Objectives: We assessed the ability of 3 clinical factors-stage, D'Amico risk stratification, and Roach formula-estimated nodal risk-to predict the pathologic outcomes in a modern cohort of prostate cancer patients.
Methods: Data were obtained from the Surveillance, Epidemiology, and End Results Program for 2004 and 2005. Men with adenocarcinoma of the prostate who had complete staging information were included. Patients were stratified according to the D'Amico criteria into low, intermediate, or high-risk groups using clinical T-stage, Gleason sum, and prostate specific antigen. Comparison was made with risk groups obtained after replacing clinical with pathologic T-stage. In addition, the Roach formula-predicted risk of positive lymph nodes was calculated and compared with the pathologic incidence of positive nodes.
Results: A total of 56,446 patients were included, of whom 13,135 had a prostatectomy with complete pathologic staging. Of 3476 patients with clinically palpable disease, who underwent prostatectomy, 2.5% were down-T-staged pathologically, whereas 34.7% were up-T-staged. Replacing clinical T-stage with pathologic T-stage in the D'Amico risk stratification resulted in the majority of men being up-risk-stratified to a higher risk group. Three hundred and forty-three patients had positive nodes. The patients with a Roach formula-predicted likelihood of positive nodes of 0% to 5%, 5.1% to 10%, 10.1% to 15%, and greater than 15% were found to have 0.2%, 0.5%, 1.2%, and 6.6% pathologic incidence of positive nodes, respectively.
Conclusions: Pathologic staging results in higher risk stratification than that predicted by clinical criteria in the majority of patients. Nodal positivity at diagnosis is uncommon in the current era, and the Roach formula overestimates the actual risk of node-positive disease.