Retrospective analysis of ECG data derived from a four-way cross over study involving a broad spectrum anti-infective agent nitazoxanide

J Taubel1, U Lorch1, J Singh1, G Ferber1, M Ayers2, J Camm3. 1Richmond Pharmacology Ltd., St George's University of London, Cranmer Terrace, London, UK, 2Romark Laboratories Ltd., 3000 Bayport Drive, Suite 200, Tampa, Florida, USA, 3Department of Cardiological Sciences, St George's University of London, Cranmer Terrace, London, United Kingdom, London, UK

A vast proportion of Thorough QT (TQT) studies use moxifloxacin (400 mg), an anti-bacterial fluoroquinolone as a positive treatment arm to assure assay sensitivity as per ICH E14 guidelines. We have shown that food is a good candidate for a positive control because it produces a QTc shortening effect (Taubel et al., 2012) that is correlated closely with the release of C peptide and blood glucose concentrations (Taubel et al., 2013). Despite these observations there is paucity in the literature when the QT effect is examined in the fed state. In this double blind placebo and positive controlled, four way cross-over Phase 1 study, 56 healthy male and female subjects were randomised to receive to one of four treatment sequences and received single doses of moxifloxacin (400 mg) or placebo or therapeutic (675 mg) or supra-therapeutic (2700 mg) doses of a broad spectrum anti-infective agent nitazoxanide. In this study, the primary ECG analysis was to compare the mean differences in change from average baseline and time-matched change of nitazoxanide compared to placebo.

Subjects’ eligibility regarding specific ECG criteria was evaluated during screening and confirmed on admission, prior to the first drug administration and subjects were randomised to one of four treatment sequences. These sequences provided balance for period and preceding treatment (Williams squares) and were performed stratified by gender and race. Each period consisted of a baseline ECG day and a treatment day. An American style breakfast was served on dosing days. The baseline ECG recordings, at the beginning of each treatment period, were treatment and period specific and the primary analysis was based on the cross-over design of the study, used the most appropriate heart rate correction (QTcIL/QTcIP/QTcF/QTcB), and was based on the change from average baseline.

The QTc analysis showed that the largest change from average baseline in QTcF between 675 mg nitazoxanide and placebo was 1.6 ms (two-sided 90% CI: -0.3, 3.6) and for 2700 mg nitazoxanide and placebo was 3.4 ms (two-sided CI: 1.4, 5.4 ms). When compared to QTcIP, the largest change from baseline was found to be 2.0 ms (two-sided 90% CI: 0.1, 4.2 ms) and 2.8 ms (two-sided 90% CI: 0.8, 4.9 ms) for 675 mg and 2700 mg nitazoxanide respectively. The categorical analyses show that almost all subjects had changes from baseline in QTcF which were <30 ms on each of the four study treatments. There were also no increases from baseline in QTcF which were greater than 60 ms at any of the measurement time-points for any of the treatments. All of the subjects had all their QTcF values ≤480 ms on each of the four study treatments. Similar findings were also observed with the other three heart rate corrections, QTcIL, QTcIP and QTcB with no increases from baseline ≥60 ms. Heart rate increased slightly (<3 beats per minute) for subjects in the 2700 mg nitazoxanide treatment group.

Taken together the results of the QTc analysis indicate that a neither a single dose of 675 mg nor 2700 mg nitazoxanide prolongs QTc in healthy male and female subjects. This study was powered to show a statistically significant result based on a formal sample size calculation in which to achieve a power of 80% (αα = 0.05) it required the inclusion of 56 subjects.