Approaches to investigate pharmacokinetic mismatch and its effect on resistance acquisition in M. tuberculosis (PKM-TB)

The burgeoning epidemic of drug-resistant tuberculosis (DR-TB) represents a public health priority that threatens to reverse any progress made in TB control. In Africa, DR-TB underpins almost 30% of TB mortality and is extremely costly to treat.

Recent data demonstrate that, even with newer drugs like bedaquiline and linezolid, ~30% of MDR-TB patients still have unfavourable outcomes.

Development of acquired resistance has traditionally been associated with poor treatment adherence and suboptimal drug dosing schedules.

However, even strict compliance under DOT fails to prevent resistance amplification, indicating that other factors are involved.

More recently, the concept of pharmacokinetic (PK) mismatch, where bacteria are exposed to sub-therapeutic antibiotic levels, has been identified as a key driver of acquired resistance.

In addition to host and mycobacteria-related factors, several lines of evidence suggest that PK mismatch can also result from poor drug penetration into thick walled cavities.

We have recently shown that several TB drugs (e.g. moxifloxacin, ethambutol) exhibit gradients across TB cavities, which are inversely related to their corresponding minimum inhibitory concentrations. This drug-specific spatial heterogeneity effectively results in monotherapy and drives resistance amplification.

However, whether this occurs for new and repurposed drugs such as bedaquiline, linezolid and delaminid, and to what extent, remains unknown.

PK mismatch also leads to the development of micro-heteroresistance i.e. minor populations of drug-resistant bacteria undetectable by conventional drug susceptibility testing (DST) methods.

We have shown that phenotypic DST using sputum is only ~50% predictive of the bacterial drug resistance profile in the lung.

Next-generation-sequencing-based techniques, such as targeted deep sequencing, can detect much smaller populations (>0.1%) of resistant bacteria. However, how well this method correlates with DST results from the site of disease requires investigation.

This project forms part of a larger bouquet of activities on anti-microbial resistance with the majority of lab consumables funded through the South African MRC.

In Aim 1, we will investigate the drug-related mechanisms of PK mismatch for new and repurposed TB drugs (bedaquiline, linezolid and delaminid) in sputum, blood, and at the site of disease.

This is critical to gain insights into how resistance may develop to these specific drugs and may inform better drug dosing and delivery strategies. We will also explore the feasibility of measuring drug levels in exhaled breath.

In Aim 2, we will determine the utility of novel sequencing-based diagnostic readouts that are reflective of the resistance profile at the disease site will facilitate rapid individualised treatment initiation, thereby minimizing resistance amplification.

There will also be a siginifcant capacity development component to the project where the fellow will (i) undergo training in whole genome sequencing, bioinformatic analysis and mass spectrometry methods to facilitate progression of the project and knowledge transfer to other UCT scientists; (ii) train and supervise a MSc student and; (iii) generate peer-reviewed publications, present at international conference proceedings and submit grant applications.

These activities will further enhance the career progression of the fellow.