Abstract
Rationale
Carbapenems are recommended for treatment of drug-resistant tuberculosis. Optimal dosing remains uncertain.
Objectives
To evaluate the 14-day bactericidal activity of meropenem, at different doses, with or without rifampin.
Methods
Individuals with drug-sensitive pulmonary tuberculosis were randomized to one of four intravenous meropenem-based arms: 2 g every 8 hours (TID) (arm C), 2 g TID plus rifampin at 20 mg/kg once daily (arm D), 1 g TID (arm E), or 3 g once daily (arm F). All participants received amoxicillin/clavulanate with each meropenem dose. Serial overnight sputum samples were collected from baseline and throughout treatment. Median daily fall in colony-forming unit (CFU) counts per milliliter of sputum (solid culture) (EBACFU0–14) and increase in time to positive culture (TTP) in liquid media were estimated with mixed-effects modeling. Serial blood samples were collected for pharmacokinetic analysis on Day 13.
Measurements and Main Results
Sixty participants enrolled. Median EBACFU0–14 counts (2.5th–97.5th percentiles) were 0.22 (0.12–0.33), 0.12 (0.057–0.21), 0.059 (0.033–0.097), and 0.053 (0.035–0.081); TTP increased by 0.34 (0.21–0.75), 0.11 (0.052–0.37), 0.094 (0.034–0.23), and 0.12 (0.04–0.41) (log10 h), for arms C–F, respectively. Meropenem pharmacokinetics were not affected by rifampin coadministration. Twelve participants withdrew early, many of whom cited gastrointestinal adverse events.
Conclusions
Bactericidal activity was greater with the World Health Organization–recommended total daily dose of 6 g daily than with a lower dose of 3 g daily. This difference was only detectable with solid culture. Tolerability of intravenous meropenem, with amoxicillin/clavulanate, though, was poor at all doses, calling into question the utility of this drug in second-line regimens.
Clinical trial registered with www.clinicaltrials.gov (NCT03174184).
Keywords: tuberculosis, carbapenem, early bactericidal activity, meropenem, phase 2A clinical trial
At a Glance Commentary
Scientific Knowledge on the Subject
Carbapenems are recommended as Class C agents for the treatment of drug-resistant tuberculosis. The optimal dosing for this class of drugs, including meropenem, which must be given together with a β-lactamase inhibitor, is uncertain.
What This Study Adds to the field
In COMRADE, a phase 2A early bactericidal activity study, reducing the dose of meropenem (given with amoxicillin/clavulanate) from the World Health Organization suggested dosing of 6 g daily (in divided doses) to 3 g daily (given all at once or in divided doses) reduced activity. Tolerability and acceptability at all doses was poor, owing to gastrointestinal side effects or intravenous delivery, with a high number of dropouts. The role of this drug combination in the context of regimens for drug-resistant tuberculosis remains uncertain. Oral or better tolerated carbapenems (with or without β-lactamase inhibitors) are needed.
Tuberculosis (TB) that is resistant to current first- and second-line anti-TB agents remains a public health burden worldwide and threatens to undo the advances made toward TB elimination. Globally, 3.4% of new TB cases and 18% of previously treated cases have resistance to at least one first- or second-line agent (1). Current World Health Organization (WHO)–endorsed treatment regimens for multidrug-resistant (MDR) TB are poorly tolerated and have significant toxicities (2). Fortunately, new drugs with novel mechanisms of action have been approved in recent years, and some combinations involving these agents are highly promising (3), albeit not yet widely available. The role of existing second-line drugs, especially group B and group C drugs, should be reexamined in the context of new and emerging therapies. Specifically, drug repurposing and optimizing the use of existing antibiotics, alongside development of novel agents, is a rational strategy for improving the treatment of TB.
Evaluation of the activity of a drug typically occurs through quantification of its early bactericidal activity (EBA), measured as the drop in bacterial load in sputum up to 14 days. The β-lactam class of antibiotics was generally considered unsuitable for the treatment of TB, until a study of the 14-day EBA in humans showed otherwise (4). It was expected that BlaC, the chromosomally encoded broad-spectrum β-lactamase of Mycobacterium tuberculosis, would hydrolyze β-lactams, rendering them useless (5). However, carbapenem antibiotics, such as imipenem and meropenem, are poor substrates for BlaC. Carbapenems exhibit in vitro activity against drug-sensitive and drug-resistant strains of M. tuberculosis (6–8). The carbapenems irreversibly inhibit L,D-transpeptidases, enzymes crucial in the formation of the peptidoglycan layer of the M. tuberculosis cell wall (9, 10). In vitro and in vivo synergy of carbapenems with rifampin against M. tuberculosis and other mycobacteria has also been reported, providing a basis for the addition of a carbapenem to rifampin-based regimens (7, 11, 12). The addition of a β-lactamase inhibitor, like clavulanate, significantly decreases the minimum inhibitory concentration of carbapenems against M. tuberculosis, lowering the exposure required to achieve bacterial killing (6, 13, 14). Amoxicillin/clavulanate itself has limited or no EBA; however, the addition of amoxicillin can potentiate the activity of meropenem/clavulanate in vitro (15, 16).
In animal models of TB, carbapenems have shown promise in reducing the M. tuberculosis burden. However, because they are rapidly inactivated by dehydropeptidases that are expressed at high levels in mouse, rabbit, and guinea pig tissues (17), carbapenems may not achieve sufficient exposures in animal models of TB to reflect their true therapeutic potential. In humans, the addition of carbapenems to both MDR and extensively drug-resistant (XDR) TB regimens resulted in significant improvement in treatment outcomes in small clinical trials and operational research (18–21).
Although meropenem is typically well tolerated, it is only available in intravenous formulations and must be given with amoxicillin/clavulanate, which may have side effects. The meropenem dose recommended by the WHO for treatment of TB is 2 g every 8 hours (TID). However, it is not known whether lower doses or less frequent administration—although likely to improve drug costs and feasibility of meropenem administration—will improve tolerability or compromise efficacy. Progress has been made in developing orally bioavailable carbapenems and penems, although oral administration is typically associated with lower exposures. We conducted this study to evaluate whether a lower total daily dose of meropenem—given in divided doses or once daily—would provide measurable bactericidal activity and improve tolerability. At the same time, we wanted to assess synergy and drug interactions between rifampin and meropenem for future comparisons among patients with MDR TB. Secondary aims were to explore dose–response relationships, safety, and tolerability, with the goal of informing the development of novel oral carbapenems. We report here results from the treatment arms conducted with participants with rifampin-susceptible pulmonary TB. Enrollment of participants with rifampin-resistant TB (arms A and B) is ongoing. Some of the results of this trial have been previously reported in the form of an abstract (22).
Methods
Study Design and Participants
The COMRADE (Coamoxiclav Meropenem Rifampin Antimycobacterial Determination of Efficacy) Study (ClinicalTrials.gov identifier: NCT03174184) is a phase 2A randomized clinical trial among adults with pulmonary TB, conducted at a single site in Cape Town, South Africa. The study received appropriate regulatory and ethics committee approval and was conducted according to international and South African Good Clinical Practice (GCP) guidelines. All study participants provided written informed consent.
Individuals aged 18 to 65 years were included in the study if they were newly diagnosed with pulmonary TB and treatment naive and had a grade of at least 1+ acid-fast bacilli on sputum smear microscopy based on the International Union Against Tuberculosis and Lung Disease scale or a semiquantitative result of “medium” or “high” on the Xpert MTB/RIF assay (Cepheid, Sunnyvale, CA). Participants with an Xpert MTB/RIF result of “rifampin resistance not detected” could participate in arm C, D, E, or F. Participants with HIV and a CD4 count ⩾100 cells/mm3 were eligible. Study participants were hospitalized at the trial site for the duration of the study. After completion of study treatment at Day 14, participants were started on standard-of-care TB treatment and referred back to their local community clinic to complete a full course of treatment as per national guidelines. A follow-up visit was performed 14 days after discharge.
Procedures
Eligible participants were randomized 1:1:1:1 to receive one of the following study treatments daily for 14 consecutive days: meropenem 2 g TID plus oral rifampin 20 mg/kg once daily (QD) (arm C), meropenem 2 g TID (arm D), meropenem 1 g TID (arm E), or meropenem 3 g QD (arm F). The duration of the intravenous meropenem infusion was approximately 30 minutes per dose for arms C–E and 60 minutes for arm F. Patients in arms C–E received oral amoxicillin/clavulanate dosed at 500 mg/125 mg orally TID at the start of each meropenem infusion. In arm F, oral amoxicillin/clavulanate dosed at 875 mg/125 mg was administered once daily at the start of the meropenem infusion.
Outcomes
Safety assessments
Adverse events were collected daily by study investigators, and severity was graded according to the Division of AIDS Table for Grading Severity of Adult Adverse Events (version 2.1, July 2017). All adverse events of severity grade 2 or higher were included in the safety analysis. Blood samples for laboratory-based safety assessments were collected at screening and on Study Days 7, 14, and 28. In the event of an early withdrawal from the study, similar clinical and laboratory safety assessments were performed.
Microbiology
Identification and genotypic susceptibility to rifampin were established with the GeneXpert MTB/RIF assay on sputum collected before randomization. Phenotypic susceptibility to rifampin was determined using the BD Bactec MGIT 960 Mycobacteria Culture System (Becton Dickinson) on baseline sputum cultures. Genotypic susceptibility to rifampin and isoniazid was confirmed using the GenoType MTBDRplus assay (Hain Lifescience GmbH). For determination of bactericidal activity, serial 16-hour overnight sputum samples were collected from each participant daily for 2 days before treatment initiation and again on treatment Days 1, 2, 3, 4, 6, 8, 10, 12, and 14. Sputum samples were processed on the day of arrival at the study laboratory. Sputum was diluted with an equal volume of dithiothreitol and homogenized, and 0.1 ml was inoculated in 10-fold serial dilutions in quadruplicate onto 7H11S agar plates and incubated at 37°C for 3–6 weeks to determine colony-forming unit (CFU) counts on solid culture. In addition, homogenized sputum and NaOH-decontaminated sputum were incubated in duplicate in the BD Bactec MGIT 960 Mycobacteria Culture system, and time to positive culture (TTP) was recorded in hours.
Pharmacokinetics
Blood samples for pharmacokinetic (PK) analysis were collected on the 13th day of study treatment at the following time points: predose, then 0.5, 1, 1.5, 2, 3, 4, 6 and 8 hours postdose, as previously described (23). For participants receiving rifampin, two additional blood samples were collected at 12 and 24 hours postdose. Blood samples were collected in K2-EDTA and lithium heparin collection tubes for meropenem and rifampin bioanalysis, respectively, and centrifuged within 30 minutes of collection in a cold centrifuge at 2,000 × g for 10 minutes. Plasma was stored at −80°C until bioanalysis. Noncompartmental analysis of steady-state PK parameters were calculated using Phoenix WinNonlin software, Version 8.1. PK parameters analyzed included overall systemic exposure in terms of area under the concentration–time curve (AUC) and maximum plasma concentration of meropenem and rifampin.
Statistical Considerations
The sample size of, on average, 15 participants per treatment arm conforms to similar exploratory EBA trials (24). For determination of bactericidal activity, we included only participants with at least one available baseline and at least one available on-treatment CFU or TTP value. Safety data are reported for all participants who received at least one dose of study drug. Demographic and safety summary data are reported in tabular form, by regimen. Median daily fall in CFU count per milliliter of sputum (solid culture) (EBACFU0–14) and increase in TTP in liquid culture (EBATTP0–14) over 14 days was assessed using a model-based approach. Separate CFU and TTP models were developed based on mono- or biexponential functions, after which covariate relationships on demographics as well as variables on disease severity were tested. Finally, the treatment regimen was tested using different functions supported by the graphical analyses. Bayesian estimations of individual EBA values were generated using the model of which the median and 2.5th–97.5th percentiles of each treatment arm were reported (see Supplementary Methods in the online supplement).
Results
Participants
Between August 2017 and July 2018, 60 participants were enrolled in the study. Median age (interquartile range) was 36 years (29–46 yr), 75% of study participants were male, and 23% were HIV positive (Table 1). Baseline mycobacterial loads represented by sputum CFU count and TTP were similar among treatment arms. Of enrolled participants, 60 were included in safety analyses, 56 met criteria for inclusion in CFU analyses, and 58 met criteria for inclusion in TTP analyses. In addition, based on visual inspection of plots of individual longitudinal data and of plots of results from replicate CFU or TTP data by the clinical, laboratory, and statistical team (while still blinded to treatment arm), 2 CFU and 11 TTP observations were treated as discordant outlying observations and were not included in the analysis dataset (see online supplement).
Table 1.
Baseline Demographics and Mycobacterial Load of All Enrolled Participants, by Treatment Arm
| Characteristic | Overall (n = 60) | Mero 2 g TID + Amx/Clv + Rifampin (Arm C) (n = 17) | Mero 2 g TID + Amx/Clv (Arm D) (n = 15) | Mero 1 g TID + Amx/Clv (Arm E) (n = 14) | Mero 3 g QD + Amx/Clv (Arm F) (n = 14) | P Value |
|---|---|---|---|---|---|---|
| Sex, n (%) | 0.18 | |||||
| F | 15 (25) | 4 (24) | 7 (47) | 2 (14) | 2 (14) | |
| M | 45 (75) | 13 (76) | 8 (53) | 12 (86) | 12 (86) | |
| Age, yr, median (IQR) | 36 (29–46) | 33 (26–47) | 36 (33–45) | 45 (30–47) | 34 (30–43) | 0.66 |
| Self-reported race, n (%) | 0.11 | |||||
| Black | 17 (28) | 3 (18) | 2 (13) | 5 (36) | 7 (50) | |
| Colored | 43 (72) | 14 (82) | 13 (87) | 9 (64) | 7 (50) | |
| BMI, kg/m2, median (IQR) | 18.7 (17.6–20.4) | 18.9 (17.9–20.1) | 18.7 (17.1–19.7) | 19.4 (18.3–20.8) | 18.1 (16.6–19.8) | 0.48 |
| Cavitary disease, n (%) | 0.92 | |||||
| No | 11 (19) | 2 (12) | 2 (13) | 4 (31) | 3 (21) | |
| Yes (<4 cm) | 13 (22) | 4 (25) | 4 (27) | 3 (23) | 2 (14) | |
| Yes (⩾4 cm) | 34 (59) | 10 (62) | 9 (60) | 6 (46) | 9 (64) | |
| Unknown | 2 (6) | 1 (6) | 0 | 1 (7) | 0 | |
| Sputum grade, n (%) | 0.75 | |||||
| Scanty | 1 (2) | 1 (6) | 0 | 0 | 0 | |
| 1+ | 14 (23) | 3 (18) | 5 (33) | 3 (21) | 3 (21) | |
| 2+ | 13 (22) | 4 (24) | 2 (13) | 5 (36) | 2 (14) | |
| 3+ | 30 (50) | 9 (53) | 7 (47) | 5 (36) | 9 (64) | |
| Test not done | 2 (3) | 0 | 1 (7) | 1 (7) | 0 | |
| MGIT TTP, h, median (IQR) | 108 (93.6–158.4) | 103.2 (91.2–117.6) | 105.6 (91.2–136.8) | 136.8 (103.2–153.6) | 117.6 (93.6–175.2) | 0.75 |
| CFUs, log10 CFUs/ml, median (IQR) | 6.3 (5.6–7.1) | 6.05 (4.8–7.3) | 6.4 (5.8–7.1) | 6.25 (5.7–7.0) | 6.3 (5.8–6.9) | 0.96 |
| HIV, n (%) | 0.10 | |||||
| Negative | 46 (77) | 16 (94) | 11 (73) | 11 (79) | 8 (57) | |
| Positive | 14 (23) | 1 (6) | 4 (27) | 3 (21) | 6 (43) | |
| CD4, cells/mm3, median (IQR)* | 321 (146–510) | 463 (375–551) | 339 (277–433) | 132 (128–212) | 397 (146–518) | 0.13 |
Definition of abbreviations: Amx/Clv = amoxicillin/clavulanate; BMI = body mass index; CFUs = colony-forming units; IQR = interquartile range; Mero = meropenem; MGIT TTP = time to positive culture in Bactec MGIT 960 liquid culture system; QD = once daily; TID = every 6 hours.
Among HIV-positive participants.
Safety and Tolerability
Adverse events that were grade 2 or higher are listed in Table 2. Of these, two were grade 3: anemia in one participant and elevated transaminases (alanine aminotransferase and aspartate aminotransferase in another, both in arm F; each resolved or improved without interruption of study treatment. No deaths or grade-4 adverse events occurred. One case of elevated transaminases of grade 2 led to prolongation of hospitalization and was thus classified as a serious adverse event (arm D). The most frequently reported adverse events of grade 2 severity or higher were gastrointestinal or hepatobiliary in nature; namely, nausea and vomiting, loose stools, and elevated transaminases (Table 2). There were 12 early withdrawals during the treatment period (Figure 1). One woman with a negative enrollment pregnancy test in arm D had a positive test on Day 14. She delivered a healthy infant at term.
Table 2.
Number of Participants with Grade 2 or Higher Adverse Events, by Arm
| Adverse Event | Mero 2 g TID + Amx/Clv + Rifampin (Arm C) (n = 17) | Mero 2 g TID + Amx/Clv (Arm D) (n = 15) | Mero 1 g TID + Amx/Clv (Arm E) (n = 14) | Mero 3 g QD + Amx/Clv (Arm F) (n = 14) |
|---|---|---|---|---|
| Gastrointestinal and hepatobiliary, n (%) | ||||
| Loose stool | 1 (5.8) | 0 | 0 | 0 |
| Nausea | 2 (11.8) | 1 (6.7) | 0 | 0 |
| Vomiting | 0 | 1 (6.7) | 0 | 0 |
| Elevated ALT or AST or both | 0 | 1 (6.7) | 0 | 1 (7.1) |
| Skin, n (%) | ||||
| Cellulitis | 1 (5.8) | 0 | 0 | 0 |
| Dermatitis of the scrotum | 0 | 1 (6.7) | 0 | 0 |
| Superficial thrombophlebitis | 0 | 0 | 0 | 1 (7.1) |
| Musculoskeletal, n (%) | ||||
| Lower back pain | 0 | 1 (6.7) | 0 | 1 (7.1) |
| Calf pain | 0 | 0 | 1 (7.1) | 0 |
| Hematological, n (%) | ||||
| Anemia | 0 | 0 | 0 | 1 (7.1) |
| Other, n (%) | ||||
| Pharyngitis | 0 | 1 (6.7) | 0 | 0 |
| Night sweats | 0 | 0 | 1 (7.1) | 0 |
| Tachycardia | 0 | 0 | 1 (7.1) | 0 |
| Headache | 1 (5.8) | 0 | 0 | 1 (7.1) |
| Sore throat | 0 | 1 (6.7) | 0 | 0 |
Definition of abbreviations: ALT = alanine aminotransferase; Amx/Clv = amoxicillin/clavulanate; AST = aspartate aminotransferase; Mero = meropenem; QD = once daily; TID = every 8 hours. For arms C–E: Amx/Clv was given at a dose of 500 mg/125 mg TID. For arm F: Amx/Clv was given at a dose of 875 mg/125 mg QD.
Figure 1.
Consolidated Standards of Reporting Trials diagram showing participant allocation and reasons for early withdrawal by treatment arm. Amx/Clv = amoxicillin/clavulanate; QD = once daily; TID = every 8 hours.
PK
Overall systemic exposures of meropenem increased in a near-linear, dose-dependent manner (Table 3) and were unaffected by concurrent rifampin administration. Giving meropenem 1 g TID or 3 g QD resulted in similar AUC0–24 values. The half-life of meropenem was short (∼1 h).
Table 3.
Meropenem PK Parameters—Median (with Range)—Estimated by Noncompartmental Analysis, by Arm
| Study Arm and Meropenem Dose | AUC0–24 (h · mg/L) | Cmax (mg/L) | Half-Life (h) | n |
|---|---|---|---|---|
| C (2 g TID, with rifampin) | 574 (377–776) | 123 (99.2–201) | 1.15 (0.92–2.17) | 12 |
| D (2 g TID) | 534 (317–937) | 129 (82.6–187) | 1.09 (0.79–1.28) | 13 |
| E (1 g TID) | 253 (135–730) | 66 (37.6–131) | 1.09 (0.84–3.25) | 12 |
| F (3 g QD) | 309 (198–423) | 174 (109–254) | 1.00 (0.64–1.19) | 12 |
Definition of abbreviations: AUC = area under the concentration–time curve; Cmax = maximum plasma concentration; n = number of patients who underwent PK sampling; PK = pharmacokinetic; QD = once daily; TID = every 8 hours.
Note that all patients received amoxicillin/clavulanate with each meropenem dose.
Bactericidal Activity
The final CFU model consisted of a monoexponential function with interindividual variability on the baseline and slope parameters. Among the tested covariates, there was a statistically significant relationship between the presence of cavitary disease on chest X-ray and baseline CFU count (see Figure E1 and Table E1 in the online supplement). The final TTP model consisted of a monoexponential function with interindividual variability on the baseline and slope parameters and an additional time-dependent function for arm C. Similar to the CFU model, a statistically significant relationship was found between cavity and baseline TTP, where a higher cavity score had a lower baseline TTP (Figure E2 and Table E2). Both regimens of 3 g daily produced similar EBA activity, whether given as a single dose or in divided doses (Figure 2 and Table 4). A total daily dose of 6 g produced a statistically significantly higher EBA than a total daily dose of 3 g when measured by CFU counting on solid media, but this difference in activity was not evident by TTP assessments in liquid media. The arm containing rifampin had a much higher EBA than arms without rifampin, as expected.
Figure 2.
Early bactericidal activity over time for colony-forming units (CFUs, top panel) and time to positive culture (bottom panel) per arm as reported by the median (solid line) and 2.5th–97.5th percentiles (shaded area, dashed lines) of individual predications based on the empirical Bayes estimates. Amx/Clv = amoxicillin/clavulanate; Mero = meropenem; QD = once daily; TID = every 8 hours.
Table 4.
Model-based Bayesian Predicted Median and 2.5th–97.5th Percentiles of Individual EBA for 0–2, 0–7, and 0–14 Days in CFU Count and TTP, by Arm
| Individual EBA* | Mero 2 g TID + Amx/Clv + Rifampin (Arm C) | Mero 2 g TID + Amx/Clv (Arm D) | Mero 1 g TID + Amx/Clv (Arm E) | Mero 3 g QD + Amx/Clv (Arm F) |
|---|---|---|---|---|
| CFUs (log10 CFUs/ml) | ||||
| 0–2 | 0.43 (0.25–0.66) | 0.24 (0.11–0.43) | 0.12 (0.066–0.20) | 0.11 (0.07–0.16) |
| 0–7 | 1.5 (0.86–2.3) | 0.84 (0.40–1.5) | 0.42 (0.23–0.68) | 0.38 (0.25–0.57) |
| 0–14 (overall) | 3.0 (1.7–4.6) | 1.7 (0.80–3.0) | 0.83 (0.46–1.4) | 0.75 (0.49–1.1) |
| 0–14 (per day) | 0.22 (0.12–0.33) | 0.12 (0.057–0.21) | 0.059 (0.033–0.097) | 0.053 (0.035–0.081) |
| TTP (log10 h) | ||||
| 0–2 | 0.13 (0.074–0.24) | 0.016 (0.007–0.052) | 0.013 (0.005–0.033) | 0.017 (0.006–0.058) |
| 0–7 | 0.24 (0.16–0.44) | 0.057 (0.026–0.18) | 0.047 (0.017–0.114) | 0.059 (0.02–0.20) |
| 0–14 | 0.34 (0.21–0.75) | 0.11 (0.052–0.37) | 0.094 (0.034–0.23) | 0.12 (0.04–0.42) |
Definition of abbreviations: Amx/Clv = amoxicillin/clavulanate; CFUs = colony-forming units; EBA = early bactericidal activity; Mero = meropenem; QD = once daily; TID = every 8 hours; TTP = time to positive culture.
EBACFU0–2, EBACFU0–7, and EBACFU0–14 are reported by overall change in CFUs over the time interval. EBACFU0–14 is also described as decrease in log10 CFUs/ml per day (averaged over 14 d) for comparability to other studies. EBA and TTP are reported by overall change in TTP over the time interval.
Individual EBA was based on the empirical Bayes estimates (see online supplement).
Discussion
Meropenem (plus amoxicillin/clavulanate) is recommended by the WHO as a group C drug for drug-resistant TB, those drugs that are used only when a reasonable regimen cannot be constructed with group A or B agents (25). It is, however, only available in intravenous formulation and must be administered multiple times per day. In our trial, lowering the total daily dose from the currently recommended 6 to 3 g reduced microbiologic activity as measured on solid media, and tolerability was not improved, with many participants withdrawing from the study early, citing mild to moderate gastrointestinal disturbances or discomfort of intravenous treatment administration. This calls into question the utility of this drug combination, even in the context of second-line treatment.
Better tolerated carbapenem–β-lactamase inhibitor combinations or carbapenems not needing additional β-lactamase inhibition will be needed for this class of drugs to have an important role in TB treatment regimens.
Giving a TB drug intravenously over the course of at least 6 months is problematic, owing to the high cost and the inconvenience to patients and programs, especially when the drug is dosed three times per day. In the dose-fractionation arms of our study, we attempted to lower the total dose, or to lower both the dose and the frequency to once daily, from the standard 2 g three times per day (total of 6 g daily) to see if those dosing strategies would be similarly efficacious to the standard dose. Although the dose of 3 g total daily produced measurable EBA, the effect was modest at best and lower than that seen with the standard recommended dose. Moreover, tolerability remained poor, with several patients dropping out of the study early, mostly because of low-grade gastrointestinal side effects or aversion to intravenous dosing. Thus, lowering the meropenem dose or giving it all at once does not appear to be a good strategy for optimizing use of carbapenem antibiotics in patients with pulmonary TB in practice.
Developing novel carbapenems that do not require intravenous delivery may be a more successful drug development strategy for TB. Upcoming PK-pharmacodynamic studies that combine data from COMRADE and other completed and ongoing studies with meropenem, faropenem, or ertapenem will help define exposure-response relationships for this class of antibiotic, including the degree to which activity is time dependent and must be present at target concentrations in the blood over the dosing interval. In this way, we hope that our trial data will be informative for drug development of novel carbapenems. Additionally, new carbapenem plus β-lactamase inhibitor combinations must aim to have a better tolerability profile than that of meropenem plus amoxicillin/clavulanate to gain wide acceptance.
Our study has limitations. First, there was no arm with meropenem dosed at 6 g once daily to compare safety and efficacy with the standard 2 g TID dosing. However, in a recent trial conducted at the same site under similar conditions, the 14-day EBA of meropenem dosed at 6 g daily and infused over 6 hours, together with amoxicillin/clavulanate dosed at 2,000 mg–125 mg, was assessed using CFU count (26). The mean daily log10 decline in CFUs was 0.094 (95% confidence interval [CI], 0.075–0.12), similar to that in arm D in our study (0.12; 95% CI, 0.057–0.21). Arms testing 6 g or 4 g infused over 1 hour are ongoing. Second, all participants reported here had drug-sensitive TB, whereas dose–response relationships may be different for drug-resistant TB (27). The COMRADE study is currently enrolling patients with rifampin-resistant TB into two arms, one in which meropenem is given at 2 g TID, together with amoxicillin/clavulanate and rifampin (arm A, same dosing as in arm C), and another in which meropenem and amoxicillin/clavulanate are given at the same doses but without rifampin (arm B, same dosing as in arm D). With those data in hand, we will be able to compare the activity of meropenem in drug-sensitive versus drug-resistant TB and also to determine whether meropenem restores, to some degree, rifampin antimycobacterial activity, as has been seen in some preclinical models (7). Last, it is unclear why dose–response relationships (e.g., 6 g > 3 g total daily dose) were seen with longitudinal CFU counting on solid media but were not present in TTP measurements on liquid media. Drugs active against peptidoglycan synthesis or remodeling can have myriad effects on viability, osmotic fragility, postantibiotic effects, and alternative cross-linking. These effects could result in laboratory artifacts or in real consequences for viability and persistence that, in turn, present as differences in cultivability on solid versus liquid media. Although CFU counting on solid culture gives an estimate of bacterial death over time on treatment, TTP on liquid culture measures both injury (resulting in slower metabolism) and death of bacteria. Additionally, there are populations of bacteria that grow in liquid media but do not grow in solid media (28). Modeling approaches for CFU versus TTP data are necessarily different, which may also be a contributing factor. Which method (CFU versus TTP) best reflects dose–response relationships that are relevant for drug development and dose optimization remains unclear.
In conclusion, the 14-day bactericidal activity of meropenem is modest when given at a total dose of 3 g daily. Although antimycobacterial activity appears to improve with doubling of the dose, this difference in activity was only evident using CFU counts on solid media and not TTP in liquid media, raising concerns about using either measure alone in dose-fractionation studies, at least in short-term EBA studies of β-lactams until this difference is better understood. Tolerability of meropenem given with amoxicillin/clavulanate in our study was poor. A lower dose did not diminish side effects but did reduce the activity of meropenem, suggesting that the poor tolerability was more due to the circumstances of intravenous treatment or to the companion amoxicillin/clavulanate combination than to carbapenem-related adverse events. In this context, the role for meropenem in programmatic TB treatment appears minimal. Therefore, the field should focus on new carbapenem–β-lactamase inhibitor pairs (or carbapenems that do not require a β-lactamase inhibitor) that are oral and better tolerated yet capable of achieving exposures sufficient for bactericidal activity.
Acknowledgments
Acknowledgment
The authors are grateful, first and foremost, to the study participants and their families. They thank all TASK recruiters and ward staff assisting with this study, as well as the Western Cape and City of Cape Town public health authorities and clinic staff for their collaboration.
Footnotes
Supported by the U.S. Food and Drug Administration Office of Orphan Products Development Grants Program. This study was supported by the U.S. Food and Drug Administration grant no. R01FD005724 and by the National Institute of Allergy and Infectious Diseases grant no. K24AI150349 (K.E.D.). The computations were enabled by resources in project SNIC 2020-5-524, provided by the Swedish National Infrastructure for Computing (SNIC) at UPPMAX, partially funded by the Swedish Research Council through grant agreement no. 2018-05973.
Author Contributions: K.E.D., A.H.D., S.E.D., and N.G. designed the study. V.D.J., C.M.U., and A.H.D. conducted the study. N.G., R.C.v.W., A.F., U.S.H.S., A.A.A., and E.M.S. performed the analyses. V.D.J. and K.E.D. drafted the manuscript. All authors provided input and reviewed and approved the final manuscript.
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.
Originally Published in Press as DOI: 10.1164/rccm.202108-1976OC on March 8, 2022
Author disclosures are available with the text of this article at www.atsjournals.org.
References
- 1.World Health Organization. 2019. https://www.who.int/publications/i/item/9789241565714
- 2.World Health Organization. 2019.
- 3. Conradie F, Diacon AH, Ngubane N, Howell P, Everitt D, Crook AM, et al. Nix-TB Trial Team Treatment of highly drug-resistant pulmonary tuberculosis. N Engl J Med . 2020;382:893–902. doi: 10.1056/NEJMoa1901814. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Diacon AH, van der Merwe L, Barnard M, von Groote-Bidlingmaier F, Lange C, García-Basteiro AL, et al. β-Lactams against tuberculosis—new trick for an old dog? New Eng J Med . 2016;375:393–394. doi: 10.1056/NEJMc1513236. [DOI] [PubMed] [Google Scholar]
- 5. Hugonnet JE, Blanchard JS. Irreversible inhibition of the Mycobacterium tuberculosis beta-lactamase by clavulanate. Biochemistry . 2007;46:11998–12004. doi: 10.1021/bi701506h. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hugonnet JE, Tremblay LW, Boshoff HI, Barry CE, III, Blanchard JS. Meropenem-clavulanate is effective against extensively drug-resistant Mycobacterium tuberculosis. Science . 2009;323:1215–1218. doi: 10.1126/science.1167498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kaushik A, Makkar N, Pandey P, Parrish N, Singh U, Lamichhane G. Carbapenems and rifampin exhibit synergy against Mycobacterium tuberculosis and Mycobacterium abscessus. Antimicrob Agents Chemother . 2015;59:6561–6567. doi: 10.1128/AAC.01158-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Dhar N, Dubée V, Ballell L, Cuinet G, Hugonnet JE, Signorino-Gelo F, et al. Rapid cytolysis of Mycobacterium tuberculosis by faropenem, an orally bioavailable β-lactam antibiotic. Antimicrob Agents Chemother . 2015;59:1308–1319. doi: 10.1128/AAC.03461-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Kumar P, Kaushik A, Lloyd EP, Li SG, Mattoo R, Ammerman NC, et al. Non-classical transpeptidases yield insight into new antibacterials. Nat Chem Biol . 2017;13:54–61. doi: 10.1038/nchembio.2237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Cordillot M, Dubée V, Triboulet S, Dubost L, Marie A, Hugonnet JE, et al. In vitro cross-linking of Mycobacterium tuberculosis peptidoglycan by L,D-transpeptidases and inactivation of these enzymes by carbapenems. Antimicrob Agents Chemother . 2013;57:5940–5945. doi: 10.1128/AAC.01663-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Kaushik A, Ammerman NC, Tasneen R, Story-Roller E, Dooley KE, Dorman SE, et al. In vitro and in vivo activity of biapenem against drug-susceptible and rifampicin-resistant Mycobacterium tuberculosis. J Antimicrob Chemother . 2017;72:2320–2325. doi: 10.1093/jac/dkx152. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ramón-García S, González Del Río R, Villarejo AS, Sweet GD, Cunningham F, Barros D, et al. Repurposing clinically approved cephalosporins for tuberculosis therapy. Sci Rep . 2016;6:34293. doi: 10.1038/srep34293. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Solapure S, Dinesh N, Shandil R, Ramachandran V, Sharma S, Bhattacharjee D, et al. In vitro and in vivo efficacy of β-lactams against replicating and slowly growing/nonreplicating Mycobacterium tuberculosis. Antimicrob Agents Chemother . 2013;57:2506–2510. doi: 10.1128/AAC.00023-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Horita Y, Maeda S, Kazumi Y, Doi N. In vitro susceptibility of Mycobacterium tuberculosis isolates to an oral carbapenem alone or in combination with β-lactamase inhibitors. Antimicrob Agents Chemother . 2014;58:7010–7014. doi: 10.1128/AAC.03539-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Gonzalo X, Drobniewski F. Is there a place for β-lactams in the treatment of multidrug-resistant/extensively drug-resistant tuberculosis? Synergy between meropenem and amoxicillin/clavulanate. J Antimicrob Chemother . 2013;68:366–369. doi: 10.1093/jac/dks395. [DOI] [PubMed] [Google Scholar]
- 16. Chambers HF, Kocagöz T, Sipit T, Turner J, Hopewell PC. Activity of amoxicillin/clavulanate in patients with tuberculosis. Clin Infect Dis . 1998;26:874–877. doi: 10.1086/513945. [DOI] [PubMed] [Google Scholar]
- 17. Rullas J, Dhar N, McKinney JD, García-Pérez A, Lelievre J, Diacon AH, et al. Combinations of β-lactam antibiotics currently in clinical trials are efficacious in a DHP-I-deficient mouse model of tuberculosis infection. Antimicrob Agents Chemother . 2015;59:4997–4999. doi: 10.1128/AAC.01063-15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Chambers HF, Turner J, Schecter GF, Kawamura M, Hopewell PC. Imipenem for treatment of tuberculosis in mice and humans. Antimicrob Agents Chemother . 2005;49:2816–2821. doi: 10.1128/AAC.49.7.2816-2821.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Keener AB. Oldie but goodie: repurposing penicillin for tuberculosis. Nat Med . 2014;20:976–978. doi: 10.1038/nm0914-976. [DOI] [PubMed] [Google Scholar]
- 20. Sotgiu G, D’Ambrosio L, Centis R, Tiberi S, Esposito S, Dore S, et al. Carbapenems to treat multidrug and extensively drug-resistant tuberculosis: a systematic review. Int J Mol Sci . 2016;17:373. doi: 10.3390/ijms17030373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Story-Roller E, Lamichhane G. Have we realized the full potential of β-lactams for treating drug-resistant TB? IUBMB Life . 2018;70:881–888. doi: 10.1002/iub.1875. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.de Jager V, Abufathi AA, Feyt H, Gupte N, Vanker N, Barnes FL, et al. 8–11.2020, Boston, MA.
- 23. Abulfathi AA, de Jager V, van Brakel E, Reuter H, Gupte N, Vanker N, et al. The population pharmacokinetics of meropenem in adult patients with rifampicin-sensitive pulmonary tuberculosis. Front Pharmacol . 2021;12:637618. doi: 10.3389/fphar.2021.637618. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Hafner R, Cohn JA, Wright DJ, Dunlap NE, Egorin MJ, Enama ME, et al. Early bactericidal activity of isoniazid in pulmonary tuberculosis. Optimization of methodology. The DATRI 008 Study Group. Am J Respir Crit Care Med . 1997;156:918–923. doi: 10.1164/ajrccm.156.3.9612016. [DOI] [PubMed] [Google Scholar]
- 25.World Health Organization. 2020. https://www.who.int/publications/i/item/9789240007048
- 26. de Jager VR, Vanker N, van der Merwe L, van Brakel E, Muliaditan M, Diacon AH. Optimizing β-lactams against tuberculosis. Am J Respir Crit Care Med . 2020;201:1155–1157. doi: 10.1164/rccm.201911-2149LE. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Cohen KA, El-Hay T, Wyres KL, Weissbrod O, Munsamy V, Yanover C, et al. Paradoxical hypersusceptibility of drug-resistant Mycobacteriumtuberculosis to β-lactam antibiotics. EBioMedicine . 2016;9:170–179. doi: 10.1016/j.ebiom.2016.05.041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Bowness R, Boeree MJ, Aarnoutse R, Dawson R, Diacon A, Mangu C, et al. The relationship between Mycobacterium tuberculosis MGIT time to positivity and cfu in sputum samples demonstrates changing bacterial phenotypes potentially reflecting the impact of chemotherapy on critical sub-populations. J Antimicrob Chemother . 2015;70:448–455. doi: 10.1093/jac/dku415. [DOI] [PubMed] [Google Scholar]