Narrow Vote Calls for Keeping Drug Safety Trials, but Many See Changes

A decade after they began, large cardiovascular outcomes trials (CVOTs) will likely remain part of the process for companies that develop type 2 diabetes (T2D) drugs. But following last week’s narrow vote of an FDA panel, it seems likely the trials may be less cumbersome.

The question asked was, “Should an unacceptable increase in cardiovascular risk be excluded for all new drugs to improve glycemic control in patients with type 2 diabetes, regardless of the presence or absence of a signal for cardiovascular risk in the development program?” The vote of the FDA’s Endocrinologic and Metabolic Drugs Advisory Committee was 10 yes and 9 no, but on both sides of that split there were calls to adjust the 2008 guidance that created the current system.

“We now have eight clinical trials conducted under the guidance,” with none uncovering excess cardiovascular risk, wrote William Chong, MD, acting director of FDA’s division of Metabolism and Endocrinology Products, in a memo to panel members. “Notably, some of the trials have shown a reduced risk for adverse cardiovascular events.”

It’s not so simple, said Steven Nissen, MD, of the Cleveland Clinic, said in an interview with The American Journal of Managed Care^® (AJMC^®) ahead of the meeting. Nissen’s 2007 metanalysis of rosiglitazone in the New England Journal of Medicine, which suggested increased risk of myocardial infarction and cardiovascular death, led to the 2-step process that he recommended, which guides T2D drug development to this day. He received backing from cardiologist Robert Califf, MD, who would go on to become FDA commissioner.

Doing CVOTs would do more than their expressed purpose of telling physicians and the FDA whether T2D drugs increased the risk of major adverse cardiovascular events (MACE), Nissen said. “We also knew that if you did outcomes trials, you would learn things you didn’t otherwise know,” he said.

For example, he said, CVOTs have caused FDA to add information to the label of saxagliptin about heart failure risk, and the trial for canagliflozin, the sodium glucose co-transporter 2 (SGLT2) inhibitor, showed an increased risk for lower limb amputations.

At the same time, other trials have shown that SGLT2 inhibitors appear to have cardiovascular benefits, as does the glucagon-like peptide-1 (GLP-1) receptor agonist (RA) liraglutide. Nissen noted that the CVOTs have shown that GLP-1 RAs don’t cause pancreatitis, which had been a concern.

FDA’s guidance sets one standard for data to allow T2D drugs to reach the market, while these giant randomized controlled trials continue post market. The trials are huge, involving thousands of patients and laboratories around the world to ensure they are adequately powered. As a result, pharmaceutical companies argue they have become too expensive and stifle innovation, while driving up the cost of drugs now on the market.

Nissen said using claims data or other “real-world” observational sources just would not do, and 2 of the 3 experts who testified before FDA agreed: cardiologist Marc Sabatine, MD, MPH, of Harvard and the TIMI Study Group, and Jennifer Green, MD, of the Duke Clinical Research Institute. “Adequately powered and randomized CVOTs of individual agents should continue,” she said. “There is no substitute.”

“I don’t think loosening regulatory standards is good public policy,” Nissen said. And he said it was “unlikely” that doing so would make T2D drugs less expensive.

But Robert Ratner, MD, the former chief scientific and medical officer at the American Diabetes Association (ADA), now with Georgetown University, and several panelists said the knowledge gained from the first decade of CVOTs could allow the FDA to recommend adjustments, which would make premarket trials stronger and more streamlined. Nissen agrees there are approaches that can achieve this goal without going back to pre-2008 standards. Among the ideas offered during the meeting:

• Strengthen the phase 2 and 3 trial requirements and only require CVOTs if a signal is detected.
• Tighten premarket requirements and use registry or observational data to detect safety signals post approval.
• Add premarket requirements beyond MACE for other safety issues, based on findings in the first decade of CVOTs.

CVOTs have caused major professional societies, including the ADA and the American College of Cardiology (ACC), to update their clinical guidelines based on findings about newer classes of drugs. And pharmaceutical companies have gone back to FDA to add cardiovascular indications after their initial approval to gain a leg up in the market.

ADA published a paper earlier this year that discussed the issues surrounding CVOTs; it rejected the idea that the trials were discouraging drug development but acknowledged their contribution to drug costs. ACC’s vice president Richard Kovacs, MD, FACC, similarly cited the FDA panel’s effort to balance competing concerns in an email to AJMC^®:

“Over the course of the last decade, our awareness of the connection between diabetes treatments and cardiovascular outcomes has increased tremendously. Outcomes trials performed since publication of the guidance provided us with a great deal of additional information, but they also raised many new questions. The Committee has appropriately highlighted the need for pausing to consider whether we have the correct information or whether we should be asking different questions and/or using different types of trials. The information generated by the guidance has forced clinicians to face the realities of cost and value consideration, generating winners and losers among the drugs under development.”