Predicting the risk of lymph node involvement using the pre-treatment prostate specific antigen and gleason score in men with clinically localized prostate cancer

Purpose: To evaluate the predictive value of an empirically derived equation for identifying patients with clinically localized prostate cancer at low and high risk for harboring occult lymph node metastasis. Methods and Materials: A simple equation for estimating the risk of positive lymph nodes was empirically derived from a nomogram published by Partin et al. demonstrating the value of combining the pre-treatment prostate specific antigen and Gleason Score in predicting the risk of lymph node metastasis for patients with clinically localized prostate cancer. The risk of positive nodes (N+) was calculated using the equation; N+ = 2 3(PSA) + (GS - 6) × 10 where PSA and GS are the pre-treatment prostate specific antigen and Gleason Score respectively, and the calculated risk is constrained between 0-65% for a PSA ≤ 40 ng/ml (as in the nomogram). To test the general applicability of this equation, we reviewed the pathologic features of 282 of our patients who had undergone a radical prostatectomy. Results: Based on 212 patients for whom the pre-operative prostate specific antigen's and Gleason Scores were available, we identified 145 patients with a calculated risk of positive nodes of < 15%, (low risk group) and 67 patients with a calculated risk as ≥ 15% (high risk group). The observed incidence of positive nodes was 6% and 40% among the low and high risk groups respectively (p < 0.001). When used alone neither clinical stage, pre-treatment prostate specific antigen nor the pre-treatment Gleason Score was as useful in identifying the largest low and high risk groups. Conclusion: Using the equation described we confirmed the general applicability of the nomogram reported by Partin et al. and identified patients at low and high risk for lymph node involvement. Based on these data we have adopted a policy of omitting whole pelvic irradiation in patients identified as low risk.