To the Editor.
—Dr Partin and colleagues1 recently updated their experience using prostate-specific antigen (PSA) level, clinical stage, and tumor grade for predicting the pathological stage of prostate cancer determined at the time of radical prostatectomy.The authors are to be commended for this herculean effort. However, in examining our own series of radical prostatectomies, our data did not correspond to the outcomes listed in the article. Focusing the analysis of tumors with a Gleason score of 6 or 7, clinical stage of T2b, and PSA levels ranging between 4 and 10 ng/mL, the chance of having organ-confined disease was greater than that reported by Partin et al (70% vs 38% for tumors with Gleason scores of 6 and 50% vs 22% for tumors with Gleason scores of 7). This may be because of the high rate of interobserver variability that exists in clinical staging based on the digital