OBJECTIVES: Our objective was to compare the efficacy and safety of three drug combinations for the treatment of late-stage human African trypanosomiasis caused by Trypanosoma brucei gambiense. DESIGN: This trial was a randomized, open-label, active control, parallel clinical trial comparing three arms. SETTING: The study took place at the Sleeping Sickness Treatment Center run by Médecins Sans Frontières at Omugo, Arua District, Uganda PARTICIPANTS: Stage 2 patients diagnosed in Northern Uganda were screened for inclusion and a total of 54 selected. INTERVENTIONS: Three drug combinations were given to randomly assigned patients: melarsoprol-nifurtimox (M+N), melarsoprol-eflornithine (M+E), and nifurtimox-eflornithine (N+E). Dosages were uniform: intravenous (IV) melarsoprol 1.8 mg/kg/d, daily for 10 d; IV eflornithine 400 mg/kg/d, every 6 h for 7 d; oral nifurtimox 15 (adults) or 20 (children <15 y) mg/kg/d, every 8 h for 10 d. Patients were followed up for 24 mo. OUTCOME MEASURES: Outcomes were cure rates and adverse events attributable to treatment. RESULTS: Randomization was performed on 54 patients before enrollment was suspended due to unacceptable toxicity in one of the three arms. Cure rates obtained with the intention to treat analysis were M+N 44.4%, M+E 78.9%, and N+E 94.1%, and were significantly higher with N+E (p = 0.003) and M+E (p = 0.045) than with M+N. Adverse events were less frequent and less severe with N+E, resulting in fewer treatment interruptions and no fatalities. Four patients died who were taking melarsoprol-nifurtimox and one who was taking melarsoprol-eflornithine. CONCLUSIONS: The N+E combination appears to be a promising first-line therapy that may improve treatment of sleeping sickness, although the results from this interrupted study do not permit conclusive interpretations. Larger studies are needed to continue the evaluation of this drug combination in the treatment of T. b. gambiense sleeping sickness.