FDA committee narrowly supports continued CV outcomes trials for diabetes drugs

An FDA advisory committee voted 10-9 Thursday in favor of supporting continued agency- recommended cardiovascular outcomes trials to demonstrate safety for all type 2 diabetes drugs, with most committee members recommending modifications to the 2008 guidance after a decade of data and no indication of increased CV risk.

The members of the Endocrinologic and Metabolic Drugs Advisory Committee (EMDAC), although nearly evenly split in their vote on whether to require CV outcomes trials for all new type 2 diabetes therapies regardless of the presence or absence of a CV risk signal during development, seemed to express similar sentiments in a discussion after the vote. Most called for modifications to the 2008 FDA-issued guidance, including broadening the population recruited for such trials and the inclusion of endpoints beyond three-point major adverse cardiac events.

“If you turn off the screen so you don’t see how people voted, everybody said exactly the same thing,” George Grunberger, MD, FACP, FACE, chairman of the Grunberger Diabetes Institute in Bloomfield Hills, Michigan, said after his “no” vote. “I voted ‘no’ not to get rid of the guidance. I voted ‘no’ because what made me nervous was [the question] says [conduct a trial] regardless of a CV safety signal. To me, that makes no sense.”

Grunberger said that as new antidiabetes therapies come through the pipeline, phase 2 and phase 3 trials should be broadened, with additional endpoints considered, including heart failure.

“We need a strong message that we need a careful assessment of the efficacy and safety of these drugs,” Yves Rosenberg, MD, MPH, chief of the atherothrombosis and coronary artery disease branch of the National Heart, Lung, and Blood Institute of the NIH, said after his “yes” vote. “However, excluding [CV] risk is not necessarily a requirement. There should be a more flexible, pragmatic approach to the evaluation of risk early on.”

The FDA issued guidance in 2008 to ensure that new antidiabetes therapies were not associated with an unacceptable increase in CV risk. The recommendations in the guidance were applied to all new drug products intended to treat type 2 diabetes, irrespective of whether a signal of concern was identified in the development program.

A decade after that guidance was issued — with no excess CV risk demonstrated in any study — the EMDAC convened a 2-day meeting to ask members whether every new diabetes drug should continue to go through the rigorous process of a formal CV outcomes trial.

‘No substitution’ for CV trials

Kenn e th D. Burman , MD, chief of the endocrine section at Medstar Washington Hospital Center, who voted “yes,” echoed the sentiments of several committee members when he noted that there is “no substitution” for a rigorous randomized controlled trial to assess CV safety. However, changes to a trial design can be made that make sense, he said, including the adoption of specific endpoints beyond three-point major adverse cardiac events, and eliminating an independent CV endpoints committee for the prospective adjudication of events.

“I agree with these appropriate modifications, which will hopefully decrease cost without detracting from the high standards expected for such a trial,” Burman said. “The CV safety of new agents is, unequivocally, the highest priority.”

Thomas J. W ang , MD, professor of medicine and physician-in-chief at the Vanderbilt Heart and Vascular Institute in Nashville, Tennessee, said CV outcomes trials have had limitations, including high cost and applicability to the broader type 2 diabetes population, but they remain an important determinant of safety.

“Rather than addressing the imperfections in the system by eliminating guidance, it seems preferable to focus on how the design of these trials and the endpoints might be improved,” Wang said.

Broader assessment needed

Still other committee members said 10 years of evidence has demonstrated that requiring CV outcomes trials for all antidiabetes therapies, regardless of CV risk signals observed during development, is unnecessary.

“I am not convinced that this two-stage approach is the right approach.” Susan S. Ellenberg, PhD, professor of biostatistics, epidemiology and informatics at the Perelman School of Medicine at the University of Pennsylvania, said after her “no” vote. “I would like to see more information developed premarket and to think about whether safety signals are seen there. I think the focus on CV safety is too narrow. A broader assessment of both safety and efficacy is needed.

“By no means is my ‘no’ vote indicating that we should go back to before 2008,” Ellenberg said. “But there are other outcomes that need to be factored into whether a drug can provide more benefits than harms to the population.”

David C. Robbins, MD, professor of medicine and director of the University of Kansas Diabetes Institute at the University of Kansas School of Medicine, agreed.

“The agency and industry should be given increased flexibility, given the knowledge we have achieved in the last 10 years, and the tools that are available to answer these questions,” Robbins said.

Mi chael Blaha, MD, MPH, associate professor of cardiology and epidemiology and director of clinical research at the Johns Hopkins Ciccarone Center for the Prevention of Heart Disease, said the CV outcomes trials have already influenced clinical guidelines. In a new joint consensus statement issued in September from the American Diabetes Association and the European Association for the Study of Diabetes, the organizations recommended new medications shown to reduce CV risk, including SGLT2 inhibitors and GLP-1 receptor agonists, be introduced earlier for high-risk patients, based on results from the CV outcomes trials.

“If a company wants a foothold [in the market] now, they will need to show a CV benefit ... but that will be their choice,” said Blaha, who voted “no” on the question.

The committee’s vote followed a full day of presentations on Wednesday from experts who spoke of the benefits and drawbacks of the CV outcomes trials since FDA’s guidance was first issued in 2008. More than 190,000 patients with type 2 diabetes have participated in CV outcomes trials in the decade since.

Researchers have gleaned valuable information on adverse events, experts noted, including a signal for congestive heart failure with DPP-IV inhibitors and an observed increased risk for lower-limb amputations with SGLT2 inhibitors. Still, experts cautioned, the trials have demonstrated an effect on secondary prevention, not primary prevention, and more than 70% of those with diabetes do not have CVD.

The advisory panel’s recommendations are nonbinding, although the FDA often follows its suggestions. – by Regina Schaffer

For more information:

FDA briefing materials. Available at: https://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/EndocrinologicandMetabolicDrugsAdvisoryCommittee/UCM623913.pdf.