A Historic View on CV Outcomes in Diabetes

John B. Buse, MD, PhD: Around 2005, there were some emerging studies that suggested potential cardiovascular risks with certain diabetes medications. And this resulted in a great deal of additional focus in the lay media, in the professional press, and eventually at the Food and Drug Administration. In 2008, there was guidance that for a drug to be approved to be marketed for the treatment of diabetes in the United States, before approval, you had to exclude a high cardiovascular risk of the drug. Specifically, you had to exclude a greater than 80% increase in cardiovascular risk to even be considered for marketing. And after marketing, you’d have to demonstrate cardiovascular safety, specifically no more than a 30% change in cardiovascular risk.

So, this resulted in a large number of so-called cardiovascular outcome trials being performed since then. The first several did as expected. They demonstrated the cardiovascular safety of the drugs. The original studies demonstrated no increased risk of cardiovascular harms and no benefit. The hazard ratio, or the risk of the drug versus placebo, was 1.0. That provided a great deal of reassurance, particularly in the cardiology community. But there were some questions as to why we were prescribing drugs that lowered glucose but didn’t really provide any cardiovascular benefit.

It’s very exciting. About a year ago, there was a presentation of a trial called EMPA-REG, which, for the first time, demonstrated cardiovascular benefit. Until I go to my grave, I’ll remember being in the lecture hall where they pointed out a statistically significant 14% reduction in the first heart attack, stroke, or cardiovascular death, and a 38% reduction in cardiovascular death. The audience was just gob-smacked that a diabetes drug could do that.

These cardiovascular outcome trials in diabetes, in order to accumulate the needed number of cardiovascular events to achieve the possibility of statistical power, have recruited high-risk patients. So, the highest risk patients included in these trials include patients who’ve recently recovered from a hospitalization for acute coronary syndrome. That has been used in a subset of these trials. The next highest risk are patients who haven’t had a recent hospitalization, but have had a prior MI, stroke, or procedure, like bypass surgery or stenting. And then, the next highest risk would be patients who don’t have clinical cardiovascular disease, but have advanced risk factors. So, not only diabetes, but generally some combination of diabetes, age, and let’s say hypertension, with evidence of left ventricular hypertrophy. These patients are higher risk than the general population of patients with type 2 diabetes.

Now that we have had a number of cardiovascular outcome trials that have shown benefit, I think there’s a much greater interest in doing these cardiovascular outcome trials in a way that answers meaningful questions. And so, 2 areas that are still pretty uncertain are: Are there cardiovascular benefits of these newer agents that have demonstrated cardiovascular risk reduction, namely from the SGLT2 inhibitor class and the GLP-1 receptor agonist class? And are there benefits in patients without prior cardiovascular disease and at more moderate risk of cardiovascular events?

To date, the signal has been unclear and, if anything, there doesn’t seem to be benefit, but the number of patients without prior cardiovascular disease has been relatively small. I believe that these drugs will have benefits in those lower risk populations, but we need studies to address that question.