In the SEER database, it is relatively easy to identify secondary cancers that follow an initial diagnosis of cancer. However, the SEER public-use data does not explicitly capture relapse of the initial cancer. One can assume that organ-confined prostate cancer treated by radical prostatectomy will show no evidence of disease after treatment. Death due to prostate cancer in such cases can be assumed to follow a relapse of previously occult metastatic disease. We devised an estimate of the timing of these relapses and tabulated rates for both new primaries and relapses following radical prostatectomy for organ-confined prostate cancer.