Objectives: To analyze reported clinical outcomes for patients in whom an agent that acts via a steroid hormone receptor was "withdrawn."
Methods: Published reports where agent(s) known to act via steroid hormone receptors were discontinued in patients with relapsing prostatic cancer were retrieved from MEDLINE listings. The trials included patients who progressed on steroidal and nonsteroidal antiandrogens, progestational agents, and estrogens. Included were the specifics of all treatments administered prior to discontinuation of the drugs, concomitant therapies, and factors that might predict a favorable response to "withdrawal."
Results: Withdrawal responses were observed following the discontinuation of the antiandrogens flutamide and bicalutamide, flutamide plus aminoglutethimide, estrogens, and progestational agents. In most responding cases, responses were seen in patients with long exposure to the drug. No specific factors were predictive for response.
Conclusions: Withdrawal responses to agents that act via steroid hormone receptors represent a generalized phenomenon that can result in palliation for patients with hormonally relapsed prostate cancer. A trial of "withdrawal therapy" is warranted in patients with relapsing disease prior to the initiation of more toxic therapies. Failure to control for this phenomenon in clinical trials may lead to false attribution of response to a study agent. The data provide support for the concept that androgen independence does not necessarily mean that a tumor is resistant to further hormonal manipulations.