Prostate Cancer Clinical Trials of the Southwest Oncology Group

The Genitourinary (GU) Cancer Committee of the Southwestern Oncology Group (SWOG) has achieved repeated successes in conducting prospective studies of prostate cancer. This article is a summary of recently completed and current trials in prostate cancer and, as such, represents an intriguing snapshot of priorities in prostate cancer clinical trials in 1997. Bilateral Orchiectomy With or Without Flutamide The most provocative recent trial, and one which can be anticipated to reverberate for decades through the prostate cancer literature, is the recently analyzed comparison of bilateral orchiectomy with or without flutamide (Eulexin) for D2 prostate cancer. This trial is of landmark importance. It represents the 27th randomized prospective trial of total androgen blockade compared to monotherapy performed since 1979. It is unique because of its large size (about 1,300 patients). Few clinical questions in all of medicine have been subjected to so many prospective clinical trials as that of combination vs monotherapy for advanced prostate cancer. Ironically, the results of these numerous trials are contradictory. In 1995, the Prostate Cancer Trialists Cooperative Group (PCTCG) subjected 22 of these trials published up to 1994 to a meta-analysis.[1] Although there was a trend toward a benefit of total androgen blockade, it was not significant at 95% confidence limits. This threw substantial doubt on the benefit of total androgen blockade. This meta-analysis has been criticized. In particular, many of the trial studies were of relatively poor quality, and a number were unpublished. For a number of these trials, there was no data regarding such important issues as concealment of randomization, (ie, the effectiveness of blinding) patient selection, and adequacy of follow-up. A sensitivity analysis, which ranked the published studies according to their quality, demonstrated that a benefit of total androgen blockade was demonstrated if the least well done studies were excluded from the meta-analysis.[2] Thus, controversy continues to rage over the benefit of total androgen blockade. The limitations of meta-analysis are well known. Meta-analysis is retrospective and subject to bias. Methodologists agree that there is no substitute for a large, well-done, randomized trial. Therefore, the results of the SWOG trial have been anticipated with eagerness. It was hoped that this trial would provide the final proof for the benefit (or lack thereof) of total androgen blockade. The trial, in fact, demonstrated a lack of benefit for the addition of flutamide to bilateral orchiectomy. This raises a difficult question. The earlier SWOG trial showed a clear benefit of flutamide plus daily leuprolide (Lupron) injection compared to leuprolide alone. Both trials were large and carefully carried out. Both trials seem conclusive. How does one explain the different outcomes? There are several possibilities. It is likely that at least part of the benefit of total androgen blockade is due to blocking of the flare effect. This benefit would not be seen with orchiectomy. In addition, it may be that compliance with daily luteinizing hormone-releasing hormone (LHRH) injections was poor. If so, it is possible that patients who frequently missed injections would have frequent minor spikes in testosterone. In that case, the use of flutamide would have blocked the repeated "mini-flares," and thus produced a durable survival benefit. This would not be an issue in patients on long-term depot LHRH analog. These studies have contributed enormously to the quality of evidence relating to the treatment of metastatic disease. It is safe to say that the magnitude of the effect of total androgen blockade is small; when orchiectomy is used, that effect appears to be negligible. Unfortunately, some issues remain unresolved. What is the significance of the finding in the LHRH-flutamide study of an increased benefit in the patients with minimal metastatic disease? Does this mean there is a subset of patients who might benefit after all? Can these individuals be