An Adaptive Trial to Find the Safest and Shortest TB Preventive Regimens.

DESIGN: The investigator proposes to assess safety, completion, and tolerability of three experimental Tuberculosis Preventive Treatment (TPT) regimens in a Phase 2 adaptive trial. The investigator will start the trial with the reference 4 months of Rifampin 10mg/kg (4R10) and first two experimental regimens (2 months of Rifampin 20mg/kg (2R20), one month Levofloxacin and Rifapentine (1LP); if either experimental regimen does not meet pre-determined selection criteria, the experimental regimen will be dropped; if acceptable, enrolment will be reduced so that a 3rd new experimental regimen may be tested. This will be either one month of Isoniazid plus rifapentine (1HP), or an entirely novel regimen; this choice will be made on the basis of all available pre-clinical evidence and clinical studies. Note that when this 3rd regimen is selected the protocol and informed consent will be revised to include a description of this regimen, including results from all pre-clinical and clinical studies of efficacy, tolerability and safety. The revised consent and protocol will be submitted, as an amendment, for ethics and regulatory review. Randomization will be adjusted as regimens are added or dropped. Including new regimens, and dropping those with unacceptable safety, completion, and/or tolerability, will improve trial efficiency and timeliness of results. Regimens will be assessed for acceptability (qualitatively), health system and patient costs, and drug exposure (based on pharmacokinetics (PK)) to understand potential trade-offs between dose, duration, acceptability, tolerability, and costs. After the first 100 persons have completed treatment for each new Tuberculosis Preventive Treatment (TPT) regimen, the investigators plan an early safety analysis to identify whether a regimen has substantially higher rates of severe Adverse Event (AE); if this were to occur, enrolment to that arm will stop. After 400 participants have completed each of the two initial experimental regimens, the investigators plan an 'Initial regimen selection analysis. Regimens will continue enrolment if the regimen meet the criteria for non-inferior safety, completion, and tolerability compFared to that of 4 months of Rifampin 10mg/kg (4R10). Margins for non-inferiority are stricter for safety - our primary outcome -than for completion and tolerability, our secondary outcomes (see section 2.10 for non-inferiority margins and justification). The investigators have planned for 24 months of enrolment to evaluate at least three experimental regimens. Enrolment will stop after 400 persons have been enrolled to experimental regimen 3; the Final regimen selection analysis will determine which regimen will continue in a Phase 3 trial. STUDY SITES: Calgary, Edmonton, Montreal, Ottawa, Toronto, Vancouver, and Winnipeg (Canada); Cotonou (Benin); Manaus (Brazil); Bandung (Indonesia); and Ho Chi Minh City (Vietnam). PLANNED ENROLMENT: The first 'regimen selection analysis' will be conducted after 400 participants have completed treatment in each experimental arm. The investigators anticipate enrolling 1000 total participants in 13-14 months on the basis of our past performance at all sites and adding three new sites in Canada, plus Brazil, and Benin, which were high enrolling sites in our 4v9 trial. After participant 1000 has been enrolled, the 3rd experimental regimen will be added, and randomization will be modified so that equal numbers are allocated to the 3rd regimen, and to all continuing regimens (i.e., a ratio of 3:1:1:1 for 3rd experimental: 4 months of Rifampin 10mg/kg (4R10), 2 months of Rifampin 20mg/kg (2R20), and one month levofloxacin and rifapentine (1LP), until the first regimen selection analysis is complete. To ensure timely initiation, this 3rd regimen must be selected by the end of Year 1 (at the latest), so that ethics and regulatory approval can be obtained for all sites. After the first regimen selection analysis is complete, enrolment to experimental regimens 1 and/or 2 may stop. The investigators expect 1800 total enrollees by mid-Year 3, providing 400 enrollees to the 3rd experimental regimen and an added 400 to be split between 4 months of Rifampin 10mg/kg (4R10) and experimental regimens 1 and/or 2. RANDOMIZATION: Randomization will be by computer-generated random sequences, in blocks of variable size, and stratified by site to account for previously observed differences between sites in rates of completion and Adverse Events (AEs). The investigator will allocate all members of the same household (for household contacts), to the same arm. The randomization program will be developed by the same group as for our prior Research Clinical Trials (RCTs) (demo: https://2r2-demo-ltb.cred.ca Identification: 2r2-demo Password: 2R2-demo123456789!). This will verify eligibility and if another household member was randomized. The investigator plan an initial randomization ratio of 1:2:2 for 4 months of Rifampin (4R10), 2 months of Rifampin (2R20), and 1 month Levofloxacin and Rifapentine (1LP); fewer participants need to be randomized to the 4 months of Rifampin (4R10) arm given the availability of results from our three prior 4 months of Rifampin (4R10) trials at the same sites, with the same inclusion/exclusion criteria, similar participant characteristics, and very consistent rates of Adverse Event (AE) and completion DURATION OF TREATMENT AND FOLLOW-UP: The investigators wish to ensure that burden on participants of in-person follow-up visits is similar in all arms, reduce effects of differential follow-up on treatment completion, and also reflect usual care follow-up procedures for the standard arm. Given the well-established safety of 4 months rifampin (4R), most patients receiving this regimen outside of a study (ie routine care) are seen in-person only once after 4 weeks of treatment and are called by telephone every month thereafter. Hence persons randomized to 4 months of Rifampin 10mg/kg (4R10) will be seen at 2 weeks, called by telephone at 4, 8, and 12-13 weeks, then come for a final end-of-treatment in person visit. Those randomized to one month levofloxacin and rifapentine (1LP) will be seen for routine follow-up visits at 2 weeks of treatment and end-of-treatment (4 weeks). Persons randomized to 2 months of Rifampin 20mg/kg (2R20) will be seen in person at 2 weeks, and end-of-treatment (8-9 weeks). To allow detection and response to poor tolerability, participants in both experimental arms will be called evaluated at 2 weeks; if symptoms of intolerability are reported and if judged not an adverse event, the participant will be offered the opportunity to switch to 4 months of Rifampin 10mg/kg (4R10). Routine blood tests (complete blood count, alanine transaminase (ALT), bilirubin) and blood sampling for population pharmacokinetics (PK) will be performed at 2 weeks for all persons. For all regimens, participants will be called 2 weeks after the last dose taken to detect possible late Adverse Events (AEs). During each in-person visit, participants will complete an interviewer-administered symptom questionnaire, pill counts will be performed, and new pills dispensed. During telephone follow-up visits, the same symptom questionnaire will be administered, and participants asked to report pills remaining. If a participant wishes to stop therapy, the benefits of Tuberculosis Preventive Treatment (TPT) will be re-emphasized, but the participant may stop without negative consequences. To reduce bias due to differential efforts to enhance adherence, staff will follow Standard Operating Procedures (SOPs) written pre-trial. In summary, safety monitoring will be the most intensive for persons allocated to one month levofloxacin and rifapentine (1LP), for which the investigator have the least safety data, still quite intensive monitoring for persons allocated to 2 month rifampin (2R20), for which the investigator have reassuring safety data from approximately 450 participants, and much less monitoring for the standard regimen, but which aligned with current practice. In addition to routine visits and calls, participants will also be encouraged to call the study coordinator (the study participant will be given the study cell phone number which is carried by study personnel 24/7), or TB clinic nurses at any time if the participant experience symptoms or other problems, the participant think might be related to the study medications. The participant will also be encouraged to come as 'walk-ins' to the clinic at any time if symptoms or concerns arise. Post-treatment follow-up to detect incident tuberculosis (TB) disease: Participants will be contacted every 3 months until 26 months post-randomization and asked about current symptoms of active tuberculosis (TB), any intercurrent diagnoses of active tuberculosis (TB), or hospitalization for any reason. Suspected active tuberculosis (TB) will be investigated. At study end, the investigators will also cross-check all participants with local nominal registries of tuberculosis (TB) disease notifications to detect other participants diagnosed with tuberculosis (TB) disease. PLANNED ANALYSES: Safety: Our primary analysis is performed within the Bayesian framework using a beta-binomial model that places a non-informative (flat) prior distribution on the rate parameter. Statistical significance is defined with respect to the posterior probability of non-inferiority. The investigators will use robust dynamic borrowing methods to incorporate the prior trial data available on the reference treatment into the analysis. The weight assigned to previous data will be determined in a sensitivity analysis to ensure controlling the type I error. Secondary analyses will include a Bayesian generalized mixed effect model to account for clustering. Regimen completion and tolerability: The investigators will use the same beta-binomial model for these two secondary outcomes. For both outcomes, the investigators specify a larger non-inferiority margin of 10% as the maximum tolerated difference, and a more permissive probability threshold of 90% for concluding non-inferiority. This means an experimental treatment will be deemed non-inferior to 4R10 with respect to either tolerability or completion if the posterior probability of non-inferiority exceeds 90%. Acceptability: Transcribed, translated, and anonymized interview recordings will be thematically analyzed, drawing on acceptability frameworks, to elucidate participant-prioritized criteria for Tuberculosis Preventive Treatment (TPT) acceptability. The investigators will use researcher reflexivity, inter-rater reliability, contextualization, and gather thick descriptions to enhance rigor. Qualitative findings will contextualize quantitative results, be triangulated to inform treatment tolerability and adherence data, and guide knowledge translation activities. TB Disease: This Phase 2 trial is underpowered for this outcome. Hence, the investigators will describe the incidence of TB disease by arm, using total person-time of follow-up of the modified intention to treat (MITT) population. INTERIM SAFETY ANALYSES: To optimize the safety of study participants and identify any major excess in Adverse Event (AE) rates, the investigators plan safety analyses after the first 100 participants are randomized to each new regimen. The data and safety monitoring board (DSMB) will be blinded to study arm during review. If needed, unblinding will be done. The data and safety monitoring board (DSMB) will "make judgments about the strength of the evidence and the absolute magnitude and seriousness of any safety signals", rather than absolute 'stopping rules'. PLANNED SUB-GROUP ANALYSES: The investigators will examine the association of severe Adverse Events (AEs) with age, sex, country, and comorbidities, and completion and tolerability by age, sex, country, and indication for Tuberculosis Preventive Treatment (TPT).