Treatment of adult chronic indeterminate Chagas disease with benznidazole and three E1224 dosing regimens: a proof-of-concept, randomised, placebo-controlled trial

Abstract

Background: Chagas disease is a major neglected vector-borne disease. In this study, we investigated the safety and efficacy of three oral E1224 (a water-soluble ravuconazole prodrug) regimens and benznidazole versus placebo in adult chronic indeterminate Chagas disease.

Method: In this proof-of-concept, double-blind, randomised phase 2 clinical trial, we recruited adults (18-50 years) with confirmed diagnosis of Trypanosoma cruzi infection from two outpatient units in Bolivia. Patients were randomised with a computer-generated randomisation list, which was stratified by centre and used a block size of ten. Patients were randomly assigned (1:1:1:1:1) to five oral treatment groups: high-dose E1224 (duration 8 weeks, total dose 4000 mg), low-dose E1224 (8 weeks, 2000 mg), short-dose E1224 (4 weeks + 4 weeks placebo, 2400 mg), benznidazole (60 days, 5 mg/kg per day), or placebo (8 weeks, E1224-matched tablets). Double-blinding was limited to the E1224 and placebo arms, and assessors were masked to all treatment allocations. The primary efficacy endpoint was parasitological response to E1224 at the end of treatment, assessed by PCR. The secondary efficacy endpoints were parasitological response to benznidazole at end of treatment, assessed by PCR; sustainability of parasitological response until 12 months; parasite clearance and changes in parasite load; incidence of conversion to negative response in conventional and non-conventional (antigen trypomastigote chemiluminescent ELISA [AT CL-ELISA]) serological response; changes in levels of biomarkers; and complete response. The primary analysis population consisted of all randomised patients by their assigned treatment arms. This trial is registered with ClinicalTrials.gov, number NCT01489228.

Findings: Between July 19, 2011, and July 26, 2012, we screened 560 participants with confirmed Chagas disease, of whom 231 were enrolled and assigned to high-dose E1224 (n=45), low-dose E1224 (n=48), short-dose E1224 (n=46), benznidazole (n=45), or placebo (n=47). Parasite clearance was observed with E1224 during the treatment phase, but no sustained response was seen with low-dose and short-dose regimens, whereas 13 patients (29%, 95% CI 16·4-44·3) had sustained response with the high-dose regimen compared with four (9%, 2·4-20·4) in the placebo group (p<0·0001). Benznidazole had a rapid and sustained effect on parasite clearance, with 37 patients (82%, 67·9-92·0) with sustained response at 12-month follow-up. After 1 week of treatment, mean quantitative PCR repeated measurements showed a significant reduction in parasite load in all treatment arms versus placebo. Parasite levels in the low-dose and short-dose E1224 groups gradually returned to placebo levels. Both treatments were well tolerated. Reversible, dose-dependent liver enzyme increases were seen with E1224 and benznidazole. 187 (81%) participants developed treatment-emergent adverse events and six (3%) developed treatment-emergent serious adverse events. Treatment-emergent adverse events were headaches, nausea, pruritus, peripheral neuropathy, and hypersensitivity.

Interpretation: E1224 is the first new chemical entity developed for Chagas disease in decades. E1224 displayed a transient, suppressive effect on parasite clearance, whereas benznidazole showed early and sustained efficacy until 12 months of follow-up. Despite PCR limitations, our results support increased diagnosis and access to benznidazole standard regimen, and provide a development roadmap for novel benznidazole regimens in monotherapy and in combinations with E1224.

Funding: Drugs for Neglected Diseases initiative.

Figure 1. Study Disposition.

ITT - intention-to-treat analysis set: all randomized patients by their assigned treatment arms (primary analysis set); PP - per-protocol analysis set: a subset of the ITT (and FAS) analysis set composed of all ITT patients without any major protocol deviations (compliance/ exposure deviation defined as less than 80% of the planned cumulative dose); FAS - full analysis as treated set: all randomized patients by their actual treatment arms (patients who did not receive treatment will not be included in this analysis set); SAF - safety (all treated) analysis set: all randomized, patients having received at least one dose of study therapy; EOT - end of treatment: all randomized patients by their actual treatment arms who did not discontinued treatment.

Figure 2

(A) Sequential quantitative PCR measurements. The mean of qPCR-Based responses was derived during the treatment phase (left panel) and over time up to 12 months post-treatment (right panel). Plc=Placebo. (B) PKPD model of predicted probability of relapse. The continuous lines in the graph are model predictions of the probability of relapse using E1224 at different treatment regimens. For clarity reasons, the 95% CI (shaded area) for these predictions is only shown for the E1224 High Dose cohort. Circles represent the observed population of patients with relapse.

Figure 3. AT CL-ELISA titers at 0, and 12 months of follow-up in placebo- and BZN-treated patients.

Positive results (red circles): titer ≥1·0; negative results (green circles): titer <0·9; inconclusive results (yellow circles): titer <1·0, ≥0·9. Placebo group: 43 positive (93·5%), 2 negative (4·3%), and 1 inconclusive (2·2%); BZN group: 37 (84.1%), 5 negative (11.4%), and 2 inconclusive (4.5%). Red horizontal line, titer=1·0; green horizontal line, titer=0·9. Black circles: serum samples from the same patient with negative AT CL-ELISA results at months 0 and 12. Only patients with serum samples at 0 and 12 months were considered for this analysis. The geometric mean (long black horizontal line) with 95% C.I. (short black horizontal lines) are indicated for each group. AT CL-ELISA titers were calculated as described. [16]