Comment on: daily 300 mg dose of linezolid for multidrug-resistant and extensively drug-resistant tuberculosis: updated analysis of 51 patients.

Sir, We read with great interest the article ‘Daily 300 mg dose of linezolid for multidrug-resistant and extensively drug-resistant tuberculosis: updated analysis of 51 patients’ by Koh et al., in which they describe retrospectively examined records of one of the largest case series of patients treated with linezolid for multidrug-resistant and extensively drug-resistant tuberculosis (MDR/XDR-TB). They describe the long-term outcomes of 51 patients, whereas the previous study only described the shortterm outcomes of 17 patients. Their objective was to evaluate the efficacy, tolerability and adverse events of a 300 mg daily dose of linezolid in the treatment of MDR/XDR-TB. Based on a favourable treatment outcome of 78%, compared with 60%– 100% in the literature albeit in smaller case series, they suggest that linezolid is effective against intractable MDR/ XDR-TB at a daily dose of 300 mg. In our opinion it is difficult to draw this conclusion from the presented data. The lack of a control group makes it impossible to attribute a favourable outcome in patients to a single drug such as linezolid as it is only a part of an expanded treatment regimen. Favourable treatment outcomes could very well be caused by other drugs of the expanded regimen the patients received. To implicate the efficacy of their linezolid-containing regimen, Koh et al. make assumptions on the MIC of linezolid for their population and the linearity of linezolid pharmacokinetics. They assume their patients are infected with Mycobacterium tuberculosis with an MIC of linezolid of 0.25 mg/L. Koh et al. base this on a recent study that described wild-type MIC distributions of linezolid and six other second-line drugs for 78 consecutive susceptible clinical isolates. Although most isolates had an MIC of linezolid of 0.25 mg/L, the wildtype MIC distribution ranged from 0.125 to 0.5 mg/L and an epidemiological cut-off (ECOFF) value of 0.5 mg/L was suggested. The fact that we found the MIC to be 0.5 mg/L in eight isolates, 1 mg/L in eight isolates and even .1 mg/L in one isolate in a previous study of 23 isolates, may raise some doubt about the assumption that all clinical isolates have an MIC value of 0.25 mg/L. Koh et al. assume the pharmacokinetics of linezolid to be linear. Unfortunately, the pharmacokinetics of linezolid are not linear in TB patients, as we demonstrated in a previous study. Besides, substantial intraand interpatient variability of linezolid in TB patients can be observed. We found the AUC0–12 of 300 mg twice daily to be 56 mg.h/L, but with an IQR of 38.5–64.2 mg.h/L. We agree with Koh et al. that for linezolid the fAUC0 – 24/MIC ratio is often used as a predictive model for efficacy. Koh et al. refer to data of Schon et al. suggesting that a daily dose of 600 mg linezolid would lead to an fAUC of 56 mg.h/L, resulting in an fAUC/MIC ratio of 100 for a wild-type MICECOFF of 0.5 mg/L and of 200 for the more common MIC of 0.25 mg/L. In both cases, the pharmacodynamic target of fAUC/MIC .100 is met. However, during the study period, drug susceptibility testing (DST) as well as plasma concentration monitoring [therapeutic drug monitoring (TDM)] were not performed for linezolid. This is very unfortunate, since linezolid is the drug of interest in their study. As a consequence, it is unknown if the pharmacokinetic/pharmacodynamic target of fAUC0 – 24/MIC .100 is met. Therefore, in our opinion, it is not correct to assume linezolid to be effective without DST and TDM for linezolid or without a control group. Finally, we also do not support the conclusion that Koh et al. draw from the presented data that a daily dose of 300 mg of linezolid may be associated with fewer neuropathic side effects than a daily dose of 600 or 1200 mg. It was necessary to cease linezolid therapy in 14 patients (27%) due to neurotoxicity, which is within the range of the highly variable incidence of neurotoxicity of 4%–89% at daily doses of 600 or 1200 mg as presented by Koh et al. in an overview of the literature. Despite the daily dose of linezolid being low, the duration of administration of linezolid is long, with a median of 278 days. This concurs with the current notion that the risk of adverse events of linezolid increases time-dependently.