New drugs and regimens for tuberculosis

INTRODUCTION

Tuberculosis (TB) is an old disease that dates back to at least 4000 years.1 Like any other infectious disease, an epidemic curve can be drawn for TB,2 which shows that it will take about 300 years for a TB epidemic to naturally die out. Since standardized rifampin-based first-line regimens were available3 and fluoroquinolone-based second-line regimens were used, unfortunately without timely modification according to the drug resistance profile, TB and drug-resistant disease have remained important public health threats worldwide, with an estimated 10.4 million TB new cases in 2016, including 600 000 rifampin-resistant TB, of which 490 000 were multidrug-resistant TB (MDR-TB, defined as TB with bacillary resistance to at least isoniazid and rifampin).4-7 The reported treatment success rate was only 54% for MDR-TB.7 The End TB Strategy has set ambitious targets that include a 90% reduction in TB deaths and an 80% reduction in TB incidence by 2030 in comparison with 2015.7 Although the last decade has witnessed advances in rapid diagnostic tools and use of repurposed and novel drugs for better managing drug-resistant TB,8 we would probably require a new TB vaccine in order to achieve the specific targets set in the End TB strategy. Short of a major breakthrough in the development of TB vaccines,7 we need an appropriate TB control strategy and a well-functioning health infrastructure to ensure optimal operational use of rapid tests, judicious use of effective TB regimens that can be rapidly customized according to the drug resistance profile and proper management of TB risk factors and co-morbidities that promote infection and its progression to disease.

The spread of MDR-TB in numerous low- and middle-income countries has significantly weakened already fragile healthcare systems. To cure TB disease and properly manage treatment failure and relapse, we need one or more treatment regimens for tackling both drug-susceptible and drug-resistant TB in the best possible control strategy. National TB control strategies may be broadly classified into standardized and individualized therapies.9 Standardized treatment enables simple operation and broadens access to care but at the expense of possible amplification of existing but undetected drug resistance. On the contrary, individualized treatment helps customize drug combinations to avoid amplification of drug resistance and toxicities, but it is heavily dependent on drug susceptibility testing (DST) services. While the World Health Organization (WHO) has recommended the longer MDR-TB regimens, which may be standardized or individualized, and the shorter MDR-TB regimens, which are largely standardized,9 an one-size-fit-all approach has also been advocated to develop an injection-free regimen that could be administered to any patient with active TB, with the assumptions that DST and new drug regimens are readily available, and bacillary resistance to new drugs has not developed to any significant levels.10

Host-directed therapies are outside the scope of this review, but the important role of host immunity in TB pathogenesis and the emergence of MDR-TB have fuelled enthusiasm in exploring their use as adjunctive treatment options for either shortening treatment or improving treatment outcomes.11 By either amplifying host immunity or dampening destructive inflammation, host-directed therapies have involved use of small molecule drugs (such as metformin, nonsteroidal anti-inflammatory drugs and statins) and immunomodulatory biologics (such as vitamins A and D3, micro RNA and interleukin-2).11

We searched the published literature to discuss (i) standardized versus individualized therapies, including the choice between a single one-size-fit-all regimen versus different options with different key drugs determined mainly by rapid DST, (ii) alternative regimens that may help facilitate management of drug-susceptible TB, (iii) evidence for using the WHO longer and shorter MDR-TB regimen and (4) evidence for using repurposed (linezolid and clofazimine) and new (bedaquiline, delamanid and pretomanid) drugs in the management of fluoroquinolone-susceptible versus fluoroquinolone-resistant MDR-TB. On the basis of these findings, we suggest the way forward for optimally controlling the TB epidemic.

NATIONAL TB CONTROL STRATEGY: STANDARDIZED VERSUS INDIVIDUALIZED THERAPIES

DRUG-SUSCEPTIBLE TB: ALTERNATIVE REGIMENS THAT MAY HELP FACILITATE ITS MANAGEMENT

Shorter treatment would facilitate management of drug-susceptible TB by improving treatment adherence and increasing the chance of treatment completion. However, there is as yet insufficient clinical evidence for shortening the standard 6-month regimen.

Rifampin is the key sterilizing drug in the standard short-course regimen. Although it has been in use for nearly 50 years, accumulating evidence suggests that currently recommended dosages are suboptimal and that higher doses could be used to safely shorten therapy or improve clinical outcomes. A dose-ranging trial has demonstrated that 2 weeks of rifampin dosed up to 35 mg/kg/day is safe and well tolerated, with a greater estimated fall in bacterial load in the higher dosing groups.35 A subsequent phase 2 clinical trial in a multiple-arm, multiple-stage design (NCT01785186) suggested that rifampin 35 mg/kg/day and rifampin 20 mg/kg/day with moxifloxacin may have the potential to shorten treatment. Four-month regimens with rifampin dosed to either 1200 mg daily or 1800 mg daily are being evaluated in the RIFASHORT trial (NCT02581527). High-dose rifampin has also been explored for treatment of TB meningitis.36, 37 Although the results of two randomized trials appear conflicting, they may indicate that initial intravenous administration is the preferred route for initial high-dose rifampin treatment in this lethal disease due to prominent first-pass metabolism and marked inter-individual variability in rifampin bioavailability.

Rifapentine is another rifamycin derivative with more potent activity in vitro and a longer half-life in vivo compared with rifampin. Although rifapentine was initially approved for use in once-weekly continuation phase therapy, this regimen had inferior efficacy in patients at high risk of relapse.29, 38, 39 Daily dosing regimens that greatly multiplied rifapentine exposures were studied in murine models, where they exhibited dose-dependent treatment-shortening effects.40, 41 Although substituting rifapentine 10 mg/kg/day for rifampin was not associated with a significant change in 8-week culture conversion rates,42 a subsequent dose-ranging study demonstrated that higher rifapentine doses of 15–20 mg/kg/day significantly improve 2-month sputum culture conversion compared with standard therapy.43 Four-month regimens containing rifapentine 1200 mg daily, with or without moxifloxacin replacing ethambutol and given throughout, are being evaluated in TBTC Study 31/ACTG A5349 (NCT02410772).

Despite encouraging findings from murine TB models and those from five randomized controlled trials and one cohort study,44-50 four randomized controlled trials failed to demonstrate the non-inferiority of 4-month regimens containing either moxifloxacin or gatifloxacin in place of ethambutol or isoniazid,31-33 or moxifloxacin and high-dose twice-weekly rifapentine in place of isoniazid and rifampin, respectively.51

Murine model studies first suggested that a novel rifamycin-sparing combination of the investigational new nitromidazole, pretomanid, with moxifloxacin and pyrazinamide could be as effective as regimens containing rifampin, moxifloxacin and pyrazinamide and superior to the standard regimen.52 After successful phase 2 trials,53, 54 4- and 6-month treatment durations of the novel regimen were evaluated in the phase 3 STAND trial (NCT02342886). However, the trial was temporarily put on hold owing to concerns about hepatotoxicity before the sponsor discontinued enrolment in favour of a new phase 2 trial SimpliciTB (NCT03338621) that will evaluate a more promising 4-month 4-drug regimen that combines bedaquiline with pretomanid, moxifloxacin and pyrazinamide. This regimen is much more potent in mice,55 and was associated with near-100% sputum culture conversion in MDR-TB patients at 2 months of treatment using the stringent MGIT-960 liquid culture system (unpublished data of NC-005 trial presented in the 47th Union World Conference on Lung Health in 2016).

Although it is still not possible to shorten treatment for all drug-susceptible TB patients, it may be possible to shorten treatment for the majority with more individualized treatment approaches. For example, a recent meta-analysis of the phase 3 fluoroquinolone trials indicated that the majority of patients had good outcomes with the 4-month regimen and that, if ‘hard-to-treat’ patients expected to require 6 months of treatment for cure (e.g. HIV co-infection, cavitation, 3+ or 4+ sputum smear grade, etc.) could be identified by baseline and/or on-treatment characteristics, other patients may do well with 4-month regimens in a stratified approach (unpublished data presented in the 47th Union World Conference on Lung Health in 2016).

In addition, more efficient supervision of therapy and improved treatment adherence may be achieved using once-weekly dosing in the continuation phase. A 6-month regimen with moxifloxacin and rifapentine 1200 mg administered once-weekly in the 4-month continuation phase was found to be non-inferior to the standard 6-month regimen.51 A caveat is the importance of excluding rifampin resistance by rapid molecular DST to avoid amplification of drug resistance.

MDR-TB: WHO LONGER AND SHORTER MDR-TB TREATMENT REGIMENS

As available therapeutic options for the management of MDR-TB are quantitatively and qualitatively inadequate, many MDR-TB patients have been treated with poorly effective, highly toxic and expensive drugs.4-6, 56

On the basis of the reviews of aggregated and individual patient data from published and unpublished studies carried out by an international scientific group,57, 58 WHO has revised its previous guidelines on the programmatic management of MDR-TB,6, 9 with recommendations for a shorter standardized MDR-TB regimen for carefully selected patients and a longer individualized regimen for those ineligible for the shorter one. Individual patient data meta-analysis showed that MDR-TB treatment success and survival were associated with use of later-generation fluoroquinolones, ethionamide or prothionamide, four or more likely effective drugs in the intensive phase and three or more likely effective drugs in the continuation phase.57 Treatment success was higher among MDR-TB patients without additional drug resistance or with bacillary resistance to second-line injectable drugs only, than those with fluoroquinolone resistance alone or XDR-TB.58 Among patients with XDR-TB, treatment success was highest when there were at least six drugs in the intensive phase and four in the continuation phase, with treatment success maximized by 6.6–9.0 months of treatment during the intensive phase and 20.1–25.0 months overall, and by including more drugs than those recommended for MDR-TB.58

MDR-TB: REPURPOSED AND NOVEL DRUGS

The past decade has witnessed frequent use of two repurposed agents (linezolid and clofazimine) and two clinically approved novel drugs (bedaquiline and delamanid) in non-trial settings, often on a compassionate basis, owing to the poor treatment response of complicated MDR-TB to conventional fluoroquinolone-based MDR-TB treatment regimens. The initial evidence based on observational studies for using repurposed agents has initiated clinical trials for further evaluating their efficacy, safety, optimal dosages and shorter treatment durations in various combinations with novel drugs in clinical trials. Pretomanid (PA-824) is a promising novel drug still being evaluated in phase 2 and 3 clinical trials but not yet approved for clinical use.

Table 1 briefly summarizes evidence for using repurposed agents (linezolid and clofazimine) and major novel drugs (bedaquiline, delamanid and pretomanid) in the treatment of MDR-TB.

A regimen containing linezolid, bedaquiline and delamanid may be adequate for the most difficult-to-treat MDR-TB, but sequential acquisition of drug resistance to bedaquiline and delamanid,19 and to linezolid and delamanid,112 has been reported. This may be anticipated, as delamanid is prone to bacillary drug resistance,117, 118 and the 120-week culture conversion rates in patients with pre-XDR- and XDR-TB given an optimized background regimen plus bedaquiline in a clinical trial setting were only 70.5% and 62.2%, respectively.121 At least three case studies have reported combined use of bedaquiline with delamanid among patients with XDR-TB,122-124 but published data may be insufficient for any informative systematic review.125

CONCLUSIONS AND THE WAY FORWARD

To substantially reduce TB morbidity and mortality worldwide, we need implementation and scale-up of a systematic diagnostic work-up and a tailored therapeutic approach in the national TB programmes.4-6, 56 The prognosis of MDR-TB and the halt of transmission of Mycobacterium tuberculosis strains to susceptible contacts can be improved with an early diagnosis, including the detailed assessment of the drug resistance pattern, and an immediate administration of an effective therapy.6, 56

We believe that the most rational approach to effectively treat TB and tackle drug-resistant TB is using different regimens with different key drugs (rifamycin vs fluoroquinolone vs linezolid vs novel drugs) determined mainly by rapid DST, rather than a single regimen. In the treatment of drug-susceptible TB, we advocate optimizing use of the rifamycin-based regimen, avoiding use of fluoroquinolones as far as practicable, and using fluoroquinolones only when rapid molecular DST results suggest a low likelihood of MDR-TB. Before we use shorter rifamycin-based regimens to promote the treatment success rate of drug-susceptible TB, once-daily dosing in the continuation phase using high-dose rifapentine and moxifloxacin may help promote treatment adherence, provided that rifampin resistance has been rapidly excluded. The WHO shorter MDR-TB regimen, which critically relies on bacillary susceptibility to fluoroquinolones and second-line injectable drugs, may be used when bacillary resistance to each key drug is unlikely according to rapid DST results, drug exposure history or local epidemiology. In case of fluoroquinolone resistance, linezolid should be used, with addition of bedaquiline or delamanid or both, when the number of likely effective drugs is inadequate. Reliable rapid DST methods should be developed to guide use of novel drugs, as well as pyrazinamide.127 We hope an easily applicable combination of biomarkers that accurately predict individual treatment outcome will soon be available to ultimately enable an individualized precision medicine approach for individually tailored therapy. Finally, it is important to concurrently optimize the management of common and major TB risk factors and co-morbidities (such as smoking, alcoholism, diabetes mellitus and HIV) to enhance TB treatment response and tolerance, and reduce the risk of TB relapse.