T2D develops slowly over years with the development of both insulin secretion impairment and insulin resistance (see Fig. 4.1 ). Acute glucotoxicity is a state of extreme hyperglycemia that is thought to be a driver of β-cell functional decline. Studies have initiated insulin at the early onset of T2D with the hope of restoring the early-phase insulin secretion and reaching remission. It has been suggested that the administration of intense insulin can lead to a “β-cell rest,” lowering the demand for insulin secretion. The ORIGIN trial randomized 12,537 participants with IGT, IFG, or early T2D with cardiovascular (CV) risk factors to receive insulin glargine (target fasting blood glucose of <95 mg per deciliter) or standard care, plus n-3 fatty acids or placebo over a follow-up of 6.2 years. Although the CV outcomes rates were similar in both groups, the incidence of new-onset T2D was reduced by 28% in those with prediabetes in the insulin arm. There were, however, higher rates of hypoglycemia and weight gain in the insulin glargine group. A significant legacy effect was noted with a 2-year passive follow-up as well.

A study from Taiwan evaluated the effect of insulin administration to patients with newly diagnosed T2D with severe hyperglycemia. Patients were initially treated with intensive insulin administration over 10 to 14 days, then were randomized (50/50) for a span of 6 months of either oral antidiabetic drugs (OADs) or insulin (basal insulin twice per day). Following the initial 6 months, the insulin group discontinued their insulin and were transitioned to OADs. Patients were then managed per national guidelines for the following 5 years. At the end of the 5 years, the HbA1c was lower in the initial insulin group, as was the rate of remission: 27.3% in the insulin group compared with 5.9% in the OAD group ( P =.048).

A meta-analysis of seven studies ( n = 839 participants) looked at newly diagnosed T2D patients who were administered a short-term intensive insulin treatment for 14 to 21 days either through multiple daily injections or continuous subcutaneous insulin infusion. Four of these trials evaluated the long-term remission rates following the short-term intervention. In these studies, the drug-free remission after 3 months of follow-up was 66.2% (292 of 441 patients), after 6 months 58.9% (222 of 377 patients), after 12 months 46.3% (229 or 495), and after 24 months 42.1% (53 of 126 patients). As evident in this data, the remission after short-term intensive insulin did diminish over time.

Insulin secretion can be stimulated by sulfonylureas and nateglinide. The now-unavailable sulfonylurea, tolbutamide, was tested for diabetes prevention in the 1960s. These studies preceded current definitions of prediabetes and T2D, so the terms used to describe these trials have slightly different definitions than those used today. The Bedford (UK) trial of tolbutamide evaluated 241 adults with IGT randomized to tolbutamide or placebo and to two dietary groups in a 2-by-2 factorial design. Over 10 years of follow-up, 15% of participants developed T2D, but neither the drug nor diet interventions reduced the progression to T2D. The Malmöhus trial of tolbutamide also studied in 147 men with IGT in Malmöhus County, Sweden. All study participants were given dietary advice and randomly assigned to tolbutamide, matching placebo, or neither medication. Tolbutamide was reported to prevent T2D when the trial was analyzed by estimated drug adherence but not when analyzed by the more rigorous intention-to-treat principle.

The Nateglinide and Valsartan in Impaired Glucose Tolerance Outcome Research (NAVIGATOR) was a 2-by-2 factorial 5-year RCT of the short-acting insulin secretagogue nateglinide of the meglitinide class and the angiotensin receptor blocker valsartan in 9306 participants with IGT, elevated fasting glucose, and CVD or CVD risk factors. Nateglinide did not affect T2D incidence (HR = 1.07; 95% CI, 1.00–1.15) and was associated with increased frequency of hypoglycemia and modest weight gain. Based on these RCTs, it appears that sulfonylureas and nateglinide, aimed primarily at enhancing insulin secretion, have limited utility for preventing or delaying T2D in high-risk adults.

The Diabetes Prevention Program enrolled 3234 patients with IFG/IGT to placebo, metformin (850 mg twice/day), or lifestyle modifications (goal of 150 minutes of activity per week, and goal of 7% weight loss) over 2.8 years of follow-up (see also Chapters 3 and 17 ). The metformin group had a 31% reduction in the incidence of diabetes (95% CI, 17%–43%) and the lifestyle group had a 58% reduction (95% CI, 48%–66%), compared to the placebo. The Diabetes Prevention Program Outcomes Study (DPPOS) evaluated continued follow-up of the DPP participants over a 10-year period. Eighty-eight percent of those enrolled in the DPP continued on to be part of the DPPOS. In this long-term follow-up cohort, diabetes incidence in the metformin group was reduced by 18% and in the lifestyle group by 34% when compared to the placebo.

A review of the literature published between 1998 and 2017 on the use of metformin for diabetes prevention suggested that metformin is an effective tool to prevent or delay the development of T2D. They demonstrated that the relative risk reduction of T2D development was highest in patients with existing diabetes risk factors such as IGT, IFG, age ≤60 years, body mass index (BMI) ≥35 kg/m ² , and gestational diabetes. Metformin was additionally noted to be cost-effective in 11 analyses and even demonstrated some cost savings. Metformin for the prevention of T2D and its associated complications in persons at risk for increased risk for the development of T2D was recently systematically reviewed and meta-analyzed, including 20 RCTs randomizing 6774 participants. One trial contributed 48% of all participants. The duration of intervention in the trials varied from 1 to 5 years. In the final analysis, metformin compared with placebo or diet and exercise reduced or delayed the risk of T2D in people at increased risk for the development of T2D (moderate-quality evidence). However, metformin compared to intensive diet and exercise did not reduce or delay the risk of T2D (moderate-quality evidence). Likewise, the combination of metformin and intensive diet and exercise compared to intensive diet and exercise only neither showed an advantage or disadvantage regarding the development of T2D (very low-quality evidence).

TZDs improve cell responsiveness to insulin and are believed to also preserve insulin secretion. It is postulated that β-cell function improves with the intake of TZDs through a variety of mechanisms, such as decreasing glucotoxicity, lipotoxicity, inflammatory cytokines, oxidative damage, and lowering of free fatty acids, leading to prolonged survival of β-cells. In the Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medication (DREAM) Trial, 5269 patients with IGT and/or IFG were randomized to receive rosiglitazone or a placebo for 3 years. β-cell function was measured using the validated assessments of insulinogenic index and proinsulin concentration. This international trial demonstrated that 25% of patients developed diabetes in the placebo arm compared with 10.6% in the rosiglitazone group (HR 0.38; P < 0.0001), a 62% reduction in the risk of progression to T2D with a daily intake of 8 mg of rosiglitazone. The effects of rosiglitazone were not influenced by baseline glucose control. The study also showed a 70% to 80% likelihood to revert from IFG/IGT to normal glucose tolerance during that time frame. This regression to normoglycemia, however, appeared to be similar to lifestyle modifications. Though the waist:hip ratio was unchanged in the rosiglitazone group, there was a weight gain of 2.2 kg during the 3 years. As expected, there did not seem to be a persistent effect of the TZD use 6 months after its discontinuation with the closure of the DREAM trial.

The ACT NOW study evaluated 602 patients with IGT randomized to receive pioglitazone 30 mg per day or a placebo and followed for 2.4 years. Although there was significant weight gain and edema in the TZD group, there was a reduction of the progression to diabetes by 72%. The PREVENT-J, Pioglitazone and Sulfonylurea Remission from Type 2 Diabetes Mellitus Development and Antiatherosclerosis in Japan, study evaluated patients with early-stage T2D and randomized them to lifestyle modifications, pioglitazone, or sulfonylurea treatment over an 18-month period (phase 1) and then subsequently discontinued the medications over the following 18-month period (phase 2). Enrolled patients had diabetes duration for less than 5 years and had either an HbA1c < 7.9% or < 7.4% (with no prior diabetes treatment, or were on either metformin or alpha-glucosidase inhibitor, respectively). Throughout phase 2, the pioglitazone group had the highest proportion of patients who were able to maintain an HbA _1c < 6.2%. In this study, the remission of diabetes was best sustained after discontinuation of the TZD in comparison to the sulfonylurea and lifestyle modifications. In a Cochrane Systematic Review, pioglitazone reduced or delayed the development of T2D in people at increased risk of T2D compared with a placebo (low-certainty evidence) and compared with no intervention (moderate-certainty evidence). It is unclear whether the effect of pioglitazone is sustained once discontinued. Pioglitazone compared with metformin neither showed advantage nor disadvantage regarding the development of T2D in people at increased risk (low-certainty evidence). The data and reporting of all-cause mortality, serious adverse events, and micro- and macrovascular complications were generally sparse. None of the included studies reported on quality of life or socioeconomic effects.

The incretins, including glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP), are released in a glucose-dependent manner and behave as insulin secretagogues. GLP-1 also contributes to the inhibition of glucagon secretion, gastric emptying delays, and the promotion of centrally mediated appetite suppression, leading to sustained weight loss. For example, high-dose injectable, once-weekly semaglutide (2.4 mg/wk) plus lifestyle intervention was shown to reduce weight by >12% over 68 weeks in individuals with obesity or overweight compared with lifestyle alone in STEP 1 (STEP 1: Research Study Investigating How Well Semaglutide Works in People Suffering From Overweight or Obesity). In the STEP 2 trial assessing the efficacy of once weekly semaglutide in adults with overweight or obesity and T2D, the proportion of individuals in each group who achieved diabetes remission with HbA _1c levels of ≤6.5% at week 68 were 72.0%, 63.1%, and 14.2% for semaglutide 2.4, semaglutide 1.0, and a placebo, respectively; and the proportion of those achieving HbA _1c <7.0% at week 68 were 82.3%, 76.0%, and 25.8% for semaglutide 2.4, semaglutide 1.0, and a placebo, respectively. Most recently, semaglutide 2.4 mg once weekly was seen to be superior to a placebo in reducing the incidence of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke (hazard ratio, 0.80; 95% CI, 0.72–0.90; P <.001) in adults with preexisting cardiovascular disease and overweight or obesity but without diabetes in the SELECT trial. Novel agents targeting multiple obesogenic pathways thought to act synergistically are also in development and may offer even greater weight reduction. SURPASS-2 (a study of tirzepatide [LY3298176] vs. semaglutide once weekly as add-on therapy to metformin in participants with type 2 diabetes), testing treatment with the combined glucose-dependent insulinotropic polypeptide receptor agonist (RA)/GLP-1 RA tirzepatide compared with semaglutide 1.0 mg subcutaneous injection (SQ) weekly in patients with T2D, showed that tirzepatide reduced body weight to a greater degree than the 1.0-mg dose of semaglutide as a secondary end point. In the SURPASS-1 study, tirzepatide treatment reduced HbA _1c to less than 6.5% in 81% to 86% of patients (compared with 10% on a placebo) and to an HbA _1c less than 5.7% in 31% to 52% of patients (compared with 1% on placebo). The SURPASS-4 trial assessed glycemic efficacy and cardiovascular safety in patients with T2D and elevated cardiovascular risk and found that tirzepatide, compared with glargine, demonstrated greater and clinically meaningful HbA _1c reduction (−2.43% [SD 0.05] with 10 mg and −2.58% [0.05] with 15 mg, vs. −1.44% [0.03] with glargine) with a lower incidence of hypoglycemia at week 52 and was not associated with excess cardiovascular risk compared with glargine (HR, 0.74; 95% CI, 0.51–1.08). Data from the pooled SURPASS clinical program in adults with T2D showed that diabetes remission was common with tirzepatide treatment, with 44% and 37% of individuals treated with tirzepatide achieving HbA _1c 5.7% to 6.5% and <5.7%, respectively. GLP-1 RA effects on body weight are most prominent on visceral and ectopic body fat, with a two to threefold greater effect on reducing these diabesogenic risk factors compared with other body fat compartments. GLP-1 RAs and related compounds are therefore highly promising agents for prevention and/or remission of T2D.

Liraglutide was evaluated in a 20-week European study where 564 patients with obesity but without T2D were randomized to placebo versus liraglutide versus orlistat. Of these patients, 31% enrolled had IFG. At 20 weeks, those treated with liraglutide showed a decrease of 84% to 96% in the prevalence of prediabetes. HbA _1c also showed a reduction of 0.14% to 0.24% in the liraglutide group and the metabolic syndrome decreased by >60%. The effect of liraglutide on fat distribution and β-cell function was assessed in a study looking at 62 metformin-treated patients with obesity and prediabetes or new-onset T2D. Patients were randomized to liraglutide or lifestyle modifications and both groups achieved comparable weight loss. Visceral adipose tissue was significantly reduced ( P =.028) and β-cell function (measured as β-index) improved ( P = 0.021) in the liraglutide group. Since weight loss and HbA _1c reduction were the same in the liraglutide and lifestyle groups, the analysis described the β-cell function improvement in the treatment group as being an independent outcome. Therefore early use of the GLP-1 agonist can diminish β-cell dysfunction in prediabetes and prevent progression to T2D.

As part of the SCALE Obesity and Prediabetes Study Group, patients with prediabetes and a BMI of 30 kg/m ² or 27 kg/m ² (with comorbidities) were randomized to either daily liraglutide 3 mg or placebo over a 160-week period. In this study, liraglutide reduced the risk of new-onset of T2D by 80% relative to placebo (HR, 0.21; 95% CI, 0.13–0.34) and 66% of individuals in the treatment group showed regression from prediabetes to normoglycemia. With liraglutide, the time to onset of T2D was 2.7 times longer than with placebo over the 160 weeks.

Many other GLP-1 RA-based compounds that combine GLP-1 RA with other targets, such as glucose-dependent insulinotropic polypeptide (GIP) and glucagon receptor (GR) are currently in research and development (phase I/II/III) and hold additional promise for substantial cardiometabolic benefits. The triple agonist (GLP-1/GIP/GR) retatrutide recently demonstrated up to 24.2% body weight loss over 48 weeks in adults with overweight or obesity without diabetes and significant reductions in HbA _1c in patients with T2D compared with dulaglutide (up to–2.02% [SD 0.11] for the 12 mg group vs.–1.41% [0.12] for the dulaglutide group, P <.001). The dual GLP-1/GR agonist, survodutide, recently demonstrated significant efficacy for weight loss in a phase II study (up to −14.9% [−16.9 to −13.0] for 4.8 mg vs. −2.8% [−4.9 to −0.7] for a placebo). These compounds are also being tested for other conditions associated with obesity and diabetes including metabolic dysfunction-associated steatohepatitis (MASH), now the leading cause for liver transplantation. Both survodutide and tirzepatide recently showed improvement in MASH without worsening of fibrosis in patients with biopsy-confirmed MASH. Other compounds in development for obesity include oral agonists of GLP-1 such as orforglipron and oral semaglutide (estimated mean body weight change from baseline to week 68%–15.1% [SE 0.5] with oral semaglutide 50 mg vs.–2.4% [0.5] with placebo), as well as the GLP-1 RA/GIP antagonist, maridebart cafraglutide.

Sodium glucose cotransporter 2 (SGLT2) inhibitors lower the plasma glucose concentration via a mechanism independent of insulin secretion and insulin action. By inhibiting renal glucose reabsorption, they cause glucosuria, leading to a decline in fasting and postprandial plasma glucose concentration. Despite the lack of a direct effect on the β-cell, preclinical studies in animal models of diabetes have demonstrated improved β-cell function secondary to reduction of the plasma glucose concentration and amelioration of glucotoxicity. SGLT2i have been demonstrated to reduce markers of visceral adiposity and improve adipose-tissue function. Therefore SGLT2i have been postulated to prevent development of T2D.

The DAPA-CKD and DAPA-HF trials evaluated the effect of dapagliflozin on reduction of chronic kidney disease and heart failure as primary outcomes in patients with or without T2D. A pooled analysis from both trials demonstrated that dapagliflozin, compared with placebo, reduced the development of new-onset T2D (HR, 0.67; 95% CI, 0.51–0.88; p =.0040) with a greater effect in younger patients and those with elevated blood pressure. Due to the minimal difference in mean HbA _1c between placebo and dapagliflozin in the trials, the mechanisms of benefit were suggested to be a potential direct positive impact of dapagliflozin on β-cell function as well as increased sensitivity of insulin in the periphery. The PRE-D trial was a multiarm study comparing 1:1:1:1 randomization of patients with overweight/obesity and prediabetes to dapagliflozin, metformin, interval-based exercises, and a control arm over 13 weeks. The evaluation of glycemic variability was conducted using a continuous glucose monitor with a measurement of mean amplitude of glycemic excursions (MAGE). As compared to metformin, dapagliflozin yielded positive improvements in glycemic variability (dapagliflozin vs. metformin, MAGE reduction of 17.2% [ p = 0.041], and lower incidence of diabetes [2.1% vs. 3.9%; HR, 0.53; 95% CI, 0.36–0.77]).

However, the prevention of T2D development may not be universal among the SGLT2i. Studies of empagliflozin in heart failure did not individually reduce new diagnosis of T2D in those without diabetes. Neither the EMPEROR-reduced nor the EMPORER-preserved trial showed a reduction in new-onset T2D with empagliflozin (HR, 0.86; 95% CI, 0.62–1.19 and HR, 0.84; 95% CI, 0.65–1.07, respectively). This might be due to inadequate power; therefore a pooled analysis might produce statistically significant results, as opposed to a difference in the SGLT2 inhibitors per se. Furthermore, much like other studies of diabetes prevention, the question remains whether the effects seen are due to masking or true prevention.

The XENDOS study was a 4-year double-blind RCT of 3305 patients with obesity (BMI ≥ 30 kg/m ² ) with normoglycemia (79%) or IGT (21%) randomized to lifestyle modification with either orlistat 120 mg or a placebo three times daily. The incidence of T2D with orlistat was 6.2% compared with 9.0% with a placebo, resulting in a 37.3% reduced risk of progression to T2D in the orlistat group (HR, 0.63; 95% CI, 0.45–0.86).

The progression of prediabetes to T2D has been shown to be preventable in patients with metabolic syndrome and obesity with weight loss of 5% to 10%. An RCT looked at overweight/obese patients with prediabetes and/or metabolic syndrome and randomized participants to Phentermine 7.5 mg/Topiramate ER 46 mg, Phen 15 mg/TPM ER 92 mg, or placebo plus lifestyle modifications for 108 weeks. The annualized incidence rate for T2D was reduced by 70.5% in those receiving 7.5/46 dosage, and by 78.7% in the 15/92 dosage, versus placebo ( P <.05). The ability to prevent diabetes development in the treatment arms was related to the degree of weight loss.

The effect of lorcaserin on the prevention and remission of T2D in patients with overweight and obesity was tested in the CAMELLIA-TIMI 61 trial. Lorcaserin decreased the risk for incident diabetes, induced remission of hyperglycemia, and reduced the risk of microvascular complications in obese and overweight patients, supporting the role of lorcaserin as an adjunct to lifestyle modification for chronic management of weight and metabolic health. Unfortunately, in February 2020, the FDA requested that the manufacturer of lorcaserin voluntarily withdraw the drug from the US market because a safety clinical trial showed an increased occurrence of cancer, and the medication is no longer available for use.

Alpha-glucosidase inhibitors have no known direct effects on insulin secretion or action, but lower postprandial hyperglycemia by slowing glucose absorption from the gut. They have been tested for T2D prevention under the hypothesis that dampening postprandial glucose excursions in persons with IGT may help preserve insulin secretion or action, thus preventing incident T2D. The STOP-NIDDM trial was a double-blind, placebo-controlled study that evaluated patients with IGT and randomized them either to acarbose or placebo for 3 years. Findings demonstrated that the intake of acarbose lowered the progression to T2D by 25% and fostered the reversion of IGT to normal glucose tolerance ( p <.0001). The voglibose Ph-3 Study Group evaluated the use of the alpha-glucosidase inhibitor voglibose in a high-risk Japanese population with IGT. The study demonstrated that in contrast to placebo, those treated with voglibose had a significantly reduced risk for progressing to T2D (50 of 897 vs. 106 of 881; HR, 0.59; 95% CI, 0.43–0.82; p =.001). Also, they noted a statistically significant achievement of normoglycemia in the voglibose group compared with the placebo.

A meta-analysis of acarbose monotherapy for prevention of T2D from prediabetes evaluated eight randomized controlled trials which included 2628 participants over at least 3 years. They demonstrated that patients randomized to acarbose had a decreased incidence of T2D (number needed to treat [NNT] to prevent one case of T2D was 6.7 over 3 years). Notably, the T2DM incidence was lower in the Eastern group (Asia) than in the Western group (North America and Europe) (NNT 5.9 Eastern, NNT 11.1 Western; P <.0001). Alpha-glucosidase inhibitors work best on individuals who consume higher carbohydrate intake. The analysis suggested that the Eastern diet contained a greater proportion of starchy carbohydrates and therefore had a greater benefit from acarbose. A recent systematic review concluded that alpha-glucosidase inhibitors may prevent or delay the development of T2D in people with IGT. There is no firm evidence that alpha-glucosidase inhibitors have a beneficial effect on cardiovascular mortality or cardiovascular events.

Vitamin D deficiency and its role in the development of T2D is a controversial topic. Studies have shown that vitamin D is associated with improvement in β-cell function, reducing the risk of the metabolic syndrome and inflammation. However, it is not known if supplementation with vitamin D actually prevents progression to T2D. The D2d Research Group randomized 2423 patients with prediabetes to 4000 IU of vitamin D ₃ or placebo, regardless of their underlying vitamin D level. After a median of 2.5 years of follow-up, there was no statistically significant lowering of the risk of T2D in the vitamin D group (HR, 0.88; 95% CI, 0.75–1.04; P =.12). A meta-analysis conducted in 2020 evaluated 10 studies with a total of 34,882 patients, analyzing both cross-sectional studies and RCTs. There was a significant association between lower vitamin D levels and T2D in the cross-sectional studies (pooled OR, 1.77 [1.51–2.07, P <.001] but a nonsignificant 11% [ P = 0.08] decrease in risk of T2DM for patients receiving vitamin D was observed in the RCTs). In contrast, a different meta-analysis evaluated eight randomized controlled studies with a total of 4896 patients with prediabetes who received vitamin D supplementation versus a placebo. They detected a statistically significant reduction in the risk of T2DM development with supplementation of vitamin D (risk ratio [RR], 0.89; 95% CI, 0.80–0.99; I ² = 0%); however, this was significant only in patients without obesity. They also noted a 48% increase in reversion of prediabetes to a normal glucose state in the vitamin D supplementation groups (RR, 1.48; 95% CI, 1.14–1.92; I ² = 0%).