The surge in global Mycobacterium tuberculosis (MTB) infections, compounded by the post-COVID-19 landscape, underscores the urgency for swift diagnostics, especially given the rise of multidrug-resistant (MDR) and extensively drug-resistant (XDR) strains. Conventional antimicrobial susceptibility testing (AST), notorious for its time-intensive nature, impedes the prompt identification of drug-resistant cases, posing a challenge in managing escalating global MTB infections. This study delves into nanomotion-based AST, lauded for its ultra-rapid phenotypical capabilities in the 2023 Tuberculosis Diagnostics Pipeline Report (2) for its high performance on established antitubercular agents (RIF, INH) (1). Our focus lies in evaluating the adaptability of this method for detecting third-line drugs, such as Macozinone (MCZ).

Employing the Phenotech AST device, we assessed its efficacy in distinguishing Macozinone profiles, utilizing MTB susceptible (H37Rv) and resistant (NTB1 with DprE1 mutations) strains (3). Changes in bacterial metabolism patterns were correlated with drug susceptibility profiles, and a machine learning approach gauged the device's accuracy, sensitivity, and specificity in predicting strain phenotypes.

Susceptible MTB strains exhibited a notable decline in cantilever oscillations, signifying reduced bacterial metabolism and eventual inactivation. Conversely, resistant strains remained unaffected by the drug in their environment. Classification models for the DprE1 inhibitor Macozinone demonstrated high training accuracy, surpassing 95%.

The nanomotion-based rapid AST protocol was effectively applied to the developmental drug MCZ. Crucially, the approach validated Macozinone's novel effects in vitro. The Phenotech AST device exhibits promise for direct deployment in endemic countries, facilitating timely, accurate treatment decisions for patients with results delivered in under a day (21 hours).