Background: For treating multidrug-resistant tuberculosis (MDR TB), the World Health Organization (WHO) recommends a regimen of at least four second-line drugs that are likely to be effective as well as pyrazinamide. WHO guidelines indicate only marginal benefit for regimens based directly on drug susceptibility testing (DST) results. Recent evidence from isolated cohorts suggests that regimens containing more drugs may be beneficial, and that DST results are predictive of regimen effectiveness. The objective of our study was to gain insight into how regimen design affects treatment response by analyzing the association between time to sputum culture conversion and both the number of potentially effective drugs included in a regimen and the DST results of the drugs in the regimen.
Methods and findings: We analyzed data from the Preserving Effective Tuberculosis Treatment Study (PETTS), a prospective observational study of 1,659 adults treated for MDR TB during 2005-2010 in nine countries: Estonia, Latvia, Peru, Philippines, Russian Federation, South Africa, South Korea, Thailand, and Taiwan. For all patients, monthly sputum samples were collected, and DST was performed on baseline isolates at the US Centers for Disease Control and Prevention. We included 1,137 patients in our analysis based on their having known baseline DST results for at least fluoroquinolones and second-line injectable drugs, and not having extensively drug-resistant TB. These patients were followed for a median of 20 mo (interquartile range 16-23 mo) after MDR TB treatment initiation. The primary outcome of interest was initial sputum culture conversion. We used Cox proportional hazards regression, stratifying by country to control for setting-associated confounders, and adjusting for the number of drugs to which patients' baseline isolates were resistant, baseline resistance pattern, previous treatment history, sputum smear result, and extent of disease on chest radiograph. In multivariable analysis, receiving an average of at least six potentially effective drugs (defined as drugs without a DST result indicating resistance) per day was associated with a 36% greater likelihood of sputum culture conversion than receiving an average of at least five but fewer than six potentially effective drugs per day (adjusted hazard ratio [aHR] 1.36, 95% CI 1.09-1.69). Inclusion of pyrazinamide (aHR 2.00, 95% CI 1.65-2.41) or more drugs to which baseline DST indicated susceptibility (aHR 1.65, 95% CI 1.48-1.84, per drug) in regimens was associated with greater increases in the likelihood of sputum culture conversion than including more drugs to which baseline DST indicated resistance (aHR 1.33, 95% CI 1.18-1.51, per drug). Including in the regimen more drugs for which DST was not performed was beneficial only if a minimum of three effective drugs was present in the regimen (aHR 1.39, 95% CI 1.09-1.76, per drug when three effective drugs present in regimen). The main limitation of this analysis is that it is based on observational data, not a randomized trial, and drug regimens varied across sites. However, PETTS was a uniquely large and rigorous observational study in terms of both the number of patients enrolled and the standardization of laboratory testing. Other limitations include the assumption of equivalent efficacy across drugs in a category, incomplete data on adherence, and the fact that the analysis considers only initial sputum culture conversion, not reversion or long-term relapse.
Conclusions: MDR TB regimens including more potentially effective drugs than the minimum of five currently recommended by WHO may encourage improved response to treatment in patients with MDR TB. Rapid access to high-quality DST results could facilitate the design of more effective individualized regimens. Randomized controlled trials are necessary to confirm whether individualized regimens with more than five drugs can indeed achieve better cure rates than current recommended regimens.