GETUG-AFU 15

GETUG-AFU 15 randomised 385 men with metastatic non-castrate prostate cancer to androgen deprivation alone or with docetaxel 75 mg/m2 every 21 days for up to nine cycles. At a median 50 months, median overall survival was 58.9 months (95 percent confidence interval 50.8 to 69.1) with docetaxel and 54.2 months (42.2 to not reached) without, hazard ratio 1.01 (0.75 to 1.36). The authors concluded that docetaxel should not be used first line in this setting.

That conclusion was wrong, and the reason is instructive. The trial had 385 men where CHAARTED had 790 and the STAMPEDE docetaxel comparison had nearly 3,000, and a substantial proportion of GETUG-AFU 15 patients had low-volume disease, the group in which docetaxel has since been shown to add least. Later analyses of GETUG-AFU 15 with longer follow-up found a benefit concentrated in high-volume disease, matching CHAARTED.

The toxicity it reported is real and still applies: 72 serious adverse events with docetaxel, most commonly neutropenia in 40 men (21 percent), febrile neutropenia in six (3 percent), abnormal liver function tests in three (2 percent) and neutropenia with infection in two (1 percent). Four treatment-related deaths occurred, two of them from neutropenia, after which the data monitoring committee recommended granulocyte colony-stimulating factor and no further treatment-related deaths occurred. None were reported in the androgen deprivation arm.

GETUG-AFU 15 is the clearest example in prostate cancer of a trial that was negative because it was underpowered, not because the drug does not work, and it is the reason the field now insists on volume-stratified analysis in this setting.