IMPACT OF COMPETING MORTALITY ON THE CANCER-RELATED MORTALITY IN LOCALIZED PROSTATE-CANCER

Objectives. To evaluate the impact of competing mortality and extended observation time on the cancer-related mortality in localized prostat e cancer (PC). Methods. A comparison was made between two theoretical populations of prostate cancer patients. Both populations had a slowly increasing mortality due to PC, corresponding to a 10-year cause-spec ific mortality of 15%, One population (A) experienced a high competing mortality reaching 50% after 10 years, corresponding to series on def erred treatment, The other population (B) experienced a low competing mortality, 10% after 10 years, corresponding to series on radical pros tatectomy. The impact of these different competing mortality rates on the absolute number of patients succumbing to PC and the effect of ext ended followup to 15 years was assessed. Results. The ultimate risk of death from PC after 10 years was 8% in group A and 12.3% in group B. When the observation time was extended to 15 years, group A had a 16.5 % risk of cancer death and group B had a 35.3% risk. The PC mortality increased twofold between 10 and 15 years in group A (8% versus 16.5%) and threefold in group B (12.3% versus 35.3%). Conclusions. Low cause -specific mortality rates at 10 years of follow-up in series on deferr ed treatment comprising older patients with high competing mortality c annot be extrapolated to younger patients with a low competing mortali ty. Long expected survival is associated with a considerable cancer-re lated mortality at 15 years (35%) despite favorable outcome at 10 year s.