To evaluate the effect of discontinuation of the antiandrogen, flutamide, in patients with metastatic prostate cancer who are progressing on hormonal therapy.

Thirty-six patients with progressive disease on hormonal treatment that included flutamide had discontinuation of the antiandrogen. Thirty-five (95%) had progressive increases in prostate-specific antigen (PSA) levels, despite castrate levels of testosterone. Twenty-five patients (69%) were treated with combined androgen blockade (orchiectomy or gonadotropin-releasing hormone [GnRH] analog plus flutamide) as initial therapy and 11 (31%) were started on monotherapy alone. Patients who had not undergone a previous orchiectomy were continued on the GnRH analog. Patients were monitored clinically and with serial PSA measurements, radionuclide scans, and radiographs as indicated to assess response.

Considering the 35 patients with increasing PSA values, 10 (29%) showed a significant decline (> or = 80% in seven, and > or = 50% in three) in PSA from baseline. All 10 had received combined androgen blockade as initial therapy. The duration of decline was short (median, 5+ months; range, 2 to 10+), but was associated with improvement in clinical symptoms, while one patient had a partial response in an epidural mass with parallel decline in PSA. None of the patients started on single hormone therapies responded.

Discontinuation of flutamide was associated with a significant decrease in PSA values in 10 of 25 patients (40%; 95% confidence interval, 21% to 59%) and clinical improvement in a subset of patients who had an initial response, but later progressive disease on combined androgen blockade. A trial of flutamide withdrawal should be considered in patients progressing on total androgen blockade before the initiation of more toxic therapies. It is likely that flutamide withdrawal has contributed to the observed responses in phase II trials of both second-line hormonal therapies and new cytotoxic agents. Future phase II trials in hormone-refractory prostatic cancer must control for this observation, and insure that progression off flutamide is documented before initiation of alternative treatment.