Efficacy and safety results in participants co-infected with HIV from TB-PRACTECAL Clinical Trial

BACKGROUND: Globally, TB/HIV coinfection accounts for 477461 notified cases and 456000 cases of rifampicin-resistant tuberculosis (RR-TB) are estimated. TB-PRACTECAL clinical trial (NCT02589782) evaluated the safety and efficacy of three 24-week all-oral regimens for the treatment of pulmonary RR-TB in adults and adolescents above 15 years from Uzbekistan, Belarus and South Africa. Participants were randomised to receive one of three investigational regimens or the control. BPaL arm consisted of bedaquiline, pretomanid and linezolid. Clofazimine was added in BPaLC or moxifloxacin in BPaLM arm. The primary efficacy outcome was the percentage of patients with a composite unfavourable outcome (treatment failure, death, treatment discontinuation, recurrence, loss to follow-up) at 72 weeks post-randomization. We present the efficacy and safety of regimens in the HIV coinfected patients.
METHODS: All patients were offered a HIV test and were eligible irrespective of CD4 count. Antiretroviral treatment was modified to minimise drug-drug interactions and co-trimoxazole prophylaxis offered. Pre-specified analyses were conducted for primary efficacy and safety outcomes at 72 weeks post-randomization for HIV status in mITT and ITT populations, respectively.
RESULTS: 552 patients were enrolled, of whom 153 were HIV positive. Median CD4 count at baseline was 330, 297, 326 and 250 cells/µL in BPaLM, BPaLC , BPaL and control arm, respectively. In mITT population 28.6%, 35.7%, 28.6% and 40% of HIV-positive patients experienced unfavourable outcomes in BPaLM, BPaLC, BPaL and control arm, respectively. The small sample size and P-values don't provide strong evidence of interaction. In ITT population grade 3 or above and/or serious adverse events (SAEs) were no more frequent in HIV-positive versus negative patients across arms.

HIV status	Unfavourable outcome Control n/N (%)	Unfavourable outcome Experimental Arm n/N (%)	Risk difference (one-sided 98.3% CI)	Interaction p-value	Grade '¥3 and/or SAEs Control n/N (%)	Grade '¥3 and/or SAEs Experimental Arm n/N (%)	Risk difference (one-sided 98.3% CI)	Interaction p-value
BPaLM versus control
Negative Positive	26/51 (51.0) 6/15 (40.0)	3/48 (6.3) 4/14 (28.6)	-44.7% (-'to -28.1%) -11.4% (-'to 25.6%)	p = 0.08	33/56 (58.9) 10/17 (58.8)	11/55 (20.0) 3/17 (17.6)	-38.9% (-'to -20.9%) -41.2% (-'to -9.2)	p = 0.89
BPaLC versus control
Negative Positive	26/51 (51.0) 6/15 (40.0)	7/50 (14.0) 5/14 (35.7)	-37.0% (-' to -18.9%) -4.3% (-'to 33.9%)	p = 0.10	33/56 (58.9) 10/17 (58.8)	17/57 (29.8) 6/15 (40.0)	-29.1% (-'to -10.1%) -18.8% (-'to 18.0%)	p = 0.60
BPaL versus control
Negative Positive	26/51 (51.0) 6/15 (40.0)	10/46 (21.7) 4/14 (28.6)	-29.2% (-'to -9.6%) -11.4% (-'to 25.6%)	p = 0.37	33/56 (58.9) 10/17 (58.8)	11/51 (21.6) 4/18 (22.2)	-37.4% (-'to -18.8%) -36.6% (-'to -3.9%)	p = 0.97

CONCLUSIONS: Current TB-PRACTECAL data supports the use of 24-week regimens irrespective of HIV status. A trend towards the shorter regimens being more efficacious in HIV-negative patients was observed. However, this trend was not seen in the safety outcomes for the BPaL and BPaLM arms. The trial is accruing more data and will update at a later date.