FDA’s Diabetes Guidance On CV Risk Will Get First Test With Pending NDAs

FDA's flexibility in applying its new guidance for cardiovascular safety of diabetes drugs will be tested in coming months as it reviews a handful of pending NDAs and enters into discussions about other drugs in late stages of development.

The ¹ guidance, outlined in a briefing Dec. 18, requires a sweeping safety review for products that:"eat type 2 diabetes, so even drugs at a late stage of review may be sent back to clinical if they can't pass statistical muster" (see chart: ² Type 2s That Might Get Through: Diabetes NDAs Pending At FDA).

Prior to publication of the new recommendations, FDA issued more than 100 letters related to applications affected by the changes. According to the guidance, the events studied should include cardiovascular mortality, myocardial infarction and stroke, and can include hospitalization for acute coronary syndrome and urgent revascularization procedures (³ (Also see "FDA Diabetes Guidance Encourages Enrollment Of Sicker Patients" - Pink Sheet, 17 Dec, 2008.)).

Higher-Risk Patients Must Be Taken Into Account

FDA has stressed the importance of studying higher risk patients who are likely to use anti-diabetic medication, such as the elderly and those with relatively advanced disease.

The emphasis on patients with co-morbidities shows FDA's increasing interest in ensuring that pre-market safety evaluations better resemble the patient population that might use a product after launch. That philosophy is also reflected in the agency's push to get firms to enroll hepato-compromised patients in trials to help test drugs' potential for liver injury (⁴ (Also see "Drug-Induced Liver Injury Workshop Offers More Questions Than Answers" - Pink Sheet, 7 Apr, 2008.), p. 22).

Phase III trial data must be available for at least 2,500 subjects exposed to the investigational product with at least 1,300 to 1,500 of these subjects exposed to the investigational product for one year or more and at least 300 to 500 subjects exposed to the investigational product for 18 months or more. Events would need to be assessed by an independent cardiovascular events committee.

The guidance allows a meta-analysis to rule out an 80 percent increased risk of CV events. CV outcome trials post-approval may not be required if a drug shows a CV event risk of less than 30 percent (risk ratio of less than 1.3) but may be required for a CV risk ratio between 1.3 and 1.8.

Trials that have already been conducted will not be subject to the prospective requirements established in the guidance, but the sponsors must still meet the goal in terms of risk ratio. For those that can't, it may be "back to the drawing board," Mary Parks, director of CDER's Division of Metabolism and Endocrinology Products, said in an interview.

Parks explained that there will be one standard for emerging drugs and that the agency does not intend to make the approval process easier for drugs in later stages of development.

"In some ways, it may be harder for those already in Phase II and III," Parks said, noting that the studies were not prospectively designed with the guidance in mind and the data may not be robust enough to pass.

Cleveland Clinic cardiologist Steven Nissen described the diabetes guidance as "moderate" in a Dec. 17 interview, arguing that it was "unlikely to greatly increase time for drug development."

Nissen was the driving force behind an advisory committee recommendation that FDA detail the statistical criteria used in the guidance, and conducted the initial meta-analysis of GlaxoSmithKline's Avandia (rosiglitazone) that touched-off the wave of concern about the safety of diabetes products (⁵ , p. 6).

Guidance Could Be Expensive To Implement

But while Nissen is upbeat, others say the guidelines could prove very onerous.

"Generally speaking, it's really hard to [deal with] new requirements that completely change how you do drug development," said Mary Pendergast, a former associate FDA commissioner and president of Pendergast Consulting in Washington D.C. "It could be very, very difficult for some companies to do something like this without phasing it in or getting a heads up."

Companies that happened to have run Phase III trials enriched with people likely to get cardiac events could fare well in the meta-analysis, Pendergast added.

If the guidance is applied strictly, developers will need to test a sicker population longer to reach a minimum number of cardiovascular events (2 percent) to do the necessary calculations.

"That's what kills you in terms of costs and time of development," said Timothy Morris, chief financial officer of Vivus which has obesity drug Qnexa in Phase III for diabetes (⁶ (Also see "Vivus’ Qnexa Combo For Obesity Needs FDA ‘Persuasion’ To Take First-Line In Diabetes" - Pink Sheet, 15 Dec, 2008.)).

To some extent, the cardiovascular outcome requirement may be satisfied with events from efficacy studies but the patient populations may not entirely match, Morris said.

It might be necessary to test 4,000 to 6,000 patients in order to satisfy the cardiovascular outcome requirement of the new guidance, he added. Costs for diabetes trials tend to be high at about $30,000 per patient and long-term follow up will incur additional expenses.

"This is going to be a hard risk to assess and will add a lot of cost and time to drug development," said Scott Gottlieb, resident fellow at the American Enterprise Institute and former deputy commissioner of medical and scientific affairs at FDA.

"You are looking for small risks that only become manifest after prolonged use, so it may often take big, long-term studies to uncover these things," he said.

Rodman & Renshaw analyst Jason Butler noted that "these trials could extend the time of development of new anti-diabetic drugs and lower the profitability of a drug once it reaches the market should a large post-marketing cardiovascular trial be necessary."

Ultimately there may be many drugs - both old drugs that have long ago gone generic and new drugs that have latent heart risks, Gottlieb said. "I think it's prudent for patients to assume that any drug could have some small, long-term risks that are hard to uncover in any reasonably-sized trial," he said.

For marketed products, the guidance only prolongs the uncertainly about what FDA's safety standards will be. A higher barrier to market entry could give Avandia, even with its new black box, and other drugs, longer sales life without additional competitors (⁷ 'The RPM Report,' October 2008). However, FDA notes that although products already on the market are not affected by the guidance, they will be targeted in separate guidance in the future.

"There is no definitive evidence that any currently marketed anti-diabetic drug increases the risk of cardiovascular disease," Parks said. "We have extensive experience with these drugs and remain confident in their safety profile when the therapies are taken as prescribed," she said in explaining why the new standards do not apply to approved products.

- Emily Hayes ([email protected])