Diabetes Drug Face-Off: New Japanese Trial Tests Whether a Popular SGLT2 Inhibitor Eats Away at Muscle

A large share of Japanese patients living with type 2 diabetes are older adults who are not obese, a population that has been largely invisible in the international evidence base for one of the fastest-growing classes of diabetes drugs. Now, a team of researchers led by Yasutaka Takeda and Naoki Kumashiro of Kanazawa Medical University has launched a randomized clinical trial designed to answer a question that has been quietly worrying diabetologists for years: do sodium-glucose cotransporter 2 (SGLT2) inhibitors, celebrated for protecting the heart and kidneys, quietly erode skeletal muscle in lean patients? The study, known as TIGHTEN, compares the SGLT2 inhibitor tofogliflozin against the older sulfonylurea glimepiride, and its protocol has just been published in the journal Diabetes Therapy.

The concern is rooted in the pharmacology of the drug class itself. SGLT2 inhibitors lower blood glucose by blocking glucose reabsorption in the kidney, flushing roughly 60 to 80 grams of glucose into the urine each day. That caloric loss drives weight reduction, which is often marketed as a benefit, particularly for patients with obesity. But evidence drawn mainly from trials in heavier patients suggests the lost weight may include not only fat but also skeletal muscle, raising the specter of sarcopenia, the age-related loss of muscle mass and strength that predisposes older adults to frailty, falls, disability, and death. In a country where most people with type 2 diabetes do not meet the Japanese threshold for obesity, the question is not academic.

Previous data have been ambiguous. The EMPA-ELDERLY trial, which tested empagliflozin in Japanese adults aged 65 and older, found that the drug improved glycemic control and promoted weight loss without compromising muscle mass or strength. Yet that trial enrolled participants with a mean body mass index of 25.6 kilograms per square meter, a figure that qualifies as obese under Japanese criteria. In other words, even the most reassuring study to date was conducted in a population that does not resemble the lean, older Japanese patient most likely to receive these drugs in everyday practice. The TIGHTEN investigators argue that this gap leaves clinicians prescribing in a zone of genuine uncertainty.

The trial itself is deliberately straightforward. It is a prospective, multicenter, open-label, randomized, parallel-group comparative study conducted at three sites in Ishikawa Prefecture: Kanazawa Medical University Hospital, Asanogawa General Hospital, and Kanazawa Nakade Clinic. Participants are adults with type 2 diabetes whose body mass index is below 25 kilograms per square meter, the Japanese cutoff for obesity, and whose glycated hemoglobin sits between 7.0 and just under 10.0 percent. Patients with an estimated glomerular filtration rate below 30 milliliters per minute per 1.73 square meters are excluded, consistent with the safety labeling of both drugs. After informed consent, participants are centrally randomized in a one-to-one ratio using the minimization method, balanced for sex and for age below or above 70 years.

One arm receives 20 milligrams of tofogliflozin orally once daily for 24 weeks. The other starts glimepiride at 0.5 to 1 milligram per day, with dose adjustments between 0.5 and 4 milligrams permitted at the treating physician’s discretion to reach glycemic targets while avoiding hypoglycemia. All participants follow a diet calibrated to their target body weight and maintain daily physical activity in line with Japan Diabetes Society guidelines. Changes to other antidiabetic medications are not permitted in principle during the study, ensuring that any differences in body composition can be attributed to the assigned treatment rather than to shifting polypharmacy.

The primary endpoint is the change in the ratio of total skeletal muscle mass to total body weight from baseline to 24 weeks. That choice is more subtle than it looks. Because body weight is the denominator, the ratio can rise simply because weight falls, even if absolute muscle mass declines. The investigators have anticipated this trap and pre-registered a battery of secondary measures to guard against a misleading result: percentage change in the muscle-to-weight ratio, the skeletal muscle mass index calculated by dividing appendicular lean mass by height squared, absolute total lean body mass, appendicular lean mass, total and relative fat mass, and dietary intake measured with the Brief-type Self-administered Diet History Questionnaire. If the ratio climbs while absolute lean mass or the muscle index falls, the increase will be interpreted as denominator-driven rather than as evidence of muscle preservation.

All body composition measurements will be performed with a single dual-energy X-ray absorptiometry system at Kanazawa Medical University Hospital, a design decision meant to eliminate the inter-scanner variability that can plague multicenter imaging studies. Blood counts, glycemic parameters, a full biochemical panel, urinary albumin excretion, blood pressure, and body weight round out the observation schedule. An independent data center and data and safety monitoring board manage the trial’s data and monitor its conduct, and adverse events, including exacerbations of preexisting conditions, are systematically captured and reported.

The sample size is modest but justified. Drawing on a prior trial of dapagliflozin versus glibenclamide that reported changes in the muscle-to-weight ratio of 0.012 versus 0.001, the investigators assumed a between-group difference of 0.011 with a standard deviation of 0.011. With a two-sided alpha of 0.05 and 90 percent power, 23 participants per group suffice; adding a 15 percent dropout allowance brings the target to 56 participants. Recruitment ran from April 2024 to June 2025, and complete findings are expected in 2027. No interim analysis is planned; the primary endpoint will be analyzed with analysis of covariance adjusting for treatment group, sex, age category, and the baseline muscle-to-weight ratio, with per-protocol analyses as sensitivity checks.

The open-label design, the restriction to Japanese patients, and the 24-week horizon all limit how far the results can travel, and the authors acknowledge each of these constraints. Still, the trial will be the first randomized study to test the effects of an SGLT2 inhibitor on DXA-measured skeletal muscle mass in patients with type 2 diabetes who are not obese, and it sets no lower limit on body mass index, meaning it will capture even very lean participants with BMI values below 20. For millions of older, lean patients with type 2 diabetes worldwide, and for the physicians weighing cardiorenal protection against the hidden cost of frailty, the answer cannot come soon enough.

Subject of Research: Effects of the SGLT2 inhibitor tofogliflozin versus glimepiride on skeletal muscle mass in patients with type 2 diabetes without obesity

Article Title: Study Protocol for the TIGHTEN Study Comparing the Effects of Tofogliflozin and Glimepiride on Skeletal Muscle Mass in Patients with Type 2 Diabetes without Obesity: A Prospective, Multicenter, Open-Label, Randomized, Parallel Group Trial in Japan

Article References: Takeda, Y., Ozawa, T., Ikoma, M., Shimada, K., Saito, R., Nakagawa, A., Sawamura, T., & Kumashiro, N. (2026). Study Protocol for the TIGHTEN Study Comparing the Effects of Tofogliflozin and Glimepiride on Skeletal Muscle Mass in Patients with Type 2 Diabetes without Obesity: A Prospective, Multicenter, Open-Label, Randomized, Parallel Group Trial in Japan. Diabetes Therapy. https://doi.org/10.1007/s13300-026-01920-1

Image Credits: AI Generated

DOI: 10.1007/s13300-026-01920-1

Keywords: SGLT2 inhibitor, tofogliflozin, glimepiride, type 2 diabetes, sarcopenia, skeletal muscle mass, randomized trial, dual-energy X-ray absorptiometry, Japan, Diabetes Therapy, clinical trial protocol, body composition