Use of tirzepatide within the first 4 years of type 2 diabetes (T2D) diagnosis produced significant improvements in glycemic control, body weight, and cardiovascular risk factors compared to usual treatment in 2-year results from Eli Lilly’s SURPASS-EARLY trial. 

“These findings support the concept that early initiation of tirzepatide could establish better and potentially more durable glycemic control and improve multiple cardiovascular risk factors,” said Stefano Del Prato, MD, PhD, professor of endocrinology and metabolism at the University of Pisa School of Medicine, Italy. 

Del Prato, who is also chief of the Section of Diabetes at the University of Pisa, presented the 2-year results of the 4-year study here at the 19th International Conference on Advanced Technologies and Treatments for Diabetes (ATTD) 2026.

Study Details 

The study included a total of 794 individuals from 10 countries with a diagnosis of T2D for 4 years or less who were inadequately treated (A1c, 7.1%-9.5%) with metformin. Study participants were randomized to tirzepatide 15 mg (n = 398) or maximal tolerated dose plus metformin with “intensified conventional care” — which could include a variety of approved medications for treating T2D along with metformin, including oral or injectable GLP-1 agonists or SGLT2 inhibitors, but not tirzepatide or insulin (n = 396). Drug additions and changes were permitted within labeling, but the majority (88%) took just one in addition to metformin. 

By week 104, 359 participants randomized to the tirzepatide group and 350 with intensified conventional care remained in the trial. 

Compared to an overall mean baseline A1c of 7.81%, by week 104, the tirzepatide group had dropped to 5.56% vs 6.35% with intensified conventional care. Results were significant for both the efficacy estimand (including all randomized participants had they all continued to receive treatment without rescue medication), with an estimated treatment difference of -0.79, and the treatment regimen estimand (all participants regardless of discontinuation or rescue treatment) of -0.68, Del Prato reported. 

The proportions achieving target A1c levels of < 7%, ≤ 6.5%, and < 5.7% were all significantly greater with tirzepatide. For < 7%, differences were 95.06% vs 77.94% for the efficacy estimand and 85.5% vs 72.14% for the treatment regimen estimand.

Change in weight from an overall mean baseline of 100.1 kg were -15.8 kg vs -6.5 kg for the efficacy estimand. The treatment regimen estimand difference was a significant -7.95 kg. The proportions achieving weight reduction thresholds of ≥ 5%, ≥ 10%, and ≥ 15% were also all significant for both estimands. 

Serum cholesterol level percent changes from baseline didn’t differ between the two groups, but there were significant differences in HDL cholesterol (estimated treatment difference 7.7%), LDL cholesterol (3.9%), VLDL cholesterol (-18.6%), and triglycerides (-18.9%), all in favor of tirzepatide. The estimated treatment difference in systolic blood pressure (-2.15 mm Hg) was also significant, but diastolic was not. 

Other significant differences between groups in percent change from baseline included C-peptide, proinsulin, insulin levels, and insulin resistance. 

Treatment-associated adverse events were greater with tirzepatide (74.6% vs 68.6%) as were treatment discontinuation due to adverse events (4.5% vs 0.3%), but serious adverse events were not (5.5% vs 6.8%). The most common adverse events were gastrointestinal, most often nausea, vomiting, and constipation. There were two deaths in the tirzepatide group and one in the intensified conventional care group. 

Likely Beneficial, but Expensive 

Asked to comment, Rozalina G. McCoy, MD, associate professor and vice chief of clinical research in the Division of Endocrinology, Diabetes, and Nutrition, at the University of Maryland School of Medicine, Baltimore, told Medscape Medical News, “my personal impression is that achieving glycemic and weight loss goals early, before obesity-related complications develop, is likely beneficial and aligns with how we should be approaching early T2D management.”

However, she added, “the elephant in the room is cost. Whether the savings associated with complication reduction from early improvement of glycemia and weight will outweigh the cost of the medication will need to be shown, but ultimately that is a problem of high medication cost, not the lack of clinical benefit to patients. We need to tackle cost and make improving health outcomes not only effective, but cost-effective.”

McCoy pointed out that the exclusion of people with baseline atherosclerotic cardiovascular disease is “an important point…tirzepatide has been shown to be comparable to dulaglutide for cardiovascular risk reduction, so would be a reasonable option for those at high CVD risk as well, though it's glucose- and weight-lowering effectiveness is much greater than that of dulaglutide."

She also noted that the study investigators “did not evaluate long-term outcomes, so we cannot answer to what extent early glycemic and weight loss goal attainment will translate into reduced complications over time. We do know that there is a legacy effect of early attainment of glycemic goals on clinical outcomes over time, so we can expect that this will be the case here, but that will need to be examined with continued follow up of these patients.”

Should “Prediabetes” Be Exchanged for T2D Stages? 

Session moderator Moshe Phillip, MD, director of the Institute for Endocrinology and Diabetes, National Center for Childhood Diabetes at Schneider Children’s Medical Center, Petah Tikva, Israel, told Medscape Medical News that the data align with a new way of thinking about T2D. 

“When is it early to start intervening in the process of diabetes? I think that one of the mistakes that we as a community have done in the past is to call people as having ‘prediabetes,’ because it implies that they are healthy, and actually many who are defined as ‘prediabetes’ have higher risks for all the diabetes complications, mainly the cardiovascular complications,” said Phillip, who is also professor emeritus of pediatrics at Gray Faculty of Medical and Health Sciences, Tel-Aviv University.

He pointed to a recent commentary he co-authored that was published in the Lancet Diabetes & Endocrinology, proposing replacing “prediabetes” with a new staging system. In this system, stage 1 would represent a gradual increase in fasting glucose but still within the normal range; stage 2 is when glucose levels are above normal but still below the threshold for T2D; and stage 3 would combine all dysglycemia levels — ie, impaired fasting plasma glucose, impaired 1-hour or 2-hour 75 g oral glucose tolerance test values, or A1c 5.7%-6.4% (39-46 mmol/L). He noted that the rate of progression through the stages varies in individuals.

The new SURPASS-EARLY data, then, “is in line with the other activities to intervene early, and it shows that early intervention, any medical intervention, has a better chance to lower the risk for late complications,” Phillip said.

Del Prato serves on advisory boards for, and/or has paid speaking engagements with Abbott, Eli Lilly, Hoffman LaRoche, Innovent, Menarini International, Nephris, Novo Nordisk, Roche, Sun Pharma, AstraZeneca, Boehringer Ingelheim, Laboratori Guidotti, and Sanofi. 

McCoy has no disclosures. 

Phillip has been serving as advisory board member of AstraZeneca, Eli Lilly, MannKind, Medtronic Diabetes, Pfizer, Sanofi, DOMPE, LifeScan, Novo Nordisk, Insulet, Provention Bio, Merck, Ascensia, Bayer, Embecta, and Tandem. He has received consulting fee from: QuLab Medical, Provention Bio. The institute he heads received research grants from: Eli Lilly, Medtronic Diabetes, Novo Nordisk, Pfizer, Sanofi, DreaMed Diabetes, NG Solutions, DOMPE, Lumos, GWAVE, OPKO, Provention Bio, AstraZeneca, MP is stock owner: DreaMed Diabetes and NG Solutions. 

Miriam E. Tucker is a freelance journalist based in the Washington DC area. She is a regular contributor to Medscape, with other work appearing in the Washington Post, NPR’s Shots blog, and Diatribe. She is on X (formerly Twitter) @MiriamETucker and BlueSky @miriametucker.bsky.social