Reference: N Engl J Med. 2022 Sep 22;387(12):1063-1074
Practice Point: For glucose lowering in patients with type 2 diabetes, what you prescribe after metformin doesn’t seem to matter as long as it’s not a sulfonylurea.
EBM Pearl: Try not to get so lost in the trees that you can’t see the forest. Selection of an appropriate and clinically useful endpoint is as important as finding a significant result.
When it comes to type 2 diabetes medication management, clinicians have been hungry for a prescribing algorithm for years — one based on high-quality evidence and devoid of big pharma’s influence. Ask and you shall receive, sort of.
The recent GRADE trial compared safety and effectiveness of four diabetes medications (glargine, liraglutide, sitagliptin, and glimepiride) in more than 5,000 fairly diverse adults with a mean baseline A1c of 7.5% (range 6.8%-8.5%) who were already taking high-dose metformin. (SGLT-2 inhibitors were not FDA-approved in the U.S. at the time of enrollment.) If you want to cut this short now, the ‘effectiveness’ goes in this order: liraglutide, glargine, glimepiride, sitagliptin. But that’s not nearly the whole story.
The primary outcome was "primary metabolic failure," defined as the cumulative incidence of an A1c ≥ 7.0% over the trial period of about five years. But is that really a treatment failure? The patient-oriented interpretation of the UKPDS trial suggests that there is a U-shaped curve with most benefit and least harm seen between an A1c of 7.5% and 8.5%, the 2008 ACCORD trial was stopped early due to increased mortality in patients with an A1c target < 6%, and the ACP recommends a target of 7-8% for most people. At the questionably-selected endpoint, rates of A1c ≥ 7% ranged from 67% to 77% among the different medications compared to one another. Lest we not forget that the primary outcome of A1c is completely disease-oriented.
Probably the most important and actionable finding was the significantly increased risk of severe hypoglycemia (requiring treatment) with glimepiride compared to the others, even moreso than insulin. Not surprisingly, liraglutide had the greatest weight loss (mean 3.5 kg lost over 5 years) and GI side effects.
The bottom line is that for patients with an A1c in the range recommended by the ACP who are already on max-dose metformin, the absolute reduction in A1c by adding a second agent appears small and of questionable clinical significance. While cardiovascular protection and weight loss may be reasons to add a second agent, trying to get an A1c lower than 7.5% in and of itself, may not be. Diabetes medication prescribing remains more of a menu than an algorithm. Metformin first, whatever they can afford next, but just not a sulfonylurea if you can help it.
For more information, see the topic Glucose-lowering Medications for Type 2 Diabetes in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Katharine DeGeorge, MD, MS, Deputy Editor at DynaMed and Associate Professor of Family Medicine at the University of Virginia. Edited by Alan Ehrlich, MD, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School; Dan Randall, MD, Deputy Editor at DynaMed; Nicole Jensen, MD, Family Physician at WholeHealth Medical; Vincent Lemaitre, PhD, Senior Medical Writer at DynaMed; and Sarah Hill, MSc, Associate Editor at DynaMed.