Androgen-deprivation therapy in men with metastatic prostate cancer: less may not necessarily be more.

I n a pivotal phase 3 randomized controlled trial, Hussain et al. tested the hypothesis that, with respect to survival, intermittent androgen deprivation therapy (ADT) is non-inferior to continuous in men with newly diagnosed metastatic prostate cancer. While the trial findings were statistically inconclusive, the study suggests, but does not prove, that intermittent may do more harm than good, although findings are not definitive. While outcomes of ongoing trials are awaited, the trial by Hussain et al., in conjunction with an earlier trial in men with non-metastatic prostate cancer by Crook et al., does provide important new guidance regarding the choice of ADT in men with androgen-sensitive prostate cancer. ADT is one of the most effective palliative therapies for patients with metastatic prostate cancer, but not without drawbacks. While not as toxic as chemotherapy, ADT carries a significant risk of morbidity, including sexual dysfunction, fatigue, anemia, accelerated bone loss and fractures, sarcopenia, increased risk of diabetes, and possibly, of cardiovascular events. In addition, despite an initial response rate of more than 90%, most patients develop resistance to ADT, resulting in a median survival of 2.5–3 years. Preclinical data suggest that continuous use of ADT may accelerate the emergence of resistance to this therapy, and that re-exposure of prostate cancer stem cells to androgens can re-induce differentiation and increase their apoptotic potential. These ADT-associated shortcomings, in addition to treatment expense, have spurred the development of strategies to minimize the exposure to ADT, including the use of intermittent ADT. While several smaller randomized controlled clinical trials have compared the use of intermittent with continuous ADT, no definitive information is available for patients with metastatic prostate cancer. To fill this evidence gap, a large multinational randomized controlled trial led by Hussain et al. was designed in 1993, and outcomes have been published recently in the New England Journal of Medicine. In a coprimary end point, the authors tested the hypotheses that (i) intermittent ADT is not inferior to continuous ADT with respect to survival in men with metastatic, hormonesensitive prostate cancer; and (ii) compared to continuous therapy, intermittent ADT improves quality of life. The primary finding from the study was inconclusive, that is, intermittent ADT was not proven to be as good as continuous ADT, and there was instead a trend to inferiority. While intermittent ADT was associated with better erectile function and mental health, this benefit did not persist beyond 3 months. Due to the inconclusive finding of this non-inferiority trial its findings may be difficult to interpret and apply to clinical practice. Therefore, we explore this trial in more detail. The study by Hussain et al. enrolled 3040 men from the Unite States, Canada and UK who had newly diagnosed prostate cancer with lymph node, visceral or bone metastases and a prostate-specific antigen (PSA) . 5 ng ml. Men received a 7-month induction course with a luteinizing hormone releasing hormone agonist and anti-androgen (goserelin and bicalutamide, or equivalent) to select androgen-dependent disease, defined by a PSAf4 ng ml at induction end. One thousand five hundred and thirty-five men fulfilled this criterion and were then randomized but not blinded to intermittent or continuous ADT, stratified by performance status, prior hormone therapy and extent of disease. Men assigned to continuous therapy continued, whereas men assigned to intermittent therapy discontinued ADT at completion of the 7-month induction course. The thresholds for re-commencement of ADT in the intermittent group were: a rise of PSA to baseline or o20 ng ml or investigator discretion (PSA .10 ng ml or symptomatic disease). For the statistical analysis, the authors assumed a median survival in men receiving continuous ADT of 35 months and considered a 7-month shorter survival with intermittent clinically unacceptable, and therefore, a hazard ratio of 1.20 was set as a one-sided test of the null hypothesis. Median age of the randomized population was 70, pre-treatment PSA 42 ng ml, 96% had an Eastern Cooperative Oncology Group performance status of 0–1 and 50% had extensive (vs. 50% minimal) disease, and one-third of men had bone pain at the beginning of the induction period, with no difference in men assigned to intermittent (n5770) and continuous (n5765) ADT. Median follow-up was 9.8 years. Median duration of protocol therapy after randomisation was only 17 months in the continuous group and 19 months in the intermittent group. Those in the intermittent group received ADT for a median 47% of time, and at 15 months, 78% of men in the intermittent group had resumed ADT. Median survival was 5.8 years in the continuous group and 5.1 years in the intermittent group, and 73% to 80% of deaths were related to prostate cancer. The hazard ratio for death was 1.10, representing a 10% increased risk of death with the use of intermittent ADT, with a 90% confidence interval of 0.99–1.23. Because the upper limit of the 90% confidence interval exceeded the predefined threshold of 1.20, the hypothesis that intermittent ADT was not inferior to continuous therapy Department of Medicine, University of Melbourne Austin Health, Heidelberg, Victoria 3084, Australia and Endocrinology, Austin Health, Heidelberg, Victoria 3084, Australia Correspondence: Professor M Grossmann (mathisg@ unimelb.edu.au) Asian Journal of Andrology (2013) 15, 445–446 2013 AJA, SIMM & SJTU. All rights reserved 1008-682X/13 $32.00