The FDA’s proposed move beyond a narrow focus on ischemic cardiovascular risk in type 2 diabetes trial requirements was largely met with acceptance as inevitable, although some cardiologists worried about potential for a drop-off in evidence on hard outcomes.
Draft guidance issued March 9 would withdraw the 2008 policy under which manufacturers must perform large cardiovascular outcomes trials (CVOTs) to rule out a 30% relative increase in cardiovascular risk for drugs newly approved to treat type 2 diabetes.
While that 2008 guidance had been based on concerns about MI risk with rosiglitazone (Avandia) raised by a meta-analysis and the lack of sufficient data to definitively answer it, William H. Chong, MD, and colleagues at the FDA wrote Wednesday in a New England Journal of Medicine “Perspective” article:
“None of the trials completed since the implementation of this guidance have shown an increased risk of ischemic cardiovascular events; in fact, several such trials have shown a reduction in risk. With this new insight, the FDA has determined that it is no longer necessary to continue narrowly focusing on establishing cardiovascular safety. Rather, it is time to move toward a broader goal of ensuring, to the extent possible, the existence of adequate premarketing safety data to support consideration of the benefits and risks of new glucose-lowering drugs for type 2 diabetes, particularly in vulnerable populations such as older patients and patients with coexisting conditions.”
FDA’s announcement of the proposed change went so far as to call its current safety assessment recommendations “outdated.”
The new plan calls for broad, representative populations of patients in phase III trials with a sufficient number and duration of treatment to rule out a doubling in cardiovascular risk. For instance, it calls for at least 4,000 patient-years of exposure, with at least 1,500 exposed to the new drug for at least 1 year and at least 500 exposed for at least 2 years. Also, the phase III program would need at least 600 participants with established CVD exposed to the new drug and at least 500 with stage 3/4 chronic kidney disease and at least 600 over age 65.
Chong’s group called it a more holistic approach to safety: “In moving away from a narrow focus on evaluating cardiovascular safety, the FDA is moving toward an approach that will yield a robust evaluation of a broad range of safety concerns.”
The endocrinology and metabolic medicine community was initially reluctant to embrace the need for CVOTs, but got a great deal of value out of them, noted Robert Eckel, MD, American Diabetes Association president for medicine and science and past president of the American Heart Association.
Not only was there reassurance of no excess ischemic risk in any of the trials but also event-driven superiority of several GLP-1 receptor agonists and SGLT2 inhibitors for people with diabetes with atherosclerotic cardiovascular disease or high risk of it and for the SGLT2 inhibitors in heart failure with reduced ejection fraction, he pointed out.
Sanjay Kaul, MD, of Cedars-Sinai Medical Center in Los Angeles, argued that continuing to mandate rule-out of unacceptable ischemic CV risk despite evidence to the contrary was “no longer prudent or tenable” with the evidence that has accrued since 2008.
“In my opinion, this guidance will raise, not lower, the evidentiary bar for approval of new diabetes medications as safety will have to be demonstrated pre-approval,” he said. “Overall, this is a step in the right direction.”
An FDA advisory committee that met in 2018 had been somewhat torn on changes to these requirements. They voted 10-9 in favor of continuing to require CVOTs for new type 2 diabetes drugs, although many of the panelists called for streamlining the process.
Marc Sabatine, MD, MPH, of Harvard University and chair of the TIMI Study Group, a member of that FDA advisory panel who had emphasized that observational and registry data would not be enough, said he was fairly satisfied with the FDA’s plan.
“I think it’s better than what existed before 2008,” he told MedPage Today. “The datasets will be smaller than in the 2008 guidance, so there’s some risk of missing safety signals. But I think the bigger issue is that the goalposts have changed in that people are now demanding efficacy signals for their drugs.”
Practically speaking, with so much competition from safe, inexpensive drugs that lower glucose levels, a new drug to market needs a strong efficacy signal for hard endpoints to drive clinicians to adopt it, he said.
“The way to show that is to do large trials, whether the FDA mandates that or not,” he said. “And those trials will have not only efficacy data but also the safety data.”
While the proposal has not yet been finalized, “upfront it appears to be more broad-reaching for safety signals and more economic for pharmaceutical companies, and likely the FDA, too,” Eckel said. But he wasn’t as optimistic that pharma would continue shelling out for CVOTs, saying, “the value of the RCTs for superiority will be lost going forward.”
Steven E. Nissen, MD, chair of cardiovascular medicine at the Cleveland Clinic, also expressed concern.
“While I agree that there is probably a need to modify the standards for approval of diabetes drugs established in 2008, I am concerned that the current proposal goes too far in relaxing the regulatory requirements for approval,” he told MedPage Today.
“It is not correct that the trials completed since 2008 have failed to show an increased risk of adverse cardiovascular outcomes,” he added. “Saxagliptin increased congestive heart failure and canagliflozin increased the risk of amputation, both of which might not have been detected without a large outcome trial. In many ways, the new guidance may prove less relevant.”
But he, too, agreed that “marketing any new diabetes drug will require a cardiovascular outcome trial, since payers now expect favorable effects.”
Kaul highlighted the role of prescribers in making sure pharmaceutical companies don’t take the “path of least resistance” in doing the bare minimum by studying low-risk patients over short periods of time.
“Clinicians should not prescribe, and payers should not reimburse a new type 2 drug that does not have evidence of reducing CV or renal risk or other complications of diabetes in longer outcomes trials,” he said. “It is incumbent upon us to see that does not happen, hopefully in partnership with the regulators.”
The public comment period for the draft guidance ended on June 8, but there is no projected date for finalizing the document, according to an FDA press officer.
Chong’s group disclosed no relevant conflicts of interest.
Sabatine disclosed grant support through Brigham and Women’s Hospital from Amgen, Anthos Therapeutics, AstraZeneca, Daiichi-Sankyo, Eisai, Intarcia, The Medicines Company, MedImmune, Merck, Novartis, and Pfizer. He disclosed consulting for Althera, Amgen, Anthos Therapeutics, AstraZeneca, Bristol-Myers Squibb, Intarcia, and Merck.
Kaul has consulted for companies that have conducted CVOTs in diabetes, including Boehringer Ingelheim, AstraZeneca, Novo Nordisk, and Sanofi.