A partially randomised trial of pretomanid, moxifloxacin and pyrazinamide for pulmonary TB

SUMMARY BACKGROUND: Treatment for TB is lengthy and toxic, and new regimens are needed. METHODS: Participants with pulmonary drug-susceptible TB (DS-TB) were randomised to receive: 200 mg pretomanid (Pa, PMD) daily, 400 mg moxifloxacin (M) and 1500 mg pyrazinamide (Z) for 6 months (6Pa200MZ) or 4 months (4Pa200MZ); 100 mg pretomanid daily for 4 months in the same combination (4Pa100MZ); or standard DS-TB treatment for 6 months. The primary outcome was treatment failure or relapse at 12 months post-randomisation. The non-inferiority margin for between-group differences was 12.0%. Recruitment was paused following three deaths and not resumed. RESULTS: Respectively 4/47 (8.5%), 11/57 (19.3%), 14/52 (26.9%) and 1/53 (1.9%) DS-TB outcomes were unfavourable in patients on 6Pa200MZ, 4Pa200MZ, 4Pa100MZ and controls. There was a 6.6% (95% CI −2.2% to 15.4%) difference per protocol and 9.9% (95%CI −4.1% to 23.9%) modified intention-to-treat difference in unfavourable responses between the control and 6Pa200MZ arms. Grade 3+ adverse events affected 68/203 (33.5%) receiving experimental regimens, and 19/68 (27.9%) on control. Ten of 203 (4.9%) participants on experimental arms and 2/68 (2.9%) controls died. CONCLUSION: PaMZ regimens did not achieve non-inferiority in this under-powered trial. An ongoing evaluation of PMD remains a priority.

; 100 mg pretomanid daily for 4 months in the same combination (4Pa 100 MZ); or standard DS-TB treatment for 6 months. The primary outcome was treatment failure or relapse at 12 months post-randomisation. The non-inferiority margin for between-group differences was 12.0%. Recruitment was paused following three deaths and not resumed. R E S U LT S : Respectively 4/47 (8.5%), 11/57 (19.3%), 14/52 (26.9%) and 1/53 (1.9%) DS-TB outcomes were unfavourable in patients on 6Pa 200 MZ, 4Pa 200 MZ, 4Pa 100 MZ and controls. There was a 6.6% (95% CI -2.2% to 15.4%) difference per protocol and 9.9% (95%CI -4.1% to 23.9%) modified intention-to-treat difference in unfavourable responses between the control and 6Pa 200 MZ arms. Grade 3þ adverse events affected 68/203 (33.5%) receiving experimental regimens, and 19/68 (27.9%) on control. Ten of 203 (4.9%) participants on experimental arms and 2/68 (2.9%) controls died. C O N C L U S I O N : PaMZ regimens did not achieve noninferiority in this under-powered trial. An ongoing evaluation of PMD remains a priority. K E Y W O R D S : tuberculosis; drug resistance; TB treatment; TB-HIV TB is the leading cause of death from an infectious disease globally. 1 TB treatment for drug-susceptible and -resistant disease is hindered by its long duration and toxicity. 2 Pretomanid (PMD, Pa) is a member of the nitroimadazole drug class 3 that has demonstrated significant bactericidal and sterilising activity against Mycobacterium tuberculosis. 4,5 PMD has recently been approved by the US Food and Drug Administration (FDA) as part of a 6-month regimen in combination with bedaquiline (BDQ) and linezolid for the treatment of extensively drug-resistant TB (XDR-TB) and treatment-intolerant or non-responsive drug-resistant TB (DR-TB).
PMD, moxifloxacin (MFX, M) and pyrazinamide (PZA, Z) were studied in a Phase 2 clinical trial in combination, with promising results for 8-week bactericidal activity. 6 The STAND (Shortening Treatment by Advancing Novel Drugs) trial investigated the efficacy and safety of PaMZ for the treatment of both drug-susceptible (DS) and rifampicin-resistant (RR) pulmonary TB.

Study design
The study was designed as a partially randomised, open-label, non-inferiority Phase 3 clinical trial comparing three experimental treatment regimens against standard TB treatment for pulmonary DS-TB. RR

Study procedures
Chest X-rays were performed during screening. At each study visit, sputum samples were obtained, a physical examination performed and information on adverse events (AEs) collected. Electrocardiograms were performed at pre-defined visits and at the site doctor's discretion (see Section 8 in the Supplementary Data for the protocol visit schedule). AEs were graded according to the Division of Microbiology and Infectious Diseases (National Institutes of Health, Bethesda, MD, USA) criteria (available at https://www.niaid.nih.gov/sites/default/files/ dmidadulttox.pdf) with an assessment of relationship to study drugs made by the investigator. AEs were considered ''treatment-emergent'' or ''on-treatment'' if they occurred in the period from first dose of trial drug up to 14 days after last dose. Follow-up was for 24 months after randomisation (see Section 13 in the Supplementary Data).

Study outcomes
The primary efficacy, ''unfavourable'' outcome, was the proportion of participants with bacteriologically or clinically defined treatment failure or relapse 12 months after randomisation (from 50 to 54 weeks). A ''favourable'' outcome was defined as having a negative culture status (defined as two consecutive negative culture results at least 1 week apart with no intervening positive result) at 12 months if not already classified as having an unfavourable outcome. Details are provided in Table 1 Secondary efficacy endpoints included time to an unfavourable outcome, time to culture-negative status, and proportions of treatment failure or relapse 24 months after randomisation. The primary safety outcome was the proportion of participants with one or more Grade 3 or 4 AEs.

Study oversight
An independent data safety monitoring committee (DSMC) of clinicians and statisticians reviewed unblinded data and oversaw the conduct of the trial.

Statistical analysis and sample size
A sample size of 300 participants recruited to each DS-TB treatment arm was calculated to provide 90% power to show non-inferiority, with a margin of 12% Table 1 Trial analysis populations and related unfavourable outcomes: list of definitions applied at end-point review for patients in the trial by analysis population* Modified intention-to-treat population All randomised patients included, except: Late exclusions due to resistance pattern, lack of culture confirmation, protocol violation at enrolment Patients who, having completed treatment, are lost to follow-up or withdrawn from the study with their last status being culture-negative Women who become pregnant during treatment and stop their allocated treatment Patients who died during treatment from violent or accidental cause Patients who died during follow-up (after the end of treatment) with no evidence of failure or relapse of their TB Patients who, after being classified as having culture-negative status, are re-infected with a strain other than that with which they had been originally infected Patients who are able to produce sputum at 12 months, but whose 12-month visit sputum samples are all contaminated or missing, who cannot be brought back for repeat culture testing † Unfavourable outcome definitions Patients not classified as having achieved or maintained culture-negative status when last seen Patients previously classified as having culture-negative status who, following the end of treatment, have two positive cultures without an intervening negative culture Patients who had a positive culture not followed by at least two negative cultures when last seen Patients dying from any cause during the 6-month treatment phase, except from violent or accidental cause (e.g., road traffic accident), not including suicide (e.g., suicide was considered an unfavourable outcome) Patients definitely or possibly dying from TB-related cause during the follow-up phase Patients requiring a restart or a change of treatment because of an unfavourable outcome with or without bacteriological confirmation, i.e., on bacteriological, radiographic or clinical grounds Patients requiring an extension of their treatment beyond that permitted by the protocol, a restart or a change of treatment for any reason except reinfection or pregnancy Patients failing to complete an adequate course of treatment, who were unassessable at 12 months Patients lost to follow-up or withdrawn from the study before the end of treatment Per protocol population All randomised patients included, except: Late exclusions due to resistance pattern, lack of culture confirmation, protocol violation at enrolment Patients who, having completed treatment, are lost to follow-up or withdrawn from the study with their last status being culture-negative Women who become pregnant during treatment and stop their allocated treatment Patients who died during treatment from violent or accidental cause Patients who died during follow-up (after the end of treatment) with no evidence of treatment failure or relapse of their TB Patients who, after being classified as having culture-negative status, are re-infected with a strain other than that with which they had been originally infected Patients who are able to produce sputum at 12 months, but whose 12-month visit sputum samples are all contaminated or missing, who cannot be brought back for repeat cultures † Patients lost to follow-up or withdrawn before the end of treatment † Patients whose treatment was modified or extended for reasons other than an unfavourable therapeutic response to treatment † Patients not meeting the definition of having received an adequate amount of their allocated study regimen (80% of treatment by selfreporting) † Patients who are classified as ''major protocol deviations'' † Unfavourable outcome definitions: Patients not classified as having achieved or maintained culture-negative status when last seen Patients previously classified as having culture-negative status who, following the end of treatment, have two positive cultures without an intervening negative culture Patients who had a positive culture not followed by at least two negative cultures when last seen Patients dying from any cause during the 6 month treatment phase, except from violent or accidental cause (e.g., road traffic accident), not including suicide (e.g., suicide will be considered an unfavourable outcome) Patients definitely or possibly dying from TB related cause during the follow-up phase Patients requiring a restart or a change of treatment because of an unfavourable outcome with or without bacteriological confirmation, i.e., on bacteriological, radiographic or clinical grounds in the upper boundary of the two-sided 95% Wald confidence interval (CI) for the proportion with unfavourable outcome and a one-sided significance level of 2.5%.
To preserve the overall type I error, a hierarchical approach to the analysis was adopted. The first comparison was 6Pa 200 MZ vs. the control arm. Comparison between the control and the 4-month arms would only be made if this was found to be noninferior. Non-inferiority would only be demonstrated if indicated in both the mITT and PP analyses. Table 1 lists the definitions for mITT and PP populations in the trial. Participants who received at least one dose of trial medication were included in the safety analysis.

Study participants
The recruitment disposition of the participants in the trial is shown in Figure 1  The first DSMC review (8 months after enrolment began) recommended a pause in trial enrolment following three deaths associated with hepatotoxicity on the experimental regimen arms. While these deaths raised concern about the contribution of the experimental regimen, it was noted that some of the deaths involved delays in the recognition and management of hepatotoxicity, and administration of concomitant potentially hepatotoxic medications. Patients already enrolled in the trial continued their allocated treatment without interruption. A full review of the safety data relating to the trial, and the earlier Phase 2 study, 6 was undertaken and included external specialists in hepatotoxicity and the DSMC. No conclusive evidence was found supporting an unduly increased risk for severe drug-induced liver injury (DILI) on the experimental arms. The DSMC subsequently recommended resuming enrolment into the trial with additional safety monitoring in place. Nevertheless, the sponsor decided instead to pursue a Phase 3 clinical trial of a combination of BDQ with the Pa 200 MZ regimen that had demonstrated very promising bactericidal activity. 7

Subgroup analyses
There were 14/43 (32.6%) HIV-positive participants with an unfavourable outcome on the experimental arms in the mITT analysis compared to 2/15 (13.3%) receiving the control regimen. More details can be found in Section 14 of the Supplementary Data.
There were 12 deaths among participants with DS-TB and 1 death among those with RR-TB (Table 5). Two deaths on the 4Pa 200 MZ arm and one death on the 4Pa 100 MZ arm were attributed to hepatotoxicity and assessed as possibly related to the study treatment; delay in recognition of DILI and withdrawing drug were identified as contributing factors (see Section 16.2 of the Supplementary Data for more details).

DISCUSSION
It was not possible to assess whether the 6Pa 200 MZ regimen was non-inferior in terms of efficacy compared to the standard TB treatment regimen for the treatment of DS-TB because the trial was stopped early after Phase 2 trial data suggested that the addition of BDQ was associated with greater efficacy.
Experimental arms had higher proportions of treatment failure or relapse than the standard TB regimen; however, because of the small numbers these results are inconclusive.  In the safety analysis, a higher proportion of TEAE and serious AEs was observed in the experimental arms than in the standard TB regimen among DS-TB participants, and were most commonly hepatic in nature. Overall, 11% of DS-TB participants receiving experimental regimens and 6% of participants allocated to the standard TB regimen demonstrated a peak ALT 5xULN. This is higher than the proportion reported in the REMoxTB trial for the standard regimen (3% for ALT 5xULN, and 6% for peak ALT 3xULN 9 ), and suggests that the inclusion criteria for this trial could have recruited patients with higher risk of hepatoxicity (for example, lower permitted haemoglobin and CD4þ count at baseline).
In the Phase 2 study that preceded this trial, there was little difference observed in liver-related withdrawals between the experimental arms and standard TB treatment: eight participants were withdrawn because of elevated aminotransferases on each of the two experimental arms (100 mg and 200 mg PMD arms), and six participants were withdrawn for the same reason from the standard TB regimen. 6 However, the numbers of patients with peak ALT/AST 3xULN on the experimental arms in the Phase 2 study were higher than the number of liver-related withdrawals, at respectively 10/60 (16.7%) and 14/88 (15.9%) on the 100 mg and 200 mg (DS-TB and DR-TB) PMD arms, compared to 7/59 (11.8%) on standard treatment. The higher incidence of clinically significant peak ALT/AST elevations seen in the current trial compared to the Phase 2 study may be due to differences in the patient population recruited, which included more HIV-positive participants with lower CD4þ counts and a lower permitted haemoglobin at baseline. Previous work has indicated that comorbidities and less favourable baseline physiological parameters are associated with higher rates of significant hepatotoxicity with TB treatment. [10][11][12] The REMoxTB study recruitment was more restrictive than STAND, and it was hoped that this would represent the wider population of TB patients and enhance the generalisability of the results.
This report highlights the need to better understand the relationship between TB treatment and hepatotoxicity. 13 Although likely to be multifactorial, oxidative stress on hepatocytes could be a major factor. 14 Individual responses can be hard to predict, for example some participants can tolerate the same regimen when reintroduced for unknown reasons. 15 We need to focus on the pathological basis of TB treatment DILI to improve prediction of potentially hepatotoxic regimens.
Study limitations were predominantly related to the early halt in recruitment. Phase 2 and 3 clinical trials are not powered to make statistical analyses relating to safety, and this is especially true when halted early. Analysis of data using small patient numbers can over-or under-estimate efficacy and/or safety 16 associated with a certain intervention, although there were approximately 60 DS-TB participants per arm. Furthermore, very few TB-HIV coinfected patients were participants in the trial.
In this Phase 3 trial, the results for the experimental regimens compared to standard TB treatment in a Phase 2 study did not appear to translate into noninferior outcomes in either the 6-month or 4-month treatment arms. However, the study failed to reach an adequate sample size, and this limits conclusions relating to the efficacy and safety data. An ongoing evaluation of the PMD safety profile remains a priority.