Comparison of Atovaquone (566C80) with Trimethoprim-Sulfamethoxazole to Treat Pneumocystis carinii Pneumonia in Patients with AIDS

Abstract

BACKGROUND: Both trimethoprim-sulfamethoxazole and pentamidine are effective as treatments for Pneumocystis carinii pneumonia, but adverse effects frequently limit their use. Atovaquone (566C80) is a new hydroxynaphthoquinone with activity against P. carinii. METHODS: We conducted a double-blind, multicenter study in patients with the acquired immunodeficiency syndrome and mild or moderately severe P. carinii pneumonia. They were randomly assigned to 21 days of orally administered treatment three times daily with either atovaquone (750 mg) or trimethoprim (320 mg) plus sulfamethoxazole (1600 mg). RESULTS: Of the 322 patients with histologically confirmed P. carinii pneumonia, 160 received atovaquone and 162 received trimethoprim-sulfamethoxazole. Of those who could be evaluated for therapeutic efficacy, 28 of 138 patients given atovaquone (20 percent) and 10 of 146 patients given trimethoprim-sulfamethoxazole (7 percent) did not respond (P = 0.002). Treatment-limiting adverse effects required a change of therapy in 11 patients in the atovaquone group (7 percent) and 33 patients in the trimethoprim-sulfamethoxazole group (20 percent) (P = 0.001). Therapy involving only the initial drug was successful and free of adverse effects in 62 percent of those assigned to atovaquone and 64 percent of those assigned to trimethoprim-sulfamethoxazole. Within four weeks of the completion of treatment, there were 11 deaths in the atovaquone group (4 due to P. carinii pneumonia) and 1 death in the trimethoprim-sulfamethoxazole group (P = 0.003). Diarrhea at entry was associated with lower plasma drug concentrations (P = 0.009), therapeutic failure (P < 0.001), and death (P < 0.001) in the atovaquone group but not in the trimethoprim-sulfamethoxazole group. CONCLUSIONS: For the treatment of P. carinii pneumonia, atovaquone is less effective than trimethoprim-sulfamethoxazole, but it has fewer treatment-limiting adverse effects.