Staff Writers

The FDA again delayed its review of cephalosporin antibiotic ceftobiprole due to unresolved "data integrity issues," an unexpected setback for Swiss biotech Basilea Pharmaceutica Ltd. and co-development partner Johnson & Johnson Pharmaceutical Research & Development LLC.

Until the data integrity issues are resolved, the FDA said it cannot review the clinical data for ceftobiprole that was submitted previously. The J&J subsidiary and Basilea have 12 months to iron out the problems identified by the agency or else the application will be deemed to have been withdrawn.

Basilea said it remains confident in the safety and efficacy of the product and vowed to work with the J&J unit and U.S. regulators to resolve the outstanding issues as soon as possible.

Shares in the Basel-Switzerland-based company fell 36 percent on news of the FDA's rejection of the skin infection treatment. The stock (ZURICH:BSLN) closed Wednesday at CHF96 (US$79.79), down CHF55.50.

The latest snag is the second to hit the ceftobiprole review in the U.S. In March, the FDA issued an approvable letter and J&JPRD recently responded to that letter. (See BioWorld Today, March 19, 2008.)

In its latest complete response letter, the FDA raised issues about the monitoring of three study sites, company spokesman Greg Panico said. Although the study site issues "may be similar" to those previously cited in the approvable letter, the most recent letter cites a different set of issues, Panico said.

Asked whether or not the company agrees with the most recent concerns raised by the agency, Panico said it was premature to respond since the company has not yet conducted its analysis.

As a result of FDA inspections, the agency has suggested that the Raritan, N.J.-based drugmaker conduct additional clinical site audits, according to the company. In addition, the FDA has requested information on J&JPRD's clinical quality assurance programs and asked for a new audit plan that also addresses deficiencies in contract research organization monitoring.

"In essence, what we know [is] there was an indication that clinical data from three sites could not be used to support the NDA [new drug application]," Anthony Man, CEO of Basilea told analysts on a conference call Wednesday.

More than 120 study sites were involved in each of the two Phase III trials that formed the basis of the NDA submission for ceftobiprole in complicated skin and skin structure infections (cSSSI), including diabetic foot infections. "The FDA undertook an audit on a minimal number of sites to reach their conclusion on this. The actual number we cannot divulge," Man said.

Initial analyses indicated that excluding the three sites identified by the FDA as having monitoring deficiencies "would have no impact on the overall outcome of the data package," Man said. Moreover, the study design, which involved more than 1,600 patients and a noninferiority margin of 10 percent, has not been called into question, he said.

"That does bring up the question, What is going on?" analyst Andrew Weiss at Bank Vontobel AG in Zurich told BioWorld Today. "It puts a bit more risk on it. Then again, the Europeans have reviewed the whole thing and the CHMP says, 'Go,'" he said, referring to the unanimous positive opinion on the dossier issued last week by the London-based European Medicines Agency's Committee for Medicinal Products for Human Use.

Ceftobiprole is a treatment for resistant bacterial infections that has been approved or has been recommended for approval in the European Union and other several other territories.

In two large Phase III trials, the product has been shown to work by achieving what's known as a "noninferiority" endpoint. That means ceftobiprole was shown to be no worse than the other antibiotics used in those trials: ceftazidime plus vancomycin.

Noninferiority is not an issue in the FDA review of ceftobiprole, Panico confirmed.

But the agency may be acting more cautiously now as a result of problems it has seen with previous antibiotics. Perhaps that explains the string of bad news in the space lately.

Last week, the FDA's scientific advisers said Reinach, Switzerland-based Arpida AG failed to show in clinical trials that its experimental antibiotic, iclaprim, was safe and effective for treating patients with cSSSIs.

The scientific panel also said it was not convinced by data from Cambridge, Mass.-based Targanta Therapeutics Corp. about the safety and effectiveness of its antibiotic, oritavancin, for the same indication, sending that company's shares falling 82.2 percent in November. (See BioWorld Today, Nov. 21, 2008.)

The FDA panel, however, supported approval of South San Francisco-based Theravance Inc.'s injectable antibiotic telavancin, despite the potential for the drug to cause birth defects and kidney problems. (See BioWorld Today, Nov. 20, 2008.)

Yet, the agency's response to Basilea came as "a big surprise," Weiss said, as he had been expecting no action until the agency action date set for early next year. "We have scientists that said, 'This is good - we want it, and we don't think it's messy data,'" he said.

At least one antibiotic, Ketek (telithromycin), was approved despite problems at some sites, but it was later restricted just to community acquired pneumonia for safety reasons.

News of fraudulent Ketek study data came to light in 2006 and was the subject of a congressional probe, sparking what some see as greater FDA scrutiny of antibiotics. Since that time, the FDA has set new guidelines for antibiotic trials, which traditionally have been tested using the noninferiority approach.

In October 2007, the FDA recommended that antibiotics for certain infections use superiority trials. The FDA's experience with telithromycin "probably provided the impetus for this change," according to an editorial that appeared in the April 2008 edition of The Lancet Infectious Diseases.

Factive (gemifloxacin, Oscient Pharmaceuticals Corp.) for sinusitis, and faropenem (Replidyne Inc.) for bronchitis, both of which had been tested in noninferiority trials, failed to gain FDA approval. (See BioWorld Today, Sept. 13, 2006, and Oct. 24, 2006.)

Published  December 1, 2008