A class of drugs already taken by millions of people with heart failure and kidney disease may be quietly accomplishing something extraordinary: stopping type 2 diabetes before it begins. In a narrative review published in Diabetes Therapy, researchers at the Medical University of Warsaw conclude that sodium–glucose cotransporter 2 (SGLT2) inhibitors—the oral agents better known as gliflozins—cut the incidence of new-onset type 2 diabetes by roughly one fifth in adults with prediabetes, with the most consistent benefit seen in patients who also live with cardiovascular or chronic kidney disease. The review, authored by Anna C. Owczarczyk-Durma, Monika Blicharska and Leszek Czupryniak, synthesizes randomized trials, prespecified subgroup analyses and meta-analyses published between January 2007 and February 2026. Its verdict is deliberately double-edged: the biological and clinical signal is strong enough to justify dedicated prevention trials, yet not mature enough to rewrite treatment guidelines. With roughly one in eight adults worldwide now carrying prediabetes, the question of whether a daily pill can intercept one of modern medicine’s largest epidemics has become one of the most consequential in endocrinology.

The scale of the problem gives the question its urgency. According to the 11th edition of the International Diabetes Federation Diabetes Atlas, the age-standardized global prevalence of prediabetes among adults aged 20 to 79 reached 12.0% for impaired glucose tolerance and 9.2% for impaired fasting glucose in 2024—substantially higher than the figures of 9.1% and 5.8% reported three years earlier. Prediabetes is far from benign: meta-analytic data indicate that 5–10% of affected individuals convert to overt type 2 diabetes each year, with rates climbing to approximately 15–19% in those carrying both glycemic abnormalities simultaneously. The condition also operates as an independent cardiovascular risk factor: a landmark meta-analysis in The BMJ linked prediabetes to increased cardiovascular morbidity and all-cause mortality, while microvascular damage—retinopathy, neuropathy and nephropathy—has been documented even before diabetes is formally diagnosed. Lifestyle modification that combines diet, exercise and weight loss remains the cornerstone of management, famously reducing diabetes incidence by 58% in the Diabetes Prevention Program, but long-term adherence is notoriously difficult. That gap between what works in trials and what patients sustain in practice is precisely why pharmacologists keep hunting for a preventive drug.

The mechanism of SGLT2 inhibitors explains why endocrinologists find the class so intriguing. These agents selectively block the SGLT2 transporter in the proximal convoluted tubule of the kidney, slashing renal reabsorption of filtered glucose and flushing the surplus into the urine, a state called glycosuria. At standard therapeutic doses, patients excrete approximately 60 to 90 grams of glucose per day—an energy loss equivalent to 200–300 kilocalories that drives a modest but sustained weight reduction. Because the pathway is insulin-independent, hypoglycemia risk is exceptionally low with monotherapy. The glycosuric drain lowers serum insulin and raises plasma glucagon, while secondary effects ripple outward: tissue insulin sensitivity improves, renal cortical inflammation attenuates and tubular oxygenation recovers. Osmotic diuresis and natriuresis also shrink plasma volume, lowering blood pressure even in people without diabetes. The result is a drug whose metabolic fingerprint—glycemic unloading, caloric loss, natriuresis and ketone-fueled myocardial energetics—maps almost perfectly onto the pathophysiology of prediabetes, a state defined by insulin resistance in skeletal muscle, liver and adipose tissue and, increasingly recognized, the brain itself.

The preventive evidence rests on two contemporary meta-analyses and a cascade of trial subgroups. A meta-analysis by Mori and colleagues pooling more than 5,500 participants found a significantly lower incidence of new-onset type 2 diabetes among prediabetic adults treated with SGLT2 inhibitors. An updated analysis by Singh and colleagues, incorporating five randomized controlled trials with 6,752 participants, confirmed a 19% relative reduction in new-onset diabetes (hazard ratio 0.81; 95% confidence interval 0.69–0.94) with no statistical heterogeneity. The most influential dataset comes from the DAPA-HF heart failure trial: in a prespecified exploratory analysis, patients with prediabetes given dapagliflozin developed diabetes at a rate of 4.9% over a median 18 months, versus 7.1% on placebo—a 32% relative risk reduction (hazard ratio 0.68; p = 0.019). Wider context arrived in 2025, when an analysis of seven cardiovascular and renal trials reported that dapagliflozin reduced new-onset diabetes by 33% in heart failure populations, and that SGLT2 inhibitor therapy lowered new diagnoses by 26% across the combined cardiorenal spectrum—figures that place gliflozins in the same conversation as metformin, the field’s long-standing pharmacological benchmark.

Perhaps the most striking findings concern the brain. Insulin resistance in prediabetes extends to the hypothalamus, the neural hub governing energy intake and peripheral glucose handling, and until recently no drug had demonstrated a robust ability to reverse this cerebral insulin resistance. In a randomized, double-blind, placebo-controlled phase 2 trial, Kullmann and colleagues gave 40 adults with prediabetes 25 milligrams of empagliflozin daily for eight weeks and used functional magnetic resonance imaging after intranasal insulin administration to probe hypothalamic signaling. The treated group showed restored hypothalamic insulin sensitivity, accompanied by reductions in fasting glycemia and hepatic fat content—positioning empagliflozin among the first agents shown to modulate central insulin action in humans. Weight effects are more modest but measurable: in a randomized study of 24 adults with prediabetes, dapagliflozin trimmed roughly 3 kilograms, or 3.7% of body weight, versus 1 kilogram on placebo, with parallel improvements in waist circumference, fasting plasma glucose, serum uric acid and insulin resistance indices.

Beyond glucose, gliflozins deliver multi-organ protection that strengthens the preventive case. In DAPA-CKD, dapagliflozin reduced the composite renal endpoint by 39% in patients with chronic kidney disease with or without type 2 diabetes, while EMPA-KIDNEY demonstrated a 28% reduction in kidney disease progression or cardiovascular death with empagliflozin; a 2026 meta-analysis confirmed clear renal benefits in patients without diabetes, with slower declines in glomerular filtration rate and reduced albuminuria. The liver tells a similar story. Metabolic dysfunction-associated steatotic liver disease, the renamed fatty liver condition tightly coupled to insulin resistance and prediabetes, responds favorably to the class: randomized trials of empagliflozin reduced steatosis and fibrosis on transient elastography both with and without diabetes, an updated 2025 meta-analysis of 16 randomized trials covering 11,300 participants confirmed significant improvements in fibrosis markers, and a large real-world cohort study associated the therapy with improved ten-year survival and markedly lower progression to advanced liver disease. Empagliflozin has even improved coronary microvascular function and cardiac contractility in prediabetic mice, anchoring the cardiovascular benefits documented across the EMPA-REG OUTCOME, CANVAS, DECLARE-TIMI 58, EMPEROR and DELIVER programs.

Yet the review is unsparing about the gaps. The EMPEROR-Reduced and EMPEROR-Preserved trials recorded only numerically lower—statistically non-significant—rates of new-onset diabetes in their prediabetes subgroups, 11.2% versus 12.6% and 12.0% versus 14.0% respectively, and empagliflozin failed to lower HbA1c in prediabetic participants. The PRE-D trial, which randomized 120 adults with prediabetes to dapagliflozin, metformin, exercise or control for 13 weeks, documented a 17.1% reduction in the mean amplitude of glycemic excursions but was never powered to detect changes in diabetes incidence. Crucially, no large, adequately powered randomized controlled trial designed primarily to evaluate diabetes prevention in isolated prediabetes has been completed with any SGLT2 inhibitor. The existing evidence derives almost entirely from post-hoc analyses of cardiovascular and renal outcome studies, and a methodological shadow hangs over all of them: because gliflozins transiently lower glucose readings, they may simply mask diabetes rather than prevent it, and whether any benefit persists after drug discontinuation—a question answerable only through formal washout periods—remains unknown. The optimal duration of preventive therapy is likewise undefined, with most studies spanning merely one to two years.

Safety and economics add further friction. The class is generally well tolerated but carries an increased risk of genitourinary infections, particularly mycotic genital and uncomplicated urinary tract infections, driven by the same drug-induced glycosuria that defines its benefit—side effects absent from metformin and pioglitazone. Severe events such as pyelonephritis and urosepsis are not significantly more frequent than with comparator agents. The most feared complication, euglycemic diabetic ketoacidosis—a metabolic acidosis with ketosis despite normal or only mildly elevated glucose—remains rare, below 0.2% per year in type 2 diabetes and presumed rarer still in prediabetic individuals who retain endogenous insulin secretion, though it is not zero and can evade diagnosis precisely because glucose levels appear reassuring. Early signals linking canagliflozin to lower-limb amputation and bone fractures were not replicated in subsequent studies and are no longer considered a class effect. Cost presents a parallel barrier: prediabetic patients typically fail reimbursement criteria for these comparatively expensive drugs, although generic dapagliflozin and empagliflozin formulations expected in several European markets between 2026 and 2028 may improve affordability.

Where does this leave gliflozins in the prevention hierarchy? Metformin, validated by the Diabetes Prevention Program, reduced diabetes incidence by 31% over a median 2.8 years with a number needed to treat of approximately 14, while pioglitazone achieved a number needed to treat of 8 in the ACT NOW study and acarbose cut diabetes incidence from 42% to 32% in STOP-NIDDM. The incretin therapies outperform them all on weight: in the SCALE trial, liraglutide produced diabetes in 2% of prediabetic participants versus 6% on placebo, while tirzepatide in SURMOUNT-1 delivered up to a 22% body-weight reduction and returned 95.3% of prediabetic participants to normoglycemia after roughly 1.4 years—although neither program was designed primarily as a prevention study. Against this backdrop, the 2026 American Diabetes Association Standards of Care stop short: SGLT2 inhibitors are not endorsed as first-line preventive therapy in isolated prediabetes, but for patients who also have heart failure, atherosclerotic cardiovascular disease or chronic kidney disease, treating those conditions with a gliflozin may simultaneously delay diabetes as a clinically meaningful secondary benefit. The authors’ conclusion is unambiguous: the class now needs a definitive prevention trial with adequate follow-up and washout before the pill that rescues failing hearts and kidneys can also be prescribed to stop diabetes in its tracks.