High Initiation but Suboptimal Completion of Tuberculosis-Preventive Treatment Among People Living With HIV on Antiretroviral Therapy, Persistent Tuberculosis Incidence, and Programmatic Implications: Findings From a 2-Year Multicountry Cohort Study.

Abstract

Background: Tuberculosis-preventive treatment (TPT) is recommended for those at risk of tuberculosis, but despite encouraging trial data, evidence of its effectiveness from routine settings with diverse TPT regimens is limited. We describe routine TPT implementation outcomes in people living with human immunodeficiency virus (HIV; PLWH) to identify bottlenecks and opportunities for improvement along the cascade of care.

Methods: From 2021-2023, we conducted a 24-month prospective cohort study of PLWH facilities in Ethiopia, South Africa and Zimbabwe. Sociodemographic and clinical data were collected at enrollment. During follow-up, we collected TPT initiation, TPT-related adverse effects, and completion data. A subgroup was allocated digital boxes as a proxy measure for assessing treatment completion. Tuberculosis diagnoses were confirmed from clinic records or symptom agnostic sputum testing using Xpert MTB/RIF assay and mycobacteria growth indicator tube (MGIT) culture at 12- and 24 months after enrollment.

Results: We enrolled 2095 participants, mostly female (n = 1489 [71.1%]), with a median age of 42 years (interquartile range, 35-50 years). Approximately 60% (n = 1312) were considered eligible for TPT, and 1110 (84.6%) initiated TPT within 60 days of antiretroviral therapy initiation or refill during follow-up (rifapentine and isoniazid [3HP] in 625 and 6 or 12 months of isoniazid [6H/12H] in 487). TPT was completed in 55.4% (406 of 733). Mild symptoms compatible with TPT adverse effects were commonly reported in 3HP versus 6H/12H (190 [31.0%] vs 51 [11.0%], respectively; P < .001]. The tuberculosis incidence rate was 1.27/100 person-years (95% confidence interval, .95-1.69) during study follow-up in a mostly long-term antiretroviral therapy population and was lower for PLWH who completed TPT (adjusted incidence rate ratio, 0.39 [.20-.78]; P = .01).

Conclusions: There were losses along the cascade, particularly at completion. In programmatic settings, we observed more potential mild symptoms from 3HP compared to 6H/12H use. Interventions to support TPT completion are needed.