancer of the Prostate Risk Assessment (CAPRA) score, which was validated previously to predict pathological and biochemical outcomes after radical prostatectomy, to predict metastases, prostate cancer-specific mortality, and all-cause mortality.

Methods: We studied 10 627 men with clinically localized prostate cancer in the Cancer of the Prostate Strategic Urologic Research Endeavor registry, who underwent primary radical prostatectomy, radiation therapy (external beam or interstitial), androgen deprivation monotherapy, or watchful waiting/active surveillance, and had at least 6 months of follow-up after treatment. CAPRA scores were calculated at diagnosis from the prostate-specific antigen level, Gleason score, percentage of biopsy cores that were positive for cancer, clinical tumor stage, and age at diagnosis. Survival was studied with Kaplan-Meier analyses. Associations between increasing CAPRA scores and bone metastasis, cancer-specific mortality, and all-cause mortality were examined by use of proportional hazards regression, with adjustment for primary treatment; for all-cause mortality, the analysis also included adjustment for age and comorbidity. Accuracy of the CAPRA score was assessed with the concordance (c)-index.

Results: Among the 10 627 patients, 311 (2.9%) men developed bone metastases, 251 (2.4%) died of prostate cancer, and 1582 (14.9%) died of other causes. Each single-point increase in the CAPRA score was associated with increased bone metastases (hazard ratio [HR] for bone metastases = 1.47, 95% confidence interval [CI] = 1.39 to 1.56), cancer-specific mortality (HR for prostate cancer death = 1.39, 95% CI = 1.31 to 1.48), and all-cause mortality (HR for death = 1.13, 95% CI = 1.10 to 1.16). The CAPRA score was accurate for predicting metastases (c-index = 0.78), cancer-specific mortality (c-index = 0.80), and all-cause mortality (c-index = 0.71).

Conclusions: In a large cohort of patients with clinically localized prostate cancer who were managed with one of five primary modalities, the CAPRA score predicted clinical prostate cancer endpoints with good accuracy. These results support the value of the CAPRA score as a risk assessment and stratification tool for both research studies and clinical practice.

Figure 1

Kaplan–Meier plots of metastasis-free interval. A) Stratified by Cancer of the Prostate Risk Assessment (CAPRA) score (ie, CAPRA scores of 0 through 8–10). B) Stratified by grouped CAPRA scores (ie, CAPRA scores of 0–2, 3–5, and 6–10). Numbers of patients at risk are given at 4-year intervals. Table 4 presents the numerical results of the same analysis, including the metastasis-free interval estimates with 95% confidence intervals at 5 and 10 years.

Figure 2

Kaplan–Meier plots of prostate cancer–specific survival. A) Stratified by Cancer of the Prostate Risk Assessment (CAPRA) score (ie, CAPRA scores of 0 through 8–10). B) Stratified by grouped of CAPRA scores (ie, CAPRA scores of 0–2, 3–5, and 6–10). Numbers of patients at risk are given at 4-year intervals. Table 4 presents the numerical results of the same analysis, including the prostate cancer–specific survival estimates with 95% confidence intervals at 5 and 10 years.

Figure 3

Kaplan–Meier plots of overall survival. A) Stratified by Cancer of the Prostate Risk Assessment (CAPRA) score (ie, CAPRA scores of 0 through 8–10). B) Stratified by grouped CAPRA scores (ie, CAPRA scores of 0–2, 3–5, and 6–10). Numbers of patients at risk are given at 4-year intervals. Table 4 presents the numerical results of the same analysis, including the overall survival estimates with 95% confidence intervals at 5 and 10 years.