New Drugs for the Treatment of Tuberculosis

INTRODUCTION

In 2014, tuberculosis (TB) surpassed HIV as the leading infectious cause of death. According to the World Health Organization (WHO), there were 10 million incident cases of TB in 2017 with 1.3 million deaths.¹ Treatment success was 82%, a number that seems to be decreasing. This is alarming given that small changes in treatment efficacy meaningfully affect population-level incidence and mortality.² The first-line regimen, developed in the 1950s to 1970s, remains lengthy (≥6 months) and is unforgiving to minor lapses in adherence.³ Meanwhile, in 2017, 458,000 people developed multidrug-resistant (MDR) TB, Mycobacterium tuberculosis (M.tb.) resistant to isoniazid and rifampicin. Despite efforts to extend access to treatment, the prognosis for patients with MDR-TB remains poor, with only 55% treatment success.¹ Treatment outcomes for patients with extensively drug-resistant (XDR) TB (MDR-TB that is also resistant to fluoroquinolones and injectable agents) are exceptionally poor, and now totally drug-resistant (TDR)-TB is documented, ushering us back to the preantibiotic era.⁴ In a study of patients in South Africa with “programmatically incurable tuberculosis,” many were discharged to the community once treatment options were exhausted, and community-level transmission occurred. Despite treatment failure, 90% of their sputum isolates were sensitive to newer drugs, namely linezolid, BDQ, and delamanid.⁵

There have been recent WHO updates to guidance for MDR-TB treatment.⁶ A short-duration regimen, commonly referred to as the Bangladesh regimen, yielded promising results for a 9- to 12-month course,⁷ and subsequent cohort studies and a phase 3 clinical trial, STREAM, confirmed these impressive results.⁸ This “short-course” MDR-TB regimen still requires use of 7 drugs and is offered as a complete regimen, with no allowance for substitutions or deletions. Thus, the application of this regimen in some settings may be limited by drug-resistance patterns.⁹

Fortunately, the second wave of antituberculosis drug development, which included the diarylquinoline, BDQ, and 2 nitroimidazoles, delamanid and pretomanid, offers the potential for simple, shorter, all-oral regimens for MDR- and XDR-TB treatment. In vitro, animal, and early clinical experience with these 3 compounds has been promising, and trials are ongoing to determine the best multidrug combinations, including among children and persons living with HIV (PLWH). Emerging results from the Otsuka phase 3 randomized controlled trial (RCT) of delamanid, the short-course MDR-TB regimen (STREAM Stage 1), and pediatric pharmacokinetic (PK) and safety studies of BDQ and delamanid, led the WHO to release a rapid communication in August 2018 with updated guidance (Table 1).¹⁰ Standard-duration treatment of MDR-TB now should include fluoroquinolone, BDQ, and linezolid plus clofazimine or cycloserine, with additional drugs such as delamanid used to comprise a 4- to 5-drug regimen.

Table 1.

Standard-duration therapy for MDR-TB recommended by the WHO

Group	Drug
Group A: Include all 3	Levofloxacin (Lfx) OR moxifloxacin (Mfx, M) Bedaquiline (BDQ, B) Linezolid (LZD)
Group B: Add one or both medicines	Clofazamine (CFZ) Cycloserine (Cs) OR terizidone (Trd)
Group C: To complete regimen OR Group A or B cannot be used	Ethambutol (Emb, E) Delamanid (DLM, D) Pyrazinamide (PZA, or Z) Imipenem-cilastatin (Ipm/Cln) OR meropenem (Mpm)a Amikacin (Am) OR streptomycin (S) Ethionamide (Eto) OR prothionamide (Pto) p-Aminosalicylic acid (PAS)

^aWith amoxicillin-clavulanic acid.

From World Health Organization (WHO). WHO Technical report on critical concentrations for TB drug susceptibility testing of medicines used in the treatment of drug-resistant TB. Geneva; 2019. Available at: https://apps.who.int/iris/bitstream/handle/10665/311389/9789241550529-eng.pdf?ua=1

This article outlines the history, mechanisms of action and resistance, preclinical studies, pharmacology, clinical evaluation, and treatment niche for new drugs for TB treatment that are licensed or in late clinical testing: BDQ, delamanid, and pretomanid. Also provided is a short summary of promising new chemical entities in the “third wave” of mycobacterial drug development, those that are in earlier phases of clinical testing but show great promise for shorter or less toxic regimens.

NEW DRUGS IN CLINICAL USE OR LATE-PHASE CLINICAL TESTING

BDQ

BDQ is a diarylquinoline that blocks the proton pump of adenosine triphosphate (ATP) synthase. It was granted approval for MDR-TB at 400 mg daily for 14 days followed by 200 mg thrice weekly for 22 weeks by the US Food and Drug Administration (FDA) in 2012 and by the European Medicines Agency (EMA) in 2013, making it the first drug approved for the treatment of TB in 40 years. Shortly thereafter, the WHO released interim guidance that BDQ could be used in combination with second-line drugs to treat pulmonary MDR-TB.¹¹

Preclinical data

BDQ is bactericidal against M.tb. and bacteriostatic against other nontuberculous mycobacteria.^12,13 In mice, BDQ has sterilizing activity superior to that of rifampicin and shortens treatment duration required for cure in multidrug regimens.^14,15 BDQ-pyrazinamide (BZ) has superior activity to rifampin-pyrazinamide (RZ).^16,17 Relapse rates after 3 months of BDQ-pyrazinamide plusrifapentine (BZP) or moxifloxacin (BZM) were similar to 6 months of standard treatment.¹⁸ Pyrazinamide-free regimens, BPM or B-pretomanid (Pa)-M (BPaM), given for 4 months showed lower relapse rate than standard 6-month treatment.¹⁶ BPaMZ cured mice in 2 months.¹⁹

Clinical pharmacology

BDQ is well absorbed, and food increases bioavailability 2-fold.²⁰ BDQ is highly protein bound (99.9%) and has a large volume of distribution (V_d), greater than 10,000 L.²¹ BDQ is primarily metabolized by cytochrome P450 (CYP) 3A4, forming a metabolically active M2 metabolite. The terminal half-life of BDQ and M2 is long, ~160 days.²² BDQ’s bactericidal activity is concentration dependent.²³ BDQ exposures are lower in patients of Black race or with low albumin.^24,25 Rifampicin and rifapentine reduce BDQ concentrations substantially (70%–80%),²⁶ as does EFV (50%).²⁷ BDQ can be given with nevirapine without dose adjustment. Ritonavir-boosted lopinavir decreases BDQ clearance by 75%, so coadministration requires caution and ECG monitoring.²⁸

Resistance

Spontaneous resistance mutations to BDQ occur at rates comparable with rifampicin with no loss of fitness.²⁹ Given BDQ’s unique target, cross-resistance was not anticipated, but a study of clofazimine-resistant isolates found coexisting BDQ resistance via upregulation of the efflux pump MmpS5-MmpL5, owing to mutations in the transcriptional regulatory Rv0678.³⁰ Clinical and murine studies have implicated Rv0678 and Rv2535c.^31,32 Although BDQ resistance generally confers cross-resistance to clofazimine, the opposite is not universally true.³³ Critical concentrations of 0.25 μg/mL on 7H11 and 1 μg/mL on MGIT have been proposed.³⁴

Clinical data

In humans, BDQ monotherapy has delayed early bactericidal activity (EBA), likely because of its mechanism of shifting ATP use to alternative pathways such as glycolysis, leading to eventual (not immediate) cell death.^35-37 In a phase 2b RCT among 160 patients with pulmonary MDR-TB receiving multidrug background regimen (MBR), time to culture conversion was 83 days with BDQ versus 125 days with placebo. Twenty-four-week culture conversion was higher (79% vs 58%), as was cure (58% vs 32%). The BDQ arm had greater QTc prolongation and mortality, although deaths were late and of variable causes.³⁸ A subsequent single-arm trial of 233 patients confirmed the efficacy of 24 weeks of BDQ.³⁹ Multiple trials involving BDQ as part of all-oral or treatment-shortening regimens are currently under way in both drug-sensitive and drug-resistant TB (Table 2).

Table 2.

Recent, ongoing, or upcoming trials involving BDQ, delamanid, or pretomanid

Trial	Status	Population	Sample	Description
Diarylquinoline
BDQ
Janssen C211 (NCT02354014)	Phase 1–2 Enrolling	India, Philippines, Russian Federation, South Africa	60	Pediatric/adolescent PK, safety, tolerability of BDQ + background regimen
IMPAACT P1108 (NCT02906007)	Phase 1–2 Enrolling	Haiti, India, South Africa	72	Pediatric PK, safety, tolerability of 24 wk BDQ + BR for MDR-TB in HIV-infected/uninfected children
STREAM stage 2 (NCT02409290)	Phase 3 Enrolling	China, Ethiopia, Georgia, India, Indonesia, Republic of Moldova, Mongolia, South Africa, Uganda, Vietnam	1155	Locally used WHO approved MDR regimen OR 40 wk (Bangladesh): MXF-CFZ-EMB-PZA + INH-PTO-KAN (1st 16 wk) OR 40 wk (all-oral): CFZ-EMB-PZA-LFX-BDQ + INH-PTO (1st 16 wk) OR 28 wk CFZ-PZA-LFX-BDQ + INH-KAN (1st 8 wk)
TRUNCATE-TB (NCT03474198)	Phase 3 Enrolling	Philippines, Singapore, Thailand	900	Multiarm, multistage (MAMS) trial of ultrashort regimens in drug-sensitive TB: 8 wk RHZE, then 16 wk RH OR 8–12 wk RHZE-Lzd OR 8–12 wk RHZE-CFZ OR 8–12 wk RPT-HZ-LZD-LFX OR 8–12 wk HZE-LZD-BDQ
NExT (NCT02454205)	Phase 3 Terminated	South Africa	300	6–8 mo KAN-MXF-PZA-ETH-TRD, then 18 mo MXF-PZA-ETH-TRD OR 6–9 mo LZD-BDQ-LFX-PZA-TRD-ETH/INH
Nitroimidazoles
Delamanid
Otsuka 213 (NCT01424670)	Phase 3 Complete	Estonia, Latvia, Lithuania, Moldova, Peru, Philippines, South Africa	511	DLM or placebo × 6 mo + BR × 18–24 mo
Otsuka 232 (NCT01856634)	Phase 1 Complete	Philippines, South Africa	37	Pediatric PK, safety of DLM + BR for 10 d DLM 100 or 50 mg; pediatric formulation DLM 25, 10, or 5 mg; or optimized BR
Otsuka 233 (NCT01859923)	Phase 2 Complete	Philippines, South Africa	37	Pediatric 6-mo tolerability, PK/PD, efficacy various doses DLM + BR (follow-up 232)
MDR-END (NCT02619994)	Phase 2 Enrolling	Republic of Korea	238	9–12 mo DLM-LZD-LFX-PZA vs 20–24 mo MDR-TB standard of care
IMPAACT 2005 (NCT03141060)	Phase 1–2 Enrolling	Botswana, India, South Africa, Tanzania	48	PK, safety, tolerability of DLM + BR for HIV-infected/noninfected children with MDR-TB 24 wk DLM + BR continued
ACTG A5356	Phase 2a In development	ACTG sites globally	120	24 wk DLM-LZD 600 mg daily + BR OR 24 wk DLM-LIN 600 + BR OR 24 wk DLM-LZD 1200 every other day
ACTG 5300/IMPAACT 2003 (PHOENIx)	Phase 3 Enrolling			Treatment of adult and child contacts of MDR cases: 26 wk DLM OR 26 wk INH
Pretomanid
NC-006 STAND (NCT02342886)	Phase 3 Terminated	Georgia, Kenya, Malaysia, Philippines, South Africa, Tanzania, Uganda, Zambia	1500	24 wk PaMZ (MDR or DS-TB) OR 17 wk PaMZ (DS-TB) OR 24 wk RHZE (DS-TB)
APT (NCT02256696)	Phase 2 Enrolling	South Africa	183	8 wk Pa-rifabutin-INH-PZA, then 4 wk Pa-Rb-INH OR 8 wk Pa-RIF-INH-PZA, then 4 wk Pa-RIF-INH OR 8 wk Pa-Rb-INH-PZA, then 4 wk Pa-Rb-INH OR 8 wk RHZE, then 4 wk RH
Combinations of new drugs
NixTB (NCT02333799)	Phase 3 Complete	South Africa	109	6–9 mo BPaLzd for XDR-TB and nonresponsive MDR-TB
DELIBERATE (ACTG 5343) (NCT02583048)	Phase 2 Complete	Peru, South Africa	84	Cardiac (QT) safety of BDQ vs DLM vs BDQ plus DLM, with BR in MDR-TB × 24 wk
NC-005 (NCT02193776)	Phase 2b Complete	South Africa, Tanzania, Uganda	240	Drug-sensitive: 8 wk BPaZ (with or without BDQ loading dose) OR 4HRZE/2RH Drug-resistant: 8 wk B-Pa-MXF-PZA
TB-PRACTECAL (NCT02589782)	Phase 2–3 Enrolling	Belarus, South Africa, Uzbekistan	630	Short-course treatment, adaptive design 24 wk BDQ-Pa-LZD-MXF OR 24 wk BDQ-Pa-LZD-CFZ OR 24 wk BDQ-Pa-LZD OR local standard of care
endTB (NCT02754765)	Phase 3 Enrolling	Georgia, Kazakhstan, Kyrgyzstan, Lesotho, Peru, South Africa	750	Adaptive design 39 wk BDQ-LZD-MXF-PZA OR 39 wk BDQ-LZD-CFZ-LFX-PZA OR 39 wk BQD-DLM-LZD-LFX-PZA OR 39 wk DLM-LZD-CFZ-LFX-PZA OR 39 wk DLM-CFZ-MXF-PZA OR 86 wk local standard of care
ZeNiX (NCT03086486)	Phase 3 Enrolling	Georgia, Republic of Moldova, Russian Federation, South Africa	180	26 wk tx for XDR or nonresponsive MDR-TB, duration/dose Lzd LZD 1200 mg × 26 wk-Pa-BDQ OR LZD 1200 mg × 9 wk-Pa-BDQ OR LZD 600 mg × 26 wk-Pa-BDQ OR LZD 600 mg × 9 wk-Pa-BDQ
SimpliciTB (NCT03338621)	Phase 2–3 Enrolling	Georgia	450	Drug-sensitive: 17 wk BDQ-Pa-MXF-PZA OR 26 wk standard (2HRZE/4HR) Drug-resistant: 26 wk BDQ-Pa-MXF-PZA
IMPAACT 2020	Phase 2 In development	IMPAACT international sites	148	Children with MDR-TB 8 wk LZD + 24 wk BDQ-DMN-LFX/FLO
BEAT TB	Phase 3 In development			BDQ-DLM-LZD-CF-LFX for 6 mo treatment of MDR-TB

Abbreviations: BR, background regimen; CFZ, clofazimine; DLM/D, delamanid; DS, drug-sensitive; EMB, ethambutol; ETH, ethionamide; INH, isoniazid; KAN, kanamycin; LFX, levofloxacin; LZD/L, linezolid; MXF, moxifloxacin; Pa, pretomanid; PTO, prothionamide; PZA, pyrazinamide; Rb, rifabutin; RIF, rifampin; TRD, terizidone.

Reassuringly, large observational cohort studies, including a meta-analysis of 12,030 patients with MDR-TB and studies among PLWH, have shown that BDQ decreases rather than increases mortality.^6,40,41 Although QT prolongation of 15 to 20 milliseconds is common, there have been no reports of clinically important cardiac toxicity, even with therapy up to 12 months.^42,43 Although most data are for BDQ added to MBR, it also confers benefit when substituted for injectables.⁴⁴

Delamanid

Delamanid (OPC-67683, DLM) is a nitroimidazole that inhibits synthesis of ketomycolate, a cell wall lipid that constitutes one-third of M.tb’s dry weight.⁴⁵ Delamanid was approved by EMA in 2014 at a dose of 100 mg twice daily for 6 months and is recommended by WHO for MDR-TB in certain circumstances. It is approved in Europe but is not yet FDA approved.

Preclinical data

DLM has an extremely low minimal inhibitory concentration (MIC) against M.tb. (0.006–0.024 μg/mL) and displays activity against replicating and dormant bacilli.⁴⁶ Animal studies are limited, although a guinea pig model of cavitary disease demonstrated activity.⁴⁷ In a mouse model, DLM-RZ had similar treatment efficacy at 4 months as RZHE for 6 months.⁴⁶

Clinical pharmacology

Delamanid’s absorption is enhanced by food and may be reduced by concomitant dosing with other medications.⁴⁸ It displays less-than-proportional PK.⁴⁹ Delamanid is catalyzed to DM-6705 by a pathway that involves albumin.⁵⁰ DLM is not a substrate, inhibitor, or inducer of CYP enzymes. Whereas the half-life of the parent drug is 38 hours, DM-6705 has a terminal half-life of 121 to 322 hours.⁵¹ There were no clinically significant interactions when DLM was coadministered with efavirenz, lopinavir/ritonavir, or tenofovir.⁴⁸

Resistance

Both delamanid and pretomanid (see later discussion) require activation involving coenzyme F₄₂₀. Resistance mutations have been identified in fbiA, fbiB, and fbiC (involved in F₄₂₀ biosynthesis) and fgd1 and ddn (prodrug activation). The F₄₂₀ cofactor is not present in mammalian cells, explaining the narrow spectrum of activity.⁵² There are limited data on treatment-induced resistance, but case reports describe emergence of fgd1, fbiA, fbiB, and fbiC mutants.^33,53 A break point of 0.016 μg/mL on 7H11 and 0.06 μg/mL on MGIT is proposed.³⁴

Clinical data

Phase 2a EBA trials demonstrated measurable but overlapping activity of different doses of delamanid.⁵¹ In a phase 2b RCT in 481 patients with MDR-TB receiving MBR, delamanid 100 mg or 200 mg twice daily given for 8 weeks had higher culture conversion at 2 months (45% and 42%) than placebo (30%).⁴⁹ Modest QT prolongation was seen. In a nonrandomized follow-on study, longer delamanid use was associated with favorable outcomes (74.5% vs 55%) and decreased mortality (1.0% vs 8.3%).⁵⁴ In the phase 3 RCT of delamanid versus placebo added to MBR, overall treatment success at 30 months was high in both arms (>80%), so a statistically significant difference was not detected.⁵⁵ Nonetheless, given the measurable activity of the drug in the phase 2 program coupled with the drug’s excellent safety profile, the WHO includes DLM among drugs that can be used for MDR-TB. Simplified regimens including delamanid are under study (see Table 2). Though marketed for use for 6 months, it has been well tolerated for up to 20 months.^56,57

Pretomanid

Like delamanid, pretomanid (Pa) (PA-824) is a nitroimidazole that inhibits mycolic acid synthesis in actively multiplying bacilli. In addition, when it is activated by the M.tb.-specific F420-dependent nitroreductase, toxic reactive nitrogen species are released, killing nonreplicating bacilli.^58,59 A new drug application to the FDA was approved in 2019.

Preclinical data

Pretomanid’s MIC against M.tb. is between 0.015 and 0.25 μg/mL for drug-sensitive strains and 0.03 to 0.53 μg/mL for drug-resistant strains.⁶⁰ Pretomanid is active against “persisters”⁶¹ and under oxygen-poor conditions.⁶⁰ Mouse models have revealed time above MIC to be the PK driver for nitroimidazoles.⁶² PaMZ in mice reduced treatment time by 1 to 2 months compared with standard therapy,⁶³ and the addition of BDQ (BPaMZ) cured mice in 2 months.¹⁹

Clinical pharmacology

Pretomanid has a half-life of 16 to 20 hours, is 94% protein bound, and has a large V_d^64,65; PK is dose proportional up to 200 mg, then less-than-dose-proportional. CYP3A is responsible for 20% of its metabolism.⁶⁶ In a healthy volunteer trial, efavirenz and rifampicin decreased pretomanid trough concentrations by 46% and 85% and exposures (area under the curve) by 35% and 66%, respectively, raising concerns for coadministration.⁶⁶ Lopinavir/ritonavir did not affect pretomanid exposures meaningfully.

Resistance

Like delamanid, resistance to pretomanid can be related to prodrug activation (fgd1, ddn) or the F420 biosynthetic pathway (fbiA, fbiB, fbiC).⁵² Resistance is present in 1 in 10⁵ organisms pre-treatment. Mutations in Rv2983 in mouse models disrupt F420 biosynthesis, a newly identified mechanism of resistance yet to be identified in clinical samples.⁶⁷

Clinical data

In the first EBA study of pretomanid, doses of 200 to 1200 mg produced similar, measurable activity.⁶⁸ In a second EBA, 50 mg was less active than 100 to 200 mg. Pretomanid causes dose-dependent increases in creatinine and modest QT prolongation.⁶⁹ In an EBA study of multidrug regimens, BPaZ had the highest microbiological activity, and 3 patients receiving regimens containing BPa had grade 3 or 4 transaminase elevations.⁷⁰ An 8-week study of PaMZ in drug-sensitive TB showed higher 8-week culture conversion (66%–71%) than HRZE (38%).⁷¹ A phase 3 RCT (STAND) opened in 2015 to investigate PaMZ, but was put on temporary clinical hold to investigate cases of fatal hepatitis. In the mean-time, a different trial showed exceptionally high culture conversion rates in patients with MDR-TB receiving BPaMZ,⁷² and for this reason the SimpliciTB phase 3 RCT was launched to explore this regimen. Remarkably, in patients with XDR-TB, thought nearly untreatable, a regimen of BDQ-pretomanid-linezolid (BPaLz) tested in the NixTB trial showed favorable outcomes in ~90% of patients. Duration-dependent myelosuppression and neuropathy from linezolid are common; therefore, ZeNix was started to evaluate optimal linezolid dose and duration in BPaLZ regimens.⁷³

PRACTICAL ASPECTS OF TREATMENT WITH NEW DRUGS

Coadministration

There is mounting evidence to support concurrent use of BDQ and a nitroimidazole, despite initial concerns for QT prolongation risk. Medecins Sans Frontières has endorsed judicious use of BDQ and delamanid since 2016 and has reported excellent outcomes (74% conversion by 6 months) and infrequent QT prolongation in a small cohort.⁷⁴ Other small series have reported similar efficacy results,^75-77 with episodes of QTc interval greater than 500 milliseconds generally occurring only with concurrent moxifloxacin or clofazimine.⁴² Interim results from DELIBERATE, an AIDS Clinical Trials Group RCT, suggest no more than additive QT prolongation when BDQ and delamanid are combined. If BDQ and delamanid will be coadministered, considerations include: (1) repletion of K⁺ and Mg²⁺ electrolytes; (2) measuring QTcF at baseline and serially; and (3) close attention to companion drugs (fluoroquinolones, clofazimine, ritonavir, methadone, and aminoglycosides all prolong QT). Pretomanid has a modest effect on the QT interval, and with moxifloxacin did not cause QTc greater than 500 milliseconds.⁷¹ QTcF was prolonged by ~20 milliseconds in patients receiving BPaZ or BPaMZ, which is only clinically significant if baseline QTcF is high.⁷⁸ Although the WHO recommends BDQ in the standard (18–24 months) duration regimen, it is licensed only for 24 weeks; guidance regarding prolonged use is awaited.

Extrapulmonary Tuberculosis

There are no data for pleural or pericardial TB and only limited data for central nervous system (CNS) TB. In one patient with MDR-TB meningitis (TBM), BDQ levels in cerebrospinal fluid (CSF) were undetectable, but collection conditions were not optimized for measurement of this highly lipophilic drug.^79,80 A study in rats using radiolabeled delamanid showed lasting distribution into the brain.⁸¹ Recent work in a rabbit model of TBM showed high brain but low CSF levels. Three patients with XDR-TBM, which is generally fatal, who were treated with delamanid containing regimens displayed marked clinical improvement despite very low delamanid CSF/plasma ratios.⁸² In rats, pretomanid seems to cross the blood-brain barrier and penetrate into the brain, although clinical correlation is needed.⁸³

Children and Pregnant Women

Initial results from Otsuka trials 232/233 demonstrated acceptable PK and safety of delamanid in HIV-uninfected children ≥3 years old; therefore, the WHO now recommends DLM for that age group. The WHO recommends BDQ in children with MDR-TB aged 6 to 17 years.⁸⁴ Trials are ongoing to optimize pediatric dosing (BDQ: Janssen C211, IMPAACT P1008; delamanid: Otsuka 233, 234, IMPAACT 2005). There is not yet a pediatric formulation of BDQ, but suspension or crushing gives similar bioavailability to whole tablets.⁸⁵ Dispersible formulations of BDQ and delamanid are in development. Safety and efficacy of these drugs in children seem similar to those in adults, although data are limited.^86-88 The pediatric investigation plan for pretomanid is progressing. There exist little data about use of these drugs in pregnancy⁸⁹; given that these drugs are life-saving in MDR-TB, PK and safety studies in this population are needed.

THE DRUG DEVELOPMENT PIPELINE

The “third wave” of TB drug development is now here (Table 3). Some compounds are from drug classes already in clinical use for TB but have been redesigned to optimize bioavailability, potency, safety, or activity against resistant strains (fluoroquinolones, diarylquinolines, riminophenazines, carbapenems, oxazolidinones, nitroimidazoles). Many are from new classes with novel mechanisms of action (eg, inhibitors of DprE1, leucyl-tRNA synthetase, cholesterol catabolism). Many of these compounds are now in phase 1 or phase 2 trials.^90-117

Table 3.

New chemical entities in development for treatment of tuberculosis

Name	Developer	Supporting Data	Preclinical/Clinical	Trial Objectives	Trial Design
aEthylenediamine: targets Mmpl3 involved in mycolate transport and processing, disrupting incorporation into cell wall
SQ109	Sequella, Inc	Phase 2 EBA/safety with gastrointestinal upset, no QT prolongation, significant decline in exposure when administered at 150 mg with rifampin (overcome at 300 mg SQ109) Decreases RIF MIC; No EBA as monotherapy or additive to others; No difference in culture conversion rate Patient with INH resistance on RHZ-SQ109 developed RIF/PZA resistance	SQ109 EBA NCT01218217 (2a, published)	EBA, safety, tolerability, PK of several doses with/without RIF	14 d: SQ109 75, 150, 300 as monotherapy OR SQ109 15, 300 + RIF OR RIF monotherapy
			MAMS-TB, NCT01785186 (2b, published)	MAMS trial to evaluate 4 regimens	14 wk INH-RIF35-PZA-EMB OR INH-RIF10-PZA-Q300 OR INH-RIF20-PZA-Q300 OR INH-RIF20-PZA-MXF OR INH-RIF10-PZA-EMB
			(2b–3, complete)	Additive for MDR-TB in Russia	HRZE-SQ109 OR HRZE-placebo
Imidazopyridine amide: inhibits qcrB subunit of cytochrome bc1 complex
Q203, Telacebec	Qurient Co. Ltd./LLC "Infectex3	Mouse models demonstrate efficacy with once daily doses <1 mg/kg; active intracellularly and extracellularly, no CYPp450 inhibition in vitro	NCT02530710 (1a, complete)	Safety, tolerability, PK of single doses in healthy adults	Doses of 10, 20, 50, 100, 150, 200, 400, 800 mg/kg
			NCT02858973 (1b, complete)	Safety, tolerability, PK of multiple doses in adults	Q203 vs placebo
			NCT03563599 (2, enrolling)	EBA of Q203 vs RHZE	Mid, high, low doses of Q203
GSK 070, GSK3036656	GlaxoSmithKline	Good oral bioavailability, low protein binding, activity comparable with INH in mice; Active at lower doses and wider intervals (48 h) than linezolid	NCT03557281 (2, pending)	EBA, safety, tolerability in 4 sequential cohorts with DS-TB	2 wk GSK3036656 or RHZE
			NCT03075410 (1, complete)	Safety, tolerability, PK of single ascending and repeat doses in health adults	Single dose at 5, 25, or 100 mg Daily dosing
Oxazolidinone: protein synthesis inhibitor, binds bacterial 23s rRNA of 50S subunit to prevent formation of 70S initiation complex
Sutezolid, PNU-100480	Sequella, Inc, TB Alliance	In mice, shortens standard treatment by 1 mo, activity superior to linezolid Safety and tolerability up to 1000 mg in humans Highly active primary metabolite (PNU-101603) at higher concentration than parent Phase 2 data: synergy with PZA, tmax 1–2 h (fasting), 3 h (fed), half-life 3.4 h, 600 mg BID consistently higher than MIC (1200 mg was not) No heme/neuro AE, 14% patients with ALT elevation	NCT01225640 (2, complete)	EBA and whole blood activity	STZ 600 mg BID, 1200 mg daily or RHZE
			SUDOCU (2, proposed)	Sutezolid EBA and dose finding study	BDQ-DLM-MXF PLUS STZ 0 mg 600 mg daily, 1200 mg daily, 600 mg twice daily, OR 800 mg twice daily
			STEP (2c, proposed)	Safety, efficacy, exposure-response of rifampin doses, STZ	4 mo RhighHZE OR 4 mo RhighHZhighE OR 4–6 mo STZ-BDM OR 6 mo HRZE
			NCT03237182 (4, enrolling)	Gene-derived individualized drug-resistant TB regimens, including sutezolid
Delpazolid, LCB01–0371	LegoChem Biosciences, Inc	Phase 1: tolerated to 1200 mg twice daily (though decline in heme values for daily doses >800 mg), rapid PO absorption, 100% oral bioavailability, tmax 0.5-1 hr, half-life 1.3–2.1 h, plasma binding 37%	NCT02836483 (2, enrolling)	EBA, safety, and PK	800 mg daily, 400 mg twice daily, 800 mg twice daily, LZD 600 mg twice daily
Contezolid, MRX-4	MicuRx Pharmaceuticals, Inc	Broad gram-positive activity, comparable with linezolid; MRX-1 in phase 3 trials in China for skin/soft tissue, MRX-4 (prodrug of MRX-1) being studied in USA	NCT03033329 (1, complete)	Safety, tolerability, PK with dose escalation (IV), crossover to PO	Single IV dose 150–1800 mg BID IV dosing for 10 d: 600, 900, 1200, 1500 mg
TBI-223	TB Alliance, Institute of Materia Medica	Lower activity on mammalian mitochondrial protein synthesis; hepatocyte stability across 5 animal species, no evidence in vitro of CYP induction, half-life 3 h in mice, 8 h in rats; success in mice with BPa-TBI223 (replace linezolid); no hematologic/marrow toxicity in 14 and 28 d rat toxicity studies	NCT03758612 (1, pending)	Single and multiple ascending dose trial
aDprE1 inhibitors: covalent (benzothiazinone): inhibits decaprenyl-phosphoribose epimerase (DprE1) involved in cell wall arabinan biosynthesis
BTZ043	University of Munich, Hans-Knöll Institute, Jena, German Center for Infection Research (DZIF)	Superior to INH at 2 mo in mice (6 mo pending) No antagonism with existing drugs, apparent synergy in vivo with BDQ-RIF Low-level CYP450 interaction	NCT03590600 (1, enrolling)	Ascending dose study	BTZ043 125, 250, 500, 1000, 2000 mg or placebo
Macozinone, MCZ, PBTZ169	iM4TB-Innovative Medicines for Tuberculosis, Bill & Melinda Gates Foundation, Nearmedic Plus LLC	Highly active against replicating bacteria; no antagonism with RHZE, synergy in vitro with BDQ, CFZ, DLM, sutezolid Prior formulation with good tolerability, bactericidal activity against DS-TB at 640 mg	NCT03776500 (1, pending)	Safety, tolerability, PK at multiple ascending doses	150 mg BID OR 300 mg BID OR 600 mg daily OR 600 mg BID
aDprE1 inhibitor: non-covalent (OPC-167832: carbostyril, TBA-7371: azaindole): see above
OPC-167832	Otsuka Pharmaceutical Development & Commercialization, Inc	Activity against replicating and dormant intracellular bacilli; active in acute and chronic murine models; no antagonism with other TB drugs; additive effect with DLM exceeding RHZE	NCT03678688 (1–2, enrolling)	Safety, tolerability, PK, efficacy of multiple oral doses in DS-TB	Stage 1: 14 d of OPC-167832 10 mg OR 30 mg OR 90 mg OR 270 mg OR RHZE Stage 2: 14 d of OPC-167832LowDLM200 OR OPC-167832HighDLM200 OR DLM200
TBA-7371	TB Alliance	Efficacy in vitro and in mice Phase 1 trial complete on food effect, optimal dose, DDI, PK, PD as single dose or multiple doses	NCT03199339 (1, complete)	Safety, tolerability of single and multiple doses among healthy adults	Single doses of 100, 250, 500 1000, 1500 mg; Multiple doses of 100, 200, 400; DDI with midazolam, bupropion
Riminophenazine: Same class as CFZ
TBI-166	Institute of Materia Medica, CAMS & PUMC	Excellent in vivo, shorter half-life, less lipophilicity, less potential for skin discoloration and other toxicities of CFZ	Unknown name (1, enrolling)	Safety, tolerability among patients in China

Abbreviations: EBA, early bactericidal activity; RIF or R, rifampin/rifampicin; MIC, minimal inhibitory concentration; PZA or Z, pyrazinamide; PK, pharmacokinetics; INH or H, isoniazid; EMB or E, ethambutol; Q, SQ109; STZ, sutezolid; BDQ or B, bedaquiline; DLM or D, delamanid; MXF or M, moxifloxacin; LZD, linezolid; PO, per oral; Pa, pretomanid; CYP, cytochrome p450; d, day; mg, milligram; DDI, drug-drug interaction; BID, twice daily; PD, pharmacodynamic; CFZ, clofazamine.

^aDenotes novel class.

Adapted from Working Group on New TB Drugs (https://www.newtbdrugs.org/pipeline/clinical) as of January 2019.

SUMMARY

BDQ is proven to reduce mortality in MDR-TB, and it is recommended for use as a first-line agent for patients with MDR-TB. Best practices for ECG monitoring, dosing in children, and HIV cotreatment are emerging. Delamanid, despite disappointing phase 3 results, has a role as a companion drug in all-oral regimens for MDR-TB. It is well tolerated, and rational dosing recommendations and formulations for children are coming soon. Pretomanid, in combination with linezolid and BDQ, displayed exceptional efficacy among patients with XDR-TB, and it is now approved for use in combination with LZD and BDQ in patients with highly resistant TB. In addition, nitroimidazoles may have brain penetration that allows for use in CNS TB, but clinical data are lacking. Now that the WHO has listed BDQ and delamanid among the options for MDR-TB in the newest guidelines, access to these drugs will improve. Lastly, the third wave of TB drug discovery offers exciting new compounds with numerous mechanisms of action that may expand the arsenal not only for MDR-TB but for ultrashort combination regimens for drug-sensitive TB.

KEY POINTS.

• Treatment success rates for drug-susceptible tuberculosis are too low, and prognosis for drug-resistant TB remains poor.

• Though shorter 9- to 12-month regimens are available for the treatment of MDR-TB, therapy is complex and often toxic.

• The diarylquinoline, bedaquiline (BDQ), and the nitroimidazoles delamanid and pretomanid, have excellent preclinical and clinical data to support their use for MDR or XDR-TB. Multiple trials are centered on use of these drugs to produce well-tolerated, all-oral, short-course regimens.

• Further study of new drugs is urgently needed among children, pregnant women, and people living with HIV to guide clinicians.

• The pipeline now contains numerous new chemical entities from 16 drug classes with encouraging results from late preclinical or early clinical testing, the “third wave” of TB drug development.