2022 Top Story in Diabetes: Comparative Effectiveness of Common Type 2 Diabetes Therapies—Long-Awaited Results From the GRADE Study

Very few studies assess the long-term comparative effectiveness of multiple type 2 diabetes (T2D) therapies. This information is critical given the ever-expanding number of therapeutic options. The Glycemia Reduction Approaches in Type 2 Diabetes (GRADE) study assessed the comparative effectiveness of four common therapies added to metformin (glargine, glimepiride, liraglutide, and sitagliptin) over a mean 5 years of follow-up.¹ Trial participants (N = 5047) had type 2 diabetes for <10 years, were taking metformin, and had an HbA1c of 6.8 to 8.5%.¹

There were modest but statistically significant differences in the primary outcome (HbA1c ≥7.0% confirmed on two separate occasions 3 months apart) between groups, which occurred in a similar number of participants assigned to liraglutide (68%) and glargine (67%), both of which were superior to glimepiride (72%), which in turn was superior to sitagliptin (77%). A poignant finding was that 71% of the patients developed treatment failure overall and there was a gradual rise in HbA1c in all groups during the study. While there were no differences in microvascular outcomes (in part due to the low number of events and relatively small differences in HbA1c), there were fewer cardiovascular events with liraglutide compared with the other three groups combined (HR, 0.71; 95% CI, 0.56–0.90) and in pairwise comparisons with sitagliptin and glimepiride.²

Findings from the GRADE trial expand upon that of the United Kingdom Prospective Diabetes Study, in which only about one-third of patients attained an HbA1c <7% by 6 years on insulin, metformin, or sulfonylurea.³ In A Diabetes Outcome Progression Trial (ADOPT), the cumulative incidence of monotherapy treatment failure (using a more liberal definition of fasting glucose >140 mg/dL) at 5 years was 15%, 24%, and 33% with rosiglitazone, metformin, and glyburide respectively.⁴ While there are differences in study design, these studies are consistent with the concept that T2D is progressive and treatment failure is influenced by baseline HbA1c, duration of diabetes, as well as therapeutic approach.

Unfortunately, the GRADE study was not able to address the long-term comparative effectiveness between liraglutide and thiazolidinediones (which had the greatest durability in the ADOPT study),⁴ SGLT2 inhibitors, or more potent GLP-1 receptor or dual GLP-1/GIP receptor agonists.

However, there is evidence that early implementation of glucose control has a fundamental enduring effect on risk of complications.⁵ Moreover, early combination therapy has been shown to reduce the rate of treatment failure compared with a stepwise approach.⁶ Professional society guidelines have recommended the consideration of early combination medication therapy in some individuals with this in mind.^7,8 In summary, the GRADE trial provides additional support for the use of early combination therapy, with a possible advantage of liraglutide in particular, but also emphasizes the need for continued vigilance and treatment intensification in patients with T2D.