The combination of a low-calorie diet with a supervised exercise regimen induced nonpharmacologic remission of early-onset type 2 diabetes (T2D) in more than half of participants in a new study, with low drop-out rates and preservation of lean mass.

“The take-home message is that in early-onset type 2 diabetes, people should ideally have a potent nondrug option to tackle [the disease],” first author Kaberi Dasgupta, MD, a professor of medicine and director of the Division of General Internal Medicine at McGill University in Montréal, Quebec, Canada, told Medscape Medical News.

In addition to inducing diabetes remission, the diet/supervised exercise intervention “shifts weight loss where we want it to be — from the fat compartment — and also increased muscle health, strength, and fitness, improving mood, lowering distress related to diabetes, and enhancing quality of life,” she said.

Dasgupta presented the findings at the European Association for the Study of Diabetes (EASD) 2026 Annual Meeting, and the study was concurrently published online in The Lancet.

RESET for Remission Trial

Prior clinical trials have shown significant benefits of low-calorie diets in inducing diabetes remission in general; however, research is lacking in the context of early onset T2D, defined as a diagnosis at age 45 years or younger, which is known to have a more aggressive disease course and increased risk for complications and mortality.

Evidence on the combination of an intensive exercise program with a low-calorie approach among those patients is further lacking, prompting the authors to conduct a prospective, randomized, open-label trial, enrolling 96 participants with early-onset T2D and obesity who were not on insulin at three centers in England and Canada.

The participants were randomized to either usual care (n = 46) or to an intervention consisting of a low-calorie diet of 800-900 calories per day, along with twice-weekly supervised aerobic and resistance exercise, along with once-weekly independent exercise for the first 12 weeks (n = 50).

The intervention group further followed a weight-maintenance diet with independent exercise in weeks 13-24.

Intervention — Diet

In the intervention’s intensive phase, participants began the low-calorie regimen in weeks 1 and 2 and added the exercise component from weeks 3 to 12.

The diet in the first 2 weeks consisted of total meal replacements with products such as Optifast shakes, with more variety provided in weeks 3-12, allowing for one food-based meal per day.

Coaching and monitoring were provided by physicians and dieticians, with participants attending the first visit in-person, with subsequent coaching sessions provided in-person or by phone.

All glucose- and blood pressure-lowering medications (in the absence of albuminuria) were discontinued and the 800- to 900-calorie per day diet consisted of 30% protein, 50% carbohydrate, and 20% fat.

Participants were additionally encouraged to consume at least 2 liters of calorie-free fluid daily and were given fiber-based laxatives for use as needed.

Intervention — Exercise

The exercise intervention consisted of twice-weekly sessions from weeks 3 to 12 that progressed to 1 hour each, with equal time for aerobic and resistance training.

The aerobic component consisted primarily of walking and high-intensity interval training, if tolerated; resistance training involved machines and/or free weights. On exercise days, participants were recommended to consume an extra meal replacement of 200-250 calories.

The participants had a mean age of 38 years and a mean BMI of 35.2, and 47% were women. The mean duration of T2D was 2.3 years, a mean A1c was 7.2%, and over three quarters were being treated with glucose-lowering medications.

Final assessments were attended by 86 participants, for a 90% completion rate.

Results Including Weight, Cardiometabolic Improvements

The study’s primary outcome of remission at 24 weeks, defined as having achieved an A1c level less than 6.5%, without the use of glucose-lowering medication for at least 12 weeks, was met, occurring among 54% of those receiving the intervention vs 4% in the control group (adjusted odds ratio [OR], 21.6).

The rates of remission were consistent in subgroup analyses based on gender, ethnicity, and country.

Those in the intervention group also had greater general improvements in A1c than those in the control group (-0.7 vs control participants), and insulin resistance was also improved, as captured by the Homeostatic Model Assessment for Insulin Resistance.

The intervention group showed greater weight reduction at 24 weeks than the control group (-7.0%). In addition, fat mass loss was greater in the intervention group, as assessed by DEXA (-16.5% vs control group), and there were no significant differences between the groups in measures of lean mass or muscle volume, as assessed by MRI.

More than half of those in the intervention group (56%) achieved a weight loss of at least 5%.

The intervention group had increases in handgrip strength overall, while step counts and moderate-to-vigorous physical activity were more improved at 12 weeks, with the differences attenuated at 24 weeks.

Other improvements observed with the intervention included cardiorespiratory fitness relative to body mass at 24 weeks (10.3%), while improvements were also notably observed in other outcomes such as blood pressure, resting heart rate, and triglycerides.

Hypertension remission occurred at 12 weeks among 63% of participants in the intervention group and 13% in the control group (adjusted OR, 8.9), with closer differences at 24 weeks (56% vs 27%).

Important decreases were also observed in the intervention group than in control group in measures of absolute liver (-9.1%) and pancreatic fat fractions (-1.3%).

Most participants in the intervention group experienced adverse events, including constipation, headache, dizziness, and fatigue, with some having musculoskeletal discomfort or injury related to exercise, but most symptoms were resolved.

No serious adverse events were reported.

Of note, patients in the intervention group also showed reductions in depressive symptoms and diabetes-related distress, based on Hospital Anxiety and Depression Questionnaire scores, with increases in quality-of-life measures significantly improved on the EuroQUOL-5 Self-rated Health index.

While the remission rates are on the higher end compared to previously reported trials, such as the DIADEM-I trial, the study “demonstrates for the first time that such effects can be achieved in a multi-ethnic population with early-onset type 2 diabetes through a combined low-energy diet and structured exercise intervention,” the authors wrote.

Furthermore, “no previous remission trial has incorporated supervised structured exercise and low-energy diet or integrated MRI to quantify muscle volume and intramuscular adiposity,” they explained.

Other studies have shown a similar magnitude of weight loss; however, the current study showed such changes while preserving lean mass and MRI-confirmed muscle volume.

Importantly, the results suggest an effective alternative to glucose control that is substantially less costly than alternatives of metabolic surgery or pharmacologic approaches such as with GLP-1 receptor agonists, while sparing the potentially life-long need for treatment and unknown reproductive and other effects with the latter.

“Some people simply want a nonpharmacological option,” Dasgupta said.

Muscle, Fat Mass

Further discussing the findings’ relevance to muscle body composition at the meeting, co-author Tom Yates, PhD, a professor of physical activity and health in the Diabetes Research Centre, University of Leicester, Leicester, England, added that exploratory findings from the study using advanced imaging techniques also showed no reductions in abdominal and upper thigh muscle volume and “substantially less fat infiltration of the muscle,” which are favorable signs of improving muscle quality.

He also highlighted the concept of accelerated biological aging in diabetes. “Our group and others have shown that in terms of simple functional measures like hand grip strength and sit-to-stand tests, people with type 2 diabetes can present at 10 to 20 years older based on normative values than people without diabetes.”

In general, they have a 2-4 times greater risk for frailty and disability, he said.

Both authors agreed that a key limitation of the study was not having a third cohort of patients who received the low-calorie intervention, without the exercise component as a comparator.

Real-World Application Could Face Challenges

The applicability of an intervention involving supervision in the real world is another question, Dasgupta noted.

“Anecdotally, there were many people in the intervention group who reported liking the exercise intervention more than a dietary intervention because it was supervised, so getting that social support when they were exercising seemed really key to motivation and enjoyment,” she said in her talk. “Rolling that out into the real world is going to be more of a challenge.”

Yet another key question raised by an audience member in the session is the sustainability of results, with very low-calorie diets well known to have potentially short-lived results. 

Dasgupta agreed that the question is important and the weight-regain effect indeed is observed in research.

However, she noted that “there is also epidemiological evidence indicating that the more time you have without type 2 diabetes, the lower your risk for cardiovascular and kidney complications.”

“So any year without type 2 diabetes is good,” Dasgupta added. “But I think the challenge here is the early-onset disease, and the risk of an accumulation of complications.”

Commenting on the study, Professor Roy Taylor, MD, of Newcastle University in Newcastle upon Tyne, England, noted that “it is highly important to find ways of achieving remission in young-onset T2D, given the appalling rates of complications that otherwise occur.”

He noted that “younger people in the NHS England Type 2 Diabetes Path to Remission National Program tended to have lower rates of remission than older people,” and that therefore a different approach was needed than those used for older patients.

That said, he agreed that the feasibility of a similar supervised intervention outside of a study setting could have challenges.

“The supervision of exercise in the RESET study was admirable for the purpose of determining biological feasibility of achieving remission,” Taylor explained. “However, reproducing this in a real-world, financially constrained setting, would be very difficult.”

Taylor was the senior author of the landmark DIRECT trial, which showed that an approximately 850 calorie-per day diet followed by stepped food reintroduction over 8 weeks resulted in T2D remission in about half of participants, and showed that removing excess fat from the pancreas and liver can reverse the disease.

Ultimately, in trials requiring such discipline, “individuals have to be motivated to want to get rid of their diabetes,” he said. “Doctors can help with motivation, but only the individual themselves can decide.”

Preserved Skeletal Muscle Is ‘Striking’

Further commenting, Juleen Zierath, MD, a professor of physiology at Karolinska Institutet in Stockholm, Sweden, added that “these findings are important because early-onset type 2 diabetes is a growing group with an aggressive disease course, and it has been underrepresented in trials.”

“To my knowledge, this is the first trial to test that combination for remission,” Zierath told Medscape Medical News.

“What is striking is that the exercise component delivered what we would hope for: fat loss with preserved skeletal muscle, greater strength, and better fitness, even during substantial energy restriction.”

Zierath underscored that, for the benefits to be sustained, “planning for maintenance after the intensive phase will be essential.”

Overall, however, “this reinforces that exercise works, and that [it] should be a cornerstone of diabetes care,” she said.

“The clinical message is a strong one: diet and exercise together can play a profound role in easing the burden of type 2 diabetes.”

The RESET for Remission trial was funded by the Medical Research Council, the Canadian Institutes of Health Research (CIHR), and the John R. McConnell Foundation, supported by the NIHR Leicester Biomedical Research Center. Dasgupta reported receiving current or prior operating grants as principal investigator from the CIHR, Diabetes Canada, The Heart & Stroke Foundation of Canada, The Lawson Foundation, the Medavie Foundation, the Fonds de recherche du Québec. Yates reported receiving investigator-initiated funding for obesity-related research from Biomea Fusion, contracted research funds from Abbott, and serving as a consultant for Regeneron. Taylor and Zierath had no disclosures to report.