New and Repurposed Drugs for the Treatment of Active Tuberculosis: An Update for Clinicians

Although tuberculosis (TB) is preventable and curable, the lengthy treatment (generally 6 months), poor patient adherence, high inter-individual variability in pharmacokinetics (PK), emergence of drug resistance, presence of comorbidities, and adverse drug reactions complicate TB therapy and drive the need for new drugs and/or regimens. Hence, new compounds are being developed, available drugs are repurposed, and the dosing of existing drugs is optimized, resulting in the largest drug development portfolio in TB history. This review highlights a selection of clinically available drug candidates that could be part of future TB regimens, including bedaquiline, delamanid, pretomanid, linezolid, clofazimine, optimized (high dose) rifampicin, rifapentine, and para-aminosalicylic acid. The review covers drug development history, preclinical data, PK, and current clinical development.

© 2022 The Author(s). Published by S. Karger AG, Basel

Introduction

In the last century, the incidence of tuberculosis (TB) has declined considerably with a setback in the past few years due to the COVID-19 pandemic. The decline in TB incidence was accelerated by the introduction of anti-TB treatment after the Second World War, which stopped in the 1980s among other factors due to the emergence of HIV and drug-resistant TB (DR-TB). At that moment, new drugs were not specifically developed for TB. However in the last 2 decades, a new impulse in drug treatment was given through the development of global coordination in the fight against the poverty-related diseases HIV/acquired immunodeficiency syndrome, malaria, and TB. The efforts resulted in the formation of several research consortia sometimes in collaboration with the pharmaceutical industry, leading to an enhanced effort to develop new compounds, repurpose already available drugs, or optimize doses of first-line TB drugs: Pan-African Consortium for the Evaluation of Antituberculosis Antibiotics (PanACEA), International Consortium for Trials of Chemotherapeutic Agents in Tuberculosis (INTERTB), Tuberculosis Trials Consortium (TBTC), Acquired Immunodeficiency Syndrome Clinical Trials Group (ACTG), Project to Accelerate New Treatments for Tuberculosis (PAN-TB), and most recently the UNITE4TB consortium. Currently, there is a portfolio of novel/repurposed/redeveloped TB drugs in preclinical and clinical development, representing the largest pipeline in history of TB drug development. This review will highlight a selection of the clinically available drug candidates that could be part of future TB regimens. We selected registered and clinically available drugs, which were selected because they are new (bedaquiline, delamanid, and pretomanid), repurposed (linezolid), revived (clofazimine), or optimized (rifampicin, rifapentine, and para-aminosalicylic acid [PAS]). The selected drugs will potentially be suitable for an optimized or shortened regimen for drug-sensitive TB (DS-TB) or DR-TB in the near future. The review describes drug development history, preclinical data, pharmacokinetics (PK), and current clinical development.

New DrugsBedaquiline

The antimycobacterial properties of diarylquinolines were patented in 2004. The first scientific report on the activity of bedaquiline (formerly known as TMC207 and R207910) was published in Science in 2005 [1]. The compound was developed by Johnson & Johnson, which obtained conditional market approval for the treatment of DR-TB from the US Food and Drug Administration (FDA) in 2012. Bedaquiline was then the first anti-TB drug with a novel mechanism of action to reach TB patients in more than 40 years (Table 1).

Table 1.

Characteristics of each drug in the treatment of TB

/WebMaterial/ShowPic/1478064

Bedaquiline targets mycobacterial ATP synthases and thereby the energy metabolism of the bacteria (shown in Fig. 1 [2]) [3]. The minimum inhibitory concentration (MIC) of bedaquiline is typically 0.012–1.0 mg/L when tested by broth microdilution with the Mycobacterium Growth Indicator Tube (MGIT) system, and a clinical cut-off at 1 mg/L has been suggested [4]. A dose-fractionation study in mice identified total exposure (AUC) as the main driver of bactericidal effect [5]. In a murine model of latent TB, bedaquiline demonstrated at least as good sterilizing activity as rifampicin [6].

Fig. 1.

Bacterial representation of M. tuberculosis with mechanism of action for each drug, adapted with permission from Hoelscher [2].

/WebMaterial/ShowPic/1478062

Bedaquiline is primarily metabolized through n-demethylation by hepatic isoenzyme CYP3A4 to the approximately 5 times less active M2 metabolite [7]. The terminal half-life (T1/2) of bedaquiline is extremely long (>5 months), and the approved dosing regimen therefore includes a loading-phase (2 weeks of 400 mg once daily) and a maintenance phase (200 mg 3 times per week) (Table 1) [7]. The plasma protein binding of bedaquiline is >99.9% [7]. Bedaquiline exposures are affected by drugs interacting with CYP3A4, and important examples are rifampicin and rifapentine (decreasing bedaquiline exposures almost 4- to 5-fold) and antiretroviral drugs such as efavirenz (decreasing bedaquiline exposures by half) and lopinavir/ritonavir (increasing bedaquiline exposures by 2- to 3-fold increase) [8-11].

Bedaquiline has limited early bactericidal activity (EBA) on its own, and the effect can only be seen after about 1 week of treatment [12]. In phase 2 trials including multi-DR-TB (MDR-TB) patients, it was shown that the addition of bedaquiline to an optimized background regimen shortened the time-to-culture conversion and increased the proportion of favorable outcomes at follow-up [13]. Model-based analyses found that the weekly average bedaquiline concentration is associated with the rate of decline in bacterial load and suggest that the currently used dosing regimen is not achieving maximal possible efficacy [14]. The main side effect of bedaquiline is QT prolongation, driven primarily by the exposure to the M2 metabolite [15]. This raised concerns regarding combining bedaquiline with delamanid, a drug that also causes QT prolongation, but recent data indicate that the combined effect of the two is clinically modest and no more than additive [16]. Bedaquiline for treatment of MDR-TB is currently being tested in a phase 3 setting through the STREAM trial (stage 2), with results expected in 2022 [17]. Bedaquiline in combination with pretomanid and linezolid has demonstrated 90% (95% CI: 83–95%) favorable outcomes in extensively DR-TB (XDR-TB) and treatment-intolerant MDR-TB patients, enabling the recent approval of this specific regimen (bedaquiline, pretomanid, and linezolid [BPaL]) [18]. Furthermore, in May 2022, the WHO published a rapid communication endorsing the 6-month bedaquiline, pretomanid, linezolid, and moxifloxacin (BPaLM) regimen based on data of the ZeNix-TB study and TB-PRACTECAL, which are described in further sections of this review [19]. If there is resistance to fluoroquinolones, moxifloxacin should not be given [19].

In the latest update of WHO guidance on treatment of DR-TB, bedaquiline was categorized as a group A drug that is considered highly effective and strongly recommended for inclusion in all regimens unless contraindicated [20]. Very limited data on the use of bedaquiline in children are available [21]. A child-friendly, 20 mg scored, dispersible tablet has been developed. Pediatric PK and safety studies are still ongoing, but FDA and European Medicines Agency (EMA) already recommend bedaquiline for use in children 5 years and older, weighing at least 15 kg. The WHO published consolidated guidelines in 2022, where it conditionally recommends bedaquiline use in all children [22]. Experience with bedaquiline for DR-TB in programmatic use is growing, and reports are mainly positive with relatively high rates of culture conversion (80–95%) and few serious adverse events [23-29]. Bedaquiline can be used during pregnancy and in a South African cohort including 108 pregnant women where about half were treated with bedaquiline, and 88% of babies exposed to bedaquiline in utero were thriving and developing normally at 12 months compared to 82% of the babies not exposed [30]. It is unclear how well bedaquiline penetrates into the central nervous system, enabling effective treatment of TB meningitis. One case report noted that concentrations of bedaquiline in cerebrospinal fluid were undetectable [31], but a recent study reported bedaquiline levels in cerebrospinal fluid in the same order of magnitude as unbound plasma concentrations [32].

Few concerns regarding emergence of resistance to bedaquiline and MIC-increasing mutations in the atpE, mmpR (Rv0678), and pepQ (Rv2535c) genes have been described [33]. Notably, cross-resistance with clofazimine has also been detected [34, 35].

Delamanid

Delamanid (Deltyba®), a nitro-dihydro-imidazooxazole derivative, formerly known as OPC-67683 [36], was patented in the 1990s. In 2014, shortly following the approval of bedaquiline, delamanid obtained EMA’s approval for early release to the market as the second new drug for DR-TB treatment in over 40 years (Table 1).

Delamanid’s exact target is yet to be determined. However, delamanid is known to inhibit mycolic acid synthesis and cellular respiration (shown in Fig. 1 [2], Table 1). The inhibitory effect probably involves the release of reactive radicals such as nitric oxide, which are crucial in the mammalian defense mechanism against mycobacterial infections [37].

The drug has a low MIC90: 0.006–0.024 μg/mL and is found to be active against DS and DR M. tuberculosis isolates [38]. It inhibits replicating and dormant bacilli, both extracellularly and intracellularly. In murine experiments, the combination of delamanid with rifampicin and pyrazinamide resulted in more rapid sterilization of lung tissue than the standard regimen of first-line anti-TB drugs. Its activity against intracellular M. tuberculosis was equivalent to that of rifampicin at a concentration of 1–3 μg/mL [37].

Delamanid has a time to peak concentration (Tmax) of 4 h, with the peak concentration (Cmax) varying from 175 to 286 ng/mL in the dose range of 100–400 mg [39]. The area under the plasma concentration-time curve from 0 to 24 h (AUC0–24) ranged from 2,500 to 5,483 h*ng/mL at the dose range of 100–400 mg, indicating that the increase in exposure is not dose-proportional, and reaches a plateau at 300 mg dose [39]. When taken with food, delamanid bioavailability increases 2- to 4-fold. The apparent T1/2 is 30–38 h, while the T1/2 of its main metabolite is approximately 150–600 h. Delamanid is metabolized to M1, a unique metabolite formed by plasma albumin. Nonhepatic formation of M1 and multiple separate pathways for metabolism of M1, including CYP3A4, suggest that clinically significant drug-drug interactions with delamanid and M1 are limited, although concurrent administration with medicines known to induce/inhibit CYP3A4 may modestly alter delamanid exposures [40]. Delamanid and its major metabolites do not inhibit or induce CYP enzyme activity and thus have little potential for interaction with antiretroviral drugs [40, 41]. Its high plasma protein binding (≥99.5%), especially to albumin, alters its volume of distribution [42].

The 14-day EBA in a phase 1 trial with a 200 mg daily dose was 0.052 log10 cfu/mL sputum per day supporting the bactericidal potential of this drug. The EBA of delamanid was monophasic and not significantly different between dosages, in line with the overlapping exposure profiles. Adverse events in this 14-day trial were of either mild or moderate severity. No serious adverse events occurred, and QT intervals were not prolonged [39]. The first randomized placebo-controlled-phase 2 trial (2012) showed an increase in 2-month culture conversion rates in patients with DR-TB treated with delamanid in combination with a backbone regimen compared to placebo [43]. However, a randomized, double-blind, placebo-controlled phase 3 multicenter trial, published in 2019, could not show any difference in efficacy between the two groups at 6 months of evaluation [44]. A multicenter observational study (n = 53) showed that 67.6% of a difficult-to-treat cohort of DR-TB patients successfully culture converted by 6 months of treatment with delamanid. A total of 31 serious adverse events were reported in 14 patients (26.4%); most common were hepatotoxicity (5), electrolyte imbalance (5), and QT prolongation (3) [45].

A recent meta-analysis of seven studies that used a bedaquiline and delamanid combination to treat 87 cases of DR-TB showed promising outcomes. Most M/XDR-TB patients were concomitantly treated with bedaquiline and delamanid, and in most cases for a duration >6 months. The sputum culture conversion rate after 6 months of treatment was considerably higher (81.4%) than in historical M/XDR-TB patient cohorts. Out of 87 patients, 23 (26.4%) had slight increases in QTc. However, only 2.3% of treatments were interrupted because of life-threatening cardiac adverse events [46]. In children (aged 3 years and above), delamanid shows an excellent safety and side effect profile [47].

Currently, there are several ongoing studies but because of paucity and weaknesses of the existing evidence the efficacy of delamanid is still challenging to define [48]. Therefore, pivotal clinical trials are needed to clarify its clinical value.

Pretomanid

Pretomanid (Dovprela®) (PA), another nitro-dihydro-imidazooxazole derivative, formerly known as PA-824, was approved in 2019 by the FDA for XDR-TB or treatment-intolerant or nonresponsive MDR-TB in the BPaL regimen, and in 2020 it received conditional approval by the EMA (Table 1) [49-51]. PA was first identified in 1995 and is co-licensed by the TB Alliance and Mylan with affordable access agreements for low- and middle-income countries.

PA inhibits mycolic acid synthesis, a necessary step in cell wall formation (shown in Fig. 1 [2]) [52, 53]. PA is metabolized to reactive nitrogen species, which permits its activity against nonreplicating bacilli. In vitro activity is shown with an MIC of <1 μg/mL and a MIC90 of 0.063 μg/mL [54]. PA was found to be highly active in mice, especially in combination with bedaquiline and linezolid [55]. Five genes are associated with the emergence of resistance (ddn, fgd1, fbiA, fbiB, and fbiC) [56, 57]. Cross-resistance with delamanid has been observed.

PA is administered orally as a tablet once daily. Following administration of a single dose of 200 mg with food, the mean Cmax is 2.0 μg/mL, the median Tmax is 5 h, and the mean AUC∞ is 51.6 μg*h/mL (Table 1) [49]. It is 86% protein-bound, and steady state is achieved after 4–6 days. PA is metabolized via several pathways, 20% of which is via CYP3A4. PA is cleared in urine and feces mainly as metabolites. PA’s co-administration with rifampicin, lopinavir/ritonavir, or efavirenz reduces its exposures, in the case of rifampicin by 53–85% [58]. PA inhibits OAT3 transporter, suggesting that it may increase exposures of OAT3 substrates such as methotrexate.

PA has been investigated in several phase 2 trials for DS-TB. A phase 2A 14-day study with PA evaluating various combinations bedaquiline, pretomanid, and pyrazinamide showed equal EBA compared to the standard control [59]. In a phase 2B study (NC002) for 8 weeks, a regimen combining moxifloxacin, pretomanid 200 mg, and pyrazinamide showed that there was a significant increase of sputum culture conversion compared to the control [60]. In a later phase 2B trial (NC005), adding bedaquiline to the regimen of moxifloxacin, pretomanid, and pyrazinamide (BPaMZ) increased the bactericidal activity over 2 months considerably, in both DS- and DR-TB [61].

Recently, two phase 3 trials for PA were performed with a focus on DR-TB. The BPaL regimen was successfully investigated in the Nix-TB trial and was licensed for XDR-TB, treatment-intolerant or nonresponsive MDR pulmonary TB [18]. TB-PRACTECAL, a phase 2/3 clinical trial, investigated the 6-month BPaLM regimen, against the locally accepted standard of care. It found that 89% of patients in the group of patients receiving a 6-month regimen of BPaLM were cured versus 52% of patients who had been prescribed the standard regimen of up to 20 months of treatment, but this has not been published yet [62]. The BPaLM regimen has now been included in the recent WHO rapid communication [19].

For both DS- as DR-TB, the TB Alliance has initiated the pivotal SimpliciTB trial. The trial is almost finished, and the results will be presented at the Union World Conference on Lung Health in November 2022. SimpliciTB is the continuation of the STAND trial that was changed from PaMZ into BPaMZ when the results came available of the NC005 trial. Further trials with PA are the phase 3 ZeNix trial (aimed to reduce linezolid toxicity) and the phase 2B APT trial evaluating PA added to a first-line regimen with either rifampicin or rifabutin, isoniazid, and pyrazinamide [63].

The most common adverse events of PA are gastrointestinal symptoms and vomiting and suggested to be dose related [64], and the following described symptoms are not dose related: transaminase increase, hepatotoxicity, and headache. The presence of hepatoxicity is unfortunate as it reduces the value of a regimen to be used in low resources settings.

Repurposed DrugsLinezolid

In 1987, when the antimicrobial activity of oxazolidinones against plant pathogens was discovered [65], one of these agents was developed into linezolid (DuP721, PNU-100766, S-N-3-3-fluoro-4-[4-morpholinyl]phenyl-2-oxo-5-oxazolidinylmethyl-acetamide). Linezolid was marketed for treating gram-positive bacterial infections of the lungs, skin, and soft tissues in 2000 and considered for the treatment of MDR-TB as a group 5 agent in 2006 by the WHO [66].

Oxazolidinones inhibit protein synthesis by decreasing mRNA translation through binding to the 23S RNA peptidyl transferase center of the prokaryotic ribosomal 50S subunit (shown in Fig. 1 [2], Table 1) [67]. The binding of linezolid to human mitochondria can cause myelosuppression, peripheral and optic neuropathy, and hyperlactatemia [68, 69].

The critical linezolid concentration for M. tuberculosis is 1 mg/L in 7H10 and MGIT medium [70, 71]. Median linezolid MIC of 4,470 strains tested in M7H9 medium is 0.5 mg/L (range 0.06–16 mg/L) [72].

In mice, 25–260 mg/kg of linezolid showed anti-tuberculous activity against replicating and nonreplicating bacilli [73, 74]. A once daily linezolid hollow fiber model demonstrated substantial activity against M. tuberculosis in acid-phase and nonreplicating persister states [75].

Linezolid has an excellent oral bioavailability approaching 100%, low plasma protein binding of 31%, and good tissue penetration including cerebrospinal fluid [76], alveolar macrophages [77], and bone [78]. Plasma Cmax is 15–27 mg/L, and Tmax is 0.5–2 h [79]. When dosed 600 mg every 12 h, linezolid has an elimination half-life of 4.8–5.5 h (Table 1). Children under 12 years have a faster clearance and a shorter elimination half-life than adults [80].

Sixty-five percent of linezolid is oxidized into two inactive metabolites. About 30% of linezolid and 50% of the metabolites are excreted via urine [81]. Although metabolism via CYP3A4 plays a minor role in linezolid elimination, rifampicin co-administration reduces linezolid exposure. Concomitant use of monoamine oxidase inhibitors with oxazolidinones is contraindicated because of the risk of serotonin syndrome. Linezolid trough plasma concentrations and AUC0-24 are linearly related, and trough concentrations are predictive of linezolid exposure [82, 83]. Linezolid’s EBA is modest in HIV-positive TB patients [84].

Currently, the WHO categorizes linezolid as a group A drug for rifampicin-resistant or MDR-TB [47]. Several studies provide evidence for this choice. The addition of linezolid to the ongoing regimen in 41 XDR-TB patients without response to earlier treatment showed sputum culture conversion after 4 months in 15 out of 19 patients (79%) in the group that started immediately and a negative sputum culture after 6 months of treatment in 34 out of 39 patients (87%) [85]. Improved 24-month treatment outcome, based on the WHO definition, was found for an oral 6-month regimen including linezolid, levofloxacin, and bedaquiline compared to the conventional empiric injection-based regimen, although it was stopped early due to a new standard of care [86]. In the Nix-TB study, where 1,200 mg linezolid daily was given with PA and bedaquiline anemia developed in 37% of patients, and peripheral neuropathy in 81%, mainly attributed to linezolid toxicity. Only 34% completed treatment, warranting careful monitoring of side effects of linezolid [18]. The phase 3 ZeNix trial investigates the efficacy and toxicity of 600 mg versus 1,200 mg linezolid daily in combination with PA and bedaquiline, and aims to reduce linezolid toxicity. Presented results, which are not published yet, showed that the lower doses and/or shorter duration treatment arms of linezolid had similar success rates and less toxicity in comparison with the treatment arm including 1,200 mg of linezolid for 6 months [87]. As described in the PA section, the BPaLM regimen was found to be efficacious in the TB-PRACTECAL trial [62, 88] and is included in the newest WHO rapid communication [19]. Another study including linezolid is the MDR-END study, which compares the effectiveness and safety of oral linezolid, delamanid, levofloxacin, and pyrazinamide for 9–12 months with the WHO 2016 MDR-TB regimen for 20–24 months [89]. [62].

Thus, a major concern and drawback in long-term linezolid use are its toxicity, which was a reason to conduct the ZeNix trial, described above. A meta-analysis of twelve studies of linezolid-containing TB regimens showed adverse events in 58.9% (63/107) of patients. Anemia was found in 38.1% (32/84) of patients, peripheral neuropathy in 47.1% (40/85), gastrointestinal disorders in 16.7% (14/84), optic neuritis in 13.2% (10/76), and thrombocytopenia in 11.8% (10/85) of patients, especially with doses above 600 mg [90]. Monitoring hemoglobin levels and neuropathy may guide linezolid dosing optimization [91]. Therapeutic drug monitoring of linezolid is helpful when the daily dose is below 300 mg [92]. A trough concentration of 2.5 mg/L should be evaluated as a target for therapeutic drug monitoring [93].

Phenotypic linezolid resistance in M. tuberculosis, defined by an MIC >1 mg/L, emerged in 2007 and is mainly linked to mutations in or close to the PTC-binding site like the ribosomal protein L3 rplC gene and the 23S rRNA rrl gene [33, 94]. In a South African study of MDR-TB patients, linezolid resistance developed after 7–32 months of therapy. Sixteen phenotypically resistant strains most frequently showed the T460C substitution in rplC, and G2814T, G2270 C/T, and A2810C mutations in the rrl gene. No mutations were detected in isolates with MICs at or below the critical concentration [95]. Mutations in rrl led to growth impairment and decreased fitness which could limit spread in clinical settings [96]. Other resistance mechanisms need further elucidation.

Revived DrugsClofazimine

Clofazimine was first synthesized and reported from Trinity College Dublin in 1957; it is the prototype riminophenazine antibiotic [97]. While initially developed for TB treatment, its preclinical development stalled as its in vivo activity proved inconsistent in particular animal models, i.e., guinea pig and primates. This lack of in vivo activity was later linked to poor absorption, low plasma concentrations, and/or high burden of extracellular mycobacteria. Thus, for decades, its use remained limited to leprosy treatment [98].

Its exact mechanism of action remains incompletely understood but includes inhibition of mycobacterial respiration via intracellular redox cycling as well as membrane disruption [99, 100]. Other proposed mechanisms of action include generation of reactive oxygen species, inhibition of mycobacterial phospholipase A2, microbial K+ transport inhibition, and efflux pump inhibition (shown in Fig. 1; Table 1) [97-99, 101, 102]. Recently, studies have shown that upregulation of the MmpS5/MmpL5 efflux system after mutations in its repressor gene (Rv0678) leads to increased MICs against clofazimine as well as cross-resistance with bedaquiline [33]. In addition, mutations in the pepQ (Rv2535c) and Rv1979c genes, encoding an aminopeptidase and a permease transporter, increase clofazimine MICs by unknown mechanisms [33, 103, 104].

In vitro, clofazimine is highly active against M. tuberculosis complex isolates with a reported MIC90 of 0.25 mg/L in wild-type isolates using broth microdilution [98, 105]. It took mouse models with long treatment durations to demonstrate that clofazimine has potent, albeit delayed and dose-independent, anti-TB activity (across the range of doses investigated) and that clofazimine adds significant activity to first- and second-line TB treatment regimens, at serum concentrations above the MIC [106, 107]. Clofazimine also exhibits a sustained antimycobacterial activity after treatment cessation, yet again most prominent if serum concentrations during treatment were ≥0.25 mg/L (the MIC90), suggesting a potential for treatment shortening with clofazimine [108]. This treatment shortening effect was further investigated in first-line regimens where rifampicin was replaced with rifapentine and clofazimine was added; in two distinct mouse models of more acute (Balb/c) and caseous-necrotic disease (C3HeB/FeJ), adding clofazimine and rifapentine increased the bactericidal and sterilizing activity of the regimens, more than either drug alone [109]. This confirmed the potential for treatment shortening by clofazimine-containing regimens [109].

Compared to administration during fasting, the oral bioavailability of clofazimine is increased by ingestion with a fatty-meal, whereas aluminum/magnesium antacid has the opposite effect [110]. Clofazimine is highly lipophilic and has peculiar PK including long time-to-steady state, low serum concentrations (0.4–1 mg/L at 100 mg once-daily dose), high protein binding, accumulation in skin, fatty tissues, reticuloendothelial organs and macrophages, and a very long half-life (up to 70 days with long-term repeating dosing) (Table 1) [99, 110]. Although clofazimine is a weak inducer of CYP3A4 itself [111], its PK is not affected by concomitant rifamycin administration [112]. Despite its long use, the optimal dose remains undetermined (Table 1). A 100 mg once-daily dose is standard, but a recent modeling study showed that 200 mg once-daily loading doses in the first 2–4 weeks shorten the time to effective steady-state concentrations by a month [113].

Clofazimine as a monotherapy showed no detectable 14-day EBA, nor did it add to EBA of regimens with other drugs [59]; subsequent modeling revealed that it did have concentration-dependent effects, primarily on the persister subpopulation, and therefore contributes sterilizing activity to regimens [114]. In a randomized controlled trial, adding clofazimine to personalized MDR-TB treatment regimens led to accelerated time-to-culture conversion and higher treatment success rates (74% vs. 54%, p = 0.035) [115]. Subsequent meta-analyses of individual patient data have presented conflicting conclusions on the efficacy of clofazimine as part of MDR-TB treatment, with 2017 meta-analyses reporting no beneficial effect of clofazimine [116]. However, a follow-up meta-analysis in 2018 reported greater treatment success with the use of clofazimine than without it, with an adjusted risk difference of 0.06 (95% CI: 0.01–0.10) [117]. Very recent results from the TB-PRACTECAL study showed that the bedaquiline, pretomanid, linezolid, and clofazimine regimen was effective and safe, but it was not the most effective and safe regimen within the trial, but this has not been published yet [88, 118]. Based on its performance in clinical trials and meta-analyses, clofazimine was categorized as a group B drug in the 2019 WHO consolidated guidelines on MDR-TB treatment [20]. With only three drugs in group A, it implicitly recommends adding clofazimine to all MDR-TB regimens, where it is available. Yet, clofazimine availability is limited and its cost further impairs access to the drug for MDR-TB treatment [119].

Clofazimine is well tolerated; in a meta-analysis, 1.6% (95% CI 0.5–5.3%) of MDR-TB patients interrupt or stop therapy because of adverse events [120]. Adverse events associated with clofazimine use include skin discoloration (reported in frequencies ranging from 3% to 94% [99, 115, 116, 120]). Informing patients about the possibility of clofazimine causing reddish-black discoloration of skin and body secretions could mitigate against unnecessary anxiety and nonadherence. Other adverse events include gastrointestinal discomfort, nausea, vomiting, and QTc interval prolongation [99, 115, 121]. QTc interval prolongation is important particularly when clofazimine is combined with other QTc interval prolonging drugs, e.g., bedaquiline, fluoroquinolones, delamanid, and oxazolidinones (linezolid, sutezolid) [99].

Optimized DrugsOptimized Dose (High Dose) Rifampicin

Rifampicin is considered to be the cornerstone in the treatment of TB. A review concluded that the current dose (10 mg/kg) for the short-course treatment regimen was chosen because the drug was expensive at the time due to fear of adverse events/toxicity and because serum concentrations were above the minimal inhibitory concentration of M. tuberculosis. Twenty years ago, Dennis Mitchison already suggested that high dose of rifampicin should be investigated because of an increased sterilizing effect [122].

According to a murine aerosol infection model and a hollow fiber model, rifampicin shows concentration-dependent killing and the most adequate parameter for the killing of bacteria is AUC/MIC [123, 124]. Also in mice, it was shown that the maximum tolerated dose was 160 mg/kg daily and a dose of 80 mg/kg per day reduced treatment duration to 9 weeks, without adverse effects [125]. Other murine studies showed that a higher dose led to faster culture conversion, a shortened treatment period, a decreased relapse rate, and an increase in bactericidal and sterilizing activity [123, 126, 127].

The concentration of rifampicin increases more than proportional with dose, up to 7-fold [128-131]. Modeling and simulations showed that higher exposures of rifampicin led to greater EBA [132]. Rifampicin induces CYP450 enzymes which leads to decreased concentrations of many drugs, but not the other standard drugs [133-135]. There is probably a ceiling in the maximal inductive capacity of rifampicin at relatively low doses of rifampicin [133]. Currently, the PHENORIF study is looking at the interaction potential of high-dose rifampicin.

A systematic review found an advantage in terms of likelihood of culture conversion in patients that received at least 900 mg rifampicin based on historical clinical trials [136]. In two phase 2B clinical trials, it was found that 20 mg/kg rifampicin for 2 months was safe and tolerated by patients [137, 138]. Findings of a dose ranging trial showed that 40 mg/kg rifampicin was safe and tolerated by patients for 2 weeks and that 50 mg/kg was poorly tolerated [139]. In another phase 2B trial, it was found that a regimen containing 35 mg/kg rifampicin for 12 weeks was safe and reduced time to stable culture conversion in liquid media [134]. Increasing rifampicin doses (up to 30 mg/kg orally) and exposures resulted in a better survival of patients with TB meningitis [130, 140], although this was not confirmed in a larger trial with an increase of 15 mg/kg orally (Table 1) [141].

Based on a change in international guidelines in 2010, rifampicin is administered at a higher dose (15 mg/kg) [142] in children than in adults because 10 mg/kg in children leads to lower serum concentrations than in adults [142-147]. The results of the SHINE trial showed that with the revised WHO doses of all drugs a 4-month treatment regimen was noninferior to a 6-month treatment regimen in children with regard to efficacy [148], which has led to the recommendation of a 4-month regimen by the WHO [149]. Furthermore, high doses of rifampicin up to 60 mg/kg in children for a period of 2 weeks were found to be safe based on the Opti-Rif trial [150].

Suboptimal concentrations of rifampicin could induce resistance at standard doses, explained by poor penetration into cavities [131, 151]. In one study, it was shown that rifampicin reaches adequate concentrations in critical lesions such as necrotic caseum after multiple doses [152]. Other drugs do not achieve adequate concentrations; thus, it was suggested that there can be monotherapy at specific time and locations, which could lead to the emergence of resistance [152].

Optimized doses of rifampicin have not been implemented in guidelines, and the WHO is awaiting additional evidence, which includes phase 3 clinical trials, and a systematic review on all available literature. A phase 2C trial is planned within the PanACEA consortium to investigate 3- and 4-month combination regimens with isoniazid, pyrazinamide in a standard dose and an optimized dose of rifampicin, pyrazinamide, and moxifloxacin [153]. Furthermore, a pragmatic randomized trial will be conducted in Europe to study the safety of an optimized dose of rifampicin in an operational setting. There is an increasing amount of evidence to suggest that the time has come to reconsider the guidelines in special situations (meningitis, severe TB, children, diabetes mellitus, HIV, etc.).

Rifapentine

Rifapentine is a cyclopentyl-substituted semi-synthetic derivative of rifampicin. Like other rifamycins, rifapentine selectively binds to bacterial DNA-dependent RNA polymerase, and inhibition will occur already after short periods of drug exposure (shown in Fig. 1; Table 1). Specific mutations in the β subunit of the RNA polymerase gene (rpoB) lead to the development of cross-resistance to all rifamycins, including rifapentine and rifampicin [154].

The MIC against M. tuberculosis of rifapentine is approximately 0.02 μg/mL [155]. MIC and the minimum bactericidal concentration of rifapentine for intracellular bacteria are 2- to 4-fold lower than those of rifampicin; for extracellular bacteria, this difference was approximately 2-fold [156]. The prolonged effect of rifapentine and its potential for intermittent use found in this study may be associated with high intracellular accumulation, which were 4- to 5-fold higher than those found for rifampicin [156].

PK of rifapentine has been investigated at various doses, frequencies, regimens, and populations [157-161]. In general, time to Cmax is observed in 4–6 h and the half-life is relatively long with about 12–15 h [154]. This is significantly longer than for rifampicin (1.5–5.0 h) and supports a more intermittent use, e.g., for latent TB treatment. Rifapentine peak exposures have been found to be dose linear [161] or less than dose linear [159]. AUCs are less than dose-proportional [162]. The latter is in contrast to the greater than dose-proportional PK observed for rifampicin. The rifapentine Cmax (15 μg/mL) at a dose of 600 mg sufficiently exceeds the MIC [154, 162]. However, rifapentine is about 98% plasma protein-bound [154], which is 4- to 10-fold higher than for rifampicin (about 8–25% protein-unbound [163]), compensating for the lower MIC of rifapentine and possibly explaining the suboptimal efficacy of historically recommended doses of rifapentine.

Recently, a systematic review was published describing rifapentine population PK in a large cohort of nine clinical studies [164]. The analysis showed that rifapentine bioavailability was clinically relevant and was affected by HIV status, food, and dose. With intermittent dosing, autoinduction of clearance was minimal to moderate. With daily dosing, maximum induction was achieved with doses of 300 mg or more. The maximum effect was a 72% increase in clearance after 21 days [164]. Month 2 culture conversion was found to be less likely in individuals with low weight (<50 kg) and in those with low rifapentine exposure. Body weight was not a clinically relevant predictor of clearance, suggesting weight-band dosing of rifapentine should be removed from latent TB infection (LTBI) dosing guidelines [164]. Of note, very limited PK data are available for LTBI: only one study is available that collected PK data in LTBI in 77 participants [160].

Similar to rifampicin, daily/intermittent rifapentine administration significantly reduces exposures of co-administered drugs [154], such as midazolam/bedaquiline (CYP3A4 substrate) [11, 165] and moxifloxacin (UGT1A1) [166]. While in vitro rifampicin seemed to have a superior drug interaction potential [167], oral midazolam clearance was reduced to significantly greater extent (74 vs. 93%) by rifapentine than rifampicin in vivo [159]. Induction of enzyme activities by rifapentine occurs within 4 days after the first dose, returning to baseline levels 14 days after discontinuing rifapentine [154]. In general, less drug-drug interactions are anticipated for once weekly dosing of rifapentine. For example, moxifloxacin exposures remained almost unaffected when rifapentine was dosed once weekly [168]. Rifapentine given once weekly also did not affect concentrations of efavirenz, an HIV drug that is a CYP2B6 substrate [169].

Several rifapentine-based short courses of LTBI treatment have been evaluated to date. A 1-month regimen of rifapentine plus isoniazid was noninferior to 9 months of isoniazid alone for preventing TB in HIV-infected patients [170]. However, the rifapentine-based LTBI regimen with the most extensive track record at present is 12 weeks weekly rifapentine/isoniazid (3HP). Multiple large-scale clinical trials and analyses have shown that 3HP is noninferior to standard INH and has significantly higher rates of treatment completion and lower rates of adverse events [171-174]. Completion of 3HP in routine healthcare settings seems even greater than rates reported from clinical trials, and greater than historically observed using other regimens among reportedly nonadherent populations [175].

For pulmonary TB treatment, more frequent rifapentine administrations seem desirable, as once weekly or less frequent use of rifapentine increases the risk of bacteriological relapse compared to intermittent use of rifampicin [176]. In phase 2 clinical trials, no obvious safety concerns have been noted with the use of daily rifapentine during the first 8 weeks of combination therapy for pulmonary TB [177], and increasing the PK exposure to rifapentine was associated with improved sterilization at the end of the intensive phase as compared to standard dose rifampicin in mice and humans [177, 178]. In a pivotal, ground-breaking recent phase III trial the threshold of shortening to 4 months was finally reached: the efficacy of this 4-month rifapentine-based regimen containing moxifloxacin was noninferior to the standard 6-month regimen in the treatment of TB, and it was similar in regard to safety [179]. This was the first study supporting the treatment shortening potential of a rifamycin when combined with moxifloxacin. The WHO now conditionally recommends the use of a 4-month regimen consisting of rifapentine, isoniazid, pyrazinamide, and moxifloxacin in patients with drug-susceptible TB that are 12 years or older (Table 1) [149].

PAS

PAS, first discovered in 1902 by Seidel and Bittner, was rediscovered in 1943 by Jorgen Lehmann, a Swedish clinician-scientist [180, 181]. PAS is arguably the first synthetic drug used in the clinic to successfully treat TB patients in 1944 [182-184].

The precise mechanism of action of PAS remains elusive, and PAS is considered an antimetabolite, which is structurally similar to para-aminobenzoic acid, a substrate of dihydropteroate synthase [185, 186]. In turn, the generation of hydroxyl dihydrofolate antimetabolite in turn inhibits dihydrofolate reductase [186]. An additional potential PAS mechanism of action is its inhibition of mycobactin, a component of the mycobacterial cell wall resulting in reduced iron uptake by M. tuberculosis (Fig. 1; Table 1) [187]. A PAS MIC ≤1 mg/L defines susceptibility in clinical isolates [188, 189]. Concentration-dependent activity of PAS was demonstrated in an in vitro study, in which PAS concentration of 100 mg/L was decidedly better than 10 mg/L in suppressing resistance emergence in the companion drugs streptomycin and isoniazid [182, 190]. Animal models of TB show the dose and concentration-dependent bacteriostatic effects of PAS [182, 191].

The PK of PAS varies with the formulation. PAS salts are readily absorbed when administered orally and achieve higher Cmax in plasma than with PAS acid or the widely available granular slow-release PAS formulation (PASER) [188, 192-194]. Under the influence of N-acetyltransferase 1, PAS undergoes first-pass metabolism to acetyl-PAS in the gut and liver, which might explain acetyl-PAS appearance in blood earlier than PAS [181, 195]. PAS is metabolized to glycine-PAS in the liver [195]. N-acetyltransferase 1 and N-acetyltransferase 2 genetic polymorphisms have been recognized to influence PAS clearance [182, 196]. Approximately 80% of the administered dose is excreted in urine as PAS and its two major metabolites, acetyl-PAS and glycine-PAS [181, 195]. PAS has an elimination half-life of 0.5–2.5 h and a modest plasma protein binding [181, 187, 193, 197].

Oral administration of PASER with food increases the oral bioavailability of PAS, Cmax, and overall systemic exposure, while reducing intolerance particularly when taken with acidic beverages [182, 197]. The gastrointestinal intolerance to PAS is primarily because of its direct irritant effect while in the gastrointestinal tract and PAS decarboxylation to meta-aminophenol toxin [182, 187].

The early appreciation that gastrointestinal intolerance to PAS could be formulation related led to the development of many formulations [181, 182, 198]. With improving intolerance in mind, PASER is enteric-coated and designed to slowly release PAS in the intestine rather than in the stomach [187]. Furthermore, the gastrointestinal intolerance is neither related to plasma concentrations nor to the conjugated metabolites [182, 194, 196, 199, 200].

Clinical trials conducted in the 1950s by the British Medical Research Council established the role of PAS in suppressing resistance emergence in companion drugs, allowing the construction of an efficacious anti-TB regimen for the next three decades [201-204]. This ability to suppress resistance emergence is clearly dose-dependent, with 20 g/day sodium PAS outperforming 10 g/day and 5 g/day doses administered in four divided doses [203]. A subsequent study found that a sodium PAS dose of 20 g/day has no advantage over 10 g/day in the ability of PAS to suppress resistance emergence to isoniazid [204]. It is however important to note that the individual PAS dosage administered in the study was 5 g because 20 g/day and 10 g/day were administered in 4 and 2 divided doses, respectively. These findings and preclinical evidence of the PAS concentration-dependent effect could indicate that Cmax and, by extension, Cmax/MIC are the parameters linked to the prevention of resistance in companion drugs [190, 202-204].

In addition, an EBA study showed PAS might have a bactericidal effect if used at a high enough dose and, consequently, a high Cmax could be reached with a single-daily dose [205]. In this study, 15 g once-daily PAS resulted in a 2-day EBA similar to that of once-daily 10 mg/kg rifampicin [205]. Furthermore, two clinical studies conducted in the early days of PAS use provided an early indication of the potential of PAS to be a bactericidal agent [206, 207].

The Cmax/MIC ratio has been suggested to be linked to PAS’ ability to prevent resistance emergence in companion drugs; if that hypothesis is correct, the plasma PAS concentrations reached with PASER are low [182, 188, 196, 208] and maybe inadequate at the current dosing regimen of 12 g/day in 2–3 divided doses. Therefore, there remains an urgent need for a prospective EBA study with a PK component to evaluate different once daily PASER regimens. This prospective study should clarify the most appropriate PK/pharmacodynamics determinant of PAS efficacy.

In its 2019 MDR-TB treatment guideline, the WHO downgraded the use of PAS (group C) to only when there is resistance or toxicity to the more preferred anti-TB agents (group A and B) [20]. A meta-analysis of observational studies at least in part was responsible for the WHO recommendation regarding PAS use [117]. The limitations inherent in observational studies strengthen the need for prospective studies to clearly outline the use of PAS, as we are likely not using PASER optimally. Thus, to win the fight against the DR-TB epidemic, all drugs including PAS in a regimen must be optimally utilized, especially in the face of resistance emergence in newly approved drugs such as bedaquiline and delamanid.

Conclusion

We have arrived at a promising stage in the fight against TB, with several academic, governmental, and nongovernmental organizations facilitating and accelerating TB drug development. There have been recommendations and communication of new treatment regimens, with progress been made in treatment shortening for DS- and DR-TB. Moreover, currently new drugs with new targets and existing or repurposed drugs are investigated in clinical trials, with another series of new and repurposed drugs in earlier stages of development. This review highlighted a selection of clinically available drug candidates for current and future TB regimens. Nevertheless, studies are still ongoing to further investigate these drugs and their optimal application. The most recent developments are a shortened treatment regimen for DS-TB in guidelines and recent guidelines and a rapid communication for DR-TB, including new definitions of multi- and extensive drug resistance. For DS-TB, a short (4 months) rifapentine and moxifloxacin-based regimen has now been adopted in guidelines for people of 12 years or older [149]. Optimized (high dose) dose rifampicin is already implemented in severe cases of TB in some clinical settings and is currently being investigated in a study with a large sample size. Furthermore, for children with nonsevere TB a short (4 months) treatment regimen has been recommended [149]. The newest WHO rapid communication on DR-TB now stresses the importance of bedaquiline, pretomanid, linezolid, and, to a lesser extent, clofazimine [19]; for rifampicin- and MDR-TB, the (6 months) BPaLM regimen is endorsed, and if there is resistance to fluoroquinolones it can be used without moxifloxacin [19].

Newer oxazolidinones with a lower toxicity profile than linezolid are being studied. Moreover, a universal regimen for both DS- and DR-TB is also currently being investigated, the BPaMZ regimen was meant to become such a regimen, although this could be questioned considering the rapid emergence of resistance to bedaquiline and the current resistance to quinolones and pyrazinamide.

In this review, we have outlined various anti-TB drugs that to a greater or lesser extent could potentially contribute to improved treatment regimens. Some drugs are new, some are repurposed, and some are optimized with increased dosages. In order to achieve the targets of the End TB strategy [209], increasing efforts still need to be made of which the development of new treatment regimens for active (and latent) TB is essential.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This research did not receive grants from any funding agency in the public, commercial, or not-for-profit sectors.

Author Contributions

Jessica Maria Aguilar Diaz: abstract, introduction, section about an optimized dose of rifampicin, conclusion, table, adaptation figure, lay-out article, general comments, references, and end-editing. Ahmed Aliyu Abulfathi: section about para-aminosalicylic acid, English edits, general comments. Lindsey Hendrika Maria te Brake: abstract, introduction, section about rifapentine, conclusion, and general comments. Jakko van Ingen: section about clofazimine, general comments, and conclusion. Saskia Kuipers: section about linezolid and general comments. Cecile Magis-Escurra: section about delamanid and general comments. Jelmer Raaijmakers: references and general comments. Elin Margareta Svensson: section about bedaquiline and general comments. Martin Johan Boeree: abstract, introduction, section about pretomanid, conclusion, and end-editing.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.

留言 (0)

沒有登入
gif