SWOG 9346 (intermittent androgen deprivation)

SWOG 9346 enrolled 3,040 men with newly diagnosed metastatic hormone-sensitive prostate cancer, gave them all 7 months of a luteinising hormone-releasing hormone analogue with an antiandrogen, and randomised the 1,535 whose PSA fell to 4 ng/mL or lower to continuous (765) or intermittent (770) androgen deprivation.

At a median 9.8 years, median survival was 5.8 years with continuous therapy and 5.1 years with intermittent, hazard ratio for death with intermittent therapy 1.10 (90 percent confidence interval 0.99 to 1.23). The upper boundary of that interval exceeded the prespecified non-inferiority threshold of 1.20, but there were too few events to establish inferiority either. The authors described the finding as statistically inconclusive, which is an unusually honest way to report a trial that cost a decade.

Intermittent therapy gave better erectile function and mental health at month 3 (p<0.001 and p=0.003) but not thereafter, and there was no significant difference in treatment-related high-grade adverse events.

The practical reading is that a man with metastatic disease should stay on continuous androgen deprivation, and that intermittent therapy is a reasonable discussion for non-metastatic disease, where the equivalent trials were less worrying. It is also a lesson in how expensive an inconclusive non-inferiority margin is.