Once-Daily Plazomicin for Complicated Urinary Tract Infections

Trial Design and Oversight

The Evaluating Plazomicin in cUTI (EPIC) trial was a multicenter, multinational, randomized, double-blind, phase 3 trial that was designed in accordance with FDA guidelines^4,19 and was conducted from January through September 2016. The primary objective of the trial was to show the noninferiority of plazomicin to meropenem in the treatment of complicated urinary tract infections, including acute pyelonephritis. Patients were enrolled at 68 sites in North America and Europe. The institutional review board or ethics committee at each site approved the final protocol, and all patients or their representatives provided written informed consent. An independent data and safety monitoring committee reviewed interim trial data. Full details of the trial design are provided in the protocol and statistical analysis plan, available with the full text of this article at NEJM.org.

The trial was designed by Achaogen and the first author. Data were collected by Achaogen and analyzed in collaboration with three of the academic authors. A data confidentiality agreement was in place between Achaogen and the investigators. All the authors vouch for the accuracy and completeness of the data and for the fidelity of the trial to the protocol. The first draft of the manuscript was written by a medical writer from a contract research organization who was paid by Achaogen, and all the authors contributed to subsequent drafts, provided final approval of the manuscript, and made the decision to submit the manuscript for publication.

Eligibility Criteria

Eligible patients were 18 years of age or older and had a creatinine clearance of more than 30 ml per minute, pyuria, and clinical symptoms of a complicated UTI or acute pyelonephritis that would require at least 4 days of intravenous therapy with an antibiotic agent. Complicated UTIs were indicated by at least two signs or symptoms (chills, rigors, or fever [temperature >38°C]; dysuria or urinary frequency or urgency; lower abdominal or pelvic pain; or nausea or vomiting) and at least one complicating factor (a history of urinary retention [in male patients], current indwelling urinary catheter, obstructive uropathy, or any functional or anatomical abnormality). Acute pyelonephritis was indicated by at least two of the following symptoms: chills, rigors, or fever; flank pain; tenderness in the costovertebral angle; and nausea or vomiting. Details of eligibility criteria are provided in Table S1 in the Supplementary Appendix, available at NEJM.org.

Randomization and Treatment

Patients were randomly assigned in a 1:1 ratio to receive plazomicin or meropenem. Randomization was performed by the site pharmacist or designated staff member according to a prespecified randomization schedule. Block randomization was automated through an interactive Web-based or voice-response system and was stratified according to region and baseline diagnosis (complicated UTI or acute pyelonephritis). The sponsor, investigators, trial staff participating in patient care or clinical evaluations, and patients were unaware of the group assignments. Designated staff members who prepared the trial drugs had knowledge of the group assignments. Patients were assigned to receive plazomicin (15 mg per kilogram of body weight once daily) or meropenem (1 g every 8 hours), administered intravenously, with the option for oral step-down therapy after a minimum of 4 days of intravenous therapy, for a total of 7 to 10 days of therapy. Levofloxacin (500 mg once daily) was the preferred oral agent, but other agents were permitted if levofloxacin was contraindicated. Therapeutic drug monitoring was not implemented. Doses of all trial drugs were adjusted on the basis of creatinine clearance, which was estimated with the use of the Cockcroft–Gault formula.²⁰ Details on the criteria for oral step-down therapy are provided in the Supplementary Appendix.

Analysis Populations, End Points, and Assessments

The intention-to-treat population included all randomly assigned patients. The safety population and the modified intention-to-treat population included all patients in the intention-to-treat population who received any amount of a trial drug. The microbiologic modified intention-to-treat population included patients in the modified intention-to-treat population who had at least one qualifying baseline pathogen (≥10⁵ colony-forming units [CFU] per milliliter) that was susceptible to both meropenem (minimum inhibitory concentration [MIC] of ≤1 μg per milliliter²¹) and plazomicin (MIC of ≤4 μg per milliliter). We also evaluated a population that included all protocol-adherent patients in the microbiologic modified intention-to-treat population who had an available urine culture at day 5, at the test-of-cure visit, or both (microbiologic per-protocol population). Isolate and susceptibility testing were conducted at a central laboratory (JMI Laboratories).

The primary efficacy end points were composite cure (clinical cure and microbiologic eradication) at day 5 and at the test-of-cure visit (15 to 19 days after initiation of intravenous therapy with the assigned trial drug) in the microbiologic modified intention-to-treat population. Clinical cure was defined as a reduction in severity (at day 5 and at the end of intravenous therapy) or complete resolution (at the test-of-cure visit) of all core symptoms with no new symptoms or as a return to the patient’s status before development of the UTI, with no use of nontrial antibiotics for the current complicated UTI. Microbiologic eradication was defined as a reduction in the baseline uropathogen from 10⁵ CFU or more per milliliter to less than 10⁴ CFU per milliliter.

Additional end points assessed in the microbiologic modified intention-to-treat population were clinical cure and microbiologic eradication at day 5, at the end of intravenous therapy (within 24 hours after the last dose of intravenous trial drug and before oral therapy), at the test-of-cure visit, and during late follow-up (days 24 to 32); and microbiologic response at the test-of-cure visit according to baseline pathogen. To evaluate consistency with the results in the overall population, composite cure at the test-of-cure visit was assessed in prespecified subgroups defined according to geographic region (region 1 or 2), baseline diagnosis (complicated UTI or acute pyelonephritis), presence of bacteremia, treatment received (intravenous only or intravenous plus oral), presence of indwelling catheter (yes or no), and previous antibiotic use (yes or no) and in post hoc subgroups defined according to age (<65 or ≥65 years), renal function (creatinine clearance ≤60 ml per minute or >60 ml per minute), and sex. Definitions of efficacy end points are provided in Table S2 in the Supplementary Appendix.

Safety analyses included assessments of adverse events and laboratory values. Serum creatinine levels were assessed at baseline, daily during intravenous treatment, at the end of intravenous treatment, at the test-of-cure visit, and at late follow-up. Decreased renal function was prespecified as an increase in the serum creatinine level of 0.5 mg or more per deciliter (≥40 μmol per liter) above baseline at any time during the trial.^22,23

Statistical Analysis

We calculated that 394 patients would need to be enrolled for the trial to have at least 85% power to show the noninferiority of plazomicin to meropenem in the treatment of complicated UTIs and acute pyelonephritis, with a noninferiority margin of 15 percentage points, at a one-sided significance level of 0.025 for each primary end point, assuming that composite cure would occur in 64.0% of patients at day 5 and in 73.2% at the test-of-cure visit in both treatment groups. Patients whose assessments did not result in definitive findings were considered to have had treatment failure with respect to all efficacy end points assessed in the microbiologic modified intention-to-treat population. Two-sided 95% confidence intervals for the observed differences in composite cure rates (plazomicin minus meropenem) at day 5 and at the test-of-cure visit were calculated with the use of the Newcombe method with continuity correction. A determination of noninferiority required that the lower limits of the 95% confidence intervals at day 5 and at the test-of-cure visit be greater than −15 percentage points (details are provided in the Supplementary Appendix).¹⁹ No corrections for multiple comparisons were performed. The confidence intervals for the other assessments are provided for descriptive purposes and have also not been corrected for multiple comparisons.

Patients

Of the 609 patients enrolled, 604 (99.2%) were included in the safety and modified intention-to-treat populations and 388 (63.7%) were included in the microbiologic modified intention-to-treat population. The most common reason for exclusion from the microbiologic modified intention-to-treat population was the absence of a qualifying baseline uropathogen. Details regarding patient disposition are provided in Figure S1 in the Supplementary Appendix.

Table 1. Table 1. Characteristics of the Patients at Baseline (Microbiologic Modified Intention-to-Treat Population).

Baseline characteristics were balanced between the treatment groups (Table 1, and Table S3 in the Supplementary Appendix). The mean age was 59.4 years, and 72.2% of patients had impaired renal function. There were more patients with complicated UTIs than with acute pyelonephritis (58.2% vs. 41.8%), and 13.1% of patients had an indwelling urinary catheter. The mean duration of intravenous therapy was 5.5 days in each group; the mean duration of intravenous plus oral therapy was 9.2 days in the plazomicin group and 8.9 days in the meropenem group. Most patients (80.6% in the plazomicin group and 76.6% in the meropenem group) received oral step-down therapy, which was primarily levofloxacin. Of the 154 patients with a uropathogen at baseline that was not susceptible to levofloxacin, 59 received oral levofloxacin (27 patients in the plazomicin group and 32 in the meropenem group). A total of 60% of the patients who had a uropathogen that was not susceptible to levofloxacin received the maximum 7 days of intravenous therapy. Common alternative oral antibiotics were trimethoprim–sulfamethoxazole, amoxicillin–clavulanate, and cefixime.

The most common uropathogen was Escherichia coli, followed by Klebsiella pneumoniae. Of the 382 patients with Enterobacteriaceae in the microbiologic modified intention-to-treat population, 107 patients (28.0%) had uropathogens with an extended-spectrum β-lactamase phenotype, 115 (30.1%) had multidrug-resistant uropathogens, and 101 (26.4%) had uropathogens that were not susceptible to other aminoglycosides (amikacin, gentamicin, or tobramycin). MICs for plazomicin and meropenem among baseline uropathogens are provided in Table S4 in the Supplementary Appendix.

End Points

Table 2. Table 2. Primary and Additional Efficacy End Points (Microbiologic Modified Intention-to-Treat Population).

Figure 1. Figure 1. Composite Cure at the Test-of-Cure Visit, According to Patient Subgroups in the Microbiologic Modified Intention-to-Treat Population.

The percentage-point difference is for plazomicin minus meropenem; some values may differ from the expected value because of rounding. All subgroups were prespecified except for age, creatinine clearance, and sex. The 95% confidence intervals have not been corrected for multiple comparisons. Post hoc analyses used the Gail–Simon test for qualitative interaction in all subgroups (qualitative interaction,0.0; P=0.5). The vertical red line indicates the noninferiority margin of 15 percentage points. Results for patients according to geographic region are not reported, since 6 of 388 (1.5%) patients were enrolled in region 1 (United States, Mexico, and Spain) and 382 of 388 (98.5%) patients were enrolled in region 2 (Bulgaria, the Czech Republic, Estonia, Georgia, Hungary, Latvia, Poland, Romania, Russia, Serbia, and Ukraine). Results for patients according to previous antibiotic use (yes or no) are not reported because of the small number of patients who received previous antibiotic therapy (2 patients in the plazomicin group and no patients in the meropenem group). UTI denotes urinary tract infection.

Plazomicin was noninferior to meropenem with respect to the primary efficacy end points (Table 2). At day 5, composite cure was observed in 88.0% of the patients (168 of 191 patients) in the plazomicin group and in 91.4% (180 of 197 patients) in the meropenem group (difference, –3.4 percentage points; 95% confidence interval [CI], –10.0 to 3.1). At the test-of-cure visit, composite cure was observed in 81.7% (156 of 191 patients) and 70.1% (138 of 197 patients), respectively (difference, 11.6 percentage points; 95% CI, 2.7 to 20.3). The lower limit of the 95% confidence interval for the between-group difference in the rate of composite cure at the test-of-cure visit exceeded zero. The directionality of the between-group difference in the rate of composite cure at the test-of-cure visits in subgroups was consistent with that in the overall population (Figure 1). The between-group difference favoring plazomicin that was observed in the subgroup that received only intravenous treatment was greater than that observed in the subgroup that received intravenous plus oral treatment (percentage-point difference, 17.5 vs. 9.6), which suggests that oral therapy did not account for the between-group difference observed at the test-of-cure visit.

Table 3. Table 3. Microbiologic Eradication at the Test-of-Cure Visit According to Pathogen (Microbiologic Modified Intention-to-Treat Population).

Rates of microbiologic eradication were similar in the two treatment groups at day 5 and at the end of intravenous therapy and were higher in the plazomicin group than in the meropenem group at the test-of-cure visit and at late follow-up (Table 2). Among patients with infections caused by extended-spectrum β-lactamase–producing Enterobacteriaceae, microbiologic eradication at the test-of-cure visit was observed in 82.4% of the patients in the plazomicin group and in 75.0% in the meropenem group. In addition, a higher percentage of patients in the plazomicin group than in the meropenem group were found to have microbiologic eradication of Enterobacteriaceae that were not susceptible to other aminoglycosides (78.8% vs. 68.6%) (Table 3).

At late follow-up, the rate of composite cure remained higher in the plazomicin group than in the meropenem group (Table 2). Clinical relapse at late follow-up occurred in 1.6% of the patients in the plazomicin group and in 7.1% in the meropenem group, and microbiologic recurrence occurred in 3.7% and 8.1%, respectively. Most clinical relapses in the meropenem group occurred in patients who had asymptomatic bacteriuria (microbiologic persistence and clinical cure) at the test-of-cure visit. Tables S5 and S6 in the Supplementary Appendix show the outcomes for the modified intention-to-treat population and the microbiologic per-protocol population.

Safety

Table 4. Table 4. Safety Analysis (Safety Population).

The most frequent adverse events in the plazomicin group were diarrhea, hypertension, headache, nausea, vomiting, and hypotension (Table 4). Adverse events associated with a decline in renal function occurred in 11 of 303 patients (3.6%) in the plazomicin group and in 4 of 301 patients (1.3%) in the meropenem group. Potentially ototoxic events were identified in 2 patients (1 in each group). Serious adverse events were reported in 1.7% of patients in each treatment group (Table S8 in the Supplementary Appendix).

During the trial, 21 of 300 patients (7.0%) in the plazomicin group and 12 of 297 patients (4.0%) in the meropenem group had an increase in serum creatinine level of 0.5 mg or more per deciliter above the baseline level (Table 4, and Table S9 in the Supplementary Appendix). Of these, the increase occurred after completion of intravenous therapy in 10 of 300 patients (3.3%) in the plazomicin group and in 3 of 297 patients (1.0%) in the meropenem group; of the 10 patients in the plazomicin group who had serum creatinine increases after completion of intravenous therapy, 9 had had moderate renal impairment at baseline. In general, increases were less than 1.0 mg per deciliter (<80 μmol per liter) from baseline and returned to less than 0.5 mg per deciliter from the baseline value by the last follow-up. Among patients in the plazomicin group, risk factors for an increase in serum creatinine level of 0.5 mg or more per deciliter were moderate renal impairment (14 of 21 patients) and receipt of plazomicin therapy for more than 5 days (12 of 21 patients); 8 of 21 patients had both risk factors.

A total of 12 patients discontinued treatment because of adverse events (6 patients [2%] in each group) (Table S10 in the Supplementary Appendix). In both groups, these adverse events were most commonly associated with renal function; in accordance with the protocol, patients with creatinine clearance of 30 ml or less per minute were to discontinue the trial drug.

No deaths related to a trial drug were observed. One patient who received a single dose of plazomicin died on day 18 from metastatic uterine cancer that had been diagnosed 48 hours after the patient underwent randomization. No deaths occurred in the meropenem group.

In the EPIC trial, plazomicin was noninferior to meropenem in the treatment of patients with complicated UTIs, including acute pyelonephritis, with higher rates of microbiologic eradication and composite cure observed at the test-of-cure visit in the plazomicin group than in the meropenem group. The lower incidence of microbiologic recurrence and clinical relapse in the plazomicin group than in the meropenem group at late follow-up suggests that the greater microbiologic eradication with plazomicin has additional clinical benefit for patients with complicated UTIs, including acute pyelonephritis.

Although the propensity for complicated UTIs to recur — necessitating frequent retreatment with antibiotics and increasing the potential for the development of resistance — is well known,³ comprehensive evaluations of the risk of recurrence of complicated UTIs and relapse are lacking. This trial assessed the sustained effects of antibacterial therapy in patients with complicated UTIs and showed that asymptomatic bacteriuria was more common after treatment with meropenem than after treatment with plazomicin and was associated with subsequent clinical relapse. Asymptomatic bacteriuria is common among patients with complicated UTIs after completion of therapy,²⁷ and treatment is generally not recommended for most patients because of the perceived low risk of clinical consequences.^3,28,29 However, findings from this trial suggest that additional data should be generated to guide approaches for managing asymptomatic bacteriuria in patients at risk for recurrent complicated UTIs.

Unlike recent phase 3 trials involving patients with complicated UTIs,^30-32 the EPIC trial excluded patients from the microbiologic modified intention-to-treat population who had pathogens that were resistant to the comparator; thus, the results were not biased toward plazomicin. Meropenem was a reliably active comparator to evaluate the efficacy of plazomicin for multidrug-resistant pathogens. The higher observed rate of composite cure at the test-of-cure visit in the plazomicin group, including in patients with important resistant pathogens, such as extended-spectrum β-lactamase–producing Enterobacteriaceae and Enterobacteriaceae that are not susceptible to aminoglycosides, was therefore unrelated to meropenem resistance.

This trial assessed possible nephrotoxicity through analyses of adverse events and changes in serum creatinine level as markers of decreased renal function. Because published reports of clinical trials of other aminoglycosides do not systematically include analyses of serum creatinine levels after completion of treatment, an important finding in this trial was that a small number of patients in the plazomicin group had increases in serum creatinine levels approximately 1 week after completion of therapy. Risk factors for decreased renal function were consistent with drug accumulation, which is known to be associated with aminoglycoside toxicity,³³ and with the general observation that chronic renal dysfunction is a predictor of acute renal dysfunction.³⁴ Conversely, among patients with normal renal function (creatinine clearance >90 ml per minute), the incidence of nephrotoxicity was lower in the plazomicin group than in the meropenem group. Renal function was determined on the basis of ideal body weight, and plazomicin doses on the basis of adjusted body weight, which may have prevented obese patients from receiving a dose that was too high and may have helped to minimize the risk of new-onset renal dysfunction.

The proportion of patients who received oral step-down therapy with levofloxacin despite having a baseline pathogen that was not susceptible to levofloxacin was similar in the two treatment groups, which therefore minimized the possible effect on the test-of-cure results. The majority of these patients received the intravenous trial drug for the prespecified maximum duration. Subgroup analyses showed that composite cure rates were higher among patients who received meropenem plus oral step-down therapy than among patients who received meropenem alone. The rates of composite cure were more consistent in the subgroups of patients who received plazomicin, which suggests that the higher rate of composite cure with plazomicin than with meropenem observed at the test-of-cure visit was not confounded by a switch to oral therapy.

A limitation of the trial is that patients from countries outside Europe and patients of nonwhite race were underrepresented, a limitation that is similar to that in previous studies of complicated UTIs.^31,32,35 However, this factor most likely did not affect the results, because the pharmacokinetics of plazomicin and meropenem are not expected to be affected by ethnic group. In addition, the analysis in the microbiologic modified intention-to-treat population excluded patients with pathogens resistant to the trial drugs, so geographic differences in resistance rates also would not have affected the results.

In conclusion, the EPIC trial met the primary objective of showing the noninferiority of plazomicin to meropenem in composite cure at day 5 and at the test-of-cure visit. These findings support the use of once-daily plazomicin in adult patients with complicated UTIs or acute pyelonephritis, including infections caused by extended-spectrum β-lactamase–producing Enterobacteriaceae and Enterobacteriaceae that are not susceptible to other aminoglycosides.