Phase 2, Randomized, Double-Blind Study of the Efficacy and Safety of Two Dose Regimens of Eravacycline versus Ertapenem for Adult Community-Acquired Complicated Intra-Abdominal Infections

Study population.Male and female patients, 18 to 75 years of age, with a body mass index of ≤30 kg/m² and a diagnosis of cIAI requiring urgent surgical or percutaneous intervention and not expected to require antibacterial therapy for longer than 14 days (7 days for patients in India) were eligible for inclusion. For preoperative enrollment, patients were required to have a sonogram or radiographic imaging results congruent with the diagnosis of cIAI; acute surgical or percutaneous intervention was foreseen; it was planned to have a specimen collected by aspiration or a tissue sample sent for culture and sensitivity; and they met the clinical diagnosis of cIAI. For intraoperative or postoperative enrollment, patients had visual confirmation of pus within the abdominal cavity, had samples for aerobic and anaerobic culture taken by aspiration or a tissue sample, and had surgical intervention in which the initial intervention was considered adequate. The acceptable diagnoses were as follows: appendiceal perforation and periappendiceal abscess, diverticulitis abscess, acute gastric and duodenal perforation if operated on >24 h after the perforation, traumatic perforation of the intestines if operated on >12 h after the perforation, and/or abscess or peritonitis due to perforated viscus or another focus of infection or other intra-abdominal abscess excluding liver and spleen from a gastrointestinal source.

Major exclusion criteria included the following: symptoms related to diagnosis of complicated appendicitis for <24 h prior to the current hospitalization, previous hospitalization within 6 months of screening, management by staged abdominal repair or other open abdominal technique, Acute Physiology and Chronic Health Evaluation (APACHE) II score of >25, anticipated survival less than the study period, rapidly progressing disease or immediately life-threatening illness, requirement for vasopressors at therapeutic dosages to maintain a systolic blood pressure of ≥90 mm Hg or diastolic blood pressure of ≥70 mm Hg, renal failure or abnormal renal function, abnormal liver function, or known or suspected inflammatory bowel disease or associated visceral abscess. Patients were also excluded if they had received systemic antibiotics for the current condition for >24 h, received ertapenem or any other carbapenem or tigecycline for the current infection, or required systemic antimicrobial agents other than a study drug.

Eligible patients were randomized in a 2:2:1 ratio to receive an intravenous (i.v.) infusion of eravacycline at 1.5 mg/kg of body weight every 24 h (q24h), eravacycline at 1.0 mg/kg every 12 h (q12h), or ertapenem at 1 g (q24h). Randomization was stratified based on the primary site of infection (complicated appendicitis versus all other diagnoses). An enrollment cap of 50% for patients with complicated appendicitis was planned.

Investigators and patients were blinded to the i.v. study antibiotic regimen. The study site's unblinded pharmacist obtained each patient's study identification number and study drug assignment from a computer-generated randomization coding scheme using an interactive web-based response system and prepared the solutions for infusion. To maintain blinding, matching placebo infusions were used and infusions and infusion lines were masked. The unblinded pharmacist provided the investigator with ready-to-use blinded infusion solutions for administration at scheduled study drug infusion times.

Patients were hospitalized while receiving the study drug. Eravacycline was administered intravenously over 60 min, while ertapenem was administered intravenously over 30 min. Due to differences between eravacycline and ertapenem infusion volumes, the timing of study drug administration (q12h versus q24h), and the length of infusion time (30 min versus 60 min), the study used a double-dummy methodology. During each 24-h period of study drug administration, the patient received three infusions.

Patients were treated for a minimum of 4 days and a maximum of 14 days, depending on clinical response. No other concomitant systemic antibiotics were permitted. Patients who required concomitant systemic antibiotics were withdrawn from the study drug and considered treatment failures. Test-of-cure (TOC) evaluations occurred 10 to 14 days after the last dose of the study drug. A follow-up visit occurred 28 to 42 days after the last dose of the study drug.

Assessments.Clinical outcome assessments included abdominal examination, surgical wound assessment (if applicable), and other pertinent examinations and were performed at the end-of-treatment (EOT), TOC, and follow-up visits. Samples from the site of intra-abdominal infection for culture and susceptibility testing were obtained at the time of the initial surgical procedure, at subsequent reinterventions, and if there were any signs and symptoms at the EOT, TOC, and follow-up visits. Other cultures were obtained during the study as clinically indicated. Blood cultures were taken if the patient had clinical signs of bacteremia.

Safety assessments included adverse events, laboratory tests, vital signs, electrocardiograms, and physical examinations. Treatment-emergent adverse events (TEAEs) were defined as adverse events that started during or after the first dose of study drug administration or increased in severity or relationship to the study drugs during the study. Serious adverse events (SAEs) were defined as those that resulted in death, were life-threatening, required hospitalization or prolonged existing hospitalization, resulted in persistent or significant disability or incapacity, were a congenital anomaly or birth defect, or were considered to be an important medical event.

Statistical methodology.The primary efficacy endpoint was the clinical response at the TOC visit in the microbiologically evaluable (ME) population, a subset of the clinically evaluable (CE) population. The CE population included all randomized patients who received any amount of a study drug and met the minimal disease definition of cIAI and for whom sufficient information was available to determine outcome with no confounding factors present that interfered with assessment of outcome. The ME population included CE patients who also had a baseline pathogen identified and a microbiological response assessed. Clinical response was classified as cure, failure, or indeterminate based on clinical outcomes. A favorable clinical response was cure, defined as complete resolution or significant improvement of signs and symptoms of the initial infection such that no additional antibacterial therapy or surgical or radiological intervention was required. Patients were classified as clinical failures based on death related to intra-abdominal infection, persisting or recurrent infection within the abdomen documented by findings at reintervention, postsurgical wound infection, or administration of effective concomitant antibacterial therapy for any indication after the start of study drug. Clinical response was also evaluated at the EOT and follow-up visits.

Clinical response rates with the associated 95% confidence intervals (CIs) using the Clopper-Pearson method (12) were calculated for each treatment group and each randomization stratum (primary site of infection). The continuity-corrected exact 95% CIs for the difference in clinical response rates between treatment groups (13) were also calculated.

Per-pathogen and per-patient microbiological responses were evaluated at the EOT and TOC visits in the microbiologically modified intent-to-treat (m-MITT) and ME populations. The m-MITT population included all randomized patients who received any amount of a study drug, met the minimal disease definition of cIAI, and had a baseline pathogen identified, regardless of susceptibility to the study drug. MIC determinations were made according to CLSI methods (23). Per-pathogen microbiological response categories were eradication, presumed eradication, persistence, presumed persistence, or assessment not possible. The categories were further classified as favorable (eradication or presumed eradication), unfavorable (persistence or presumed persistence), or indeterminate (assessment not possible). Per-patient microbiological response categories were eradication, presumed eradication, persistence, presumed persistence, superinfection, recurrent infection, entry culture not obtained, no baseline pathogen identified, or unknown.

Per-pathogen and per-patient microbiological response rates with the associated 95% CIs were calculated for each treatment group. The number of favorable responses was determined for each pathogen at each MIC level available for that pathogen. Data were summarized for each individual pathogen and the broad pathogen types for both the EOT and TOC visits.

An exploratory efficacy endpoint was time to defervescence, defined as the first time after the start of study drug treatment that a patient's temperature was ≤38°C and remained ≤38°C for all evaluations over the following 24-h period. Patients who discontinued treatment with the study drug prior to attaining defervescence were censored at the last evaluation at which the temperature was >38°C. Time to defervescence was displayed graphically using Kaplan-Meier methodology (14). The Wilcoxon test (15) was used for pairwise comparisons of the Kaplan-Meier curves between treatment groups.

The safety population included all randomized patients who received any amount of study drug. Patients were evaluated as treated. The number and percentage of patients who experienced TEAEs and SAEs were summarized by system organ class and preferred term levels using the Medical Dictionary for Regulatory Activities (MedDRA) (16) and tabulated by treatment group.

This study was not statistically powered to demonstrate noninferiority of either eravacycline treatment group to the comparator, ertapenem, but was intended to provide an estimate of efficacy and safety. Approximately 150 randomized patients were expected to provide at least 115 patients (46 in each of the eravacycline groups and 23 in the ertapenem group) in the ME population.