Our results show that by using the nerve-sparing radical retropubic prostatectomy, potency can be preserved in the majority of appropriately selected patients without compromising the adequacy of tumor excision. However, proper patient selection is important. Patients with focal, well-differentiated tumors, especially young patients with stage A or B1 tumors, are ideal candidates. In patients with more extensive and less well-differentiated tumors, there is a higher risk of incomplete tumor excision. Although we suspect that the adequacy of tumor excision is determined more by the extent of the tumor than by the technique of radical prostatectomy used, we believe that nerve-sparing surgery should be used with great caution, if at all, in patients with extensive or high-grade tumors. In these patients, microscopic extracapsular tumor extension is extremely common, can be impossible to detect at the time of operation, and is less likely to be adequately encompassed by nerve-sparing techniques. On the other hand, our current data provide little evidence that excision of the neurovascular bundles is beneficial. It is possible that more extensive resections will not materially alter the incidence of positive margins or cure rates. Finally, it might be argued that all forms of radical prostatectomy are inappropriate for patients with poorly differentiated clinical stage B2 prostate cancer for whom there are no really effective treatment options. We continue to recommend radical prostatectomy for these patients based on the finding that patients with clinical stage B2 disease who have organ-confined tumors can be expected to have excellent long-term disease-free survival rates similar to those of clinical stage B1 patients. In the remaining patients who are clinically understaged, the prospects for the minimal microscopic tumor remaining being controlled with adjunctive radiation therapy may be better than those of controlling the bulky primary tumor with radiation therapy alone. This hypothesis will need to be tested in a randomized clinical trial.