Targeted Universal TB Testing With Simultaneous TPT Prescribing Among People Living With HIV in South Africa

1 Background HIV and TB in South Africa. South Africa is home to over 7.5 million people living with HIV (PLHIV), more than any other country globally. In addition, South Africa has one of the highest incidence rates of tuberculosis (TB) worldwide, estimated at 513 cases per 100,000 individuals in 2021. In recent years, several studies in high TB-burden settings have evaluated the prevalence and clinical features of asymptomatic, or subclinical, TB using universal TB testing, irrespective of symptom status. In a 2021 review of TB prevalence surveys conducted in 23 countries across Africa and Asia, a median of 50.4% of TB disease identified in population-based surveys was identified as subclinical.15 Studies have also reported on the prevalence of subclinical TB among PLHIV in South Africa.

The new approach for universal TB testing is a paradigm shift for South Africa. In the 2023 ART Clinical Guidelines, the South African National Department of Health (NDoH) called for targeted universal TB testing (TUTT) for people living with HIV as a component of the clinical evaluation for ART and TPT initiation, irrespective of symptom status. However, guidelines regarding the timing of TPT initiation are conflicting - recommending either (1) delaying TPT initiation for all patients until a negative TB test result is returned2 or (2) delaying TPT only for those patients with a positive TB symptom screen. The new TUTT approach along with ambiguity in the timing of TPT initiation has the potential to substantially reduce timely TPT initiation - missing the period of highest TB risk. In modelling approaches for TB screening and TPT delivery, the investigators found that waiting for a sputum result could decrease TPT prescribing from the current level to approximately 17-31%. A novel alternative is to provide TPT at the time of TB testing to all patients initiating or re-initiating ART (TUTT-PT).

There is a knowledge gap regarding the potential viability, safety, and effectiveness of TUTT-PT for PLHIV initiating ART. The investigator's model suggests that the TUTT-PT strategy of universal TB testing with universal TPT (and switch to anti-TB treatment for those who test positive for TB) could ensure that almost 90% of patients receive TPT and are tested for TB simultaneously. The viability of such an approach, however, depends on demonstrating the safety and effectiveness of TUTT-PT vs the standard of care. Guiding the optimal implementation strategy for TPT delivery - balancing high-levels of TPT prescribing with rapid diagnosis and treatment initiation for active TB disease - is essential to ensure the continued success of South Africa's TPT program.

The Fedisa PreventTB study has a currently active platform to compare these two strategies to continue to optimize TPT delivery. The investigators will conduct a one-year study built into Fedisa PreventTB to compare the novel approach of TUTT-PT to the standard approach to inform further TPT policy and optimize the effectiveness of the Fedisa PreventTB behavioural economics approach to TPT prescribing.

Early identification and treatment of active TB disease is critical to reduce transmission of TB. However, universal TB testing for newly initiating or re-initiating ART clients may result in delayed and/or decreased TPT prescribing at a time of high TB risk for PLHIV. Universal TB testing combined with universal TPT prescribing would ensure simultaneously high rates of TB prevention, testing, diagnosis, and treatment initiation for this high-risk group.

Study design The investigators are proposing a non-randomized pilot evaluation of a universal TPT initiation strategy within the context of TUTT implementation for PLHIV initiating or re-initiating ART. Clinics will be assigned to the novel TUTT-PT arm or standard TUTT implementation arm with the primary analysis focused on the proportion initiated on TPT within 7 days of ART start. This study will also assess the rate of new TB diagnoses, TB treatment initiation for those testing positive for TB, the prevalence of TB drug resistance at the time of TB treatment initiation, time to TPT initiation, and provider experiences with TUTT implementation and TPT provision, among others.

Study setting This study will be conducted by Perinatal HIV Research Unit (PHRU) in 4 or more public sector health facilities in the Dr Kenneth Kaunda district, North West province, South Africa that are participating in the Fedisa PreventTB study. As of 2016, the Dr Kenneth Kaunda district has a population of 742,820 with HIV prevalence estimated as 12.9%.22 The Fedisa PreventTB study was conducted in 18 public-sector health facilities providing HIV and TB related care. Thus, the PHRU team is very familiar with the study setting and has strong relationships with the provincial Department of Health, the district and sub-district offices, and the clinic managers and staff.

A primary goal is to test the safety of universal TPT with TUTT. To maximize safety in terms of immediate receipt of appropriate anti-TB therapy, each clinic, irrespective of arm, will be supported by a PHRU study team tracer. Any participant who tests positive for TB will be immediately contacted by the tracer telephonically. Those who cannot be reached via telephone will be visited at the household by the tracer to alert the participant of the TB test result and facilitate return to the clinic for switch from TPT to anti-TB treatment (or to initiate anti-TB treatment if not on TPT). All participants who test positive for TB will have two additional spot sputum specimens collected on the date of anti-TB treatment initiation. The first specimen will be used as per routine care. The second specimen will be sent for study-specific culture and drug-susceptibility testing.

Participants who test positive for TB will be referred to clinic staff for termination of TPT and initiation of anti-TB treatment. Those with trace results on Xpert Ultra will be managed as per routine care. No further interaction between the study nurse and the participant will take place thereafter.

The investigators hypothesize that TPT prescribing will be 65% in routine TUTT implementation and 80% in TUTT-PT implementation. Thus, the investigators will enroll 200 PLHIV in TUTT clinics and 200 PLHIV in clinics where the participants will receive TUTT-PT (400 total) in order to detect this difference with 80% power at a 5% level of significance while accounting for clustering by clinic (a design effect of approximately 1.3). A sample size of 400 will also enable the investigators to identify an estimated 20 new ART clients diagnosed with TB (assuming prevalence of 5%) in order to evaluate time to TB treatment initiation. In addition, throughout study implementation, the investigators will conduct ongoing monitoring of intervention safety, defined as the number of participants diagnosed with TB who remain on TPT after diagnosis by study arm.