CardioExchange Archive

CardioExchange’s Harlan M. Krumholz interviews Benjamin M. Scirica about his study group’s analysis of data from the SAVOR-TIMI 53 randomized, placebo-controlled trial of saxagliptin in patients with type 2 diabetes. The new observations are published in Circulation.

Krumholz: Please describe your major findings how they add to those from the main trial.

Scirica: This analysis provides more detail on the initial finding of an increased risk of hospitalization for heart failure among patients treated with saxagliptin in the SAVOR-TIMI 53 trial that we reported last year in NEJM. Here are our major observations:

• Treatment with saxagliptin increased the risk of hospitalization for heart failure from 2.8% to 3.5% over two years, a relative 27% increased risk in the 16,492 patients randomized to the trial.
• The greatest risk of hospitalization for heart failure appeared to occur within the first 12 months after treatment initiation.
• Although the relative risk of heart failure was not greatest in any subgroup, the absolute greatest excess in hospitalizations for heart failure occurred in patients who had the overall highest risk for heart failure (those with chronic kidney disease, prior heart failure, or elevated levels of natriuretic peptide).

Krumholz: In your discussion, you argue that this risk is likely to be real in part because it was prespecified. Why did you design the study to assess hospitalization for heart failure? Were you anticipating a problem, or did you expect a benefit?

Scirica: We prespecified hospitalization for heart failure as a component of the secondary endpoint for several reasons. First, diabetic heart failure is a growing, yet poorly understood disease, so we felt it was an important clinical endpoint to adjudicate. Moreover, heart failure is a challenging event to capture appropriately, and adjudication was the only way to ensure we were identifying true episodes of heart failure. Second, preclinical studies and some small human studies actually suggested potential benefit from incretin-based therapy, so we felt this was an important setting in which to test that hypothesis.

Krumholz: Do you think that this increased risk is a class effect, given what you know from the literature?

Scirica: It is too early to tell. There was a consistent, albeit not statistically significant, signal with alogliptin in the EXAMINE study. I think the results of the large, ongoing TECOS study of sitagliptin are likely to tell us whether this is a class effect. Literally tens of thousands of patients have completed or are participating in cardiovascular outcome studies of DPP-4 inhibitors and GLP1 agonists. These trials will provide an enormous amount of safety and efficacy data on hyperglycemic agents and will hopefully guide diabetes therapy with much greater certainty.

Krumholz: What do you say to your patients who are taking saxagliptin?

Scirica: Our finding in SAVOR-TIMI 53 was that for the great majority of patients, saxagliptin was safe from a cardiovascular standpoint and therefore provides a valuable option for improving glycemic control. For a small proportion of users of this drug, the risk for heart failure is high. In those patients, if needed, I would use DPP-4 inhibitors cautiously and closely monitor symptoms, in particular during the initial months of therapy.

JOIN THE DISCUSSION

Share your thoughts on Dr. Scirica’s new analysis of data from SAVOR-TIMI 53.