A decade of <scp>SGLT2</scp> inhibitors: Lessons learned and the road ahead in type 2 diabetes

The therapeutic journey of sodium–glucose cotransporter 2 (SGLT2) inhibitors in type 2 diabetes reflects a remarkable transformation in clinical medicine. When these agents were first introduced more than a decade ago, skepticism prevailed. Their mechanism of action, independent of pancreatic β-cell stimulation, raised doubts about the durability of glycemic efficacy. Safety concerns were also voiced, particularly regarding dehydration, genitourinary infections, and the possibility of euglycemic diabetic ketoacidosis. Moreover, some speculated that metabolic change might lead to impaired amino acid utilization, muscle catabolism, and sarcopenia in vulnerable patients. Over 10 years later, however, most of these initial fears have been largely dispelled: with appropriate use, SGLT2 inhibitors have proven to provide durable glucose lowering together with systemic benefits. Experimental studies have also demonstrated protective effects on pancreatic β cells, suggesting that SGLT2 inhibition may contribute to the preservation of endogenous insulin secretion capacity1. What was once regarded as a controversial class of glucose-lowering drugs is now widely recognized as a cornerstone of diabetes management, offering multi-organ protection that extends to the cardiovascular, renal, hepatic, and even neurological systems (Figure 1). A line of evidence has reduced safety concerns. Long-term observational data from East Asia demonstrated that the overall incidence of adverse events with SGLT2 inhibition was below 5% even in older patients2. A large prospective multicenter study in Japan reported that adverse events were mainly limited to mild or moderate urinary/genital infections and volume depletion, and importantly, unexpected serious safety signals were observed3. A recent systematic review and meta-analysis with over 77,000 older or frail patients with type 2 diabetes and heart failure demonstrated that SGLT2 inhibitor therapy was associated with an approximately 20% reduction in all-cause mortality and a 30% reduction in hospitalization for heart failure4. These findings suggest that SGLT2 inhibitors consistently lower the risk of death and heart failure hospitalization even in frail or older individuals with type 2 diabetes. In a large 3-year post-marketing surveillance study conducted in Japan, empagliflozin was well tolerated across a broad age range, with no major new safety concerns identified5. The magnitude of weight reduction varied by baseline body mass index (BMI): the greatest decrease was observed in patients with BMI ≥30 kg/m2, whereas almost no change was seen in patients with BMI <20 kg/m2. These findings suggest that SGLT2 inhibitors are generally safe and effective even in lean and elderly patients. An interesting analysis using the Japan Medical Data Center (JMDC) Claims Database compared the healthcare resource utilization of the widely used dipeptidyl peptidase-4 (DPP-4) inhibitors with that of SGLT2 inhibitors6. The study demonstrated that the advantages of SGLT2 inhibitors are not limited to individuals but extend to healthcare systems, a critical factor in aging societies with escalating medical expenditures. The liver has emerged as a rapidly expanding area of research. A recent trial in China evaluated dapagliflozin in patients with biopsy-proven metabolic dysfunction-associated steatohepatitis (MASH)7. In this trial, dapagliflozin was associated with significantly greater histological benefit compared with placebo, and MASH improvement was achieved in 53% of patients vs. 30% with placebo. A 5-year follow-up study of ipragliflozin showed improvement of glycemic control, hepatic steatosis, and aminotransferases in patients with type 2 diabetes and metabolic dysfunction-associated steatotic liver disease (MASLD)8. Reductions in body weight, visceral fat, and fibrosis markers were maintained throughout the 5 years. In the Japanese Study of Diabetes Treatment by Long-Term Empagliflozin (J-STEP/LT study), tofogliflozin significantly reduced alanine aminotransferase (ALT) levels in patients with elevated baseline ALT across all BMI categories9. Notably, ALT improvements were observed even in lean patients without weight loss. These findings strongly support a possibility of SGLT2 inhibitors as a promising therapeutic option for MASLD and MASH. Renal protection is a well-established effect of SGLT2 inhibitors, while recent reports on renal composite outcomes with semaglutide have drawn increasing attention to glucagon-like peptide-1 (GLP-1) receptor agonists10. In a propensity score–matched real-world cohort from the TriNetX network, combination therapy with GLP-1 RAs and SGLT2 inhibitors further reduced 5-year risks of major adverse kidney events, end-stage kidney disease, and mortality compared with SGLT2 inhibitor monotherapy, underscoring the potential of dual therapy in type 2 diabetes11. The RESTORE-nephropathy study demonstrated that tofogliflozin reduced urinary albumin excretion and improved liver enzymes in patients with type 2 diabetes and diabetic nephropathy12. Notably, urinary albumin levels declined during treatment, rose after withdrawal, and decreased again with re-administration, highlighting the reversible nature of its renoprotective effects and the importance of continuous therapy. Beyond renal protection, microvascular effects on the eye are now being recognized. In a nationwide claims database analysis from Japan including 2,412 propensity score–matched patients with diabetic macular edema, treatment with SGLT2 inhibitors significantly reduced the need for intravitreal triamcinolone acetonide injections, with incidence rates of 63.8 vs. 94.9 per 1,000 person-years and hazard ratios of 0.66–0.44 for repeated injections13. Thus, SGLT2 inhibitors may serve as a novel, low-cost adjunctive option for diabetic macular edema. Perhaps, the most exciting frontier involves cognition. In a population-based retrospective cohort of older adults with type 2 diabetes, initiation of SGLT2 inhibitors compared with DPP-4 inhibitors was associated with about a 20% lower risk of incident dementia (aHR 0.80), based on propensity score–weighted Cox models14. Genetic studies using Mendelian randomization also support a neuroprotective role of SGLT2 inhibition, linking genetically proxied inhibition to longevity, better cognition, and brain structural markers of aging15. A prospective cohort study of henagliflozin significantly improved montreal cognitive assessment (MoCA) scores and reduced plasma p-tau181 levels in patients with type 2 diabetes and cognitive impairment, independent of glycemic control16. These results highlight the potential of SGLT2 inhibitors to expand their organ-protective profile to the brain. Integrating these diverse findings, SGLT2 inhibitors have clearly transcended their original role as glucose-lowering drugs. They are now widely recognized as systemic organ-protective therapies, with benefits across cardiovascular, renal, hepatic, ophthalmic, and neurological domains. Nevertheless, important gaps remain. Evidence in frail elderly with sarcopenia, cognitive impairment, or functional dependence is still limited. While early data are promising, dedicated large-scale randomized controlled trials and mechanistic studies on microvascular and neuroprotective pathways are needed. Finally, combination approaches, particularly with GLP-1 receptor agonists, represent an exciting frontier for synergistic organ protection. In conclusion, after more than a decade of clinical use, early concerns have been supplanted by consistent evidence of safety and multi-organ benefit. Together, findings from numerous studies position SGLT2 inhibitors as transformative therapies, capable of reshaping diabetes care in an era of population aging, where systemic protection is increasingly vital. K.K. has been an advisor to, received honoraria for lectures from, and received scholarship grants from Novo Nordisk Pharma, Sanwa Kagaku, Taisho Pharma, Sumitomo Pharma, Mitsubishi Tanabe Pharma, Nippon Boehringer Ingelheim. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.