Randomized Phase II Trial Evaluation Of Erectile Function After Attempted Unilateral Cavernous Nerve-Sparing Retropubic Radical Prostatectomy
The difficulties were several-fold: the nerve-sparing alone may account for potency, patients requesting graft may have different baseline function, and patients requesting grafts may have different rates of "penile rehabilitation" compliance. Despite these uncertainties, many patients were willing to pay out-of-pocket for a plastic surgeon to place a graft with the rationale that they had that one opportunity to improve their potency preservation, while still having a nerve resected if their surgeon felt exptraprostatic disease was a significant risk. In fact, our center has experience with another 180 bilateral/unilateral graft cases in addition to the 107 randomized to this trial.
The trial was designed to enroll potent men 65 years and younger who were recommended to have a unilateral nerve sparing (UNS) RP due to the extent of biopsies. This was a popular management option for the late 1990's as the concept of sparing a nerve on the side of a positive biopsy (sextant) was very early. The trial was randomized with a 3:2 ratio graft to control, with the incentive that if randomized to the graft, the plastic surgery fees were waived. Many men continued to request an off protocol graft and pay the out of pocket expenses, and in the later years we stopped offering this option due to the importance of finishing this trial. Both groups of men were counseled to participate in an aggressive penile rehabilitation program to include Sildenafil, vacuum erection device, and injections. The statistical design assumed a control rate of potency of 40% at 2 years, and a graft potency rate of 60% at 2 years - 200 patients would provide 80% power. The trial was designed to be a phase II single institution trial that if possible would be expanded into a phase III multi-institutional trial.
At an interim analysis of 107 patients, enough patients had 2 year follow-up to perform a planned statistical test for futility, which in fact showed a > 95% chance that the trial would be negative. In addition, accrual to the trial decreased to 2/month with the rising feasibility of performing a bilateral nerve sparing procedure.
The results reported in this paper show the results of these 107 patients plus an additional year of follow-up from stopping the trial. Because of the multiple events possible after enrolling a patient in this trial that affect the denominator, we reported the results in 3 ways. Regardless, there were 32 potent patients in the graft arm and 14 in the control arm. If you compare strictly by intent to treat, the comparison is 32/66 (48.5%) vs. 14/41 (34%), p-0.271. If you only look at patients completing the protocol to 2 years, the comparison is 32/45 (71%) versus 14/21 (67%), p=0.777. If you compare all patients still on the protocol, including those not potent with < 2 years follow-up, the comparison is 32/56 (56%) vs. 14/32 (44%), p=0.164. The differences in denominator are created by patients with less than 2 years follow-up but also by patients leaving the protocol because they refused surgery, or required hormonal therapy, or withdrew. Additional findings included no difference in IIEF scores and no differences in 1 year urinary function/bother scores. Both groups had surprisingly poor compliance with penile rehabilitation that declined further with follow-up. The trial was not designed to test our ability to predict extra capsular extension, but it was interesting to note that many patients either had organ confined disease (69%) or locally advanced disease (17%). Thus only 14% had pT3a disease potentially curable and needing a nerve bundle resected, with the remaining either not needing it, or being at high risk for relapse and hormonal therapy.
In summary, our randomized trial did not find a significant benefit from unilateral sural nerve grafting, using the state of the art technique in 2000: autologous sural grafts sewn with fine sutures. A larger trial would certainly have power to detect smaller differences, and different techniques such as cadaveric nerves, nerve sheaths, robotic techniques, etc. can certainly be tested. Our trial demonstrated that a randomized design is feasible, although future trials will almost certainly need to be multi-institutional, and should account for a 20-30% attrition rate.
Written by John W. Davis, MD, as part of Beyond the Abstract on UroToday.com
UroToday - the only urology website with original content written by global urology key opinion leaders actively engaged in clinical practice.
To access the latest urology news releases from UroToday, go to: www.urotoday.com
Copyright © 2008 - UroToday