Although multiple studies have addressed the prognostic importance of tumor differentiation in patients with clinically localized prostate cancer, few data are available in patients with metastatic disease. We evaluated and compared survival data in two groups of men with Whitmore stage D2 metastatic prostate cancer initially treated with hormonal therapy. A series of 76 patients with D2 metastatic disease were evaluated and treated at the National Cancer Institute (NCI) in conjunction with an additional cohort of 141 patients from the Louisiana State University School of Medicine (LSU). Pathological specimens were classified according to the Gleason score. Fifty-two (25%) of the combined NCI/LSU specimens had a Gleason score of 6 or less, 71 (34%) had a value of 7, and remaining 87 (41%) had scores between 8 and 10. The median PSA at the time of diagnosis for the NCI patients was 294.2 ng/ml. Time to treatment failure was defined as the time that a greater than 50% increase above nadir PSA was noted. In neither group was Gleason score correlated with overall survival. There was no association between the time to progression following hormone therapy and primary tumor Gleason score. The PSA concentration at the time of diagnosis was not correlated with the Gleason score for the NCI patients; however, there was an inverse correlation between pretreatment PSA level and time to progression following hormonal ablation. Gleason score does not appear to impact survival in metastatic prostate cancer. PSA as a marker of the biological behavior in metastatic disease may also be limited. These findings should be reevaluated in larger, better matched cohorts. Novel techniques such as serum proteomics, microarrays, and metastatic cell isolation methods may better predict outcome in advanced prostate cancer.
