A post hoc analysis delved further into the EXAMINE trial to evaluate the risk of cardiovascular death in all of the type 2 diabetic participants.
A post hoc analysis of the EXAMINE trial evaluated the risk of cardiovascular (CV) death in all EXAMINE participants and in those who experienced an on-study, major non-fatal CV event. The results were presented by Simon Heller, MD, Professor at the University of Sheffield, UK and co-member of the EXAMINE Steering Committee during the ADA’s Scientific Sessions.
The EXAMINE trial1,2 was a cardiovascular outcomes study. Patients (n=5380) had type 2 diabetes and an HbA1c of 6.5% to 11.0% (7.0% to 10.0% if on insulin). They were currently receiving treatment for their diabetes and had experienced an acute coronary syndrome within 15 to 90 days. Patients were randomized to standard of care plus placebo (n=2679) or standard of care plus alogliptin 25 mg QD (n=2701). The primary endpoint was time to first occurrence of major adverse cardiac event (MACE): CV death, nonfatal myocardial infarction (MI), or nonfatal stroke. The median time in the study was 18 months, with the longest time being up to 3.3 years.
For the post hoc analysis, deaths and non-fatal CV events, and hospitalization for unstable angina were adjudicated. The non-fatal CV events included MI, stroke, and hospitalized heart failure. Patients were followed until censoring or death, regardless of a prior post-randomized non-fatal CV event. Time updated multivariable Cox models were used to estimate the risk of death in the absence of or following each non-fatal event.
In the EXAMINE trial, cumulative incidence of CV death was 4.9% for the placebo group (n=130) and 4.1% for the alogliptin group (n=112) (HR, 0.85; 95% CI, 0.66-1.10). The cumulative incidence of sudden cardiac death is new information that was reported. Rates were 2.7% in the placebo group (n=73) and 2.2% (n=59) in the alogliptin group (HR, 0.08; 95% CI, 0.57-1.12).
In the post hoc analysis, researchers looked at CV mortality during EXAMINE following the adjudicated non-fatal events. For unstable angina, the cumulative incidence of death was no different from those who didn’t have an event. Non-fatal stroke and non-fatal MI accumulated at an incidence of CV death at a significantly higher rate, about twice overall. “The startling thing in this study was the effect for those who were hospitalized for heart failure because either they had a really high cumulative incidence of CV death, such that 1 in 10 had died within 3 to 4 months of hospitalization for heart failure, and around one-third had died just after 20 months.”
Next, Dr. Heller shared post-randomization mortality rates according to treatment assignment. “Because of the issues regarding alogliptin and other DPP-4 inhibitors, we thought it important to show these mortality rates according to treatment assignment.” In the group with no non-fatal event, the patients who took alogliptin had a mortality rate of 4.5% and the placebo group had a rate of 5.8% (HR, 0.81; CI, 0.63-1.05). In the group hospitalized for heart failure, the patients who took alogliptin had a mortality rate of 22.7% and the placebo group had a rate of 34.1% (HR, 1.02; CI, 0.52-2.02).
In conclusion, in type 2 diabetes patients with a recent acute coronary syndrome, the risk of death was increased following a post-randomization, non-fatal event, especially heart failure, compared to those who did not experience an event. Also, the risk of CV death was similar in the alogliptin and placebo groups, including in those patients who had a non-fatal event post-randomization.
Moderator Robert H. Eckel, MD of the University of Colorado School of Medicine added two points about the post hoc findings. The first was that a third of patients with a history of a CV event died, so “Let’s prevent cardiovascular disease in diabetes the best way we can.” The second was a question as to whether or not the caution now included with alogliptin will be relaxed based on these extended findings.
1. White WB, et al. Alogliptin after acute coronary syndrome in patients with type 2 diabetes. N Engl J Med. 2013 Oct 3;369(14):1327-1335.