The impact of subsequent metastasis on survival in Medicare patients with prostate cancer.

Abstract

94 Background: This study assessed the impact of subsequent metastasis on survival in Medicare prostate cancer (PC) patients initially diagnosed with locoregional disease. Methods: Using data from the United States Surveillance, Epidemiology, and End Results (SEER) cancer registry linked to Medicare claims, we identified patients diagnosed with locoregional PC between 2000 and 2011, age ≥ 66 at diagnosis, and who first had a diagnosis of metastasis ≥ 4 months after PC diagnosis (cases). Cases were matched to controls (patients without metastasis) in a 1:4 ratio to assess the incremental impact of developing metastasis. For each control, the timeline to develop metastasis was matched to the cases. Kaplan-Meier (K-M) analysis was used to compare all-cause, cancer-specific, and other-cause mortality between cases and controls. Cox proportional hazards regression was used to adjust for other factors associated with all-cause mortality. Results: There were 10,370 cases and 39,200 controls. Mean age at baseline was 79 years in each group. Among the cases, the median time to first metastasis was 37 months, 85% had bone metastasis, and the median survival time (months) after metastasis was 18 months compared to 118 months for controls (P < 0.0001: Log-Rank). In K-M analysis, metastasis was associated with significantly (P < 0.0001: Log-Rank) higher all-cause, cancer-related, and other-cause mortality. In multivariable survival analysis, metastasis was associated with increased all-cause mortality (HR = 4.6, 95% CI = 4.4-4.7, P < 0.0001). High risk disease (based on Gleason score, comorbidity Index, and performance status) and delayed diagnosis of metastases were associated with worse survival. Conclusions: Development of metastasis in elderly PC patients diagnosed with locoregional disease significantly increases mortality.