In vitro studies have shown that the nonsteroidal antiestrogen tamoxifen can suppress deoxyribonucleic acid synthesis and cell proliferation in cultured human gliomas. This growth suppression is independent of its antiestrogenic properties. Tamoxifen may act through the inhibition of the enzyme protein kinase C, which transduces mitogenic signals from the cell surface to the nucleus. Based on these preclinical studies, we initiated a clinical trial of orally administered tamoxifen, 20 mg twice daily, to patients with recurrent, progressive malignant gliomas who were not candidates for other "failed" protocols, such as brachytherapy. No limits were placed on age, Karnofsky Performance Score (KPS), or expected survival. Thirty-two patients were entered in the study, 29 with a glioblastoma multiforme and 3 with an anaplastic astrocytoma. The mean age of the group was 48 years, and the mean KPS was 65. Median survival of the entire cohort from the onset of tamoxifen therapy was 17 weeks; the median survival of those patients with an initial KPS of 70 or more was 21 weeks. Seven patients survived for more than 6 months with no change in their baseline computed tomographic scans or KPS on tamoxifen, including 2 patients with computed tomographic evidence of regression during the course of therapy. There were no significant patient-reported side effects of the treatment. Three patients had thromboembolic complications during tamoxifen administration. We conclude that tamoxifen can be administered safely to these patients and may show some efficacy against glial neoplasms.