Knowledge gaps impeding Mycobacterium tuberculosis's fast and accurate phenotypic drug-susceptibility testing

Recent World Health Organization (WHO) reports estimate that ⅓ of the world’s population is infected by Mycobacterium tuberculosis complex (MTBc), inflicting 10.6 million new cases of tuberculosis (TB) and leading to 1.6 million deaths in 2021. Moreover, due to inappropriate drug use, the burden of drug-resistant TB (DR-TB) is also estimated to have increased between 2020 and 2021, with 450 000 new cases of rifampicin-resistant TB (RR-TB).

Prompt and accurate drug-susceptibility testing (DST) is essential for effective TB management and is a critical component of global efforts to combat TB, particularly DR-TB. The WHO recommends universal DST for all diagnosed TB cases.

Owing to the time-consuming culture-based phenotypic DST (pDST) methods that require sophisticated infrastructure and skilled labor, genotypic DST (gDST) methods for the MTBc have become increasingly popular. However, the most accessible rapid gDST methods, only rule in resistance as they have a limited sensitivity and do not cover the novel and re-purposed drugs. Currently available targeted next-generation sequencing (tNGS) methods that focus on specific known resistance genes still don’t include delamanid and pretomanid, two new drugs for RR-TB treatment. Whole-genome sequencing (WGS), which provides a more comprehensive view of the entire genome has limited capacity to classify all identified mutations as causing resistance or not, especially in the context of newer and repurposed drugs with a more diverse and less well-defined genetic basis of resistance, and requires specialized bioinformatics tools to interpret the results. In addition, WGS still relies on culture isolates and the process of sequencing, data analysis and interpretation can take several days to weeks, thus may not always improve turnaround time compared to current pDST methods.

For the majority of anti-TB drugs, the reference standard for determining drug susceptibility is pDST which mostly involves testing at the critical concentration (CC), the lowest concentration of a drug that inhibits growth of drug-naïve, wild-type MTBc strains, thus differentiating likely drug-sensitive strains from likely drug-resistant strains. For certain drugs there is also a clinical breakpoint (CB) established to differentiate between mutant isolates that may still respond to the drug at higher concentrations (low-level resistance) and those that are unlikely to respond to the drug at any concentration (high-level resistance). This contrasts with other bacterial pathogens, for which usually quantitative DSTs are performed by determining minimal inhibitory concentrations (MICs) and is fraught with challenges.

First, as pDST relies on growth inhibition in/on drug-containing media, fastidious MTBc strains may result in invalid results due to lack of growth in/on the drug-free control tubes/wells, or in unreliable (false-susceptible) results due to too slow growth in/on drug-containing media. Secondly, only a single CC value is used for most drugs, yielding a binary classification of resistance or susceptibility which may not always capture nuances in the level of resistance, thus not effectively guiding treatment decision-making. Thirdly, current knowledge on in-vitro susceptibility to anti-TB drugs has been largely biased toward globally dominant MTBc lineages (L2 and L4). However, other MTBc lineages, such as L1, L3, L5, and L6, are prevalent in specific geographic areas where they can co-circulate with L2 and L4. Some recent evidence suggests lineage-specific drug susceptibility and even treatment outcomes.

Using the classical pDST methods applied for MTBc (proportion method on solid medium or in the liquid-medium based BACTEC 960 system) for quantitative DST is labor-intensive and costly. As an alternative approach, the WHO has endorsed 96-well plate-based broth microdilution testing (BMD) to determine the MICs of MTBc. MIC testing offers several advantages such as,
- MIC values can quantify the level of resistance, which is not (always) possible by classic pDST or rapid molecular tests.
- MIC testing allows for defining the critical concentrations (and clinical breakpoints), especially for novel anti-TB drugs as for such drugs a knowledge gap remains on the association of potential resistance-conferring mutations and the level of resistance.
- MIC testing helps in resolving discordant gDST/pDST results for those mutations causing a MIC around the critical concentrations.
- MIC testing allows laboratories to detect systematic/technical errors more rapidly.
- Serial MIC testing may be a useful tool for treatment outcome monitoring.

Serial MIC testing provides a dynamic and comprehensive way to monitor the susceptibility of MTBc to specific drugs throughout the course of treatment. This allows healthcare providers to observe any shifts in drug susceptibility, which can guide treatment adjustments. Tracking changes in MIC values over time allows for identifying the early emergence of drug resistance, enabling timely interventions, such as modifying treatment regimens to effectively manage drug-resistant strains and prevent their further acquired resistance and spread.