# Efficacy and Safety of AI-enabled PRS Regimen VI (Clofazimine, Bedaquiline and Pyrazinamide) as Ultra-Short Course Therapy of LTBI in Non-Human Primates in a setting mimicking HIV co-infection > **NIH NIH R01** · UNIVERSITY OF CALIFORNIA LOS ANGELES · 2024 · $934,898 ## Abstract PROJECT SUMMARY/ABSTRACT The great majority of people who are infected with Mycobacterium tuberculosis (Mtb) do not develop active disease but contain the bacterium in a dormant state, a condition referred to as latent tuberculosis infection (LTBI). Many of these people reactivate tuberculosis (TB) later in life, often in association with an immunocompromised status, such as co-infection with HIV, immunotherapy for cancer or other diseases, aging, etc. An estimated 2 billion people on earth have LTBI and constitute a huge reservoir of people at risk of reactivation TB unless treated and the persistent Mtb state eliminated. Current treatment regimens for LTBI are long and burdensome, negatively impacting treatment completion. The study proposed herein seeks to examine a potentially much shorter regimen requiring as little as one or two weeks. If successful, and then replicated in humans, such a short-term regimen could change clinical practice. Our group pioneered the use of an artificial intelligence-enabled parabolic response surface (AI-PRS) platform allowing rapid identification of the most effective drug-dose combinations for treating active TB by testing only a small fraction of the total drug-dose efficacy response surface. This approach determined that bedaquiline (BDQ), clofazimine (CFZ), and pyrazinamide (PZA) at optimal dose ratios were highly synergistic and either by themselves or with a fourth drug were much more effective than standard treatment, achieving relapse-free cure in mouse models of active pulmonary TB within 3 weeks - an ~85% reduction in time versus the standard regimen for treating drug-sensitive TB. Here we propose to evaluate this 3-drug core regimen (BCZ) in non-human primates (NHPs) for treatment of LTBI in a setting mimicking co-infection with HIV. As a fourth drug is not needed to prevent emergence of resistance as in active TB therapy, the three core BCZ drugs should be more than sufficient for treatment of LTBI; current approved regimens comprise 1 or 2 drugs. To achieve our goals, we shall leverage the two MPIs’ expertise with AI-PRS technology and the NHP LTBI model. We shall initially perform limited pharmacokinetic (PK) studies of BCZ to optimize drug doses in NHPs such that blood levels are equivalent to the optimal human doses determined in a PK evaluation of these 3 core drugs as part of a similar AI-PRS derived ultra-short 4-drug regimen for treating active TB in a just initiated human study. We shall then infect NHPs with a low dose of Mtb by aerosol to establish LTBI infection; then not treat (Negative control - all expected later to reactivate TB with SIV co-infection), treat with BCZ for 1, 2, or 4 weeks, or treat with the approved regimen of isoniazid and rifapentine for 3 months (Positive control – none expected later to reactivate with SIV co-infection); and finally co-infect with SIV and monitor for reactivation TB. If short-term BCZ treatment prevents reactivation TB, this study will pave the... ## Key facts - **NIH application ID:** 10918871 - **Project number:** 1R01AI183978-01 - **Recipient organization:** UNIVERSITY OF CALIFORNIA LOS ANGELES - **Principal Investigator:** MARCUS AARON HORWITZ - **Activity code:** R01 (R01, R21, SBIR, etc.) - **Funding institute:** NIH - **Fiscal year:** 2024 - **Award amount:** $934,898 - **Award type:** 1 - **Project period:** 2024-03-01 → 2027-01-31 ## Primary source NIH RePORTER: https://reporter.nih.gov/project-details/10918871 ## Citation > US National Institutes of Health, RePORTER application 10918871, Efficacy and Safety of AI-enabled PRS Regimen VI (Clofazimine, Bedaquiline and Pyrazinamide) as Ultra-Short Course Therapy of LTBI in Non-Human Primates in a setting mimicking HIV co-infection (1R01AI183978-01). Retrieved via AI Analytics 2026-05-23 from https://api.ai-analytics.org/grant/nih/10918871. Licensed CC0. --- *[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*