Dose-related adverse events in South African patients prescribed clofazimine for drug-resistant tuberculosis

Abstract

BACKGROUND: Optimal drug levels and minimal toxicity are critical factors in improving treatment outcomes for patients prescribed new and repurposed medicine for drug-resistant (DR) tuberculosis (TB). The optimal dose and dose-related safety of clofazimine (CFZ), a repurposed medicine for DR TB, in the South African (SA) population are unknown. OBJECTIVES: To report on dose-related adverse events in patients receiving CFZ plus a background regimen for DR TB. METHODS: In a retrospective review of patient folders from 2012 to 2014, adverse events documented for patients receiving high- (≥200 mg) and low-dose (100 mg) CFZ in a centralised DR TB hospital in KwaZulu-Natal Province, SA, were investigated for an association between dose-weight interactions and adverse events. RESULTS: Of 600 patients included, 78.7% (n=472) weighed ≥50 kg. Of these, 17.4% (n=82) received 100 mg CFZ and 82.6% (n=390) received >200 mg. Of 128 patients (21.3%) who weighed <50 kg, 68.0% (n=87) received 100 mg CFZ and 32.0% (n=41) received ≥200 mg. Of 463 patients (77.2%) who were HIV-positive, 94.0% were on antiretrovirals. There was no difference between the dose-weight cohorts in the background regimen given in addition to high- or low-dose CFZ. The frequency and types of adverse events observed were similar to the published literature. When analysed per dose-weight cohort, patients weighing <50 kg and receiving high-dose CFZ (≥200 mg) had a 2.6 times higher risk of any adverse event (adjusted odds ratio (aOR) 2.57; 95% confidence interval (CI) 1.02 - 6.05; p=0.05: reference category <50 kg and 100 mg). Patients weighing <50 kg and receiving high-dose CFZ had a 3.3 times higher risk of gastrointestinal adverse events than patients weighing <50 kg and receiving 100 mg CFZ (aOR 3.30; 95% CI 1.51 - 7.19; p=0.003). A high risk of chest pain was observed in patients receiving high- and low-dose CFZ, irrespective of weight. Patients weighing <50 kg receiving high-dose CFZ had a slightly higher risk of adverse events related to the skin (aOR 1.2; 95% CI 0.55 - 2.62; p=0.7) There were no documented reports of the CFZ dose being reduced or the drug being stopped due to adverse events in the sample population. CONCLUSIONS: There is an association between dose-weight interaction and adverse events. The odds of any adverse event occurring were higher when low-weight patients (<50 kg) received high-dose CFZ (≥200 mg). Gastrointestinal and skin-related adverse events were more common when high-dose CFZ was used in patients weighing <50 kg. Chest pain was reported in patients receiving high- and low-dose CFZ, irrespective of weight, and may be a symptom of cardiac toxicity. Plasma concentrations of CFZ may be affected by drug-drug interactions, so active drug safety monitoring including electrocardiograms is recommended routinely when CFZ is part of the regimen.