LISBON – A new meta-analysis offers some tantalizing hints that DPP-4 inhibitors may offer some level of protection against major cardiovascular events in type 2 diabetes patients.

The analysis, which included data on more than 33,000 patients, found a consistent 31% reduction in major cardiovascular events among those who took dipeptidyl peptidase–4 (DPP-4) inhibitors, compared with those taking placebo or other treatments, Dr. Edoardo Mannucci reported at the annual meeting of the European Association for the Study of Diabetes.

Despite the positive findings, Dr. Mannucci, of the Careggi Teaching Hospital in Florence, Italy, warned against taking the data as gospel.

"The limitations here are clear," he said. "Cardiovascular events were not end points in any of these trials, and were [reported only] as major adverse events. Also, since they were not prespecified end points, there is a possibility that some of the cases could have been misclassified."

Additionally, none of the trials enrolled patients who were at a high risk of cardiovascular events. "These subjects are not very likely to be included in trials with metabolic outcomes, so we really don’t know the effect on those people."

Still, he said, the consistency of the findings "suggests the possibility of some unconventional cardiovascular protective effect that deserves further investigation."

Dr. Mannucci and his colleagues examined 42 phase III trials completed through March 2011 that included one of four DPP-4 inhibitors: saxagliptin, sitagliptin, alogliptin, and vildagliptin. (The latter two are not available in the United States.) Among placebo-controlled trials, 137 events occurred, and in comparator trials, 120 occurred.

Overall, the hazard ratios for a major cardiovascular event were 0.63 for alogliptin, 0.66 for saxagliptin, 0.74 for sitagliptin, and 0.67 for vildagliptin, compared with placebo. All the comparisons were statistically significant, he said.

In the comparator trials, none of the difference in cardiovascular events between drug regimens reached statistical significance, but the trend in favor of the DPP-4 inhibitors was similar to the findings in the placebo-controlled studies, he said.

Dr. Mannucci speculated that the lack of significance could be related to the low incidence of a major cardiovascular event in these trials, ranging from one event in an acarbose study to five each in studies with metformin and sulfonylureas.

Considering all of the 26 placebo-controlled and 16 comparator studies, the odds ratio in favor of the DPP-4 drugs was 0.69, "easily reaching statistical significance," with a P value of .006, Dr. Mannucci said.

When the trials were divided by duration, those less than 52 weeks (30) were significantly in favor of the drugs (OR, 0.62). For those 52 weeks or longer, the result was not significant (OR, 0.83).

Despite the analysis’s limitations, the unexpected observation raises intriguing questions, Dr. Mannucci said, including the possible rapidity of a cardiovascular benefit in the shorter trials. "Usually when we treat any risk factor for cardiovascular disease, we need several years to see any effect on major cardiovascular events. This is true for blood pressure, blood lipids, and blood glucose, and here we see agents that work on blood glucose and show an effect in a few months. This points to speculation that some unconventional cardioprotective effect occurs, which does not require a lot of incubation."

There are no proven mechanistic actions for the observed risk reductions, but Dr. Mannucci said that preclinical studies of the DPP-4 drugs provide some hints, including a direct myocardial effect, GLP-1 stimulation, myocardial protection from ischemia, and recruitment of endothelial cell progenitors."

"We are somewhat stretching the data here, because these trials were not designed for this end point," he acknowledged. "These findings are not fully reliable and we must admit that. But we won’t have any data with credible reliability for the next 4-5 years, so in the meanwhile, even data with some problems could give us some further information."

Dr. Carolyn Deacon, a medical physiologist at Panum Institute in Copenhagen, said that the meta-analysis offers some reassurance to those who are concerned about the cardiovascular safety of any drug. "When we get all the data for the DPP-4–inhibitor trials and the GLP-1 analogues, and examine cardiovascular events for these outcomes, we’ll know much more. But we are getting preclinical data that look positive, so we’re moving in the right direction," she noted.

The preclinical data suggest that the DPP-4 inhibitors may act directly on the heart, induce vasodilation, and improve endothelial function, Dr. Deacon added. "But right now, we cannot separate those benefits from the improved metabolic profile that the drugs induce, so I don’t want to raise hopes too much."

Dr. Mannucci disclosed relationships with Astra Zeneca, Boehringer Ingelheim, Eli Lilly Merck, Bristol-Myers Squibb, Novartis, and Takeda. Dr. Deacon disclosed a number of relationships with pharmaceutical companies involved in diabetes research.

DPP-4 Inhibitors May Be Cardioprotective