An FDA panel's votes on July 16 to support Johnson & Johnson's Doribax (doripenem) reflected more the difficulty of new antibiotic drug development than the merits of J&J's particular application.

FDA's Anti-Infective Drugs Advisory Committee's votes of 7-6 to support doripenem's efficacy and 8-5 to support its safety were given with many caveats and concerns - not the least of which being a 20 percent non-inferiority (NI) margin that the panel deemed by a 10-3 vote to be scientifically unjustified.

Behind the unjustified NI margin stood major flaws in the design and conduct of the two Phase III NI trials to support doripenem for indications in nosocomial pneumonia (NP) and ventilator-associated pneumonia (VAP).

But the committee members supporting doripenem noted the need for more options to treat antibiotic-resistant pathogens and the plausibility that the antibiotic is similarly efficacious to others in its class, despite the shaky grounds of its proposed non-inferiority.

"I voted yes with great reservation. … My level of confidence in my 'yes' decision would be very low," panel member James Stoller, Cleveland Clinic, said. "I say that on the strength of the methodological shortcomings."

"I hope that perhaps more important than my yes/no vote are the issues around which I give qualification to my 'yes,'" Stoller added.

The panel and FDA pointed to shortcomings in the two Phase III, open-label NI trials - such as lack of blinding in both studies, unsubstantiated assumptions of the active comparator's treatment effect, broad exclusions of data in the final analyses, excessive use of adjunctive therapies in one of the trials, and incomplete chest x-rays confirming the NP and VAP diagnoses among trial subjects.

Committee member Margo Smith, Washington Hospital Center, echoed Stoller's sentiment, voting yes, but qualifying that she was conflicted. "I'm just thinking practically that as far as resistance is concerned out in the real world - when you don't have options and you're pulling out your last gray hair - we need to have something else," she said.

Doripenem is an injectable, synthetic, broad-spectrum carbapenem in the beta-lactam class of antibacterial agents. Its approval last year made it the fourth in the carbapenem antibiotic class for the treatment of complicated intra-abdominal infections and the third to also carry an indication for complicated urinary tract infections.

Doribax is marketed by J&J's Ortho-McNeil division, which acquired rights to market doripenem through its 2005 acquisition of Peninsula Pharmaceuticals.

The advisory panel was generally in agreement that the microbiological data for doripenem was similar to that of other beta-lactams.

"Is doripenem totally different from other carbapenems, totally different from beta-lactams?" asked temporary voting member James Leggett, Oregon Health Sciences University. "The answer is, it seems very plausible that it should be just as effective as anything else."

The point did not sway temporary voting member Thomas Fleming, University of Washington. "When you look similar, are you similar[ly] effective or similar[ly] ineffective? [The sponsor] is claiming similar[ly] effective based on historical data that is highly likely to not be relevant to this context."

Fleming acknowledged the difficulty of designing NI trials for NP and VAP - indications for which no placebo-controlled trials or clinical response data exist for the approved active comparators - but would not support the doripenem application.

"We are not going to get perfection, but this study would not be the poster child for coming close," he asserted.

Fleming pointed to a statistically significant increase in all-cause mortality in the doripenem treatment arm in one of the studies. "Maybe these are safety concerns, maybe they are efficacy concerns - in the end mortality is both. … If it does not prove that we have excess mortality, it's clearly a signal, and signals such as this in other disease settings have led to the need for large-scale trials to rule out that you would have such excesses."

"To me, [the mortality data] is the most reliable interpretable evidence we have, and it's not pointing in the right direction," he added.

A Quest For A Superior Non-Inferiority Design

FDA is in the midst of re-evaluating the appropriate use of NI trials for antibacterial drug development and has published a series of guidances attempting to shed light on its evolving expectations.

A draft guidance published last year specified that sponsors should re-evaluate ongoing or completed NI studies to ensure that the effect size of the active control and the proposed NI margin is based on "adequate scientific rationale" (¹ [A#00690430011], p. 17). It was on this point that the panel rejected J&J's proposed 20 percent NI margin for the doripenem trials.

Panelists were hesitant, however, to advise FDA on an appropriate NI margin, instead urging a public workshop on NI trial design for development of drugs to treat NP and VAP (see ² following story).

The user fee date for the NP and VAP indications is later this month, and comes after a three-month review extension during which the agency processed additional information in the application.

There are four FDA-approved drugs for NP: ciprofloxacin, levofloxacin, linezolid and piperacillin/tazobactam. But there are no FDA-approved drugs for the specific indication of VAP, giving doripenem, if approved, a market edge for the indication.

However, the VAP indication concerned several committee members, including Susan Rehm, Cleveland Clinic Foundation, whose vote against doripenem's efficacy hinged on inclusion of the indication. "Since the number of patients with VAP was small and since we have concerns about the methodology, I find the effect to be unfounded."

Temporary voting member John Bennett, National Institutes of Health, added that "VAP is very difficult to study and very difficult to define. … We may not have reached the point where we know how to design entry criteria for VAP that will have broad enough success that we could actually do a study in a group of patients."

- Jamie Hammon (j.hammon@elsevier.com)

FDA Panel Supports J&amp;J’s Antibiotic Doribax With Votes, Not Confidence

More from Archive

More from Pink Sheet

FDA Panel Supports J&J’s Antibiotic Doribax With Votes, Not Confidence