Sensitivity analyses may help smooth the approval pathway for ceftaroline fosamil under the Food and Drug Administration's evolving efficacy requirements for skin infections and community-acquired bacterial pneumonia, even though the antibiotic's pivotal data were developed under the agency's old standards.
Ceftaroline (Forest Laboratories) was shown to be noninferior to comparator agents pursuant to the company's prespecified efficacy analyses as well as sensitivity analyses applying the FDA's new efficacy criteria, according to agency briefing materials released ahead of the agency's Anti-Infective Drugs Advisory Committee's Sept. 7 review of the beta lactam cephalosporin.
If endorsed by the committee and approved by the FDA, ceftaroline could become a poster child for how to survive the "shifting goalposts" created by the agency’s evolution in antibiotic development standards – the message being if the data are truly robust, sensitivity analyses can get you through the new efficacy hurdles.
Forest acquired ceftaroline through its January 2007 acquisition of Cerexa, which in-licensed rights from Takeda. The phase III trials for complicated skin and skin structure infections (cSSSI) and community-acquired bacterial pneumonia (CABP) began in 2007, prior to FDA-initiated public discussions about use of noninferiority trials for these indications.
New Skin Infections Guidance
In the case of the cSSSIs, the FDA issued new draft guidance on Aug. 26 that proposes changes in the nomenclature, primary efficacy end point, and study population. The draft forgoes the terms cSSSI and uncomplicated SSSI in favor of acute bacterial SSSI and milder skin infections. The document states that noninferiority trials are acceptable for acute bacterial SSSIs, although it does not set a specific margin that should be used.
During a November 2008 meeting on drug development standards for cSSSI, the Anti-Infective Drugs Advisory Committee generally recommended a 10% margin.
The guidance recommends a new primary efficacy end point based on cessation of lesion spread and resolution of fever at 48-72 hours after enrollment. This is a dramatic change from the end point that has generally been used in cSSSI studies: resolution of signs and symptoms at the test-of-cure visit 7-14 days after completion of treatment. The guidance also recommends that patients with wound infections and major cutaneous abscesses be enrolled only if they have coexisting cellulitis.
Forest's phase III studies tested ceftaroline against vancomycin plus aztreonam in adults with cSSSI requiring initial treatment with an intravenous antimicrobial agent in a hospital or urgent care setting. Cellulitis was present in about 35% of patients. Duration of therapy was 5-14 days.
The primary efficacy end point was clinical response rate at the test-of-cure visit, with clinical cure defined as total resolution of all signs and symptoms or improvement such that further antimicrobial therapy was unnecessary.
Ten Percent Noninferiority Margin Draws No Objections
The FDA's briefing documents raise no objections to the prespecified noninferiority margin of 10% in the cSSSI studies, for which the agency had asked Cerexa to provide supporting justification at the end-of-phase II meeting in October 2006.
In both studies, the 10% margin was met for both the modified intent-to-treat and clinically evaluable populations in the test-of-cure visit 8-15 days after the last dose of study drug, as well as at an end-of-therapy visit, a secondary end point.
FDA reviewers conducted a sensitivity analysis using the draft guidance's early end point of lesion spread cessation at day 3 in a study population that includes only patients with cellulitis. As a result, this reanalyzed population represented only about 56-58% of the originally randomized trial populations, reducing the statistical power to demonstrate noninferiority.
Nevertheless, in both studies the responder rate in the ceftaroline group was higher than in the comparator group. "These findings supported the noninferiority of ceftaroline to vancomycin plus aztreonam for a noninferiority margin of less than 4% for both trials," the agency said.
Additional sensitivity analyses that took into account investigator measurements of lesion size and consistency of clinical response at later timepoints also confirmed ceftaroline's noninferiority. "Overall, findings of noninferiority of ceftaroline to vancomycin plus aztreonam based on key sensitivity analyses of Day 3 responder rates in FDA-MITT [modified intention to treat] subjects appeared to be robust," the document stated.
Justifying the Margin in CABP
The noninferiority margin was the issue in the CABP indication. FDA reviewers were concerned about the absence of historical data to justify the noninferiority margin used in the two phase III trials for that indication.
In studies P903-06 and P903-09, ceftaroline was shown to be noninferior to ceftriaxone on the primary end points – clinical cure rate at the test-of-cure visit in the clinically evaluable and in the modified intent-to-treat efficacy populations – with a prespecified noninferiority margin of 10%. Patients received the study drugs for 5-7 days; the test-of-cure visit was 8-15 days after the last drug dose.