Abstract
Background
Treatment of rifampin-resistant and/or multidrug-resistant tuberculosis (RR/MDR-TB) requires multiple drugs and outcomes remain suboptimal. Some drugs are associated with improved outcome, however whether particular pharmacokinetic-pharmacodynamic relationships predict outcome is unknown.
Methods
Adults with pulmonary RR/MDR-TB in Tanzania, Bangladesh and Russian Federation receiving therapy with local regimens were enrolled from June, 2016 to July, 2018. Serum was collected after two, four, and eight weeks for each drug's area under the concentration-time curve (AUC
0-24) and quantitative susceptibility of the
Mycobacterium tuberculosis isolate measured by minimum inhibitory concentrations (MIC). Individual drug AUC
0-24/MIC targets were assessed by adjusted odds ratios (OR) for association with favorable treatment outcome and hazard ratios (HR) for time to sputum culture conversion. K-means clustering algorithm separated the cohort of the most common multidrug regimen into four clusters by AUC
0-24/MIC exposures.
Results
Among 290 patients, 62 (21%) experienced treatment failure, including 30 deaths. Moxifloxacin AUC
0-24/MIC target of 58 was associated with favorable treatment outcome [OR 3.75 (1.21, 11.56), p = 0·022], while levofloxacin AUC
0-24/MIC of 118.3, clofazimine AUC
0-24/MIC of 50.5, and pyrazinamide AUC
0-24 of 379 mg*h/L were associated with faster culture conversion [HR > 1·0, p < 0.05]. Other individual drug exposures were not predictive. Clustering by AUC
0-24/MIC revealed those with lowest multidrug exposures had slowest culture conversion.
Conclusion
Amidst multidrug regimens for RR/MDR-TB, serum pharmacokinetics and
M. tuberculosis MICs were variable, yet defined parameters to c ertain drugs – fluoroquinolones, pyrazinamide, clofazimine – were predictive and should be optimized to improve clinical outcome.