The new endpoint proposed in FDA’s recent acute bacterial skin and skin structure infections guidance – cessation of lesion spread – needs further refinement to be a reliable assessment of antibiotic efficacy, according to members of the agency’s Anti-Infective Drugs Advisory Committee.

During the committee’s Sept. 7 review of Forest Laboratories’ ceftaroline for complicated skin and skin structure infections (the previous nomenclature for ABSSSI), panel members said simply measuring the length and width of lesions to assess a drug’s efficacy could be a complicated endeavor, making measurements difficult to standardize or objectively confirm.

Concerns over reliability and variations in lesion measurements in the ceftaroline clinical trials did not derail the committee’s enthusiasm for the beta lactam cephalosporin, though. The committee voted 18-0 that safety and efficacy had been demonstrated in cSSSI (Also see "Forest's Ceftaroline Sets Standard For Skin Infection Drugs Under New FDA Guidelines" - Pink Sheet, 7 September, 2010.).

Earlier in the day, the committee voted 21-0 for a ceftaroline indication in community-acquired bacterial pneumonia (Also see "Forest Must Resolve Breakpoint Issues On Road To Ceftaroline Approval" - Pink Sheet, 13 September, 2010.).

Lesion measurement is expected to be the key focus of serious skin infection trials going forward. FDA’s draft guidance on ABSSSI, released Aug. 26, recommends cessation of lesion spread and resolution of fever at 48-72 hours after enrollment as the primary efficacy endpoint in such studies (Also see "FDA Skin Infections Guidance Spurs Debate On Endpoint Relevance" - Pink Sheet, 6 September, 2010.).

This is a significant change from the primary endpoint that has traditionally been used for drugs to treat serious skin infections, previously referred to as cSSSI, which was resolution of signs and symptoms at the test-of-cure visit 7-14 days after completion of treatment. The guidance also proposes that all patients enrolled in ABSSSI studies have a minimum degree of cellulitis.

Comments on the draft guidance are expected to question the clinical relevance of the new earlier endpoint and may suggest that other symptoms, such as pain reduction, be included as well.

Phase II Validation

The guidance provides little instruction for how lesion size should be measured, other than to say the 72-hour visit should include “evaluation and measurements of the infection site (e.g., objective characterization of cessation of (or absolute and percent reduction in length, width and area from baseline) the amount of redness, edema, and/or induration).”

Instead, the guidance advises sponsors to perform additional work in Phase II to develop how this endpoint will be measured and evaluate its performance. This research should incorporate assessment of the accuracy and reliability of the proposed lesion size measurement method to help identify the best way to minimize potential errors.

A working group of the Foundation of the National Institutes of Health’s Biomarkers Consortium has been focused on developing endpoints for clinical trials of drugs to treat ABSSSI and CABP.

While the group’s efforts in the skin infections area have initially focused on what the primary endpoint in ABSSSI trials should be and when it should be assessed, the issue of how lesions should be measured is expected to be a consideration, George Talbot, the group’s co-chair, said in an interview. Talbot is an industry consultant who formerly served as Cerexa’s chief medical officer. Forest picked up ceftaroline through its 2007 acquisition of Cerexa.

Ceftaroline Case Study

Although the ceftaroline clinical trials were not conducted under the provisions of FDA’s new ABSSSI guidance, the advisory committee review provided an opportunity for panel members to weigh in on the proposed new primary endpoint.

The ceftaroline review provided an opportunity for panel members to weigh in on the proposed new primary endpoint

Conducted in 2007, the two pivotal trials demonstrated ceftaroline was non-inferior to vancomycin plus aztreonam pursuant to a pre-specified NI margin of 10%. The primary endpoint was total resolution of signs and symptoms of infection, or improvement to the extent that antibiotics were no longer required, at the test-of-cure visit.

While ceftaroline met the efficacy markers per the pivotal trial protocols, which were designed under FDA's previous recommendations, the agency also evaluated the results under the new standards laid out in the draft guidance. In briefing documents and at the panel review, FDA presented a series of sensitivity analyses showing that ceftaroline satisfied the new efficacy criteria, thereby suggesting a path toward approval amid the agency's evolving regulatory standards (Also see "Forest's Ceftaroline Data Robust Even Under New Antibiotic Efficacy Standards, FDA Says" - Pink Sheet, 2 September, 2010.).

FDA's analyses defined a responder at Day 3 of treatment as cessation of spread of the lesion, measuring both length and width, and absence or resolution of fever. The agency examined a modified intent-to-treat population of subjects with lesion sizes of at least 75 cm. Major abscesses were included only if they were accompanied by surrounding erythema of at least 5 cm.

The agency's sensitivity analysis population included 58% of the patients in the original MITT population. FDA also analyzed rates for reduction in lesion size, consistency of responder rates with assessment of clinical cure at the end of therapy and use of prior antibiotics.

The key sensitivity analysis showed ceftaroline's non-inferiority to comparator fell within a 3.6 percent NI margin for both studies. The efficacy findings were robust with regard to the percentage of lesion size reduction, earlier and later treatment evaluation points and the individual components of cessation of lesion spread and fever.

Measurement Limitations

FDA statistical reviewer Christopher Kadoorie cited numerous limitations to the agency's sensitivity analyses. In particular, the trials were not designed to assess treatment difference in lesion sizes, and measurements were not standardized or verified to ensure accuracy.

Committee members were concerned with how lesions, which are often irregularly shaped, could be accurately and reproducibly measured in future clinical programs

Concern about a lack of standardized measurements struck a chord with many committee members, who were concerned with how lesions, which are often irregularly shaped, could be accurately and reproducibly measured in future clinical programs. Their comments also suggest the difficulties sponsors will face as they attempt to validate a method for lesion measurement.

“I can’t recall the last time I saw a rectangular cellulitis,” temporary voting member James Steckelberg, Mayo Clinic, said to laughter in the meeting room. As cellulitis resolves, the margins of the infection are not always clear, and measuring the length and width of the affected area can be a subjective assessment, particularly when dealing with curved body surfaces, he said.

“Thankfully in your analysis you didn’t come up with any real surprises that challenged the primary endpoint, but I raised these questions in the chance that this might be used as a precedent in the future where such issues really could be problematic,” he said to FDA. “I find the concept of lesion size very attractive, but I think we have a ways to go yet.”

Several committee members suggested a role for digital photography, perhaps coupled with computer analysis, in assessing lesion size.

However, in an interview Talbot cited potential problems with photographic assessments, including variations in lighting, the photographer’s expertise and difficulty photographing lesions on certain parts of the anatomy.

Evaluation of lesion size is only one component to assessing improvement of a patient with cellulitis, temporary voting member Bernhard Wiedermann, Children’s National Medical Center, said. "It might be possible to evaluate the degree of erythema, temperature of the area, swelling and things like that, so we ought to take into account those other signs of inflammation, not just size. All of them have pitfalls, of course."

Vancomycin Monitoring

The committee’s discussions on vancomycin therapeutic dose monitoring also could impact the design of future clinical trials for ABSSSI.

The initial vancomycin dosing regimen used in the ceftaroline trials was consistent with the drug’s labeling and called for 1 g every 12 hours administered as one-hour I.V. infusions. Subsequent doses could be decreased for subjects with moderate renal impairment or increased for subjects who weighed 90 kg or more, according to Forest’s briefing documents for the meeting.

Cerexa President Dirk Thye said unblinded pharmacists at treatment sites were allowed to use local prescribing guidelines to change the vancomycin dose, but there was no routine monitoring of vancomycin serum levels.

Committee members questioned whether weight-based vancomycin dosing would have been more appropriate and wondered if potential underdosing may have reduced the comparator drug’s efficacy in the ceftaroline studies.

“Because we didn’t collect serum levels of vancomycin, it was difficult to do any detailed analyses of the effects of vancomycin dosing on outcome, but I think one point to keep in mind is the vancomycin arm did very well, both with respect to the primary endpoint at the test-of-cure visit and at the exploratory earlier timepoint in the FDA analysis,” Thye said. “With respect to the cure rates, they were similar to what has been seen in other contemporary trials … so the vancomycin dose appears to have worked.”

Nevertheless, some panelists said that if vancomycin is to be used as a comparator in future studies, sponsors should find a way to monitor serum levels in a manner that does not break the study’s blind.

“I think just in terms of study design, if that’s going to be your comparator you want to make sure that people are adequately dosed,” committee member Kemper Alston, University of Vermont, said, recommending weight-based dosing and serum level monitoring in future studies.

“Recognizing there are study difficulties with blinding and so forth, I think that there needs to be a way to have therapeutic drug monitoring for vancomycin when it’s used as a comparator,” said temporary voting member Jan Patterson, University of Texas Health Science Center at San Antonio.

By Sue Sutter

FDA's New Skin Infection Endpoint Needs Refinement, Advisory Panel Says

More from Archive

More from Pink Sheet