What's known on the subject? and What does the study add? The additional use of anti-androgen (deferred combined androgen blockade [CAB] therapy) for patients with castration-resistant prostate cancer (CRPC) initially treated with androgen deprivation monotherapy (ADMT) can provide a clinical response, although the reported response rates vary widely. Our previous study, which reported a response rate of 66% to deferred CAB therapy, suggested that deferred CAB responders would also respond better to subsequent therapies than non-responders because the difference in cancer-specific survival between the deferred CAB responders and the non-responders was much larger than the progression-free survival rates for the responders. The present study showed that PSA response to deferred CAB therapy predicts clinical outcomes after subsequent oestrogen and docetaxel therapy. We propose that PSA response to deferred CAB be used for planning individualized treatment that includes secondary hormonal therapy and chemotherapy.

Objective: To evaluate whether there is any association between prostate-specific antigen (PSA) response to deferred combined androgen blockade (CAB) therapy using bicalutamide in patients with castration-resistant prostate cancer (CRPC), initially treated with castration monotherapy, and the clinical outcomes after subsequent oestrogen and docetaxel therapies.

Patients and methods: Fifty-six patients with advanced prostate cancer, who were refractory to both initial castration monotherapy and subsequent deferred CAB, received oestrogen therapy (estramustine phosphate 140 or 280 mg/day in 50 patients, diethylstilbestrol diphosphate 100 mg/day orally in six patients). Of the 56 patients, 33 underwent docetaxel therapy (median dose 55 mg/m(2), every 4-8 weeks, median 6 courses) when they became refractory to oestrogen therapy. A deferred CAB response was defined as a decrease of >50% in PSA levels after the addition of bicalutamide. The difference in cancer-specific survival (CSS) after confirmation of resistance to initial castration monotherapy between the deferred CAB responders and the non-responders was evaluated, and outcomes after oestrogen and docetaxel therapies were also compared between the two groups.

Results: A response to deferred CAB was observed in 27 (48%) of the 56 patients. There was no significant difference between the responders and the non-responders in pretreatment clinical variables, including Gleason score, metastatic sites, PSA level at diagnosis, and PSA nadir during castration monotherapy. A deferred CAB response was a significant predictor of CSS after confirmation of resistance to initial castration monotherapy. The deferred CAB responders had significantly longer progression-free survival (PFS) (median 3.2 months in the responders, 2.1 month in the non-responders, P = 0.04) and CSS (median 3.0 years in the responders, 1.5 years in the non-responders, P = 0.04) after oestrogen therapy. Likewise, PFS (median 8.2 months in the responders, 2.2 months in the non-responders, P < 0.01) and CSS (median not reached in the responders, 1.4 years in the non-responders, P < 0.01) after docetaxel therapy was significantly longer in the deferred CAB responders.

Conclusion: PSA response to deferred CAB predicts clinical outcomes after subsequent oestrogen and docetaxel therapies in patients with CRPC, and provides useful information for planning individualized optimum treatment courses that include secondary hormonal therapy and chemotherapy.