Results of a phase III randomized controlled trial of the safety and efficacy of atrasentan in men with nonmetastatic hormone-refractory prostate cancer (HRPC)
5018 Introduction: Atrasentan is a potent oral, selective endothelin A receptor antagonist with biological activity in patients with HRPC. This randomized phase III trial of atrasentan vs placebo was conducted in 941 men with nonmetastatic HRPC whose prostate-specific antigen (PSA) was rising despite adequate androgen suppression. Methods: 467 patients were randomized to receive atrasentan 10 mg and 474 patients were randomized to receive placebo daily. Primary efficacy endpoints were time to disease progression (TTP) and progression-free survival, with progression defined as the onset of metastases. Secondary endpoints were overall survival, time to PSA progression, change from baseline in bone alkaline phosphatase (BAP), and PSA doubling time (PSADT). Results: In the intent-to-treat population, median TTP was 764 days with atrasentan and 671 days with placebo (hazard ratio [HR], 0.915; G1,1 P=0.288). More placebo- treated patients (267, 56.3%) than atrasentan-treated patients (227, 48.6%) experienced disease progression. Overall, 278 patients (29.5%) discontinued before reaching a protocol-defined endpoint (155, 40.8% US vs 123, 21.6% non-US). More placebo-treated patients (210, 44.3%) than atrasentan-treated patients (169, 36.2%) experienced new skeletal lesions. Median survival was 1,477 days for the atrasentan group and 1,403 days for the placebo group (HR, 0.909; G1,1 P=0.176). Mean change from baseline to final BAP was -1.51 IU/L with atrasentan compared with +2.20 IU/L with placebo (P<0.001). PSADT was lengthened with atrasentan relative to placebo (P<0.031). The incidence of peripheral edema, nasal congestion, headache, dyspnea, anemia, and heart failure was significantly greater in the atrasentan group (P<0.05). Most of these events were mild and nonserious and did not result in study discontinuation. Conclusions: Data from this study did not show a statistically significant difference in time to progression between atrasentan and placebo. However, the high discontinuation rate, especially among US patients, may have compromised the ability of the study to accurately determine the treatment effect of atrasentan in this population. No significant financial relationships to disclose.