tatectomy. Based on our data, 66% of similarly staged patients remained disease free 5 years after surgery, with approxi¬ mately 50% of the patients who had cancer extending beyond the prostate capsule also remaining free of cancer recurrence 5 years after surgery.It has been shown that patients with lower-grade pros¬ tate cancer (Gleason score <6) can benefit from radical prostatectomy even in the presence of seminal vesicle inva¬ sion or locally advanced disease.3 The problem with the TNM staging system is that tumor volume, the status of the surgical margins, and the amount of extracapsular exten¬ sion are not factored into the cancer stage, all of which are additional prognostic factors in determining the risk of re¬ lapse following radical prostatectomy.4 In addition, it has been shown that patients with similar TNM staging who have a biochemical recurrence are not at equal risk of devel¬ oping a subsequent clinical recurrence (and presumably subsequent death).5 In summary, we recommend caution in the application of these nomograms to clinical practice, since no adverse prognostic factor guarantees failure of definitive therapy. Furthermore, we would not deny a patient poten¬ tially curative therapy based on these nomograms since our data show that many patients, even in the presence of ex¬ tracapsular extension, will remain clinically and biochemi¬ cally free of evidence of disease. WJ. Interexaminer variability of digital rectal examination in detecting prostate cancer. Urology. 1 9 9 5 ; 4 5 : 7 0 -7 4 .3. Naitoh J, Dorey F, de Ker ni on JB.PSA recurrence patterns following radical prostatectomy for stage pT3c disease. J Urol. 1997;157:803A. 4. Stamey TA, McNeal JE, Yemoto CM, Sigal BM, Johnstone IM. Gleason sums of 7 lose valuable prognostic information in comparison to estimates of percent grade 4 and 5 cancer. J Urol. 1997;157:794A. 5. Patel A, Dorey F, deKernion JB. Biochemical and clinical recurrence after radical retropubic prostatectomy: utility of the log slope PSA. J Urol. 1997;157:447A.To the Editor.\p=m-\ Dr Partin and col l eagues1 have ignored the race of the patient in reporting their nomograms that are intended to predict the pathological stage of clinically localized prostate cancer. Not only does this limit the generalizability of the nomograms, but it also influences their accuracy. Black race has been shown to be an adverse prognostic factor for recurrence following radical prostatectomy, even after multivariate adjustment for the 3 factors that the nomograms use in predicting stage (serum PSA level, clinical stage, and tumor grade in biopsy).2 The prostate cancer death rate is 52% higher for black Americans than white Americans, and the 5-year relative survival rate is 16% lower, despite the fact that the percentage of blacks that present with "localized" disease (53%) is similar to whites (59%).3 Compared with white men, the percentage of black men with margin-positive disease is significantly higher (50.5% vs 38.4%; P=.