Protecting households on exposure to newly diagnosed index multidrug-resistant tuberculosis patients: study protocol for the PHOENIx phase 3 clinical trial

Abstract

Introduction:

Data to guide tuberculosis (TB) preventive treatment (TPT) of close contacts of people with multidrug-resistant tuberculosis (MDR-TB) are limited. While levofloxacin-based TPT has been shown safe and efficacious, alternatives are needed for those exposed to fluoroquinolone-resistant Mycobacterium tuberculosis (M. tb). The PHOENIx trial evaluates whether using a novel nitroimidazole, delamanid, in high-risk household contacts (HHCs) of patients with MDR-TB reduces their risk of developing active TB.

Methods/Design:

PHOENIx is a phase 3, open-label, multicenter clinical trial with a cluster-randomized superiority design (households form the clusters). The study objectives are to compare efficacy and safety of 26 weeks of delamanid versus isoniazid for preventing confirmed or probable TB during 96 weeks of follow-up among HHCs of adults with pulmonary MDR-TB. HHCs are defined as young children <5 years, people living with HIV or non-HIV immunosuppression, or people with evidence of M. tb infection. The study was originally designed to enroll 3,452 HHCs to provide 90% power to detect a 50% reduction in the cumulative proportion of HHCs developing confirmed or probable TB during 96 weeks of follow-up from 5% in the isoniazid arm to 2.5% in the delamanid arm. The design included a sample size re-evaluation to address uncertainty in study design assumptions.

Discussion:

Preventing MDR-TB is a global priority. Alternatives to levofloxacin-based TPT are needed since fluoroquinolone resistance is growing. PHOENIx, a phase 3 trial evaluating delamanid, is poised to inform WHO guidelines.

Introduction

Multidrug-resistant tuberculosis (MDR-TB) disease is caused by M. tb bacteria resistant to isoniazid and rifampicin, key drugs in first-line TB treatment. The World Health Organization (WHO) estimated 400,000 new MDR-TB diagnoses globally in 2023[1]. In 2021, 68% of people treated for MDR-TB had successful treatment outcomes[1]. With increased availability of rapid molecular diagnostic tools, the number of new MDR-TB diagnoses is growing, with associated increases in identification of household contacts. Most of these household contacts are likely to become infected with MDR-TB bacteria[2,3,4]; some will progress to active TB disease and possibly die without successful treatment[5,6,7,8,9,10]. Children <5 years, people living with HIV or non-HIV immunosuppression, and people with proof of M. tb infection (TBI) are at higher risk of developing active disease[4,8,11,12,13,14].

Here, we describe the PHOENIx study protocol, a phase 3, open-label, multicenter, cluster-randomized, superiority trial of delamanid (a nitro-dihydro-imidazooxazole derivative mycobacterial cell wall synthesis inhibitor with high potency against M. tb[15]) versus an active control (isoniazid) for prevention of active TB disease in high-risk household contacts (HHCs) of MDR-TB patients. PHOENIx started enrollment in 2019 when WHO guidelines recommended against treating contacts of MDR-TB patients for presumed MDR-TB infection in favor of close monitoring for at least two years and treating those who progressed to disease[14]. Some high-burden countries had guidelines for TPT with isoniazid[16] and either levofloxacin or rifampicin[17] for contacts of MDR-TB patients. In 2018, WHO guidelines conditionally recommended offering TPT to HHCs of MDR-TB patients with documented TBI using drugs tailored to the drug susceptibility pattern of the bacteria causing disease in the index patient[18]. However, given lack of high-quality evidence from randomized trials, WHO also recommended continuation of planned or ongoing randomized trials to provide evidence to inform MDR-TB TPT policy[18].

PHOENIx was designed to evaluate whether delamanid taken orally daily for 26 weeks reduces the risk of developing active TB in HHCs of MDR-TB patients (index participants). Delamanid has been approved by multiple regulatory agencies to treat MDR-TB; WHO issued guidance on its use in treating adults (2020)[19] and children and adolescents (2022)[20]. With potent anti-tuberculous activity, including activity against non-replicating organisms, and a manageable adverse event (AE) profile[21], delamanid was considered suitable for evaluation as TPT. A 26-week TPT duration was based on expecting nearly all index participants to be culture-converted by 6 months[22]. Furthermore, 6 months of isoniazid preventive treatment (IPT) is effective for drug-susceptible TB (DS-TB) contacts[23].

This paper describes key design features of the PHOENIx delamanid TPT trial and a brief discussion of how it will provide needed data for WHO guidelines, especially given recent trials of fluoroquinolone-based TPT in household contacts of people with drug-resistant TB.

Methods

Protocol Team and Sponsors

ACTG A5300B/IMPAACT I2003B Protecting Households On Exposure to Newly Diagnosed Index Multidrug-Resistant Tuberculosis Patients (PHOENIx) is jointly conducted by two clinical trials networks: ACTG (Advancing Clinical Therapies Globally for HIV/AIDS and Other Infections, formerly AIDS Clinical Trials Group) and IMPAACT (International Maternal Pediatric Adolescent AIDS Clinical Trials Network). ACTG and IMPAACT also jointly conducted a feasibility study to support development of this interventional study protocol, including assessment of needed site infrastructure[24,25,26]. PHOENIx is funded by the National Institute of Allergy and Infectious Diseases (NIAID) and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). NIAID is the sponsor for regulatory purposes. Otsuka Pharmaceutical Company, Ltd provides delamanid for the study; NIAID purchased isoniazid from Macleods.

Study Objectives

The primary study objectives are to compare efficacy and safety of 26 weeks of delamanid versus isoniazid for preventing confirmed or probable active TB during 96 weeks of follow-up among HHCs of adults with MDR-TB.

Use of an Active Control Arm

Choosing the control intervention was complex. Using a placebo versus an active control was reviewed and included a National Institutes of Health (NIH) ethics consultation. Although a placebo control was reasonable based on WHO guidelines when the trial started, there were several factors suggesting a 26-week IPT control arm might be preferred. First, IPT was low-cost and well-established as beneficial for preventing DS-TB; household contacts living in TB-endemic areas may be infected with DS-TB through community exposure[7]. Second, IPT was recommended for people living with HIV and household contacts <5 years regardless of known exposure to drug-resistant TB and generally initiated well before index patient drug susceptibility results might be available[23,27]. Finally, IPT may be effective in preventing MDR-TB[28]. PHOENIx follow-up is 96 weeks to ensure any benefit in preventing active TB is sustained beyond the TPT administration period as has been shown with IPT for DS-TB in some studies, including those in people living with HIV[29].

Study Setting

Thirty-one sites were included based on their affiliation with the NIH/NIAID/Division of AIDS ACTG and/or IMPAACT networks: 1 site each in Botswana, Brazil, Philippines, Tanzania, and Zimbabwe; 2 sites each in Haiti, India, Kenya, Uganda, and Vietnam; 3 sites each in Peru and Thailand; and 10 sites in South Africa. The sites are in countries with an estimated MDR-TB incidence of >1.5/100,000 people/year (Table 1), of which six are high MDR-TB burden countries[1]. Participating sites had demonstrated capacity to enroll MDR-TB index patients and exposed household contacts in our feasibility study[25].

Table 1.

MDR/RR-TB incidence in countries where the PHOENIx trial enrolled

Country (Number of PHOENIx Sites)	Number of MDR/RR-TB Cases (95% Uncertainty Interval)[1] (K=1000)	Estimated MDR/RR-TB Incidence per 100,000 Population/Year (95% Uncertainty Interval)[1]
Botswana (1)	250 (140–370)	10 (5.6–15)
Haiti (2)	730 (0–1.8K)	6.3 (0–15)
Zimbabwe* (1)	760 (480–1K)	4.6 (2.9–6.4)
Kenya (2)	1.2K (570–1.9K)	2.2 (1–3.5)
Uganda (2)	1.2K (650–1.8K)	2.5 (1.3–3.6)
Tanzania (1)	1.3K (310–2.3K)	2 (0.47–3.5)
Thailand (3)	2.9K (1.9K–3.8K)	4 (2.7–5.3)
Brazil (1)	3.3K (470–6.2K)	1.6 (0.22–2.9)
Peru* (3)	4.9K (2.9K–6.8K)	14 (8.7–20)
Vietnam* (2)	9.9K (5.7K–14K)	9.9 (5.7–14)
South Africa* (10)	13K (7.6K–19K)	21 (12–30)
Philippines* (1)	29K (9.6K–48K)	25 (8.4–42)
India* (2)	110K (82K–130K)	7.4 (5.7–9.1)

^*Denotes countries among the 30 high MDR/RR-TB burden countries per WHO [1]

The protocol and informed consent forms were approved by each site’s local institutional review board/ethics committee (IRB/EC) and any other applicable regulatory entity.

Study Population and Eligibility

PHOENIx enrolls both adults with MDR-TB and their HHCs. Restricting to adult MDR-TB index participants was practical; they needed to provide consent for their households to be approached for participation. Table 2 has eligibility criteria.

Table 2.

Eligibility criteria

Table 2A Index case eligibility criteria
Inclusion criteria
Individuals must meet the following inclusion criteria for study participation: Men and women age ≥18 years. Pulmonary MDR-TB defined as: Confirmation of RIF resistance and INH resistance by Adequate source documentation (including date of testing, test methodology, and test results) of RIF and INH resistance from a licensed/nationally approved* referral program, OR If either or both results are unknown or not adequately documented (as noted above), then confirmation must be obtained using a DAIDS-approved laboratory that operates according to Good Clinical Laboratory Practices (GCLP) guidelines and participates in an appropriate external quality assurance (EQA) program. Started MDR-TB treatment within the past 90 days. Ability and willingness of the index case to provide informed consent to access the household and approach household members for evaluation. Household of index case has at least one reported household contact.
Criteria for exclusion from study participation
Individuals meeting the following exclusion criteria are not eligible for study participation: Has previously enrolled into the A5300B/I2003B/PHOENIx trial as an index case or high-risk household contact, or is a member of a household which has previously enrolled into the A5300B/I2003B/PHOENIx trial.
Table 2B HHC eligibility criteria
Inclusion criteria
Individuals must meet the following inclusion criteria for study participation. If any member(s) of the household is/are not eligible or do not want to participate, all other eligible high-risk household contacts (HHCs) within the household can still participate. Currently lives or lived in the same dwelling unit or plot of land and shares or has shared the same housekeeping arrangements as the index case and who reports exposure within 90 days prior to the index case starting MDR-TB treatment. Also, shared >4 hours of indoor airspace with the index case during any 1-week period prior to the index case starting MDR-TB treatment. HHCs must be in one of the following high-risk groups: All children 0 to <5 years old at the time of enrollment, regardless of TBI or HIV status; Adults, adolescents, and children ≥5 years of age who are HIV-infected or are non-HIV immunosuppressed (defined as receiving anti-tumor necrosis factor (TNF) treatment, or being solid organ or hematologic transplant recipients), regardless of TBI test status; Adults, adolescents, and children ≥5 years of age who are TBI test positive {either TST-positive (≥5 mm) or IGRA-positive} and whose HIV status is negative, indeterminate, or unknown, and who are not non-HIV immunosuppressed. HIV-1 infection status must be documented as positive, negative, indeterminate or unknown for all HHCs. Unknown status includes those who previously tested HIV-negative but the test was performed more than one year ago. HIV testing will be offered to all HHCs with negative or unknown status. The following specific laboratory values for infants, children and adults obtained within 30 days prior to study entry. Absolute neutrophil count (ANC) ≥750 cells/mm3 Hemoglobin ≥7.4 g/dL Platelet count ≥50,000 /mm3 Creatinine ≤2 × upper limit of normal (ULN) Potassium level ≥3.0 mEq/L Aspartate aminotransferase (AST) (SGOT), alanine aminotransferase (ALT) (SGPT), ≤3 × ULN Total bilirubin ≤ 2.5 × ULN (Note: If on atazanavir (ATV), total bilirubin >2.5 × ULN permitted if direct bilirubin ≤2.5 × ULN) Albumin >3 g/dL For females of reproductive potential, negative serum or urine pregnancy test within 7 days prior to study entry. Female participants who are participating in sexual activity that could lead to pregnancy must agree to use one reliable form of contraceptive (i.e., hormonal contraceptive, condoms, IUD, diaphragm with spermicide, or cervical cap with spermicide) while receiving study treatments. For infants (0 to 1 year of age), weight ≥2.5 kg at screening. Ability and willingness of participant or legally-authorized representative (legal guardian or biological parent) to provide informed consent or assent as appropriate. Chest radiograph without evidence of active TB performed within 70 days prior to study entry for HHCs ≥2 years of age and within 30 days prior to study entry for HHCs <2 years of age. QTcF interval ≤460 ms within 30 days prior to study entry as confirmed by the central ECG reading center. Enrollment of the HHC within 30 days after the index case is enrolled. In the event that a HHC is suspected of having TB, then this window for enrollment may be extended from within 30 days to within 70 days to allow for TB testing of the HHC.
Criteria for exclusion from study participation
Individuals meeting the following exclusion criteria are not eligible for study participation: Current confirmed or probable or possible pulmonary or extrapulmonary TB, based on the criteria given in Table 4. Receipt of more than 30 cumulative days of isoniazid, rifamycin, fluoroquinolone, or delamanid in the 90 days prior to study entry. History of or current liver cirrhosis at any time prior to study entry. Evidence of acute hepatitis, such as abdominal pain, nausea and vomiting, jaundice, dark urine, and/or light stools within 90 days prior to study entry. Peripheral neuropathy ≥Grade 2 within 90 days prior to study entry according to the DAIDS Table for Grading the Severity of Adult and Pediatric Adverse Events (DAIDS AE Grading Table), corrected Version 2.1, July 2017. Known allergy/sensitivity or any hypersensitivity to components of study drugs or their formulation. Current cardiovascular disorder that is clinically relevant in the opinion of the site investigator, including but not limited to heart failure, coronary heart disease, second or third degree atrioventricular (AV) block, prolongation of the QRS complex over 120 ms (in both male and female participants), arrhythmia, or tachyarrhythmia. History of major psychiatric disorders including schizophrenia spectrum and other psychotic disorders, dementia, major depressive disorder, bipolar disorder, severe sleep disorders (e.g., narcolepsy), and severe neurodevelopmental disorders (e.g., autism spectrum disorder, intellectual disability) at any time prior to study entry. Active drug or alcohol use or dependence that, in the opinion of the site investigator, would interfere with adherence to study requirements. Serious illness requiring systemic treatment including parenteral therapy (e.g., antibiotics) and/or hospitalization within 30 days prior to study entry. Currently receiving other medication with potential for adverse drug-drug interactions, including QT prolongation. Taken an investigational drug or vaccine within 30 days prior to study entry. Has a clinical condition that in the site investigator’s opinion would interfere with study participation. Has enrolled into a TB vaccine or TB preventive therapy or TB therapeutic trial, including the A5300B/I2003B/PHOENIx trial in the two years prior to study entry. Not expected to be able to complete 96 weeks of study follow-up (e.g., seasonal or migrant workers or students who may not stay in the area).

Index participants:

An index participant is an adult (18+ years) with pulmonary MDR-TB who started MDR-TB treatment within the past 90 days and reported at least one other member in his/her household. The index participant must have proof of pulmonary MDR-TB (resistant to at least rifampicin and isoniazid); resistance to fluoroquinolones and/or second-line injectable drugs is allowed.

High-risk household contacts:

An HHC is a person who lives or has lived in the same household (i.e., a dwelling unit or plot of land with shared housekeeping arrangements) as the index participant. The HHC must report exposure to the index participant, sharing over 4 hours of indoor airspace during any 1-week period before the index participant started MDR-TB treatment. High-risk group for HHCs are defined hierarchically.

1. Children 0 to <5 years

2. Adults, adolescents, and children ≥5 years who are living with HIV or non-HIV immunosuppression

3. Adults, adolescents, and children ≥5 years with TBI (i.e., tuberculin skin test reaction ≥5 mm and/or interferon gamma release assay positive)

A key exclusion criterion for HHCs is evidence of active TB, which includes microbiology, chest x-ray abnormalities, and/or clinical sign/symptoms suggestive of TB. The screening process allows up to 70 days to rule out active TB.

Recruitment Process

With their community advisory boards, sites conducted extensive stakeholder engagement and community outreach as trial preparation. Outreach activities included community education/awareness efforts regarding MDR-TB and risk of developing TB after exposure; forging collaborations with local TB programs and hospitals; and preparing standard operating procedures for recruitment, infection control, and trial procedures. Our feasibility study informed these processes[24].

Figure 1 contains the schematic for screening and enrolling index participants and their HHCs. Household enumeration occurs at the household by site personnel working in teams of at least two people, typically at least one female and one male. An adult household member is asked to enumerate all household members and answer questions about their characteristics, including food insecurity. HHC consent s is done individually. Not all eligible HHCs need to participate. Site personnel make up to five attempts to contact potential HHCs absent at the time of initial household visit. Once a potential HHC has been identified, details of the study are carefully discussed. The potential HHC (or, when necessary, the parent or legal representative for HHCs <18 years or under guardianship) is asked to read and sign the approved protocol consent form. If under the age of consent, assent is also obtained according to the local IRB/EC requirements and regulations.

Figure 1.

Study schematic for the trial “Protecting Households on Exposure to Newly Diagnosed Index Multidrug-Resistant Tuberculosis Patients” (A5300B / I2003B/ PHOENIx): a phase 3, open-label, multicenter clinical trial with a cluster-randomized superiority design

Abbreviations: MDR, multi-drug resistant; TB, tuberculosis; HH, household; HHC, high-risk household contact; TBI, TB infection

PHOENIx was anticipated to enroll during 2019–2023. Because of the COVID-19 pandemic and lower-than-anticipated accrual rate, additional sites were added in 2023–2024. Accrual closed in 2025 and follow-up will end in 2027.

Randomization and Treatment Allocation

After an eligible index participant consents to participate and site personnel approach their household, the index participant is enrolled by site personnel using an internet link to the ACTG/IMPAACT Data Management Center’s computerized Subject Enrollment System; the household is randomized to study treatment when the first eligible HHC is enrolled. Site personnel attempt to screen potential HHCs from a household closely together in time and enroll all eligible HHCs before disclosing treatment assignment because a potential HHC might decide not to participate based on the treatment assigned to that specific household. Delaying household randomization until all potential HHCs had been screened would delay TPT initiation, which was considered inappropriate. Delays in HHC enrollment are typically for extensive testing to rule out active TB or because the potential HHC was absent at the time of enumeration. The 1:1 randomization ratio is achieved through permuted blocks and balanced by site, which is expected to help maintain balance in risk factors since the distribution of household size and high-risk group prevalence is expected to vary across sites. Because of household-level randomization, stratification by HHC high-risk group is impossible. There are no restrictions on number of households/site, HHCs/household, and HHCs/high-risk group.

Study Treatment

All enrolled HHCs within a household are assigned the same study treatment once daily for 26 weeks.

• Arm A: Delamanid

• Arm B: Isoniazid and pyridoxine (Vitamin B6; to prevent isoniazid-related nervous system disorders)

Weight-based dosing is provided in Table 3. Weight-based delamanid doses were modeled using existing adult and pediatric pharmacokinetic (PK) data. The modeled data supported using higher once-daily delamanid doses compared to twice-daily dosing used to treat TB disease. This was an important novel feature of PHOENIx as once-daily TPT is more easily implemented than a twice-daily regimen. PK analyses, explained below, have been developed to determine if the modeled dose reaches acceptable PK targets.

Table 3.

Doses of study medications by body weight

Table 3A Delamanid doses by body weight
Weight Band	Children Receiving Dispersible Tablets (5 mg and 25 mg)		Children/Adolescents/Adults Receiving Adult Tablets (50 mg)
	Daily Dose	Number of dispersible tablets to administer orally	Daily Dose	Number of adult tablets to administer orally
2.5 to <5 kg	30 mg	1 × 5 mg AND 1 × 25 mg	n/a	n/a
5 to <8 kg	30 mg	1 × 5 mg AND 1 × 25 mg	n/a	n/a
8 to <12kg	30 mg	1 × 5 mg AND 1 × 25 mg	n/a	n/a
12 to <16kg	50 mg	2 × 25 mg	50 mg	1 × 50 mg
16 to <20kg	50 mg	2 × 25 mg	50 mg	1 × 50 mg
20 to <25kg	100 mg	4 × 25 mg	100 mg	2 × 50 mg
25 to <30kg	150 mg	6 × 25 mg	150 mg	3 × 50 mg
≥30 kg (adult)	200 mg	8 × 25 mg	200 mg	4 × 50 mg

Table 3B Isoniazid doses by body weight
Pyridoxine (vitamin B6) was administered orally to all HHCs with each dose of Isoniazid, in a form appropriate for local dose recommendations and weight requirements.
If a child is unable to swallow the isoniazid tablets, they can be cut up, crushed, and mixed in a liquid or soft food.
Weight Band	Daily Dose	Number of tablets to administer orally
2.5 to <6 kg	50 mg	½ × 100 mg
6 to <10 kg	100 mg	1 × 100 mg
10 to <14 kg	150 mg	1 ½ × 100 mg
14 to <20 kg	200 mg	2 × 100 mg
20 to <24 kg	250 mg	2 ½ × 100 mg
≥24 kg (adult)	300 mg	1 × 300 mg

All study medications are taken orally, seven days per week, throughout the treatment period. The parent/caregiver of a young child administers study medications. HHCs assigned to delamanid are instructed to take it with high-fat food and at least one hour from other medications. Isoniazid can be taken with or without food. To account for missed doses (for any reason), the 26-week course of study treatment must be completed within 34 weeks.

Study treatment is not blinded since an open-label intervention is logistically easier to implement. Site personnel were explicitly asked to not disclose study drug when consulting with the protocol team in order to minimize potential bias.

Adherence to study treatment is encouraged and measured through the widely used, validated Wisepill electronic drug monitoring (EDM) device[30,31,32,33]. It provides a more objective measure of adherence patterns than self-report and offers real-time monitoring of adherence as poor adherence might affect the study’s ability to achieve its objectives. When an HHC does not open the EDM device for 3 out of 7 consecutive days, site personnel intervene by providing adherence support either over the phone or in person (at clinic or home visits). Pill counts, self-reported adherence questionnaires, and sparse PK samples are also being collected.

Primary Outcome Measures

The primary efficacy outcome measure is occurrence of confirmed or probable TB (see Table 4) at any time between study entry and the week-96 study visit.

Table 4.

Definitions of confirmed, probable, and possible pulmonary or extrapulmonary TB

For children ≥15 years old and adults Confirmed Pulmonary TB: Compatible clinical symptoms with or without abnormal chest radiograph consistent with TB And either sputum culture positive for M. tuberculosis or nucleic acid amplification test positive for M. tuberculosis or M. tuberculosis complex Confirmed Extrapulmonary TB: Systemic illness plus clinical features of involved organs And either positive culture for M. tuberculosis or nucleic acid amplification test positive for M. tuberculosis or M. tuberculosis complex Probable Pulmonary TB: Does not meet the criteria for Confirmed Pulmonary TB And compatible clinical symptoms And at least one of the following Positive sputum smear for acid fast bacilli (AFB) Abnormal chest radiograph consistent with TB Evidence of granulomata with organisms positive for AFB or caseating granulomata on lung tissue biopsy or at autopsy Positive tuberculin skin test or interferon gamma release assay And without concurrent illness that would explain the findings Probable Extrapulmonary TB: Does not meet the criteria for Confirmed Extrapulmonary TB And systemic illness plus clinical features of involved organs And clinical decision to treat for TB

For children <15 years old Confirmed Pulmonary TB: Either compatible clinical symptoms or abnormal chest radiograph consistent with TB And either sputum culture positive for M. tuberculosis or nucleic acid amplification test positive for M. tuberculosis or M. tuberculosis complex Confirmed Extrapulmonary TB: Either compatible clinical symptoms or abnormal imaging consistent with TB And either sputum culture positive for M. tuberculosis or nucleic acid amplification test positive for M. tuberculosis or M. tuberculosis complex Probable Pulmonary TB: Does not meet the criteria for Confirmed Pulmonary TB And compatible clinical symptoms And at least one of the following Abnormal chest radiograph consistent with TB with or without a course of antibiotics Evidence of granulomata with organisms positive for AFB or caseating granulomata on lung tissue biopsy or at autopsy Pleural aspirate with WBC counts, protein, or increased adenosine aminase and glucose levels suggestive of TB And clinical decision to treat for TB Probable Extrapulmonary TB: Does not meet the criteria for Confirmed Extrapulmonary TB And compatible clinical symptoms And at least one of the following Abnormal imaging consistent with TB with or without a course of antibiotics Evidence of granulomata with organisms positive for AFB or caseating granulomata on biopsy or at autopsy Ascitic tap with WBC counts, protein, or increased adenosine aminase and glucose levels suggestive of TB And clinical decision to treat for TB Possible Pulmonary TB: Does not meet the criteria for Confirmed or Probable Pulmonary TB And compatible clinical symptoms OR at least one of the following (but not both) Abnormal chest radiograph consistent with TB with or without a course of antibiotics Evidence of granulomata with organisms positive for AFB or caseating granulomata on lung tissue biopsy or at autopsy Pleural aspirate with WBC counts, protein, or increased adenosine aminase and glucose levels suggestive of TB And clinical decision to treat for TB Possible Extrapulmonary TB: Does not meet the criteria for Confirmed or Probable Extrapulmonary TB And compatible clinical symptoms OR at least one of the following (but not both) Abnormal imaging consistent with TB with or without a course of antibiotics Evidence of granulomata with organisms positive for AFB or caseating granulomata on biopsy or at autopsy Ascitic tap with WBC counts, protein, or increased adenosine aminase and glucose levels suggestive of TB And clinical decision to treat for TB

The primary safety outcome measure is permanent discontinuation of study drug due to treatment-related AEs.

Duration of Follow-Up and Study Evaluations

All index participants and HHCs are followed for 96 weeks; see Table 5 for schedules of evaluations. Index participants provide information supporting their MDR-TB diagnosis at entry (at clinic or home) and updates, particularly TB treatment outcome, at weeks 26 and 96 (abstracted from medical charts). HHC follow-up (at clinic or home) is intensive during the 26-week treatment period (weeks 2, 4, 8, 12, 16, 20, and 26), focusing on evaluations to identify AEs, incident TB, and treatment non-adherence. After week 26, less intensive follow-up (weeks 36, 48, 60, 72, and 96) focuses on identifying incident TB. More frequent measurement of liver function, serum creatinine, and QT intervals were undertaken in the first 500 HHCs enrolled, which was not expanded to the remaining HHCs following safety data review by the study’s independent Data and Safety Monitoring Board (DSMB).

Table 5.

Schedules of evaluations

Table 5A Index participant schedule of evaluations
Evaluation		Screen	Entry	Post-Entry Evaluations (Weeks)		Premature Study D/C
				26	96
Visit Window		−90 Days	Day 0	±28 Days
1	Identify Potential Index Participant & Obtain Permission to Approach HH	X
2	Documentation of MDR-TB Status	X
3	Medical & Medication History		X
4	Documentation of Chest Imaging		X
5	Documentation of HIV Status		X
6	HIV Testing if Not Tested or Previous HIV Test Result Negative (>1 year), Indeterminate, or Unknown		X
7	Mycobacterial Isolate from the Routine Program TB Lab for Storage & Other Analyses	X1	X2
8	Sputum for Mycobacteriology Testing & Storage	X
9	Chart Extraction for Vital Status, HIV Status, TB Test Results, and Treatment Status			X	X	X
10	Documentation of TB Treatment Outcome			If Indicated
1 The mycobacterial isolate from the routine program TB lab can be sent to the DAIDS-approved TB lab to perform DST using a protocol-approved molecular or phenotypic test to confirm MDR-TB if needed for index case eligibility at screening.
2 If not needed to confirm MDR-TB for eligibility, obtain the isolate at Entry.

Table 5B HHC schedule of evaluations
Evaluation		Screen A (for ≥5 years old only)	Screen B (for all HHCs)	Entry	Post-Entry Evaluations (Weeks)											Suspected Active TB	Neuropsychiatric Adverse Event	Premature Study D/C
					2	4	8	12	16	20	26	36	48	52–90 (see below for required visit weeks)	96
Visit Window		−30 Days*		Day 0	±7 Days		±4 Days					±14 Days
							Unsafe Conditions ±10 Days
HOUSEHOLD EVALUATIONS
11	HH Enumeration & Characteristics & Food Availability Questionnaires			X
12	HH Follow-up Questionnaire										X				X
HOUSEHOLD CONTACT EVALUATIONS
13	TB Exposure History			X
14	Medical & Medication History			X
15	Clinical Assessments			X	X	X	X	X	X	X	X	X	X	60, 72, 84	X	X	X	X
16	Targeted Neuropsychiatric Signs & Symptoms			X	X	X	X	X	X	X	X	X					X
17	TB Signs or Symptoms		X	X	X	X	X	X	X	X	X	X	X	60, 72, 84	X	X		X
18	Height ≥15 Years at Enrollment			X
19	Height <15 Years at Enrollment			X		X		X			X		X		X
20	Weight		X (if aged 0–365 days)	X		X	X	X	X	X	X		X		X	X		X
21	Targeted Physical Exam			X	X	X	X	X	X	X	X	X	X	60, 72, 84	X	X	X	X
22	IGRA (Unless Available from Medical Record within 90 Days)	X		a1
23	TST (Unless Available from Medical Record within 90 Days)	X		a1
Note: Both TST and IGRA testing at screening are required for HHCs ≥5 years old unless TST is not available due to global shortages or in-country supply challenges, but only one positive test is required for eligibility.
24	Documentation of HIV Status from Medical Records	X		a1
25	HIV Test if Not Tested or Previous HIV Test Result Negative (>1 year), Indeterminate, or Unknown	X		a1	If Indicated											X		X
26	Plasma HIV-1 RNA Medical Record Abstraction (HIV+ ONLY)			X
27	CD4+ Count and Percentage Medical Record Abstraction (HIV+ ONLY)			X
28	CD4+ Count and Percentage (HIV+ ONLY) Unless Available from Record within Last 6 Months			X
29	Hematology		X		If Indicated
30	Liver Function Tests for at Least First 500 HHCs Enrolled		X		X3	X3	X3	X3	X3	X3	X3	If Indicated3				a3	a3	a3
31	Liver Function Tests after at Least First 500 HHCs Enrolled		X		a3	X3	a3	X3	a3	a3	X3	If Indicated3				a3	a3	a3
32	Blood Chemistries		X	If Indicated
33	Serum Potassium		X		a3	X3	a3	X3	a3	a3	X3	If Indicated3				a3		a3
34	Serum Creatinine for at Least First 500 HHCs Enrolled		X		X3	X3	X3	a3	X3	X3	X3	If Indicated3						a3
35	Hemoglobin A1c (≥15 years ONLY)			X
36	Pregnancy Test		X		If Indicated
37	Respiratory Specimens for Mycobacteriology Testing		If Indicated													X
38	Extrapulmonary Specimens for Mycobacteriology Testing		If Indicated													X
39	Storage of MTB Isolate if Culture Positive for TB		If Indicated													X
40	Chest Radiograph, Chest CT, or Other Imaging Study		X													X
41	ECG for At Least First 500 HHCs Enrolled		X		a2, 3	X3	X3	X3	X3	a2, 3	X3	If Indicated2, 3						X If on Treatment
42	ECG After At Least First 500 HHCs Enrolled		X		a2, 3	a2, 3	X3	a2, 3	a2,3	a2, 3	X3	If Indicated2, 3						X If on Treatment
43	Stored Whole Blood in PAXgene RNA Tube			X			X				X					X
44	Stored Serum			X			X				X					X
45	Plasma and/or Dried Blood Spot for Sparse PK (HHCs on DLM ONLY)					X4		X4			X4					X If on Treatment	X If on Treatment	X If on Treatment
46	Urine for Sparse PK (HHCs on INH ONLY)					X4		X4			X4
47	DLM Intensive Plasma PK (Birth to <5 Years of Age ONLY) (HIV+ or HIV-)						X4
48	Plasma and/or Dried Blood Spot for Sparse PK for Breastfeeding Mothers Whose Child is Participating in Intensive PK Visit (HHCs on DLM ONLY)						X4
49	Whole Blood for Pharmacogenetic and Future GWAS Testing						X
50	Study Treatment Dispensed			X4		X4	X4	X4	X4	X4	a4
51	EDM Device Dispensed with Education and Regimen Planning			X4
52	EDM Data Feedback & Adherence Counseling				X4	X4	X4	X4	X4	X4	X4
53	EDM Functionality Check				X4	X4	X4	X4	X4	X4	a4
54	Quantitative Assessment of EDM Acceptability										X
55	Factors Affecting Adherence Questionnaires										X
56	Self-Reported Adherence				X4	X4	X4	X4	X4	X4	X4
57	Retention Phone Calls													52, 66, 78, 90
58	TB/Vital Status Follow-up (ONLY IF Unable/Unwilling to Have Study Visits)							X			X	X	X	60, 72, 84	X			X
59	Cost and Quality of Life Assessment			X					X						X			X If not done within last 8 wks
60	Optimizing and Evaluating the Performance of the EDM Device/RPI	Sites will use Rapid Process Improvement (RPI) methodology to optimize acceptance and operational performance of the EDM device.
* = up to 70 days is allowed to rule out active TB
a = if indicated
a1 = to be performed on HHC <5 years old only
If Indicated2/a2: perform if albumin < 3.4 g/dL or Grade ≥1 QTcF on prior visit (QTcF ≥450 ms), and/or participant has palpitations, presyncope, syncope, or chest discomfort.
3 = for HHCs who do not start study treatment or prematurely discontinue study treatment, only perform if clinically indicated
4 = do not perform for HHCs who do not start study treatment or prematurely discontinue study treatment

Pharmacokinetic Sampling and Interim Analysis in Children <5 Years

All HHCs have sparse PK samples collected at weeks 4, 12, and 26: plasma for delamanid and its metabolite delamanid-6705 and urine for isoniazid’s metabolite, acetylisoniazid. In the first 500 HHCs enrolled, sparse PK samples are analyzed to support DSMB monitoring adherence to study treatment. Thereafter, sparse PK samples are analyzed from a 10% random sample of HHCs. After study completion, a case-cohort analysis will evaluate PK in relation to outcomes of interest (e.g., active TB and AEs).

Intensive delamanid PK sampling is undertaken in 40 children <5 years at a subset of sites with relevant experience, with an interim analysis in the first 20 children to consider a study-wide dose adjustment. Sampling is taken at pre-dose (hour 0), and 2-, 4-, 6-, and 8-hours post-dose. We aim to confirm the initial model-based once-daily delamanid dosing strategy in children.

Run-In Phase for Each Site

A run-in phase assesses the feasibility of implementing delamanid and isoniazid TPT for MDR-TB-exposed HHCs and the quality of study conduct. A site’s run-in phase evaluates data through to when the site’s tenth household has been followed for at least four weeks. There is a formal review by protocol team leadership with recommendations provided to the site if any concerns are identified. Data collected during the run-in phase will contribute to the final analysis.

Suspected New Active TB: Identification and Review

At every visit, HHCs are screened for the following TB signs and symptoms.

• Cough, fever, night sweats, involuntary weight loss, sputum production, hemoptysis, chest discomfort, dyspnea, headache or stiff neck that may suggest meningitis

• Lymphadenopathy and/or abnormalities on lung examination, such as increased respiratory rate, use of accessory muscles, tracheal deviation, stridor, fremitus, rales, rhonchi, wheezing, pleural rub, or decreased breath sounds

• In children, particular attention is paid to any cough (with or without sputum production even if of acute duration), fever, night sweats, lethargy and/or reduced playfulness, altered level of consciousness, enlarged lymph nodes, involuntary weight loss or failure to gain weight, loss of appetite, or convulsions

HHCs suspected of active TB undergo chest radiograph (those in children <15 years are digitized for blinded centralized standardized interpretation by two expert pediatric pulmonologists to support independent TB outcome review) and collection of two respiratory specimens for smear microscopy, Cepheid Xpert^® MTB/RIF Ultra or other approved molecular diagnostic testing, culture (MGIT and solid) with species identification, and phenotypic drug susceptibility testing. Sputum induction is used when HHCs cannot produce sputum spontaneously. In children <5 years, appropriate respiratory sample collection methods are used (e.g., gastric aspiration).

An Independent Outcomes Review Committee (IORC) reviews site-identified suspected active TB diagnoses and any others identified in the study database based on signs/symptoms, abnormal or indeterminate chest radiographs, or TB diagnostic tests positive for M. tb. The IORC includes experts in the diagnosis of TB in adults, adolescents, and children and excludes protocol team members and anyone directly involved with the care of study participants. The IORC is blinded to study treatment assignment and site of enrollment. Only confirmed and probable active TB diagnoses (defined in Table 4) as determined by the IORC will be included in the final analysis. The IORC also reviews children according to NIH classification of intrathoracic TB[34] and deaths to assess whether the HHC had undiagnosed TB at the time of death.

Study Oversight

Besides protocol team monitoring (including medical officer review of serious adverse events) and clinical site monitoring contracted by NIAID, an independent NIAID-appointed DSMB provides oversight. The DSMB membership includes clinical trials experts, clinicians knowledgeable about TB in all age groups, and statisticians. The DSMB reviews unmasked safety and efficacy reports prepared by the unblinded protocol statisticians.

Analysis of the Primary Outcome Measures

Estimands for the primary efficacy and safety objectives are provided in Table 6, including details about analysis sets.

Table 6.

Estimands

Primary Objective 1: To estimate the efficacy of 26 weeks of delamanid versus 26 weeks of isoniazid for preventing confirmed or probable active TB during 96 weeks of follow-up
Estimand description	Ratio comparing delamanid for 26 weeks versus isoniazid and pyridoxine for 26 weeks of the probabilities of confirmed or probable active TB through 96 weeks after initiation of TB preventive therapy among child, adolescent, and adult household contacts of adult MDR-TB patients who are at high risk of developing TB because they are under 5 years of age, HIV-infected, or have latent TB infection
Treatment	Delamanid for 26 weeks versus isoniazid and pyridoxine for 26 weeks
Target population		Analysis set
Child, adolescent, and adult household contacts of adult MDR-TB patients who are at high risk of developing TB because they are under 5 years of age, HIV-infected, or have latent TB infection		All household contacts who are enrolled, irrespective of whether or not they initiate study treatment
Variable		Outcome measure
Confirmed or probable active TB through 96 weeks after initiation of TB preventive therapy		Confirmed or probable active TB at any time between Day 0 and the Week 96 study visit
Handling of intercurrent events		Handling of missing data
The following intercurrent events are relevant to the estimand: Failure to start TB preventive therapy All observations will be used to determine the variable (treatment policy strategy) Death due to any cause All observations will be censored at the last TB symptom screen (hypothetical strategy) Premature discontinuation of TB preventive therapy for any reason All follow-up through to 96 weeks will be used to determine the variable irrespective of premature discontinuation of TB preventive therapy (treatment policy strategy)		HHCs who discontinue follow-up before week 96 or who complete 96 weeks of study follow-up without an occurrence of confirmed or probable active TB will have their outcome censored at the date of their last TB symptom screen
Population-level summary measure		Analysis approach
Ratio comparing delamanid for 26 weeks versus isoniazid and pyridoxine for 26 weeks of the probabilities of confirmed or probable active TB through 96 weeks after initiation of TB preventive therapy		Ratio (delamanid to isoniazid) of the Week 96 Kaplan-Meier estimates of the probabilities of confirmed or probable active TB To obtain confidence intervals, the variance of the log10 ratio of probabilities will be calculated using a robust estimator to take account of the cluster randomized design [36]; this method is an extension of Greenwood’s formula to the setting of clustered data Individual times will be calculated as days between Day 0 and the date of the efficacy outcome as determined by the independent outcome review committee or censoring To avoid instability in the Kaplan-Meier estimates at 96 weeks due to variability among HHCs in the timing of week 96 visits, including possible late visits, efficacy outcome and censoring times will be imputed as occurring at exactly 96 weeks (i.e., 672 days) for HHCs having a week 96 visit (or death or lost to follow-up if no prior week 96 visit) within the window in the Schedule of Evaluations in the protocol (i.e., 658–686 days) or by 8 weeks after the scheduled time of exactly 96 weeks (i.e., 658–728 days) Efficacy outcomes and follow-up beyond 728 days will be excluded from the analysis
Primary Objective 2: To estimate the safety (permanently stopping study treatment due to treatment-related AEs) of 26 weeks of delamanid versus 26 weeks of isoniazid for the treatment of presumed TB infection (TBI) with MDR-TB
Estimand description	Ratio comparing delamanid for 26 weeks versus isoniazid and pyridoxine for 26 weeks of the probabilities of permanently discontinuing TB preventive therapy due to treatment-related AEs within 26 weeks of initiation among child, adolescent, and adult household contacts of adult MDR-TB patients who are at high risk of developing TB because they are under 5 years of age, HIV-infected, or have latent TB infection
Treatment	Delamanid for 26 weeks versus isoniazid and pyridoxine for 26 weeks
Target population		Analysis set
Child, adolescent, and adult household contacts of adult MDR-TB patients who are at high risk of developing TB because they are under 5 years of age, HIV-infected, or have latent TB infection		All household contacts starting study treatment
Variable		Outcome measure
Permanent discontinuation of TB preventive therapy due to treatment-related AEs within 26 weeks of initiation		Permanent discontinuation of randomized drug (delamanid or isoniazid) due to a treatment-related AE (i.e., requiring discontinuation as defined in Protocol Section 8.0, or in the opinion of the site investigator is a treatment-limiting AE)
Handling of intercurrent events		Handling of missing data
The following intercurrent events are relevant to the estimand: Non-treatment-related death Observations will be censored at the last known dose of TB preventive therapy (hypothetical strategy) Note: Treatment-related deaths are included in the outcome measure. Premature discontinuation of TB preventive therapy for any reason other than a treatment-related AE or death Observations will be censored at the last known dose of TB preventive therapy (hypothetical strategy)		HHCs who complete 26 weeks of study follow-up without permanently discontinuing study drug due to treatment-related AEs will have their outcome determined based on data available through Week 26 (i.e., censored at the last known dose of study treatment)
Population-level summary measure		Analysis approach
Ratio comparing delamanid for 26 weeks versus isoniazid and pyridoxine for 26 weeks of the probabilities of permanently discontinuing TB preventive therapy due to treatment-related AEs within 26 weeks of initiation		Ratio (delamanid to isoniazid) of the Week 26 Kaplan-Meier estimates of the probabilities of permanently discontinuing TB preventive therapy due to treatment-related AEs To obtain confidence intervals, the variance of the log ratio of probabilities will be calculated using a robust estimator to take account of the cluster randomized design [36]; this method is an extension of Greenwood’s formula to the setting of clustered data Individual times will be calculated as days between Day 0 and the date of the safety outcome or censoring To avoid instability in the Kaplan-Meier estimates at 26 weeks due to variability among HHCs in the timing of the week-26 visits, safety outcome and censoring times falling within 161 and 216 days will be imputed as occurring at exactly 26 weeks (i.e., 182 days) Safety outcomes and follow-up beyond 216 days will be excluded from the analysis

The primary efficacy analysis will use Kaplan-Meier estimates of the cumulative proportion of HHCs developing confirmed or probable active TB during 96 weeks of follow-up. HHCs who do not develop confirmed or probable active TB will be censored at their last TB symptom screening. The ratio of cumulative proportions of HHCs with active TB will be used to estimate efficacy of TPT with 26 weeks of delamanid versus isoniazid. Estimation of the ratio is on the log scale with an estimated 95% confidence interval calculated using a robust estimator of the variance to account for the cluster-randomized design[35].

The primary safety analysis is similar but examines cumulative proportion of HHCs permanently stopping study drug due to treatment-related AEs during 26 weeks of study treatment. Follow-up of HHCs who permanently stop study drug for reasons other than treatment-related AEs will be censored at their study treatment discontinuation date.

Sample Size Assumptions

The sample size of 3452 HHCs was chosen to provide 90% power to detect a 50% relative reduction in the cumulative proportion of HHCs developing confirmed or probable TB during 96 weeks of follow-up from 5% for isoniazid to 2.5% for delamanid using a two-sided test with a Type I error rate of 5%. The calculations assumed up to 15% loss to follow-up (LFU) and for interim monitoring using Haybittle-Peto stopping boundaries, with the sample size inflated by 1%. Allowance for the cluster-randomized design (versus an individually randomized design) incorporated information about the anticipated mean (m=2) and coefficient of variation (CV=0.6667) for the number of HHCs/household, and the intra-cluster correlation (ICC=0.15; AC Hesseling, personal communication[36])[37]. Instead of using the variance inflation factor (VIF; 1+ [m−1] × ICC) for when each cluster (household) has the same number of individuals, we used (1+ [m × (1+CV²) − 1] × ICC). Since the assumed ICC is positive, the sample size is larger than if we enrolled independent individuals. Since CV² is positive, the sample size is again larger than if each cluster contained the same number of individuals.

Because there was a reasonable level of uncertainty in some assumptions underlying the sample size calculation, the study design included a formal sample size re-evaluation at about 1.5 years before the proposed sample size was expected to be accrued. This was done by a statistician blinded to outcome results to preserve trial integrity, focusing on the assumed LFU rate and factors affecting the VIF. The DSMB recommended incorporating the rate of active TB in the isoniazid arm (without knowing the rate in the delamanid arm) into a second re-evaluation consistent with the Food and Drug Administration (FDA) Guidance for adaptive designs[38].

The blinded sample size re-evaluation was based on study estimates of TB incidence in the isoniazid arm, LFU, mean and CV for number of HHCs/household, and ICC. The revised sample size requires 3834 HHCs to be enrolled, which provides 80% power to detect a 50% relative reduction in confirmed or probable active TB for delamanid versus isoniazid. The decision to require 80% power reflected cost considerations and, more importantly, the evolving trial landscape with trials of levofloxacin demonstrating efficacy for TPT and attendant changes in WHO guidelines[39].

Dissemination of Trial Findings

Upon completion of the trial, the protocol team will share results at conferences and publish results in peer-reviewed journals in accordance with the Consolidated Standards of Reporting Trials (CONSORT) statement’s extension to cluster-randomized trials[40,41]. A summary of study results will be disseminated to investigators and participants before public presentation of results.

Discussion

TB is again the leading cause of death due to a single infectious agent[1]. In 2023, there were 10,800,000 TB diagnoses, of which 400,000 were due to MDR- or rifampicin-resistant (RR)-TB[1]. MDR/RR-TB is associated with worse treatment outcomes, increased risk of death, and more household-level catastrophic costs compared to DS-TB. While treatment outcomes for MDR-/RR-TB have improved in recent years, 32% of patients still have unfavorable outcomes; over 80% of households face catastrophic costs. Furthermore, an estimated 19,100,000 persons have been exposed to MDR-TB and have MDR-TBI[42]. Thus, preventing MDR/RR-TB remains a global priority.

WHO has called for randomized clinical trials with newer agents with good sterilization properties[23]. These trials need to be adequately powered and include high-risk populations such as contacts of people with MDR/RR-TB, especially children and PLWH[23]. In 2016, two trials of levofloxacin, a third-generation fluoroquinolone with good efficacy against M. tb in vitro and a core component of MDR-TB treatments for active disease, were initiated: TB-CHAMP (ISRCTN92634082) in South Africa and V-QUIN (ACTRN12616000215426) in Vietnam. These trials completed follow-up in 2023; pooled results published in December 2024 showed a 60% relative reduction in TB incidence[43,44,45]. In 2024, updated WHO guidelines made strong recommendations for levofloxacin as TPT for contacts of MDR/RR-TB[23,39]. While these results are exciting and provide robust new evidence for levofloxacin as TPT for MDR-TBI, there are important considerations and rationale for evidence to support alternatives to levofloxacin for managing MDR/RR-TB exposure. First, an estimated 18% of MDR/RR-TB patients are resistant to fluoroquinolones; levofloxacin would not be expected to be efficacious in these patients[1]. Furthermore, in some regions of the world, fluoroquinolone resistance is growing[46]. For example, India has the most MDR-TB diagnoses and has extensively used fluoroquinolones for the last decade; prevalence of fluoroquinolone-resistant TB is now 24%–33% or higher in several regions[23,47,48].

Levofloxacin and the fluoroquinolone class are critical oral antibiotics used for other clinical indications (e.g., pneumonia, urinary tract infections, and diarrheal diseases). There is concern widespread fluoroquinolone-based TPT could increase resistance to this class of drugs, rendering them less effective for treatment or prevention of MDR-TB and other more common bacterial infections[46,48]. Lastly, while fluoroquinolone TPT trials showed a 60% relative reduction in TB incidence, an increased risk of musculoskeletal events was also observed[45]. Assessing additional TPT regimens with good sterilizing properties, such as delamanid, is still very much needed.

While the protocol included multiple exclusion criteria for HHCs (Table 2B), many are required in a clinical trial involving an investigational drug. For routine TPT administration, WHO recommends exclusion of active TB in exposed household or close contacts and alignment of TPT with the drug susceptibility profile of the index patient with MDR/RR-TB [23,49].

Several difficult design decisions were made. Using an active control was based on the NIH ethics consultation recommendation given data showing potential benefit with IPT even for MDR-TB exposure and other data showing not all incident TB among exposed contacts is drug-resistant. The selected duration of exposure to index patients was based on review of programmatic guidelines and other TPT trials and a survey from our site investigators. A common definition was practical and feasible to implement: the contact had to have recent exposure within 90 days before the index patient started MDR-TB therapy. The decision to evaluate TPT in high-risk contacts of all ages was based on the recommendation at the time PHOENIx was being designed of TPT for children <5, people living with HIV, and those with evidence of TBI. We also showed in our feasibility study[50] that these were the HHCs at greatest risk of developing TB.

In summary, PHOENIx is a well-powered phase 3 registrational cluster-randomized clinical trial examining delamanid, a mycobacterial cell wall synthesis inhibitor with high potency against M. tb that is being used in newer MDR-TB regimens[15], in adult, adolescent, and child HHCs of adult MDR-TB patients and is poised to inform WHO guidelines. A dispersible formulation was approved by European Medicines Authority for children weighing ≥10 kg. Additional work has shown sugar or sugar-free suspensions can be stored at room temperature for 15 days or 30°C for 30 days, respectively[51]. Delamanid is a promising new agent to prevent MDR-TB and, if shown to be efficacious, will provide an alternative to levofloxacin in those intolerant or exposed to fluoroquinolone-resistant M tb.

Data Availability

No data were used for the research described in the article.