Oral Lefamulin vs Moxifloxacin for Early Clinical Response Among Adults With Community-Acquired Bacterial Pneumonia: The LEAP 2 Randomized Clinical Trial

Key Points

Question  Is 5-day oral lefamulin noninferior to 7-day oral moxifloxacin in the management of community-acquired bacterial pneumonia?

Findings  In this randomized clinical trial that included 738 patients, the early clinical response at 96 hours (within a 24-hour window) after the first dose of study drug was 90.8% in the lefamulin group and 90.8% in the moxifloxacin group, a difference that met the noninferiority margin of 10%.

Meaning  This study demonstrated the noninferiority of oral lefamulin to oral moxifloxacin for the treatment of community-acquired bacterial pneumonia.

Importance  New antibacterials are needed to treat community-acquired bacterial pneumonia (CABP) because of growing antibacterial resistance and safety concerns with standard care.

Objective  To evaluate the efficacy and adverse events of a 5-day oral lefamulin regimen in patients with CABP.

Design, Setting, and Participants  A phase 3, noninferiority randomized clinical trial conducted at 99 sites in 19 countries that included adults aged 18 years or older with a Pneumonia Outcomes Research Team (PORT) risk class of II, III, or IV; radiographically documented pneumonia; acute illness; 3 or more CABP symptoms; and 2 or more vital sign abnormalities. The first patient visit was on August 30, 2016, and patients were followed up for 30 days; the final follow-up visit was on January 2, 2018.

Interventions  Patients were randomized 1:1 to receive oral lefamulin (600 mg every 12 hours for 5 days; n = 370) or moxifloxacin (400 mg every 24 hours for 7 days; n = 368).

Main Outcomes and Measures  The US Food and Drug Administration (FDA) primary end point was early clinical response at 96 hours (within a 24-hour window) after the first dose of either study drug in the intent-to-treat (ITT) population (all randomized patients). Responders were defined as alive, showing improvement in 2 or more of the 4 CABP symptoms, having no worsening of any CABP symptoms, and not receiving any nonstudy antibacterial drug for current CABP episode. The European Medicines Agency coprimary end points (FDA secondary end points) were investigator assessment of clinical response at test of cure (5-10 days after last dose) in the modified ITT population and in the clinically evaluable population. The noninferiority margin was 10% for early clinical response and investigator assessment of clinical response.

Results  Among 738 randomized patients (mean age, 57.5 years; 351 women [47.6%]; 360 had a PORT risk class of III or IV [48.8%]), 707 (95.8%) completed the trial. Early clinical response rates were 90.8% with lefamulin and 90.8% with moxifloxacin (difference, 0.1% [1-sided 97.5% CI, –4.4% to ∞]). Rates of investigator assessment of clinical response success were 87.5% with lefamulin and 89.1% with moxifloxacin in the modified ITT population (difference, –1.6% [1-sided 97.5% CI, –6.3% to ∞]) and 89.7% and 93.6%, respectively, in the clinically evaluable population (difference, –3.9% [1-sided 97.5% CI, –8.2% to ∞]) at test of cure. The most frequently reported treatment-emergent adverse events were gastrointestinal (diarrhea: 45/368 [12.2%] in lefamulin group and 4/368 [1.1%] in moxifloxacin group; nausea: 19/368 [5.2%] in lefamulin group and 7/368 [1.9%] in moxifloxacin group).

Conclusions and Relevance  Among patients with CABP, 5-day oral lefamulin was noninferior to 7-day oral moxifloxacin with respect to early clinical response at 96 hours after first dose.

Trial Registrations  ClinicalTrials.gov Identifier: NCT02813694; European Clinical Trials Identifier: 2015-004782-92

In the United States, pneumonia is among the most common causes of hospitalization and a leading cause of infectious death.^1,2 Patients who recover from pneumonia experience long-term mortality substantially higher than in age- and sex-matched controls, primarily due to comorbidities.³ Older adults are particularly vulnerable to poor outcomes from pneumonia.^4,5 Common causative pathogens of community-acquired bacterial pneumonia (CABP) include Streptococcus pneumoniae, Staphylococcus aureus, Haemophilus influenzae, Moraxella catarrhalis, and the atypical pathogens Mycoplasma pneumoniae, Chlamydophila pneumoniae, and Legionella pneumophila. Current first-line CABP treatment includes macrolides, β-lactams, or fluoroquinolones.^6,7

Quiz Ref IDSurveillance programs have observed trends of generally decreasing susceptibility among S pneumoniae and S aureus isolates to antimicrobials used to treat CABP, including oral penicillin, macrolides, and folate-pathway inhibitors in S pneumoniae and macrolides and fluoroquinolones in S aureus (particularly methicillin-resistant S aureus).⁸ High rates of bacterial resistance, combined with increasing fluoroquinolone-associated safety concerns, have created a need for new treatment options.^6,8,9

Lefamulin is the first pleuromutilin antibiotic approved for intravenous and oral use in humans.^10-12 Lefamulin is active against the most common CABP-causing pathogens, including some strains resistant to other antimicrobial classes.^8,13,14 A previous phase 3 trial (Lefamulin Evaluation Against Pneumonia 1 [LEAP 1]) in adults with moderate to severe CABP (Pneumonia Outcomes Research Team [PORT] risk class ≥III) demonstrated noninferiority of lefamulin to moxifloxacin when both groups initiated intravenous therapy with an optional switch from intravenous to oral treatment.¹⁵ Given these results, LEAP 2 was conducted to compare a 5-day course of oral lefamulin twice daily vs a 7-day course of oral moxifloxacin once daily and the findings from the second trial are reported herein.

This phase 3, double-blind, double-dummy, parallel-group randomized clinical trial was conducted from August 30, 2016, to January 2, 2018, at 99 sites in 19 countries throughout Europe, North America, South America, Asia, and Africa. The study was designed to evaluate lefamulin treatment over 5 days (minimum treatment duration recommended by CABP treatment guidelines)¹⁶ vs moxifloxacin treatment over 7 days (minimum approved treatment duration according to the moxifloxacin prescribing information).¹⁷ The trial protocol and statistical analysis plan appear in Supplement 1.

The trial protocol and the informed consent form were approved by the ethics committee or institutional review board at each participating site. The trial was conducted in compliance with the ethical principles of the Declaration of Helsinki¹⁸ and the Good Clinical Practice guidelines of the International Conference on Harmonisation. An informed consent form was signed by the patient before initiating any study-related procedures.

Adults with PORT risk class II, III, or IV radiographically documented pneumonia, acute illness (≤7 days), and 3 or more CABP symptoms (dyspnea, new or increased cough, purulent sputum production, and chest pain) were eligible for inclusion. Exclusion criteria included receipt of more than 1 dose of a short-acting (having a dosing interval more frequent than every 24 hours) oral or intravenous antibacterial for CABP within 72 hours before randomization, hospitalization for 2 days or longer within 90 days, confirmed or suspected methicillin-resistant S aureus, being at risk for major cardiac events or dysfunction (eg, known prolonged Q-T interval, clinically significant hypokalemia, clinically unstable cardiac disease, complete left bundle branch block), and having significant hepatic disease (eg, known acute hepatitis, history of cirrhosis, manifestation of end-stage liver disease). Complete details regarding the inclusion and exclusion criteria appear in Supplement 1.

Patients were randomized 1:1 using a computer-generated central randomization schedule with a block size of 6 to receive either 600 mg of oral lefamulin every 12 hours for 5 days or 400 mg of oral moxifloxacin every 24 hours for 7 days (Figure). Patients in the lefamulin group received an oral moxifloxacin placebo every 24 hours for 7 days and patients in the moxifloxacin group received an oral lefamulin placebo every 12 hours for 5 days.

Randomization was stratified by PORT risk class (II vs III or IV), geographic region (United States vs outside the United States), and prior receipt of a single dose of short-acting antibiotic therapy for CABP (yes vs no). Per protocol, 25% or less of the study population could have received a single dose of a short-acting antibiotic, and 50% or more was to have had a PORT risk class of III or IV. Study personnel, patients, and the study sponsor were blinded to treatment allocation unless unblinding was necessary for medical management. A data and safety monitoring committee reviewed tolerability and adverse event data on an ongoing basis.

Race/ethnicity was entered into the case report by the study site using fixed categories to comply with regulatory guidance for a clinical trial.¹⁹ The exact method by which patient race/ethnicity was identified (eg, self-reported vs solicited by investigator) and whether the categories were used in a closed-ended or open-ended fashion is unknown and may have varied by study site.

The intent-to-treat (ITT) population included all patients who were randomized. The randomized patients who received any amount of study drug were included in the modified ITT population. The clinically evaluable population included patients who did not have an indeterminate clinical response, received study drug for a total duration of 48 hours or longer (unless patient died prior to 48 hours), did not receive a nonstudy antibacterial potentially effective against CABP pathogens (unless administered due to clinical failure), and had no additional factors that may have confounded efficacy assessment.

The microbiological ITT population included all patients in the ITT population with 1 or more CABP pathogens detected at baseline (defined as any pathogen identified by ≥1 diagnostic method). Detailed microbiological detection and testing methods and baseline pathogen definitions appear in Supplement 1.

Analyses in the ITT population, in the modified ITT population, in the clinically evaluable population, and in the microbiological ITT population were based on the treatment group to which the patient had been randomized. The safety population included all randomized patients who received any amount of study drug, and the analyses were based on the study drug actually received. All patients received the study drug to which they had been randomized.

Quiz Ref IDThe study used CABP end points defined by regulatory bodies in the United States and in Europe.^20,21 The US Food and Drug Administration primary efficacy end point was early clinical response at 96 hours (within a 24-hour window) after receipt of first dose of either study drug in the ITT population. Patients were programmatically classified as responders if they were alive, showed improvement in 2 or more of the 4 CABP symptoms, had no worsening of any CABP symptom, and did not receive a nonstudy antibacterial for the current CABP episode. If these criteria were not met, patients were classified as nonresponders. If lost to follow-up or if there were missing data for these criteria, patients were classified as indeterminate.

The European Medicines Agency coprimary end points (US Food and Drug Administration secondary end points) were predefined based on the European Medicines Agency guidelines^21,22 as investigator assessment of clinical response at test of cure (5-10 days after last dose of study drug) in the modified ITT population and in the clinically evaluable population. Investigator assessment classified patient responses per protocol as a success if CABP was improved or resolved without additional antibacterials, as a failure if the patient died from any cause or a nonstudy antibacterial was required for the current CABP episode, or as indeterminate if the patient was lost to follow-up or these data were missing.

Other secondary analyses reported herein include early clinical response in the microbiological ITT population; investigator assessment of clinical response at test of cure in the microbiological ITT population; early clinical response (ITT population) and investigator assessment of clinical response at test of cure (modified ITT and clinically evaluable populations) in patient subgroups defined by baseline characteristics; and 28-day all-cause mortality in the ITT population. Secondary analyses not reported in this article include investigator assessment of clinical response (patients in the microbiological ITT population who met clinically evaluable criteria [microbiologically evaluable population]) and by pathogen microbiological response (microbiological ITT and microbiologically evaluable populations) at test of cure.

Samples of urine, blood, sputum, oropharyngeal, nasopharyngeal, pleural fluid, or bronchoalveolar lavage were collected at the screening or baseline visit after obtaining informed consent and within 24 hours of the first dose of study drug for eligibility assessment and microbiological testing. Additional assessments included vital signs, 12-lead electrocardiograms, laboratory tests, and treatment-emergent adverse events in the safety population.

A sample size of 738 patients randomized 1:1 provided 90% power to establish lefamulin noninferiority to moxifloxacin for early clinical response, using a 79% responder rate in both treatment groups for the ITT population, a 10% noninferiority margin, and 1-sided α level of .025. Assuming an investigator assessment of clinical response success rate of 80% (for the modified ITT population) and 85% (for the clinically evaluable population at test of cure) and a clinical evaluability rate of 80%, this study had 91% power to demonstrate noninferiority of lefamulin to moxifloxacin for investigator assessment of clinical response at test of cure using a 10% noninferiority margin and a 1-sided α level of .025.

Lefamulin noninferiority vs moxifloxacin was concluded if the lower limit of the 1-sided 97.5% CI for the treatment difference (calculated using a continuity-corrected z statistic for early clinical response and the Miettinen and Nurminen method with adjustment for the randomization stratification factors for investigator assessment of clinical response) exceeded –10%. The noninferiority margin of 10% was based on analysis of observational studies comparing no treatment vs antibacterial therapy,²³ from which a conservative treatment effect of 20% was estimated. A 10% noninferiority margin preserves 50% of the treatment effect and is consistent with US Food and Drug Administration and European Medicines Agency guidelines.^20,21 Because early clinical response and investigator assessment of clinical response were considered primary end points for 2 different regulatory agencies, no adjustment for multiplicity was required.

In the microbiological ITT population, the percentages of patients with an early clinical response or an investigator assessment of clinical response success were determined for each baseline pathogen. Early clinical response (in the ITT population) and investigator assessment of clinical response success (in the modified ITT population and in the clinically evaluable population at test of cure) were also evaluated in subgroups defined by baseline patient characteristics. For these analyses, 2-sided 95% CIs were determined for differences in rates of early clinical response or investigator assessment of clinical response success using a continuity-corrected z statistic.

No inferential testing of these outcomes was completed and the results were interpreted as exploratory descriptive analyses. Because this study was conducted at multiple study sites, post hoc analyses of early clinical response and investigator assessment of clinical response were conducted using hierarchical modeling to evaluate the potential for site effects. Statistical analyses were conducted using version 9.2 or higher of SAS software (SAS Institute Inc).

Among 738 randomized patients (370 to lefamulin and 368 to moxifloxacin), 685 (345 in the lefamulin group and 340 in the moxifloxacin group) completed treatment, and 707 (95.8%) completed the trial (Figure). The mean duration of exposure to study drug was 5.0 days for lefamulin and 6.7 days for moxifloxacin, which reflects the intended duration of active treatment for each drug per the study design.

The demographics and baseline characteristics were well balanced between treatment groups (Table 1 and eTable 1 in Supplement 2). The ITT population was broadly representative of the patient population with CABP. Overall, 37.5% of the patients were aged 65 years or older, 52.4% were male, and 50.1% had some degree of kidney impairment. Per protocol, the 50% maximum cutoff for patients with PORT risk class II was met; 50.4% of the patients were in PORT risk class II, 37.7% in PORT risk class III, and 11.1% in PORT risk class IV. Common comorbid conditions included hypertension (36.2%), asthma or chronic obstructive pulmonary disease (16.7%), and diabetes (13.4%).

In the microbiological ITT population (n = 391 [205 in the lefamulin group and 186 in the moxifloxacin group]), the most commonly isolated baseline pathogens were S pneumoniae (63.7% [249/391]) and H influenzae (26.6% [104/391]), followed by the atypical pathogens M pneumoniae, L pneumophila, and C pneumoniae (22.3% [87/391]; eTable 2 in Supplement 2). Most infections were monomicrobial (70.8% [277/391]). The remaining 29.2% (114/391) of infections were polymicrobial. Across all pathogens, the resistance rates to moxifloxacin were low (eTable 3 in Supplement 2).

The minimum inhibitory concentrations of lefamulin required to inhibit 50%/90% of baseline isolates were as follows: 0.25/0.25 μg/mL for overall, penicillin-resistant, multidrug-resistant, and macrolide-resistant strains of S pneumoniae; 1/2 μg/mL for H influenzae, 0.12/0.12 μg/mL for S aureus, and 0.001/0.001 μg/mL or lower for M pneumoniae.

The early clinical response rate in the ITT population was 90.8% with lefamulin vs 90.8% with moxifloxacin (difference, 0.1% [1-sided 97.5% CI, −4.4% to ∞]; Table 2). The rate of investigator assessment of clinical response at test of cure in the modified ITT population was 87.5% with lefamulin vs 89.1% with moxifloxacin (difference, −1.6% [1-sided 97.5% CI, –6.3% to ∞]). Similar rates of investigator assessment of clinical response at test of cure were recorded for the clinically evaluable population (89.7% with lefamulin vs 93.6% with moxifloxacin; difference, −3.9% [1-sided 97.5% CI, −8.2% to ∞]).

Post hoc hierarchical modeling analyses identified 1 study site as having a potential effect on study results; however, exclusion of this site from the analyses did not affect early clinical response in the ITT population or investigator assessment of clinical response in the modified ITT and clinically evaluable populations, and lefamulin remained noninferior to moxifloxacin (eTable 4 and eText in Supplement 2).

Lefamulin and moxifloxacin demonstrated high rates of early clinical response and investigator assessment of clinical response at test of cure across all baseline CABP pathogens (microbiological ITT population; Table 3), including multidrug-resistant pathogens. Specifically, the early clinical response rate for multidrug-resistant S pneumoniae was 100% (8/8) in the lefamulin group vs 83.3% (10/12) in the moxifloxacin group and the investigator assessment of clinical response success rate was 100% (8/8) in the lefamulin group vs 91.7% (11/12) in the moxifloxacin group. Among difficult to treat pathogens, the early clinical response rate for L pneumophila was 81.3% (13/16) in the lefamulin group vs 94.1% (16/17) in the moxifloxacin group and the investigator assessment of clinical response success rate was 81.3% (13/16) in the lefamulin group vs 88.2% (15/17) in the moxifloxacin group.

Among patients with penicillin-susceptible S pneumoniae, clinical response rates were lower with lefamulin vs moxifloxacin (early clinical response rate: 76.9% [20/26] in the lefamulin group vs 94.7% [36/38] in the moxifloxacin group; and investigator assessment of clinical response rate: 76.9% [20/26] in the lefamulin group vs 100% [38/38] in the moxifloxacin group). Of note, within this patient subgroup, a greater proportion of patients in the lefamulin group had PORT risk class severity scores of III or IV (69.2% [18/26]) compared with the moxifloxacin group (36.8% [14/38]).

Among the 1.6% (6/370) of patients randomized to lefamulin with baseline bacteremia (Table 1), 4 patients had baseline pathogens covered by lefamulin (3 with S pneumoniae and 1 with S aureus); of these, 2 patients achieved early clinical response and had an outcome of investigator assessment of clinical response success at test of cure. Of patients randomized to moxifloxacin, 2.4% (9/368) had baseline bacteremia (Table 1); of the 7 patients who had common CABP pathogens at baseline (5 with S pneumoniae and 2 with S aureus), all achieved early clinical response and 6 had an outcome of investigator assessment of clinical response success at test of cure.

Although the study was not designed for statistical inference in subpopulations, lefamulin and moxifloxacin demonstrated high rates of early clinical response and investigator assessment of clinical response across all CABP severity indices (eFigure in Supplement 2).

The overall incidence of treatment-emergent adverse events was 32.6% with lefamulin and 25.0% with moxifloxacin; most reported treatment-emergent adverse events were mild or moderate in severity (Table 4). The most common treatment-emergent adverse events were diarrhea (12.2%), nausea (5.2%), and vomiting (3.3%) in the lefamulin group and nausea (1.9%), headache (1.6%), and urinary tract infection (1.6%) in the moxifloxacin group.

Gastrointestinal-related treatment-emergent adverse events occurred in 17.9% of patients who received lefamulin and 7.6% of patients who received moxifloxacin. The higher incidence of gastrointestinal-related treatment-emergent adverse events with lefamulin was driven primarily by diarrhea (median duration, 2 days). No cases of diarrhea led to study drug discontinuation. Vomiting led to study drug discontinuation in 2 patients randomized to lefamulin and in 1 patient randomized to moxifloxacin. One case of Clostridium difficile infection was reported. The event occurred in a patient successfully treated with lefamulin who remained hospitalized, with onset approximately 1 week after completing 5 days of active lefamulin treatment. The event resolved after treatment with oral vancomycin.

Both treatment groups reported low treatment-emergent adverse event rates for hepatobiliary events (1.1% in the lefamulin group and 0.5% in the moxifloxacin group) and cardiac events (2.2% in the lefamulin group and 2.4% in the moxifloxacin group). Changes in laboratory values, including liver enzyme tests, are reported in eTable 5 in Supplement 2. No patient met criteria for Hy’s law.²⁵

On day 4 postdose (steady state), the mean change from baseline in QTcF interval was 9.5 ms with lefamulin and 11.6 ms with moxifloxacin. One patient randomized to lefamulin discontinued the study drug because of a mild treatment-emergent adverse event of electrocardiogram Q-T prolongation; the patient’s QTc interval increased from 440.3 ms at baseline to a maximum of 490 ms on day 5. No associated cardiac arrhythmias were observed.

Study drug discontinuation due to treatment-emergent adverse events occurred in 3.3% of patients who received lefamulin and 2.4% of patients who received moxifloxacin, and serious treatment-emergent adverse events were experienced by 4.6% and 4.9% of patients, respectively (Table 4). One patient in the moxifloxacin group had a serious treatment-emergent adverse event considered by the investigator to be possibly related to the study drug (acute non-Q wave anterolateral and septal myocardial infarction); the patient recovered from the event.

Serious treatment-emergent adverse events resulted in the death of 5 patients in the lefamulin group (3 died within the 28-day window, 1 each due to acute respiratory distress syndrome [study day 2], myocardial infarction [study day 3], and pulmonary edema [study day 1]; 2 died beyond the 28-day window, 1 due to subacute aortic valve endocarditis [day 57] and 1 due to acute myeloid leukemia [day 271]) and in the death of 3 patients in the moxifloxacin group (all within the 28-day window; 1 each due to respiratory failure [study day 4], natural causes [study day 12], and cerebral infarction [study day 18]). None of these events was considered related to the study drug.

Quiz Ref IDThis trial demonstrated the noninferiority of 5 days of oral lefamulin to 7 days of oral moxifloxacin among adults with CABP. Both agents were associated with a high clinical response, including analyses by PORT risk class, typical and atypical pathogens, polymicrobial infections, and demographic and baseline characteristics.

These efficacy results are consistent with those obtained in the LEAP 1 trial in which lefamulin was noninferior to moxifloxacin when both groups initiated intravenous therapy with an option to switch from intravenous to oral treatment.¹⁵ The early clinical response rates were high in both trials (90.8% in the lefamulin group and 90.8% in the moxifloxacin group in this trial vs 87.3% in the lefamulin group and 90.2% in the moxifloxacin group in LEAP 1), and notably higher than those reported in other antimicrobial trials of CABP.^26,27 In terms of CABP severity, the greatest overlap between LEAP 1 and this trial was among patients with PORT risk class III, in whom early clinical response rates were high (91.0% in the lefamulin group and 90.2% in the moxifloxacin group in this trial vs 89.3% in the lefamulin group and 93.0% in the moxifloxacin group in LEAP 1).

Quiz Ref IDOral lefamulin was generally well tolerated; however, a greater percentage of patients receiving lefamulin reported gastrointestinal-related treatment-emergent adverse events compared with moxifloxacin. All gastrointestinal-related treatment-emergent adverse events in the lefamulin group were mild to moderate in severity, manageable, and rarely led to study drug discontinuation. Mild or moderate diarrhea was reported in 12.2% of patients in the lefamulin group and in 1.1% of patients in the moxifloxacin group. Gastrointestinal-related treatment-emergent adverse events are the most notable adverse event category associated with lefamulin, particularly when starting with oral therapy.

Quiz Ref IDLefamulin has a 5-day oral treatment option, has no cross-resistance to other classes,⁸ covers typical and atypical CABP pathogens, and the option to switch from intravenous to oral administration may provide an alternative approach for the treatment of vulnerable patients (eg, older patients with comorbid conditions).

The strengths of this study include its randomized design, evaluation of oral only short-course therapy, a high proportion of patients with more severe forms of CABP, limited use of prior antibiotics, and low drug and study discontinuation rates. The use of moxifloxacin, a highly effective antibiotic for treatment of CABP, resulted in assay sensitivity such that noninferiority evaluations were valid. The use of multiple diagnostic modalities, including real-time polymerase chain reaction with conservative cutoff limits, standard culture, urinary antigen testing, and serology, increased the yield of baseline pathogen identification.

This trial has several limitations. First, the extensive list of exclusion criteria may have limited the generalizability of these results to patient subpopulations with major diseases. Second, many baseline pathogens were identified using nonculture methods, limiting the collection of minimum inhibitory concentration data. Third, samples were not tested for viral co-pathogens. Fourth, patients with suspected methicillin-resistant S aureus were excluded per protocol due to poor coverage with moxifloxacin (although 3 patients with a baseline pathogen of methicillin-resistant S aureus were enrolled). Fifth, the overall recovery of resistant pathogens was low. Sixth, race/ethnicity designation may have been misclassified, given that the methods (eg, self-reported vs solicited by investigator in an open-ended or closed-ended fashion) to collect these data may not have been consistent across sites.

Among patients with CABP, 5-day oral lefamulin was noninferior to 7-day oral moxifloxacin with respect to early clinical response at 96 hours after first dose.

Corresponding Author: Jennifer Schranz, MD, Nabriva Therapeutics US Inc, 1000 Continental Dr, Ste 600, King of Prussia, PA 19406 (jennifer.schranz@nabriva.com).

Accepted for Publication: September 5, 2019.

Published Online: September 27, 2019. doi:10.1001/jama.2019.15468

Author Contributions: Dr Moran had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Alexander, Goldberg, Das, Gasink, Spera, Sweeney, Paukner, Wicha, Gelone, Schranz.

Acquisition, analysis, or interpretation of data: Alexander, Goldberg, Das, Moran, Sandrock, Gasink, Spera, Sweeney, Paukner, Gelone, Schranz.

Drafting of the manuscript: Alexander, Das, Sandrock, Paukner, Gelone, Schranz.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Goldberg, Das, Sandrock, Gasink, Spera, Sweeney, Gelone.

Obtained funding: Gelone.

Administrative, technical, or material support: Moran, Paukner.

Supervision: Alexander, Gasink, Paukner, Gelone, Schranz.

Conflict of Interest Disclosures: Drs Alexander, Spera, Paukner, Gelone, and Schranz and Ms Goldberg and Mr Wicha are employees of and own stock in Nabriva Therapeutics plc. Dr Das reported serving as a consultant to AntibioTx, Archaogen, Boston Pharmaceuticals, Cempra, ContraFect, IterumTx, Nabriva Therapeutics, Paratek, Tetraphase, Theravance, UTILITY, Wockhardt, and Zavante. Dr Moran reported receiving grants from Contrafect and Nabriva Therapeutics. Dr Sandrock reported serving as a consultant to Allergan and Nabriva Therapeutics; receiving grants from the National Institutes of Health and the Health Resources and Services Administration; and receiving nonfinancial support from the State of California. Dr Gasink was an employee of and held stock options in Nabriva Therapeutics plc when the study was performed. Ms Sweeney was an employee of and held stock options in Nabriva Therapeutics plc when the study was performed; and reported serving as a consultant to Nabriva Therapeutics and VenatoRX Pharmaceuticals. No other disclosures were reported.

Funding/Support: The study was funded by Nabriva Therapeutics.

Role of the Funders/Sponsor: Nabriva Therapeutics was responsible for the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation and review of the manuscript. Nabriva Therapeutics did not have the right to veto or suppress publication. All authors were responsible for data interpretation and drafting of the manuscript. The approval and decision to submit the manuscript for publication were the responsibility of the coauthors, led by Dr Schranz.

Meeting Presentations: These data were presented, in part, as an oral presentation for abstract LB6 at IDWeek 2018 (October 4, 2018; San Francisco, California) and as mini-oral eposter O1068 at the 29th European Congress of Clinical Microbiology and Infectious Diseases (April 16, 2019; Amsterdam, the Netherlands).

Data Sharing Statement: See Supplement 3.

Additional Contributions: We thank Werner Heilmayer, PhD, an employee of Nabriva Therapeutics GmbH, for coordinating drug dose schedules during the study. We thank Lycely del C. Sepulveda-Torres, PhD, and Michael S. McNamara, MS (both with C4 MedSolutions LLC, a CHC Group company), for the medical writing and editorial support, which were contracted and paid for by Nabriva Therapeutics.