This comforting evidence that sitagliptin (Januvia) doesn't increase heart risk comes from a recent large randomized trial in the New England Journal of Medicine comparing the DPP-4 inhibitor with placebo in patients with a history of cardiovascular disease.
But aside from showing that sitagliptin doesn't trigger cardiovascular events, the study doesn't address the more basic and still unsettled question: does the medication actually work?
The Merck-funded study was designed as a non-inferiority trial for the primary purpose of showing, quite simply, that Merck's blockbuster diabetes medication did not increase rates of cardiovascular events -- a concern with many of the newer diabetes agents.
The results were hailed by the researchers -- many of whom reported receiving consulting fees, salary support, and honoraria from Merck -- as a win for sitagliptin. They concluded: "Our study results showed that sitagliptin may be used in a diverse group of patients with type 2 diabetes who are at high cardiovascular risk without increasing rates of cardiovascular complications."
But when viewed from the patient's perspective, the results do not shed such a favorable light on the drug.
In this carefully designed study, the medication failed to reduce one of the most important complications of diabetes -- cardiovascular disease. And while there are many other important complications from diabetes -- retinopathy, nephropathy, and neuropathy to name a few -- the study did not report data on any of these other outcomes.
One has to wonder whether Merck is withholding additional data about their blockbuster drug -- which consistently ranks among the top 25 selling pharmaceuticals in the U.S. Alternatively, perhaps the company was afraid to collect the critical data needed to understand whether or not the drug has clinically meaningful benefits.
Why did Merck fund a study for the purpose of demonstrating that sitagliptin is no more dangerous than placebo?
When one considers the approval process of new diabetes medications by the FDA, the rationale becomes clear. For many conditions like diabetes, the FDA has set a surprisingly low bar for the approval of new medications. To win approval for a new diabetes agent, pharmaceutical companies do not need to demonstrate that their product reduces clinically relevant complications; rather the firm only needs to demonstrate that the new therapy improves surrogate markers, like hemoglobin A1c, and that it doesn't kill you while doing it.
While the rationale for streamlined approval processes for new medications may seem appealing -- such rules enable pharmaceutical companies to bring new products to market more quickly -- they also encourage pharmaceutical companies to conduct studies like this one that provide little useful information for patients and their physicians.
As a result, we only have limited data to guide us in using many of the newer diabetes agents, including DPP-4 inhibitors such as sitagliptin, thiazolidinediones like pioglitazone (Actos), and GLP-1 agonists such as exenatide (Bydureon). In fact, the only medications clearly proven to prevent clinically important diabetic complications -- such as blindness, kidney failure, heart disease, and deaths -- are our time-tested favorites: metformin, sulfonylureas, and insulin.
We believe that there is rarely a role for many of the newer diabetes agents like sitagliptin in our patients' medicine cabinets. Until we see studies showing that these newer agents improve real clinical outcomes we care about, we'll stick to the treatments that we can be confident are better than placebo.
Extra: Summary of Diabetes Drug CV Outcomes Trials
Since a landmark study raised concerns about the cardiovascular safety of rosiglitazone (Avandia) in 2007, experts have increasingly questioned the impact of the newer diabetes agents on "hard clinical outcomes" -- specifically microvascular and macrovascular complications.
The older classes of diabetes agents -- metformin, sulfonylureas, and insulin -- have been shown to reduce microvascular risk and some evidence suggests that they reduce macrovascular risk with long term follow up. But as noted in this article, most of the newer diabetes agents -- including DPP-4 inhibitors such as sitagliptin, thiazolidinediones, SGLT2 inhibitors, and GLP-1 agonists -- have received regulatory approval based on evidence that they reduce HbA1C levels, not microvascular or macrovascular risk.
A recent MedPage Today article about top-line results from a cardiovascular outcomes study of empagliflozin (Jardiance) nicely sums up the evidence on the impact of many of the newer diabetes agents on cardiovascular risk. Aside from a couple of industry-funded trials (including this one announced last week), the majority of high-quality studies of the newer diabetes agents have failed to demonstrate a cardiovascular benefit. Even studies on the impact of the newer diabetes agents on microvascular disease are scant.
"Updates in Slow Medicine" applies the latest medical research to support a thoughtful approach to clinical care. It is produced by Pieter Cohen, MD, of Harvard Medical School, and Michael Hochman, MD, of AltaMed Health System in Los Angeles. To learn more, visit their Facebook page. Rachael Bedard, MD, is a palliative care fellow at the Mount Sinai Hospital in New York.