ATS/CDC/ERS/IDSA Clinical Practice Guidelines for Treatment of Drug-Resistant Tuberculosis: A Two-edged Sword?

To the Editor:

On the basis of individual patient data (IPD) meta-analysis of observational studies (1), the World Health Organization released the consolidated guidelines on drug-resistant tuberculosis (TB) treatment in 2019 (2). Shortly afterward, using a data set modified from the aforementioned IPD, the American Thoracic Society (ATS)/CDC/European Respiratory Society (ERS)/Infectious Diseases Society of America (IDSA) published their official clinical practice guidelines for the treatment of drug-resistant TB (3). The ATS/CDC/ERS/IDSA have recommended the use of linezolid and bedaquiline to treat all patients with multidrug-resistant TB (MDR-TB), regardless of the drug-susceptibility testing results. Although the present guidelines have substantiated the role of linezolid and bedaquiline in the treatment of fluoroquinolone-resistant MDR-TB, the IPD meta-analysis findings might have been overextrapolated (4), with findings regarding the use of linezolid and bedaquiline for the management of fluoroquinolone-resistant MDR-TB applied to fluoroquinolone-susceptible MDR-TB. Retrospective analysis of linezolid in better-defined cohorts with MDR-TB have suggested that linezolid would be useful largely in the treatment of more complicated MDR-TB (5). Whether adding bedaquiline to fluoroquinolone would improve treatment outcomes of fluoroquinolone-susceptible MDR-TB is still being evaluated in stage 2 of the STREAM (Evaluation of a Standardised Treatment Regimen of Anti-Tuberculosis Drugs for Patients with Multidrug-Resistant Tuberculosis) trial. Furthermore, selection bias and inadequate control of confounding in the IPD meta-analysis might have yielded some findings that cannot be readily explained on a biologically plausible basis. Although the ATS/CDC/ERS/IDSA have explicitly stated that their guidelines were based on evidence of very low certainty (3), their categorical recommendation regarding use of linezolid and bedaquiline may pose a two-edged sword for TB control programs worldwide.

Intuitively, the pros of including linezolid and bedaquiline in a standard regimen for all types of MDR-TB may be greater simplicity for programmatic implementation and lesser need for drug-susceptibility testing. However, the major cons probably lie in the concern for patient safety and tolerance, especially when the standard regimen is universally applied to many patients with MDR-TB worldwide. The first global report of surveillance of adverse events in the treatment of drug-resistant TB has suggested a substantial risk of serious adverse events related to the use of linezolid and, possibly, bedaquiline (6). The underlying mechanism, clinical impact, and optimal monitoring of some potentially serious toxicities, such as those pertaining to the cardiovascular and neurological systems, are not yet fully understood. Furthermore, the expertise and resources required for monitoring such adverse drug reactions likely overwhelm capacity in a large number of MDR-TB programs with high disease burdens, particularly when comorbidities such as diabetes mellitus and HIV infection prevail. It cannot be overemphasized that suboptimal management of drug toxicities significantly contributes to poor treatment adherence and eventually contributes to unfavorable treatment outcomes.

Linezolid resistance is now mounting in many parts of the world. Rapid emergence of resistance against bedaquiline would be formidable for global TB control. Use of linezolid and bedaquiline in selected patients with MDR-TB may facilitate the optimal use of resources in a programmatic setting for management of drug adverse reactions and curtailment of drug resistance.

Directly observed treatment in the holistic patient care package likely contributed to the high treatment success rates of optimized background regimens in Trial 213 (7) and in the STREAM trial (8). With advances in rapid detection of drug resistance, optimized background regimens or shorter World Health Organization MDR-TB regimens may still have a place in the programmatic treatment of fluoroquinolone-susceptible MDR-TB in some parts of the world (8), at least currently.