Vildagliptin in HF Patients With Diabetes Meets Echo End Point, but Questions Remain

May 28, 2013

LISBON, Portugal — In patients with ventricular dysfunction and type 2 diabetes, treatment with the dipeptidyl peptidase 4 (DPP-4) inhibitor vildagliptin (Galvus, Novartis) does not adversely affect LVEF compared with placebo, according to the results of a new study. While the study met its primary end point of statistical noninferiority in terms of change in LVEF vs placebo, investigators did observe an increase in left ventricular end-diastolic volume (LVEDV) and end-systolic volume (LVESV), as well as increased stroke volume.

"About one-third to half of all patients with reduced ejection fraction have diabetes, yet despite this, we still know very little if anything about the effect of most glucose-lowering treatments on left ventricular function and/or heart-failure status," said lead investigator Dr John McMurray (University of Glasgow, Scotland). "In fact, most trials of diabetes drugs have specifically excluded patients with heart failure. This is all the more remarkable now that we know there are drugs to treat diabetes that cause problems in patients with heart failure. Most recently we recognized this with the glitazones, which can cause worsening heart failure."

Presenting the results of the study here yesterday at the Heart Failure Congress 2013 of the European Society of Cardiology Heart Failure Association, McMurray said the finding of increased LVEDV and LVESV was surprising, and while it is currently unknown why this occurred, investigators hypothesize that vildagliptin might increase distensibility and compliance of the left ventricle. Normally, an increase in left ventricular mass is associated with declines in systolic function and an increase in brain-natriuretic peptide (BNP), but researchers reported no changes in LVEF and a reduction in BNP.


The study, known as the Vildagliptin in Ventricular Dysfunction Diabetes Trial (VIVIDD), was presented during the late-breaking clinical-trials session. It included men and women aged 18 to 65 years with type 2 diabetes and NYHA class 1-3 heart failure receiving guideline-recommended therapy. Baseline LVEF was 30.6% in the 128 patients randomized to vildagliptin and 29.6% in the 126 patients randomized to placebo. Mean HbA1c was 7.8%.

The study met its primary end point of statistical noninferiority compared with placebo in terms of change in LVEF from baseline to 52 weeks. In addition, HbA1c levels were significantly reduced. Compared with placebo, there was a statistically significant increase in LVEDV (a difference of 17.06 mL vs placebo), and a trend toward an increase in LVESV (a difference of 9.44 mL vs placebo).

Overall, there was no difference in time to any first cardiovascular event between the two groups (35 events in the vildagliptin arm vs 31 in the placebo arm), but there were numerically more cardiovascular deaths in the treatment arm vs placebo (seven vs four deaths, respectively). In addition, there were 11 deaths in the vildagliptin arm and four deaths in the placebo arm, a difference that raised concerns about safety.

Is It Correct to Even Target HbA1c?

Speaking during the session, Dr Wolfram Doehner (Charité Universitätsmedizin, Berlin, Germany) said that the VIVIDD data are welcomed by physicians who treat patients with heart failure and diabetes, but echocardiographic end points such as LVEF do not provide enough assurances that the drug is safe to use. For example, Doehner cited a randomized multicenter study of rosiglitazone (Avandia, GlaxoSmithKline) vs placebo in heart failure patients that also showed no impairment of any echocardiographic outcome measurements.

"And yet we all know that glitazones are not to be used in heart failure because of their clinical signal," said Doehner. "The message clearly here is that while it's important to look at echocardiographic measurements for safety, it is a mere surrogate. It's the clinical signal that will decide if [vildagliptin] will make it into the clinical setting or not."

Doehner noted that while the drug also reduced HbA1c, the relationship between HbA1c and clinical outcomes in heart-failure patients is currently unclear. Some studies have suggested a U-shaped relationship between HbA1c and clinical outcomes, while others have shown a J-shaped relationship. Still other studies have suggested there might be an inverse relationship between HbA1c and clinical outcomes.

"While we can't conclude which is the right [relationship], we can conclude that it's fully unclear what the correct association is between glycemia and outcomes," said Doehner. "And if you can't decide, then it might not be suitable as a therapeutic target."

The VIVIDD study was sponsored by Novartis.


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.