Background: Second-line injectable drugs (SLIs), core agents in the treatment of multidrug resistant tuberculosis (MDR-TB), are associated with substantial toxicity and treatment discontinuations. Bedaquiline is being widely used as a substitute in MDR-TB regimens for patients unable to tolerate SLIs, but the efficacy and safety of this strategy is unknown.
Methods: We conducted a retrospective cohort study to evaluate outcomes at 12-months for MDR-TB patients who substituted bedaquiline for SLIs. We included consecutive adult MDR-TB patients who had bedaquiline substitutions in the Western Cape Province of South Africa between May 2015 and May 2016, as well as MDR-TB controls who did not receive bedaquiline, matched for location and time of treatment initiation. Data were extracted from the electronic TB register. The composite primary outcome measure was the proportion of patients with death, loss to follow up, or failure to achieve sustained culture conversion at 12 months of treatment.
Results: Data from 330 patients with laboratory-confirmed pulmonary MDR-TB were analyzed; 162 with bedaquiline substitution and 168 controls. Baseline characteristics were similar between the groups, except for CD4 cell count which was lower in the bedaquiline group (Table 1). SLIs were stopped at a median of 54 days (interquartile range, IQR 25 - 82), with a 44 day (IQR 29 - 71) delay to starting bedaquiline. The primary outcome, ascertained in 200 individuals, occurred in 63
(55.3%) patients in the bedaquiline group versus 54 (62.8%) patients in the control group (odds ratio, 0.73; 95% confidence interval [CI], 0.41 to 1.23; P = 0.285). Rates of sustained culture conversion (48.6% vs. 47.8%), loss to follow up (10.5% vs. 12.5%), and death (6.8% vs. 6.6%) at 12 months were similar between the groups. There was a trend towards earlier sputum culture conversion in the bedaquiline group (hazard ratio, 1.33; 95% CI, 0.94 to 1.88; P = 0.104; Figure 1).
Conclusions: Substituting bedaquiline for SLIs in the treatment of MDR-TB does not result in inferior outcomes at 12 months compared with patients who remain on SLIs, supporting the use of this strategy in MDR-TB therapy. The substantial delay between interrupting SLIs and initiating bedaquiline needs to be addressed.