Performance of a broth microdilution assay for routine minimum inhibitory concentration determination of 14 anti-tuberculous drugs against the Mycobacterium tuberculosis complex based on the EUCAST reference protocol

Abstract

This comparative study aimed at qualifying a broth microdilution (BMD) assay for phenotypic drug susceptibility testing (pDST) of Mycobacterium tuberculosis complex (MTBC) strains for implementation in a routine DST workflow. The assay was developed based on the EUCAST (European Committee on Antimicrobial Susceptibility Testing) reference protocol for determination of the minimum inhibitory concentration (MIC) of 14 anti-tuberculous drugs (isoniazid [INH], rifampicin [RIF], ethambutol [EMB], amikacin [AMI], moxifloxacin [MFX], levofloxacin [LFX], bedaquiline [BDQ], clofazimine [CFZ], delamanid [DLM], pretomanid [PA], para-aminosalicylic acid [PAS], linezolid [LZD], ethionamide [ETH], and cycloserine [CS]). Forty MTBC strains with various drug resistance profiles were tested to determine the agreement between MIC results and genotypic drug susceptibility testing (gDST) results derived from whole-genome sequencing (WGS). The agreement between the BMD and gDST results was solid for the majority of the drugs (average agreement 98%, range 90%-100%), including key drugs such as INH, RIF, MFX, LFX, BDQ, DLM, and PA. Ten discrepancies were identified (corresponding to 1.8% of the total number of MIC determinations) and most (8/10) were characterized by MICs equal or close to the potential critical concentration (pCC) applied in the BMD assay. Importantly, the assay can be adjusted to new drug recommendations and concentrations, tailored to local needs. We conclude that the BMD assay provides reliable results, and its implementation in our MTBC routine workflow will produce valuable data that improve our understanding and management of MTBC drug resistance.

Keywords: MIC determination; antimicrobial susceptibility; broth microdilution; tuberculosis.

Fig 1

Layout of the broth microdilution (BMD) assay for the determination of minimum inhibitory concentration of 14 anti-tuberculous drugs. The assay is distributed in three 96-well plates. Plates 1 and 2 include water-soluble drugs isoniazid (INH), rifampicin (RIF), ethambutol (EMB), amikacin (AMI), moxifloxacin (MFX), levofloxacin (LFX), para-aminosalicylic acid (PAS), linezolid (LZD), ethionamide (ETH), and cycloserine (CS). Plate 3 includes non-water-soluble drugs bedaquiline (BDQ), clofazimine (CFZ), delamanid (DLM), and pretomanid (PA). The concentration of each drug is displayed in its corresponding well in milligrams per liter (mg/L). All plates include negative controls (NC) and growth controls (GC1% and GC100%). Plate 3 also contained 0.5% dimethyl sulfoxide (DMSO).

Fig 2

Minimum inhibitory concentration (MIC) of a panel of 40 Mycobacterium tuberculosis complex (MTBC) strains against 14 anti-tuberculous drugs in the presence/absence of resistance mutations as determined by whole-genome sequencing, as determined by the broth microdilution (BMD) assay developed based on the EUCAST reference method. Susceptible genotypes and presence of specific resistance mutations or MTBC lineages are indicated by colors for each drug. Eight technical replicates of reference strain H37Rv ATCC27294 are included alongside the panel of 40 strains. The 14 drugs included isoniazid (INH), rifampicin (RIF), ethambutol (EMB), amikacin (AMI), moxifloxacin (MFX), levofloxacin (LFX), bedaquiline (BDQ), clofazimine (CFZ), delamanid (DLM), pretomanid (PA), para-aminosalicylic acid (PAS), linezolid (LZD), ethionamide (ETH), and cycloserine (CS). Drug concentrations are given in milligrams per liter. Potential critical concentrations (pCCs) are indicated by black vertical dot-dash lines. For PA, the purple vertical dotted line indicates an alternative MGIT test concentration (0.5 mg/L) according to the WHO policy statement (33).