A Rapid Phenotypic Drug Susceptibility Testing System for Tuberculosis

Abstract Tuberculosis (TB) is second only to COVID-19 as the most lethal cause of death from a single infectious agent. In 2020, an estimated 10 million people developed TB, nearly half a million of which were infected with drug- resistant tuberculosis (DR-TB). Early detection of infection and drug resistance is critical to controlling

DR-TB as this enables rapid engagement into effective care. Unfortunately, only 71% of newly diagnosed TB patients are ever tested for rifampicin resistance, and even fewer receive more comprehensive testing. Additionally, despite improved treatment success rates for DR-TB globally, these success rates do not reflect

upstream losses resulting from undiagnosed (missing cases) and untreated patients. Currently, bacterial culture and nucleic acid testing remain the primary methods for diagnosing infection, with smear microscopy being phased out. However, these methods present significant limitations for diagnosing drug resistance such as

lengthy time-to-result for phenotypic tests, as well as the need for a priori knowledge of resistance mutations and prohibitive cost for molecular tests. Clearly, there remains a critical need for a fast, accurate, and cost-effective DST, particularly for resource-limited settings. To address this, we propose to design and develop a rapid

phenotypic drug susceptibility test that can be easily adapted in TB endemic regions. The trehalose- based DST, termed Tre-DST, is based on novel trehalose probes, which require metabolic conversion to emit fluorescent signals, giving them their unique ability to specifically detect live Mycobacterium tuberculosis

(Mtb). Agnostic to mechanism(s) of drug resistance, Tre-DST can be used with all WHO-recommended DRTB drugs, as well as any future TB drugs as a companion diagnostic. In Aim 1, we will develop, characterize, and optimize a family of novel fluorescent trehalose probes (3HC-Tre and RMR-Tre) that are specifically

designed to improve performance over DMN-Tre and to distinguish live Mtb, making them ideal as biomarkers of drug susceptibility. We will also evaluate probe specificity to TB and probe performance across a variety of bacteria typically present in oral mucosa. In Aim 2, we will develop and optimize Tre-DST as a multi-drug DST

for first- and second-line TB drugs. We will evaluate Tre-DST’s performance in accurately determining drug resistance using drug susceptible and drug resistant Mtb strains, treated singly and in combination with the anti- TB drugs that fulfill the WHO Target Product Profile (TPP) for next generation DST. We will also evaluate several

point-of-care (POC)-friendly detection methods to optimize efficiency and cost-effectiveness. Lastly, in Aim 3, we will perform preliminary evaluation using banked clinical isolates in Johannesburg, South Africa to assess the performance of Tre-DST in the field. We will evaluate pre-validated readout methods in this clinical study,

benchmarking the performance of Tre-DST against bacterial culture.