Abstract: PD03-05
Sources of Funding: None

Introduction

Accurate assessment of life expectancy (LE) is critical to appropriate case selection for men with prostate cancer. We previously reported the age-adjusted Prostate Cancer Comorbidity Index (PCCI), a LE prediction tool that uses a weighted score incorporating age and comorbidities to estimate 2, 5, and 10-year mortality in men with prostate cancer. We sought to operationalize the PCCI for clinical application using claims data and externally validate it across a nationally representative sample. We then compared its ability to identify patients at risk for overtreatment with the age-adjusted Charlson comorbidity index.

Methods

We sampled 181,209 men with prostate cancer diagnosed from 2000 to 2011 in the Veterans Affairs healthcare system. We used claims data within 12 months of biopsy to determine comorbidities at diagnosis. We used Kaplan-Meier analysis to plot overall survival and multivariable Cox proportional hazards analysis to assess risk discrimination between PCCI and Charlson score subgroups. We then compared the number of men with <10-year LE who were treated with surgery or radiation between the two indices.

Results

Kaplan-Meier analysis showed a stepwise increase in risk of overall mortality with increasing PCCI score (Figure). Ten-year mortality among men with PCCI scores of 1-2, 3-4, 5-6, 7-9, and 10+ was 26%, 36%, 41%, 52%, and 69%, respectively. Multivariable models showed excellent risk discrimination with hazard ratios of 1.22 (95%CI 1.18-1.27), 1.69 (95%CI 1.61-1.76), 2.08 (95%CI 2.00-2.17), 2.88 (95%CI 2.76-3.00), 4.50 (95%CI 4.32-4.69) for PCCI scores of 1-2, 3-4, 5-6, 7-9, and 10+, respectively. The PCCI identified 30,610 men with LE <10 years (<50% median survival at 10 years) vs. 25,455 men in the Charlson index. Furthermore, the PCCI identified significantly more men with <10-year LE who were overtreated with surgery or radiation compared with the Charlson index: 12,531 (41%) vs. 7,098 (28%) (p<0.0001).

Conclusions

The age-adjusted Prostate Cancer Specific Comorbidity Index (PCCI) showed excellent prognostic utility across a nationally representative sample of men with prostate cancer. It was superior to the Charlson index in identifying men at risk for overtreatment due to limited LE. _x000D_ _x000D_

Funding

None