Novel Tetracyclines Versus Alternative Antibiotics for Treating Acute Bacterial Infection: A Meta-Analysis of Randomized Controlled Trials

This meta-analysis assessed the efficacy and safety of novel tetracyclines for treating acute bacterial infections. Data from PubMed, Web of Science, EBSCO, Cochrane databases, Ovid Medline, and Embase databases were accessed until 11 July 2019. Only randomized controlled trials (RCTs) comparing the efficacy of novel tetracyclines with that of other antibiotics for treating acute bacterial infections were included. Primary outcomes included the clinical response, microbiological response, and risk of adverse events (AEs). A total of eight RCTs were included, involving 2283 and 2197 patients who received novel tetracyclines and comparators, respectively. Overall, no significant difference was observed in the clinical response rate at test of cure between the experimental and control groups (for modified intent-to-treat [MITT] population, risk ratio [RR]: 1.02, 95% confidence interval [CI]: 0.99–1.05; for clinically evaluable [CE] population, RR: 1.02, 95% CI: 1.00–1.04; and for microbiological evaluable [ME] population, RR: 1.01, 95% CI: 0.99–1.04). No significant difference in the microbiological response at the end of treatment was observed between the experimental and control groups (for ME population, RR: 1.01, 95% CI: 0.99–1.03; for microbiological MITT population, RR: 1.01, 95% CI: 0.96–1.07). No difference was observed concerning the risk of treatment-emergent adverse events (TEAEs), serious adverse events, and discontinuation of treatment due to TEAEs and all-cause mortality between the two groups. In conclusion, clinical efficacy and safety profile for novel tetracyclines in the treatment of acute bacterial infections were found to be similar to those for other available antibiotics.


Introduction
Antibiotics are crucial for treating acute bacterial infections, and the prompt use of appropriate antibiotics can save the life of a patient with sepsis [1]. However, the emergence and dissemination of antibiotic resistance among commonly encountered bacteria in many types of infections, including pneumonia and intra-abdominal, urinary tract, and skin/skin structure infections, have drastically reduced the efficacy of most antimicrobial drugs [2][3][4][5][6][7]. Therefore, searching new antimicrobials to combat the threat of antibiotic-resistant bacteria is urgent.
Recently, two novel tetracyclines, omadacycline, (Nuzyra ® , Paratek Pharmaceuticals, Boston, MA, USA) and eravacycline (Xerava ® , Tetraphase Pharmaceuticals, Watertown, MA, USA), have been developed and approved by the Food and Drug Administration in 2018 [8]. Additionally, they are broad-spectrum antibiotics such as conventional tetracyclines that act against gram-positive, gram-negative, anaerobic, and atypical pathogens. Furthermore, they exhibit potent in vitro activity against multidrug-resistant organisms [9,10]. The clinical efficacy of omadacycline and eravacycline for treating acute bacterial infections is being evaluated in several randomized controlled trials (RCTs) since their development [11][12][13][14][15][16][17][18]. However, no consensus on the efficacy and safety of novel tetracyclines has been reached due to the lack of a systematic analysis and an updated meta-analysis. Therefore, we conducted this meta-analysis to provide a real-time evidence on the efficacy and safety of omadacycline and eravacycline for treating acute bacterial infections.

Study Search and Selection
All RCTs were identified through a systematic literature review of PubMed, Web of Science, EBSCO, Cochrane databases, Ovid Medline, and Embase until July 2019 by using the following search terms: "eravacycline", "Xerava™", "TP-434", "omadacycline", "Nuzyra", "PTK-0796", and "infection". The inclusion criteria included (1) randomized controlled studies and (2) sturdy directly compared the clinical efficacy and safety of novel tetracyclines with those of other antimicrobial agents for treating adult patients with acute bacterial infections. Exclusion criteria included: (1) case reports, and abstracts presented at scientific conferences; (2) those including individuals younger than 18 years of age; (3) studies that only reported in vitro activity, animal studies, or pharmacokinetic-pharmacodynamic assessments; (4) case series without a control group; (5) trials that lacked randomized-control design. Two authors (S.P.C. and S.H.L.) searched and examined publications independently. A third author (C.C.L.) resolved any disagreement in time. The following data were extracted: year of publication, study design, type of infections, antimicrobial regimens, clinical and microbiological outcomes, and adverse effects. This systematic review and meta-analysis were conducted according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement.

Outcome Measurement
The primary outcomes of this meta-analysis included clinical response assessed at the test of cure (TOC) and end of treatment (EOT) visits, which was calculated as the portion of the patients with clinical response among analyzed populations. Clinical response was defined as the signs/symptoms of infection being sufficiently resolved and no further antibacterial therapy was required. Patients were categorized based on the occurrence of primary outcomes as follows: modified intent-to-treat (MITT), clinically evaluable (CE), and microbiologically evaluable (ME) populations. The intention-to-treat (ITT) population included all randomized patients, and the MITT population included all ITT patients who received any amount of the study drug. The CE population included all MITT patients who met the minimal disease definition of acute bacterial infections and had their clinical response assessed at the TOC visit. The ME population included all CE patients who had the baseline pathogen identified and microbiological response assessed. The microbiological MITT (mMITT) population included all MITT patients who met the minimal disease definition of clinical infection and had the baseline pathogen identified. The safety population included all patients who received any study therapy. Treatment-emergent adverse events (TEAEs) were defined as adverse events (AEs) that started during or after the first dose of the study drug administration or increased in severity or were associated with the study drug during the study.

Data Analysis
The Cochrane risk of bias assessment tool [19] was used to assess the quality of enrolled RCTs and the associated risk of bias. Review Manager, version 5.3, with the random-effects model was used for statistical analyses. The heterogeneity degree was assessed using the Q statistic generated from the χ 2 test, and the heterogeneity proportion was assessed using the I 2 measure. Heterogeneity was considered significant at p < 0.10 or I 2 > 50%. Pooled risk ratios (RRs) and 95% confidence intervals (CIs) were calculated for outcome analyses.

Data Analysis
The Cochrane risk of bias assessment tool [19] was used to assess the quality of enrolled RCTs and the associated risk of bias. Review Manager, version 5.3, with the random-effects model was used for statistical analyses. The heterogeneity degree was assessed using the Q statistic generated from the χ 2 test, and the heterogeneity proportion was assessed using the I 2 measure. Heterogeneity was considered significant at p < 0.10 or I 2 > 50%. Pooled risk ratios (RRs) and 95% confidence intervals (CIs) were calculated for outcome analyses.

Discussion
Data from eight RCTs with 4480 patients were collated to compare the efficacy and safety of novel tetracyclines and other antibiotic regimens for treating acute bacterial infections, including cIAIs, ABSSSIs, CABP, and cUTIs. In the present study, we demonstrated that these novel tetracyclines could achieve a similar clinical response as other comparators, which is supported by the following evidence. First, the clinical response rate for novel tetracyclines, namely omadacycline and eravacycline, was similar to other comparative antibiotics. This similarity between novel tetracyclines and comparators was observed in various population analyses, MITT, CE, ME, and mMITT populations, and at different timings of assessment, TOC and EOT. Second, in subgroup analyses, both omadacycline and eravacycline exhibited non-inferior clinical efficacy than comparators. This finding is consistent with those of previous studies [20,21]. In the pooled analysis of OASIS-1 and OASIS-2, Abrahamian et al. [20] demonstrated that omadacycline was non-inferior to linezolid in early clinical response (86.2% vs. 83.9%; difference 2.3, 95%

Discussion
Data from eight RCTs with 4480 patients were collated to compare the efficacy and safety of novel tetracyclines and other antibiotic regimens for treating acute bacterial infections, including cIAIs, ABSSSIs, CABP, and cUTIs. In the present study, we demonstrated that these novel tetracyclines could achieve a similar clinical response as other comparators, which is supported by the following evidence. First, the clinical response rate for novel tetracyclines, namely omadacycline and eravacycline, was similar to other comparative antibiotics. This similarity between novel tetracyclines and comparators was observed in various population analyses, MITT, CE, ME, and mMITT populations, and at different timings of assessment, TOC and EOT. Second, in subgroup analyses, both omadacycline and eravacycline exhibited non-inferior clinical efficacy than comparators. This finding is consistent with those of previous studies [20,21]. In the pooled analysis of OASIS-1 and OASIS-2, Abrahamian et al. [20] demonstrated that omadacycline was non-inferior to linezolid in early clinical response (86.2% vs. 83.9%; difference 2.3, 95% CI: 1.5-6.2) for treating ABSSSIs, and clinical responses were similar across different infection types-cellulitis or erysipelas and major abscess. Lan et al. [21] revealed that eravacycline had a clinical cure rate (88.7%, 559/630) similar to that of comparators (88.7% vs. 90.1%, RR: 0.99, 95% CI: 0.95-1.03) for treating cIAIs. Unlike these two reports [20,21], the present study included more RCTs and infection types, cUTI and CABP, to augment the knowledge regarding the usefulness of eravacycline and omadacycline. Third, the clinical efficacies of novel tetracyclines were similar to those of comparators across infections caused by different pathogens, even MRSA. A previous pooled analysis [20] of OASIS-1 and OASIS-2 revealed that omadacycline had similar efficacy to that of linezolid for treating infections caused by gram-positive anaerobes, including S. aureus, MRSA, Streptococcus pyogenes, and S. anginosus, gram-negative aerobes, and gram-negative anaerobes. In summary, all these findings indicate that novel tetracyclines, eravacycline, and omadacycline, can be as effective as other antibiotics for treating acute bacterial infections.
In addition to the clinical response, this meta-analysis demonstrated that the microbiological response rate for novel tetracyclines was comparable to that of comparators. This similarity in terms of the microbiological response between the experimental and control groups did not change with the timing of the outcome measure and study populations. These findings regarding the favorable microbiological response of novel tetracyclines have been supported by many in vitro studies [22][23][24][25][26][27][28][29]. Several global surveillance investigations [22][23][24][25] have revealed that omadacycline exhibited potent in vitro activity against gram-positive and gram-negative pathogens as well as was active against antibiotic-resistant organisms, such as MRSA, penicillin-resistant S. pneumoniae, and extended-spectrum β-lactamase (ESBL)-producing Escherichia coli. The potency of eravacycline was at least equivalent or 2-to 4-fold greater than that of tigecycline against Enterobacteriaceae, including ESBL-producing, carbapenem non-susceptible strains, and gram-positive cocci isolates [26][27][28][29]. Therefore, these findings regarding the microbiological response in this meta-analysis and previous in vitro studies can support the use of novel tetracyclines for acute bacterial infections.
Finally, the risk of AEs for novel tetracyclines was assessed. Nausea was the most common AE for novel tetracycline users, and novel tetracyclines were associated with a higher risk of nausea and vomiting compared with comparators. Further subgroup analysis revealed that high risks of nausea or vomiting were noted for eravacycline but not for omadacycline. However, compared with other antibiotics, novel tetracyclines had a similar risk of AEs in TEAEs, serious AEs, treatment discontinuation due to TEAEs, and all-cause mortality. All these findings indicated that gastrointestinal intolerance was the most common side effect of novel tetracyclines, especially eravacycline. However, novel tetracyclines were found to be as tolerable as other antibiotics.
This meta-analysis has some limitations. First, although we aimed to investigate the use of novel tetracyclines for treating all types of acute bacterial infections, we found only one study for cUTI as well as for CABP. Additional studies investigating the use of novel tetracyclines for various infection types are warranted. Second, we could not assess the association between in vitro activity and clinical response for each specific pathogen due to the unavailability of data. However, this deficit could be partially compensated by the results of several in vitro studies [22][23][24][25][26][27][28][29] that demonstrated the potent in vitro activity of novel tetracyclines.

Conclusions
In conclusion, clinical and microbiological responses for novel tetracyclines in the treatment of acute bacterial infections were similar to those for other available antibiotics. In the present analysis, eravacycline was associated with higher risks of gastrointestinal AEs, nausea, and vomiting, but overall, novel tetracyclines had a safety profile similar to that of other antibiotics. However, further research is warranted to investigate the role of novel tetracyclines in the treatment of antibiotic-resistant bacteria-associated infections.

Conflicts of Interest:
The authors declare no conflict of interest.