WASHINGTON -- A novel aminoglycoside antibiotic called plazomicin got a tepid reaction from FDA staff in advance of an , with a briefing document pointing to safety signals and no clear evidence that the agent is better than existing antibiotics.
The agency's Antimicrobial Drugs Advisory Committee will evaluate plazomicin for two indications:
- Complicated urinary tract infections (cUTIs) involving E. coli, Klebsiella pneumoniae, Proteus spp, and Enterobacter cloacae
- Bloodstream infections and hospital-acquired or ventilator-associated bacterial pneumonia (HABP/VABP) involving E. coli and K. pneumoniae
Plazomicin is an intravenous drug developed by Achaogen Inc. and is a semisynthetic agent derived from sisomicin, which appears active against bacteria expressing aminoglycoside modifying enzymes that confer resistance to other agents in this class. It is effective against Gram-negative bacteria but not Streptococcus or Enterococcus species.
Achaogen's main registration study for cUTIs was structured as a randomized non-inferiority trial comparing the drug to meropenem in 609 patients, with composite microbiological-clinical cure and clinical response at day 5 as co-primary endpoints. Those rates were similar in both groups, in the vicinity of 90% for both. They were also similar for "test of cure" conducted after a mean of 17 days (SD 2 days). Both the sponsor and the FDA agreed that non-inferiority was established.
For bloodstream infections and HABP/VABP, Achaogen had hoped to demonstrate superiority for plazomicin versus colistin, but this fell short.
A randomized, multi-center open-label trial dubbed Study 007 began in 2014 and took 2 years to complete. In addition to plazomicin and colistin, patients received other antimicrobial agents both prior to randomization and as adjunctive treatment during the study period. The latter could include tigecycline or meropenem.
Originally, the primary endpoint was set as all-cause mortality at 28 days, but problems with patient accrual (there was a long list of exclusion criteria) led Achaogen to request a change to a composite of 28-day mortality plus "significant disease-related complications." The planned sample size was 286 patients; instead, only 37 were enrolled, of whom 17 received plazomicin. Subsequently, another cohort of 20 patients were recruited and treated with open-label plazomicin plus other agents, with no control group.
Because of all these protocol changes and the small number of patients treated, it was "difficult" to demonstrate superiority for plazomicin even though there was a statistically significant survival advantage in the patients with bloodstream infections, the FDA's briefing document noted. In fact, active infection wasn't conclusively demonstrated for most of the patients in the initial cohort at treatment day 1, further compromising the interpretation.
With regard to safety, "a signal for nephrotoxicity was noted," the staff review said, although this effect was generally not severe and appeared reversible. The defects in Study 007 meant that no meaningful safety comparison with colistin was possible.
As with other aminoglycosides, ototoxicity was a concern for plazomicin going into the clinical studies. Three such events in total were noted among more than 400 patients treated in the phase II and phase III trials, which the FDA review suggested was reassuring. Still, the document indicated, "there is no definitive evidence that aminoglycoside-associated ototoxicity does not occur with plazomicin."
Achaogen itself acknowledged the limitations in its trial data, agreeing to indications that stipulate, "As only limited clinical safety and efficacy data for plazomicin are currently available, plazomicin should be reserved for use in patients who have limited or no alternative treatment options."
Advisory committee members will be asked to vote on whether safety and efficacy have been adequately established for plazomicin and whether the risk-benefit balance supports approval. The FDA is not required to follow committee recommendations but it usually does.