Shorter regimens fail to be non-inferior to the standard tuberculosis treatment plans
One third of the world’s population is currently infected with tuberculosis (TB), and, in 2012, there were 1.3 million TB-related deaths (Centers for Disease Control and Prevention). Moreover, in 2012, 450,000 people worldwide developed multi-drug resistant TB (MDR TB), which is resistant to at least rifampin and isoniazid, two of the first line antimicrobials for TB (World Health Organization). Although many cases of MDR TB are acquired from patients harboring these organisms, among the mechanisms for inducing drug resistance in TB are being on incorrect treatment or not completing a full antibiotic course. Inadequate compliance is particularly challenging in TB, because the standard treatment course lasts at least 6 months. Finding new treatment approaches that shorten the duration could help decrease the development of drug resistance and lower costs.
In this week’s NEJM, three groups published phase III trials investigating the efficacy of using fluoroquinolones in combination with other anti-TB drugs to treat patients with a four-month regimen, two months shorter than the standard of care. Merle et al. performed a randomized, open-label, controlled trial in which they compared the standard six-month treatment regimen (isoniazid, rifampin, pyrazinamide, and ethambutol) to a four-month regimen that replaced ethambutol with gatifloxacin, a fourth-generation fluoroquinolone. The trial enrolled just over 1800 patients across five African countries, and, despite the positive results from phase II trials and mouse studies, the four-month treatment regimen failed to demonstrate non-inferiority, with higher TB recurrences than that of the standard treatment (14.6% vs. 7.1%). Gillespie et al. found similar results when they randomized just over 1900 patients across nine countries to either one of two fourth-month regimens or the standard of care. In one of these shorter regimens, they used a combination of rifampin, isoniazid, pyrazinamide, and moxifloxacin, a fluoroquinolone, while, in the other, they used ethambutol in place of isoniazid. Both moxifloxacin-containing regimens failed to show non-inferiority compared to the control arm.
Finally, Jindani et al. performed a randomized controlled trial in which they enrolled just over 800 patients across four African countries. Compared to the control six-month regimen, they tested two treatment plans for non-inferiority: one four-month and one six-month regimen in which isoniazid was replaced by moxifloxacin. Similar to the other two trials, the four-month treatment plan also failed to show non-inferiority. The six-month moxifloxacin-containing treatment plan, in which treatment in the final four months was administered weekly, was as effective as the standard daily regimen. While this regimen does not decrease the overall treatment duration, it does decrease the frequency at which patients need to take medications. This advance may increase treatment adherence and is a step towards helping to decrease the development of MDR TB.
While these trials are steps in the right direction, the failure of these three trials to show non-inferiority highlights a major challenge in the field. Even though fluoroquinolones were effective at decreasing the treatment time in mouse models infected with TB, the discrepancy in human trials demonstrates again that differences in biology between mice and humans are major hurdles in the drug discovery pipeline. As Digby Warner, PhD, and Valerie Mizrahi, PhD, commented in an accompanying editorial, more effort will need to be focused on how to more effectively develop and test new therapies so that fewer drugs and regimens fail in phase III and IV trials.