September 26, 2013

BARCELONA, Spain — New data from EXAMINE, the cardiovascular-safety trial with the type 2 diabetes drug alogliptin (Nesina, Takeda Pharmaceuticals), has shown no significantly increased risk for heart failure (HF) with use of this dipeptidyl peptidase-4 (DPP-4) inhibitor, or "gliptin."

But numerically, HF occurrences were higher among those taking alogliptin compared with placebo, and part of this post hoc analysis, reported today here at the European Association for the Study of Diabetes (EASD) 2013 Meeting, did involve altering somewhat the definition of prior heart failure, the presenter, cardiologist William B. White, MD, from the Pat and Jim Calhoun Cardiology Center at the University of Connecticut Health Center, Farmington, acknowledged.

And coming on the heels of an unexpected signal for HF hospitalization reported with another drug in this class, saxagliptin (Onglyza, Bristol-Myers Squibb/AstraZeneca), from the SAVOR-TIMI 53 trial at the European Society of Cardiology (ESC) 2013 Congress earlier this month, the findings have caused some concern among both the diabetes and cardiology communities.

In fact, some at the EASD meeting said they would no longer prescribe DPP-4 inhibitors to diabetes patients with preexisting heart failure, at least not until more data are available. Others however, were much less concerned.

Differing Opinions

Naveed Sattar, MD, PhD, professor of metabolic medicine, University of Glasgow, Scotland, was asked to deliver a commentary on both the EXAMINE and SAVOR trials at a special session at the EASD meeting.

"I would be reluctant to give DPP-4 inhibitors to diabetes patients with preexisting heart failure at this time, until we have more data," he pronounced.

And cochair of the session Cornelius J. Tack, MD, from the Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands, told Medscape Medical News: "I think [heart failure] is a class effect, because [EXAMINE] more or less shows the same [as SAVOR]. And intuitively, I think it has to do with the mechanism of action of the drug. No one can prove it yet. Is it big? It's probably not big, but I would like not to have had [this problem]."

However, Dr. Tack added that it doesn't worry him that much, since he hardly ever prescribes this newer drug class for type 2 diabetes, because in the Netherlands, therapy is generally limited to "metformin, sulfonylureas, and insulin, the regular old stuff."

But the other cochair of the session, Jens J. Holst, MD, PhD, from University of Copenhagen, Denmark, said he is not concerned, telling Medscape Medical News that the morning's presentations "would not deter him" from using DPP-4 inhibitors in those with heart failure.

"If anything, I would say this applies to saxagliptin and not to alogliptin; I would keep an eye on saxagliptin. But [heart failure] didn't contribute to the mortality rates at all," he observed. And in any case, he said he doesn't use saxagliptin much: "I like some of the other [gliptins] better."

The 2 trials, Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus — TIMI 53 in 16,492 type 2 diabetes patients with or at high risk for cardiovascular disease and Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care in Patients with Type 2 Diabetes Mellitus and Acute Coronary Syndrome in 5380 participants, were both reported September 2 at the ESC meeting and were simultaneously published online (N Engl J Med 2013;DOI:10.1056/NEJMoa1307684 and DOI:10.1056/NEJMoa1305889).

These were some of the first results to report from these US Food and Drug Administration–mandated cardiovascular-safety studies for diabetes drugs; the findings showed no benefit of either drug in terms of cardiovascular events, but many observers point out that the intent of the trials was simply to rule out cardiovascular harm.

Although this was the case for the most part, there was a significantly increased risk for hospitalization for heart failure with saxagliptin compared with the control group (3.5% vs 2.8%; hazard ratio [HR], 1.27; P = .007).

Presenting these findings again to the EASD audience, cardiologist Deepak Bhatt, MD, MPH, from the Cleveland Clinic, Ohio, explained that although this was "an admittedly unexpected signal" it "seemed to be occurring in isolation." Nevertheless, further analysis is ongoing and more details with regard to the HF outcome in SAVOR will be presented at the American Heart Association meeting in November, he said.

New HF Data With Alogliptin

Meanwhile, Dr. White presented HF data from EXAMINE for the first time at the EASD meeting. The primary composite end point of the trial — CV death, nonfatal myocardial infarction, and nonfatal stroke — was not significantly different between the alogliptin and placebo groups.

The trial did have an exploratory prespecified cardiovascular composite end point that included first occurrence of HF hospitalization, he noted, and on analyzing the components of this separately, there was no increased risk for first occurrence of HF hospitalization with alogliptin vs placebo (HR, 1.07; P = .657).

"But due to the media and questions we have been asked [since the ESC meeting]," the group then performed a post hoc evaluation, he said, "in which we at least removed the confounding of death, and using the 'standard' HF composite that is typically used in HF trials," the researchers found an HR of 1.19 for HF hospitalization for alogliptin vs placebo (P = .220) for this end point, he noted.

And finally, they reanalyzed patients in the trial with a history of heart failure prior to their index (ACS) event, which constituted around 800 people, or 15% of the study population, Dr. White noted. They then reexamined the primary end point of the trial in these patients vs those without a history of chronic heart failure.

In those 419 patients with a history of chronic heart failure taking alogliptin, 18.1% experienced a primary-end-point event, compared with 22.3% of the 404 on placebo (HR, 0.82, P = .203). In the remainder without history of chronic heart failure the HR was 1.01, with "admittedly fewer events" in these patients, said Dr. White (10.1% on alogliptin had a primary-end-point event compared with 10% on placebo).

A doctor from the audience questioned the numbers regarding "history of heart failure," however, noting that in the New England Journal of Medicine paper for EXAMINE, around 750 patients in each group, so a total of around 1500, were listed as having "previous heart failure, but now this current analysis only describes some 400 patients in each arm. Did you change the definition?" he wondered.

Dr. White said the numbers included in the paper for previous heart failure "could have included post-ACS heart failure," so they limited the present analysis to those who had a history of chronic heart failure. This "made more sense than the acuity of an event" that occurs in the few weeks after an ACS event," Dr. White noted.

So pressed, then, to say if they had indeed changed the definition for this current analysis, Dr. White said that they had. "We haven't done" the exact analysis for the "previous heart failure" definition included in the New England Journal of Medicine paper, he said.

If Heart-Failure Signal Is True, It's Weak

Dr. Sattar presented his own mini–meta-analysis of the HF hospitalizations in SAVOR and EXAMINE together, coming up with an HR of 1.24 for the gliptins [95% confidence interval [CI], 1.07–1.45].

He also reminded the audience of the VIVIDD trial in diabetes patients with heart failure with another DPP-4 inhibitor, vildagliptin (Galvus, Novartis), reported earlier this year, which raised some questions.

"The heart-failure hospitalization is something we need to take seriously. The rates [in SAVOR] went up particularly in patients who looked to have prior heart failure [they were in the top quartile of brain natriuretic peptide (BNP)], and we cannot exclude this could be a class effect at the moment," he commented.

However, if there is a true effect of gliptins in patients with heart failure, it is nowhere near as strong as that seen with the "glitazone" class of diabetes drugs, such as rosiglitazone and pioglitazone, which cause fluid retention, he commented in a press conference yesterday.

Diabetologist Itamar Raz, MD, from Hadassah University Hospital, Jerusalem, Israel, co–primary investigator of SAVOR, agreed. "With the DPP-4 inhibitors we don't see any sign of fluid retention… The signal [for heart failure] is much lower than with [the glitazones]. If it is a true signal, we don't yet know the pathogenesis."

Dr. Sattar and Dr. Holst have reported no relevant financial relationships. Dr. Bhatt and Dr. Raz are co–primary investigators of SAVOR. Dr. White is an investigator in EXAMINE. Dr. Tack reports having participated on advisory boards of AstraZeneca/Bristol-Myers Squibb, Johnson & Johnson, MSD, and Takeda and having received research grants from and lectured for Novo Nordisk.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.