BACKGROUND: Treatment options for visceral leishmaniasis (VL) in East Africa are far from satisfactory due to cost, toxicity, prolonged treatment duration or emergence of parasite resistance. Hence there is a need to explore alternative treatment protocols such as miltefosine alone or in combinations including miltefosine, sodium stibogluconate (SSG) or liposomal amphotericin B. The aim of this trial is to identify regimen(s) which are sufficiently promising for future trials in East Africa. METHODS/DESIGN: A phase II randomized, parallel arm, open-labelled trial is being conducted to assess the efficacy of each of the three regimens: liposomal amphotericin B with SSG, Liposomal amphotericin B with miltefosine and miltefosine alone. The primary endpoint is cure at day 28 with secondary endpoint at day 210 (6 months). Initial cure is a single composite measure based on parasitologic evaluation (bone marrow, spleen or lymph node aspirate) and clinical assessment. Repeated interim analyses have been planned after recruitment of 15 patients in each arm with a maximum sample size of 63 for each. These will follow group-sequential methods (the triangular test) to identify when a regimen is inadequate (<75% efficacy) or adequate (>90% efficacy). We describe a method to ensure consistency of the sequential analysis of day 28 cure with the non-sequential analysis of day 210 cure. DISCUSSION: A regimen with adequate efficacy would be a candidate for treatment of VL with reasonable costs. The design allows repeated testing throughout the trial recruitment period while maintaining good statistical properties (Type I & II error rates) and reducing the expected sample sizes. TRIAL REGISTRATION: ClinicalTrials.gov: NCT01067443.