OR09

Drug resistant Tuberculosis (TB) continues to be a major global concern. The ability to predict emergence of antibiotic resistance is clinically important, yet still illudes us after years of molecular and evolutionary assessment of M. tuberculosis. INH resistance (INHᴿ) is commonly first to emerge among anti-TB drugs. As such, the hope is that prognosis of imminent INHᴿ and changing treatment course, would prevent emergence of resistance to INH but also to other subsequent drugs.

To assess the feasibility of INHᴿ prognostics, I considered 21,595 clinical isolates in two projects. The first set was curated by my systematic review in 2021 (PMCID: PMC8092511). It contained 9,306 clinical isolates (5,804 INHᴿ, 3,502 INHˢ) from 31 countries. The second set was collected globally by the CRyPTIC consortium (PMCID: PMC9363010) and included 12,289 clinical isolates from 27 countries on five continents. In both data sets, the three most frequently mutated loci continue to be katG315, inhA-15, and inhA-8. Time-course samples from the CRyPTIC dataset was used to develop prognosis hypotheses. The combination of the two sets were used to estimate the prognostic power of each variant.  While the prognostic power of most mutations are low, I report two mutations that combined provide acceptable prognostic value. When observed in INH-susceptible isolates, they provide a 71% probability of imminent emergence of katG315, and 40% for emergence of inhA-15. This information can be used to predict emergence and change treatment regimens in time to avoid INHᴿ emergence.