Reply to van Deun and Decroo Free

To the Editor—We thank Drs. van Deun and Decroo for their letter regarding our article on aminoglycosides and capreomycin in the treatment of multidrug-resistant (MDR) tuberculosis (TB) [1, 2]. In 2019, the World Health Organization [3], as well as the American Thoracic Society/U.S. Centers for Disease Control, and Prevention/European Respiratory Society/Infectious Diseases Society of America [4] issued new guidelines for the treatment of MDR-TB that demoted amikacin and streptomycin from first-choice to third-choice drugs, removing kanamycin and capreomycin from the list of recommended drugs, based on the results of a large individual patient data meta-analysis (IPDMA) [5]. Experienced clinicians everywhere, including ourselves, were surprised these drugs had such limited effect on successful treatment and mortality, as they had been mainstays of treatment for decades. Indeed, that was the reason for reporting our extended analysis of these data.

Van Deun and Decroo posit these drugs should be evaluated not by treatment failure, relapse, or death, but by preventing acquired resistance to core drugs (defined as those having sterilizing activity), such as fluoroquinolones, because aminoglycosides work mainly against rapidly dividing mycobacteria at alkaline (to neutral) pH. While this argument is valid, acquired resistance to core sterilizing drug(s) has a devastating effect on final treatment outcomes [6]. Those final treatment outcomes—cure versus treatment failure, relapse, or death—are the outcomes ultimately important to patients and, therefore, should be primary endpoints in evaluating the utility of any antimicrobial agent.

Moreover, the risk of acquired fluoroquinolone resistance as a function of one specific drug is difficult to measure in MDR-TB regimens because that risk increases exponentially with the extent of pretreatment drug resistance and decreases with the number of other effective drugs in the regimen; it also differs sharply by program characteristics [7]. Our analysis controlled for other drugs in the treatment regimens and susceptibility test results, as well as for disease severity and program differences.

Except for streptomycin, these drugs have not been properly evaluated for tuberculosis in randomized, controlled clinical trials (RCT). Similarly, our analysis was based on observational data, albeit from 50 different studies in 25 countries. The RCT data for streptomycin showed clearly that it had a beneficial impact on treatment outcomes, as well as on acquired resistance to other drugs in the regimens [8]. The pH dependency and activity against rapidly dividing mycobacteria are not binary, but matters of degree, leading Fox et al. [8], and Iseman [9] to conclude it has modest (Fox) or intermediate (Iseman) sterilizing activity.

Further, based on their work developing a highly successful standardized short-course regimen for MDR-TB, including 4 months of kanamycin, Van Deun and Decroo assert kanamycin remains important in the treatment of selected MDR-TB patients [10]. The standardized shorter regimen is now recommended (and widely used) for patients who have not been previously treated with second-line medicines or in whom resistance to fluoroquinolones and second-line injectable agents has been excluded. These criteria exclude approximately two-thirds of MDR-TB globally. On the other hand, our IPDMA included a large heterogeneous body of patients without such exclusions. In that context, the data speak for themselves: the benefit of kanamycin was not evident in the context of regimens containing newer, superior, oral agents, especially bedaquiline and linezolid.

Lastly, aminoglycosides have substantial nephro-, auditory, and vestibular toxicity, some of which is irreversible. Deafness is a horrible outcome. Linezolid also has substantial toxicity at higher doses in prolonged use, but it can be mitigated easily and is generally reversible. In addition, daily intramuscular injections for 4–8 months are painful, not to mention inconvenient for both patients and programs. All-oral regimens circumvent this problem.

In summary, aminoglycosides had a more important role in the past, but relative to newer oral agents, the utility of amikacin and streptomycin is more limited, while kanamycin and capreomycin have minimal if any identifiable benefit in a modern context. Our goals are the same: safer, more effective treatment, and better treatment outcomes. Discussion such as this leading to further research and improvements is vital.

Note

Potential conflicts of interest. The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

References