Low versus high dose of antimony for American cutaneous leishmaniasis: A randomized controlled blind non-inferiority trial in Rio de Janeiro, Brazil

Abstract

Background: Although high dose of antimony is the mainstay for treatment of American cutaneous leishmaniasis (ACL), ongoing major concerns remain over its toxicity. Whether or not low dose antimony regimens provide non-inferior effectiveness and lower toxicity has long been a question of dispute.

Methods: A single-blind, non-inferiority, randomized controlled trial was conducted comparing high dose with low dose of antimony in subjects with ACL treated at a referral center in Rio de Janeiro, an endemic area of Leishmania (Viannia) braziliensis transmission. The primary outcome was clinical cure at 360 days of follow-up in the modified-intention-to-treat (mITT) and per-protocol (PP) populations. Non-inferiority margin was 15%. Secondary objectives included occurrence of epithelialization, adverse events and drug discontinuations. This study was registered in ClinicalTrials.gov: NCT01301924.

Results: Overall, 72 patients were randomly assigned to one of the two treatment arms during October 2008 to July 2014. In mITT, clinical cure was observed in 77.8% of subjects in the low dose antimony group and 94.4% in the high dose antimony group after one series of treatment (risk difference 16.7%; 90% CI, 3.7-29.7). The results were confirmed in PP analysis, with 77.8% of subjects with clinical cure in the low dose antimony group and 97.1% in the high dose antimony group (risk difference 19.4%; 90% CI, 7.1-31.7). The upper limit of the confidence interval exceeded the 15% threshold and was also above zero supporting the hypothesis that low dose is inferior to high dose of antimony after one series of treatment. Nevertheless, more major adverse events, a greater number of adverse events and major adverse events per subject, and more drug discontinuations were observed in the high dose antimony group (all p<0.05). Interestingly, of all the subjects who were originally allocated to the low dose antimony group and were followed up after clinical failure, 85.7% achieved cure after a further treatment with local therapy or low dose of antimony.

Conclusions: Compared with high dose, low dose of antimony was inferior at the pre-specified margin after one series of treatment of ACL, but was associated with a significantly lower toxicity. While high dose of antimony should remain the standard treatment for ACL, low dose antimony treatment might be preferred when toxicity is a primary concern.

Fig 1. Flowchart of participants through each stage of the study “Low versus high dose of antimony for American cutaneous leishmaniasis.”

Fig 2. Different stages of the wound healing process in American cutaneous leishmaniasis.

Fig 3. Non-inferiority plot of low dose antimony versus high dose antimony for American cutaneous leishmaniasis.

Data are shown as point estimates and 90% confidence interval (CI) for absolute risk difference in clinical cure for intention-to-treat and per-protocol populations. Non-inferiority can be established if confidence intervals are within the prespecified boundary for non-inferiority (Δ).

Fig 4. Bubble plot of -log10(Raw p-value) by risk difference with a two-sided 95% confidence interval sized by number of subjects with adverse events in high dose versus low dose antimony groups.

Note: Grey zone area corresponds to area of p-value<0.05; X stands for point estimates; point estimates within the grey zone area have p-values<0.05; circles are sized by the number of patients with adverse events; circles located to the right of the vertical solid line favor more adverse events in the high dose antimony group.

Fig 5. Subgroup analysis: Forest plot of absolute risk difference (RD) and relative risk (RR) with two-sided 95% confidence interval (CI) for major adverse events according to low versus high dose antimony treatment.

Squares located to the right of the vertical solid line favor more major adverse events in the high dose antimony group.