Background:
endTB (NCT02754765) was an open-label Phase 3 randomized, controlled clinical trial to evaluate the efficacy and safety of five 9-month, all-oral regimens for fluoroquinolone-susceptible rifampin-resistant TB, compared to the WHO-recommended standard of care, in people 15 years of age or older. In the primary analysis, three experimental regimens (9BLMZ, 9BCLLfxZ, and 9BDLLfxZ) had noninferior efficacy compared to the control and were safe. Here, we present efficacy and safety results among people with HIV (PWH).
Methods:
endTB inclusion was irrespective of HIV status or CD4 count. The safety population included all randomized participants who started study treatment, and the modified intention-to-treat (mITT) population included those in the safety population who had a positive pre-randomization TB culture and no resistance to study drugs. The primary efficacy endpoint was favorable outcome at Week 73 post-randomization, defined as either [1] two consecutive, negative cultures (one between Weeks 65 and 73); or [2] favorable evolution. Unfavorable outcomes included death, treatment failure, drug addition/replacement, and retreatment. Safety outcomes were Grade 3-4 adverse events (AEs); serious AEs (SAEs); deaths; AEs of special interest (AESIs) defined as Grade 3-4 hepatotoxicity, myelosuppression, optic neuritis, peripheral neuropathy, or prolonged QTcF; and AEs leading to permanent discontinuation of at least one drug.
Results:
From 2017-2021, we randomized 754 participants in 7 countries; 104 (13.8%) were PWH. The safety population included 103 (99.0%) and the mITT 98 (94.2%). Median age was 39 years (range 19-70 years); 46 (46.9%) were female; median CD4 count was 296 (range 5-1294); and 61 (62.2%) were on antiretrovirals at baseline. Two of the noninferior experimental arms from the endTB trial demonstrated high efficacy in PWH, with favorable outcomes in 93.3% (9BLMZ) and 100.0% (9BCLLfxZ). The remaining three experimental arms and the control also performed well but showed relatively lower efficacy: 70.6% (9BDLLfxZ), 83.3% (9DCLLfxZ), 73.3% (9DCMZ), and 89.5% (control). Grade 3-4 AEs, SAEs, AESIs, and AEs leading to drug discontinuation were more common in the control than in experimental arms.
Conclusions:
Among regimens that were noninferior to the WHO control in the primary analysis, 9BLMZ and 9BCLLfxZ appeared to be particularly efficacious and safe for PWH. Additional research is needed to establish optimal all-oral shorter regimens in PWH.
