FDA's approach toward non-inferiority trials for cancer drugs is "excessively rigid," especially for targeted therapies and agents that seek to demonstrate improved safety, Bristol-Myers Squibb said

"The recent FDA position on non-inferiority trials has raised concerns among sponsors, investigators and patients," ^FDL-1 Bristol's comments on an agency draft guidance on cancer drug clinical endpoints indicate.

"The development of individualized treatment that could improve tolerance but not necessarily affect activity might be penalized by an excessively rigid approach to non-inferiority, and particularly so as the size of populations to be studied will be smaller, and not larger, than the historical norm."

FDA's "Clinical Trial Endpoints for the Approval of Cancer Drugs and Biologics" ^FDL-2 draft guidance issued in April states that "the goal of non-inferiority trials is to demonstrate the effectiveness of a new drug showing that it is not less effective, by a predefined amount, than standard regimen known to have the effect being investigated."

"In most cases the NI margin is not set at the control drug's full effect, but at some fraction of it (e.g. 50%), so that the study seeks to show at least 50% of the control drug effect is preserved," the guidance notes.

FDA acknowledges, however, that an NI study against a standard drug associated with a small proven survival benefit can be "prohibitively large."

Non-Inferiority Margin Of 50% Not Realistic

Bristol called the non-inferiority margin of 50% retention of control drug effect "not realistic for prospectively conducted trials in the era of targeted, individualized therapy."

"More realistic estimates of the effect of the control regimen should be obtained from consensus estimates of the absolute control effect relative to no treatment. Preservation of a fraction of this larger effect would lead to more realistically sized prospective clinical trials," Bristol said.

Such alternatives for determining non-inferiority margins, "or use of an absolute margin (e.g. .8 for hazard ratio) would allow for prospective clinical trials to address important scientific questions for new compounds that demonstrate important safety advantages over current standards of care."

Bristol markets the targeted cancer agent Erbitux (cetuximab), co-developed with ImClone. The company is studying the epidermal growth factor inhibitor, currently approved for colorectal cancer, in various other cancers.

FDA has said that it is considering less conservative approaches to non-inferiority trials, including narrowing the confidence intervals for certain measures, which could reduce the number of patients required for enrollment in a trial (¹ Also see "FDA Considering “Less Conservative” Approaches To Non-Inferiority Trials" - Pink Sheet, 27 June, 2005. p. 10)

Office of Medical Policy Director Robert Temple suggested that the agency may be somewhat more forgiving in establishing non-inferiority intervals for drugs that are pharmacologically similar (see ² Also see "Drug Similarities Provide Evidence Latitude, Temple Says; Diovan Example?" - Pink Sheet, 15 August, 2005.).

Advisory Cmte. On Non-Inferiority Trials Urged

Bristol suggested FDA convene a workshop or an Oncologic Drugs Advisory Committee meeting to develop a separate guidance for non-inferiority issues.

The company also suggested that in such a guidance "FDA indicate openness to multiple methods of non-inferiority trial designs...particularly in areas where new therapies offer safety advantages."

FDA Oncology Drug Products Division Director Richard Pazdur expressed concerns about non-inferiority trials at a 2004 advisory committee meeting.

At an advisory committee review of Lilly's Alimta in July 2004, Pazdur said superiority trials should be favored over non-inferiority trials.

At the meeting, the advisory committee found that Alimta's favorable toxicity profile outweighed uncertain survival effects from a non-inferiority trial in the second-line treatment of non-small cell lung cancer (³ Also see "Oncology Drug Development Not Advanced By Non-Inferiority Trials – Pazdur" - Pink Sheet, 2 August, 2004. p. 38).

^FDL-3 Millennium's comments on the guidance state that FDA should clarify whether the non-inferiority margin "is 50% of the point estimate of the control drug effect or 50% of the lower or upper limit of the confidence interval of the control drug effect. The latter has been generally accepted," the company said.

Since non-inferiority trials rely on historical data to establish the expected treatment effect size, Millennium recommended that "FDA encourage the development and use of registries for retrospective data collection in order to provide usable historical data."

^FDL-4 Merck's comments on the guidance address a statement by FDA that "a single NI trial seldom provides sufficient evidence of efficacy to support approval."

"We believe that the adequacy of the evidence should be driven by the indication and suggest that this sentence be modified to reflect that, in some cases, a single study based on NI should be acceptable, especially if the endpoint is overall survival and the value of the comparator agent is well established," Merck said.

^FDL-5 AstraZeneca's comments on the guidance suggest that FDA should make clear that only one study may be necessary to gain approval of a cancer agent.

Single Trials And Single Arms Suggested

AstraZeneca recommended that the draft guidance note that the FDA Modernization Act "established that data from one well-controlled clinical trial, together with confirmatory evidence obtained either before or after the trial, are sufficient to establish effectiveness."

Single arm studies should receive greater emphasis in the draft guidance, several companies said.

Merck noted that the FDA draft guidance calls for controlled oncology trials in advanced disease using add-on design to standard therapy, while European guidelines allow for more single arm trials.

The Committee on Human Medicinal Products draft guideline "is calling for controlled oncology trials in less advanced disease against a standard comparator with support for single arm trials in more advanced disease," Merck observed. The company urged FDA to harmonize its guidance with the CHMP document.

Bristol noted that many accelerated approval drugs have been approved on the basis of single arm studies followed by clinical benefit shown in full randomized trials after approval.

"We propose increased emphasis on the utility of single arm trials in accelerated approval, especially for solid tumors."

Several comments found that the draft guidance overemphasized "overall survival" as an endpoint.

Survival Overemphasized, QoL Underappreciated

^FDL-6 Abbott's comments encourage "reconsideration of the emphasis on survival as a gold standard" due to the difficulty of performing studies large enough or long enough to detect survival improvement, the confounding effects of subsequent cancer therapy and concern that the drug may affect only a small subpopulation.

The American Society of Clinical Oncology's comments and ^FDL-7 Lilly's comments urge FDA to consider Health Related Quality of Life (HRQoL) as an acceptable endpoint in oncology studies. "To date FDA has not accepted as an appropriate endpoint 'broader measures of HRQoL,'" ^FDL-8 ASCO's comments state.

ASCO urged FDA to consider the "widely-accepted" Functional Assessment of Cancer Therapy-General (FACT-G) scale as a measure of HRQoL. FDA should include patient-reported outcome measures in cancer drug labeling, even if the agency is not prepared to accept them as endpoints, ASCO said.

FDA is preparing a separate guidance on "Patient-Reported Outcome Measures: Use in Medical Product Development to Support Claims," that will discuss quality of life. The agency expects to issue the guidance in the summer of 2005.

With SPA, No AC Needed, Millennium Says

"A general statement should be made that if there is an agreement with the FDA on the protocol design endpoints through a Special Protocol assessment (SPA) that this should be sufficient for approval (accelerated or full approval) and not be subject to an advisory committee review," Millennium said.

"Committee 'votes' have fluctuated over time as noted in the guidance document," the company said. "Approvals generally should not be decided by an advisory committee post-submission after the endpoints have been achieved with full agreement with the agency," Millennium said.

Millennium's Velcade was approved in 2003 for third-line treatment of multiple myeloma without full advisory committee review despite concerns about response rate as a surrogate endpoint. Velcade was approved for second-line treatment in April.

FDA Approach To Non-Inferiority Trials For Oncologics Too Rigid, Bristol Says

More from Archive

More from Pink Sheet