Type 2 Diabetes Mellitus Remission, Dream or Reality? A Narrative Review of Current Evidence and Integrated Care Strategies

Abstract

Type 2 diabetes mellitus (T2DM) is a global health priority, with an estimated 629 million people projected to be affected by the year 2045. T2DM significantly increases the risk of atherosclerotic cardiovascular disease and other complications. Hyperglycaemia imprints early molecular and cellular changes, often termed “metabolic memory”, predisposing individuals to long-term microvascular and macrovascular complications, even after glycaemic normalisation. T2DM remission is increasingly recognised as an achievable target, offering substantial benefits such as reduced morbidity, improved quality of life, and preservation of beta-cell function. Among therapeutic options, metabolic surgery (MS) demonstrates the most significant impact, particularly for long-term outcomes. MS induces profound hormonal changes, including increased glucagon-like peptide 1 (GLP-1) levels and improved bile acid metabolism, alongside reductions in ectopic fat in the liver and pancreas, which improve insulin sensitivity and secretion. However, intensive lifestyle and pharmacological interventions, such as GLP-1 receptor agonists and glucose-dependent insulinotropic polypeptide/glucagon-like peptide 1 dual agonists like tirzepatide, also show promise, particularly when implemented early in the disease course. Predictors of sustained remission include younger age, shorter diabetes duration, lower baseline HbA1c, absence of insulin use, fewer medications and greater total weight loss percentage. Emerging tools such as the DiaRem score, machine learning models, and biomarkers like FGF-21 enhance patient stratification and predict remission likelihood. This narrative review explores the mechanisms and therapeutic options for T2DM remission, evaluates their impact on long-term outcomes and highlights the importance of early, multidisciplinary, and personalised interventions to optimize remission and improve metabolic health.

Key Summary Points

A recent consensus report, published jointly by the Endocrine Society, the European Association for the Study of Diabetes, Diabetes UK, and the American Diabetes Association, defined type 2 diabetes mellitus (T2DM) remission as achieving and maintaining glycated haemoglobin levels below 48 mmol/mol (6.5%) for at least 3 months, without the need for glucose-lowering medications.

T2DM remission is increasingly recognised as an achievable target, offering substantial benefits such as reduced morbidity, improved quality of life, and preservation of beta-cell function.

Thanks to advances in metabolic surgery, intensive lifestyle interventions, and novel pharmacological agents such as dual incretin receptor agonists, remission has become a realistic and achievable outcome.

Personalised and durable therapeutic solutions and patient-centred treatments are essential to enhancing long-term diabetes management, remission and sustained metabolic benefits.

Introduction

Type 2 diabetes mellitus (T2DM) is a global health priority, with an estimated 629 million people projected to be affected by the disease by 2045, according to Captieux et al. [1]. Diabetes itself poses an independent risk for atherosclerotic cardiovascular disease—including coronary heart disease, cerebrovascular disease, and peripheral artery disease—as well as chronic kidney disease (CKD) [2]. Furthermore, diabetes and its related health complications create a significant financial burden on individuals and society. According to the American Diabetes Association, the annual cost of diagnosed diabetes in the USA reached $413 billion in 2022, which includes direct healthcare expenses and decreased productivity [3]. Diabetes remission is increasingly recognised as a top priority for healthcare professionals and patients. It is also becoming a common outcome measure in research studies. Diabetes is now widely acknowledged as a highly heterogeneous disease, with variability in clinical characteristics, progression, therapeutic responses, and susceptibility to complications [4]. Hyperglycaemia can sometimes arise as a result of transitory events in predisposed individuals, such as gestational diabetes, glucocorticoid-induced diabetes, and stress hyperglycaemia. These conditions are often associated with T2DM through transient insulin resistance (IR) and an increased risk of developing T2DM later. Conversely, voluntary or unintended weight loss may lead to discontinuation of glucose-lowering therapies. While early intensive insulin therapy is well documented for reversing glucose toxicity, recent attention has shifted to antidiabetic agents, such as glucagon-like peptide 1 (GLP-1) receptor agonists and sodium-glucose cotransporter 2 inhibitors (SGLT2-Is), owing to their pleiotropic effects, efficacy in promoting weight loss, and ability to restore glycaemic control with minimal adverse effects [5–9]. A recent consensus report, published jointly by the Endocrine Society, the European Association for the Study of Diabetes, Diabetes UK, and the American Diabetes Association, proposed terminology and a framework to guide future research and clinical guidelines on diabetes remission [10]. Specifically, remission is characterised by achieving and maintaining glycated haemoglobin (HbA1c) levels below 48 mmol/mol (6.5%) for at least 3 months, without the need for glucose-lowering medications [11]. When HbA1c measurement is unreliable—such as in cases of haemoglobin variants or altered erythrocyte survival, estimated HbA1c derived from continuous glucose monitoring can be used as an alternative.

Diabetes remission signifies the absence of active T2DM; therefore, patients should not exhibit symptoms or develop new complications. This concept challenges the traditional belief that T2DM is a permanent, inevitably progressive condition. Remission can only be diagnosed after all glucose-lowering medications have been discontinued for a sufficient period to allow the effects of the drugs to wear off and to assess HbA1c levels, reflecting blood glucose exposure over the prior 3 months. Certain clinical contexts necessitate clarification. Glucose-lowering medications, such as GLP-1 receptor agonists (for weight control), SGLT2 inhibitors (for heart failure or CKD), or metformin (for weight maintenance), may be prescribed for conditions other than diabetes. In such cases, diabetes remission cannot be diagnosed, even if the medications are not intended for glycaemic control. This restriction does not apply to lifestyle interventions or metabolic surgery (MS), whose effects extend beyond glycaemic management [10]. Lifestyle changes, including improvements in diet, physical activity, and stress management, can lead to diabetes remission. However, long-term remission is not guaranteed. In contrast, MS offers more profound and sustained effects. Structural changes to the gastrointestinal tract reduce nutrient absorption and alter hormonal pathways, leading to increased circulating GLP-1 levels, decreased ghrelin levels, and subsequent modulation of hunger and reward mechanisms related to food intake [12].

The optimal mechanisms, therapeutic approaches, and long-term risk–benefit profile of achieving diabetes remission remain incompletely understood. The aim of our narrative review is to thoroughly examine the literature on diabetes remission, identify the most effective therapeutic options, analyse the influence of remission on clinical outcomes, and determine predictors of sustained long-term remission.

Methods

The literature review included basic science studies, cohort studies, interventional and observational trials, and review articles indexed in PubMed, Embase, and the Cochrane Library. The following search string was used: ((“diabete”[All Fields] OR “diabetes mellitus”[MeSH Terms] OR (“diabetes”[All Fields] AND “mellitus”[All Fields]) OR “diabetes mellitus”[All Fields] OR “diabetes”[All Fields]) OR “diabetic”[All Fields] OR “diabetics”[All Fields] OR “diabets”[All Fields]) AND (“remission”[All Fields] OR “remissions”[All Fields])) OR (“T2DM”[All Fields] AND (“remission”[All Fields] OR “remissions”[All Fields])). We searched systematic reviews, meta-analyses, and randomised controlled trials (RCTs) as publication types.

According to the authors’ judgment, inclusion of studies was based on their relevance. The selected studies were grouped into the following main topics:

1. Metabolic dysfunction and T2DM remission.

2. Behavioural and lifestyle interventions inducing T2DM remission.

3. Pharmacological interventions inducing T2DM remission.

4. Metabolic surgery inducing T2DM remission.

5. Long-term outcomes after T2DM remission.

6. Predictors of T2DM remission.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Metabolic Dysfunction and T2DM Remission

It is well known that genetically predisposed individuals who accumulate excess body fat and neglect lifestyle modifications are at increased risk of developing cardiometabolic diseases. Robust scientific evidence suggests that dysfunctional adipose tissue (the so-called adiposopathy [13]) significantly alters glucose metabolism, blood pressure regulation, and lipid profiles. However, while excessive body fat undoubtedly increases the risk of T2DM, it is not sufficient on its own. It is reasonable to propose that the combination of adiposopathy, genetic predisposition, gut microbiota and lifestyle factors contributes to adipose tissue dysfunction and the subsequent development of T2DM [13]. Inflammation plays a pivotal role in the pathogenesis of IR. Once initiated, inflammation and IR can perpetuate each other in a vicious cycle [14]. Obesity is characterised by chronic low-grade inflammation, which is commonly associated with IR and T2DM [15]. Free fatty acids (FFAs), elevated in obesity, are implicated as key contributors to IR and the activation of inflammatory pathways in adipose tissue, liver, muscle, and pancreas [16]. Immune cells are activated to produce cytokines and other factors that disrupt insulin signalling, ultimately contributing to the development of IR. Under physiological conditions, adipose tissue contains anti-inflammatory macrophages (ATM2) and natural killer cells, which help to maintain an insulin-sensitive state. Anti-inflammatory macrophages secrete interleukin-10, a cytokine that promotes insulin sensitivity. In obesity, however, increased lipolysis and the release of pro-inflammatory FFAs and factors, such as chemokines and tumour necrosis factor alpha (TNFα), recruit circulating monocytes into adipose tissue. These monocytes differentiate into pro-inflammatory macrophages (ATM1), which produce mediators like TNFα, interleukin-1β, interleukin-6 (IL-6), leukotriene B4, nitric oxide, and other factors that disrupt insulin signalling, induce adipokine dysregulation and drive the progression of IR [16–20].

The accumulation of ectopic fat in non-adipose tissues, such as the liver and skeletal muscle, contributes to impaired insulin signalling and systemic metabolic dysfunction [21]. Adiponectin, a marker of visceral fat accumulation, plays a protective role in metabolic health [22]. Our previously published data identified adiponectin as a predictor of increased left ventricular mass in individuals with visceral obesity, both normotensive and hypertensive [23]. Low adiponectin levels may help explain the development of cardiac damage, particularly in hypertensive subjects. Additionally, hypoadiponectinemia is likely associated with a higher prevalence of cardiovascular and metabolic comorbidities [18]. Furthermore, emerging evidence suggests that microRNA dysregulation within adipocytes may act as a critical mediator of these processes, amplifying the interplay between inflammation and IR [24].

Weight management interventions typically improve metabolic balance. However, in patients with advanced metabolic disease, dietary and lifestyle interventions alone (without pharmacologic treatment or MS) may not significantly reduce long-term cardiovascular disease (CVD) risk. In contrast, the early metabolic benefits observed after MS appear to exceed those expected from the initial weight loss alone. Improvements in metabolic balance following MS often coincide with enhanced diabetes outcomes, as evidenced by a compelling study linking diabetes remission to changes in microRNA expression after MS [25].

In recent years, the gut microbiota has emerged as a key player in the pathogenesis of metabolic diseases, including T2DM. Microbial dysbiosis can increase intestinal permeability and promote systemic low-grade inflammation, a known trigger of insulin resistance and adipose tissue dysfunction. In particular, reduced abundance of butyrate-producing bacteria (e.g. Faecalibacterium prausnitzii, Roseburia spp.) and enrichment of pro-inflammatory species have been consistently reported in individuals with T2DM [26–28]. Moreover, microbial metabolites—such as short-chain fatty acids, secondary bile acids, and branched-chain amino acids—interact with host metabolic pathways, modulating glucose and lipid metabolism, satiety signalling, and energy balance [27]. These interactions may explain part of the heterogeneity in treatment response and suggest new avenues for microbiome-targeted strategies in T2DM management. Importantly, recent insights have linked gut microbial signals to central appetite regulation mechanisms. Fasano has described how the gut microbiota can influence hunger through neuroimmune pathways, distinguishing between homeostatic and hedonic hunger, and highlighting the possibility that microbiota-driven signals contribute to overeating and metabolic derangement [29]. This understanding paves the way for microbiome-based precision interventions in obesity and T2DM. Notably, specific microbial signatures have been associated with a greater likelihood of achieving diabetes remission following MS. For instance, higher preoperative abundance of taxa such as the family Lachnospiraceae and the genus Roseburia has been identified as a predictor of remission post intervention [30]. In a study by Davies et al., gut microbial composition was shown to correlate with postoperative glycaemic improvement, supporting its potential role in outcome stratification [30]. Similarly, Kinoshita et al. demonstrated in a Japanese population that distinct gut microbiota profiles were linked not only to the presence of T2DM but also to its remission status, suggesting that microbiota composition may serve both diagnostic and prognostic functions [31]. Chong et al. further confirmed in a systematic review that taxa such as Akkermansia, Bifidobacterium, Bacteroides, and Faecalibacterium were consistently underrepresented in individuals with T2DM [28]. Finally, Hernández-Montoliu et al. found that a specific gut microbiota signature was associated with enhanced GLP-1 and GLP-2 secretion and improved metabolic control following Roux-en-Y gastric bypass, reinforcing the mechanistic relevance of microbial signals in metabolic surgery outcomes [32].

Behavioural and Lifestyle Interventions Inducing T2DM Remission

The primary CVD benefit of weight management through behavioural and lifestyle interventions may lie in its preventive effects, as individuals who avoid developing metabolic diseases are expected to have a reduced CVD risk. Conversely, in cases where metabolic diseases and CVD are already established, nutritional and physical activity interventions targeting obesity may have limited efficacy [33]. Emerging evidence indicates that the success of diabetes remission through dietary and lifestyle changes is strongly linked to the extent and speed of weight loss, particularly in reducing ectopic fat accumulation in vital metabolic organs such as the liver and pancreas [34]. Studies have shown that even moderate weight loss (approximately 10% of body weight) can significantly enhance beta-cell function and insulin sensitivity by decreasing intrahepatic and intrapancreatic fat. The reversal of glucolipotoxicity following rapid weight loss seems to restore first-phase insulin secretion and modify inflammatory pathways and metabolic flexibility [35]. Studies indicate that low-calorie diets have been found to reduce pro-inflammatory cytokines, such as TNFα and IL-6, which are involved in the development of IR [36]. Nevertheless, recent trials suggest that behavioural and lifestyle interventions may induce T2DM remission; a comparative summary of key studies is reported in Table 1.

Table 1.

RCTs on behavioural and dietary strategies for type 2 diabetes remission: a comparative summary of key studies

	DiRECT, extension study (2024) [38]	Look AHEAD (2024) [81]	Yang et al. (2023) [41]	DIADEM-I (2020) [37]
Study design	RCT	RCT	RCT	RCT
Patients enrolled (n)	298	4503	72	147
Intervention (I)	Withdrawal of antidiabetic drugs. Total diet replacement with a nutritionally complete formula (825–853 kcal/day) for 12–20 weeks, followed by stepped food reintroduction (2–8 weeks) and then structured weight maintenance support	Strict counselling program, with total caloric intake of 1200 to 1800 kcal/day, through reductions in total and saturated fat intake and by increasing physical activity levels to 175 min/week	3-month period of intermittent calorie-restricted diet (three meals per day) with 840 kcal/day for 15 days, alternating with 10 days of an ad libitum diet, followed by ad libitum diet (according with Dietary Guidelines for Diabetes in China (2017 Edition))	Withdrawal of antidiabetic drugs. 12-week total diet replacement phase (800–820 kcal/day), followed by a 12-week structured food reintroduction phase. Subsequently, participants managed their own energy-restricted diets. Physical activity: 10,000 steps/day and progressed to at least 150 min/week
Comparator (C)	Best-practice routine care	Group sessions (3/year) focusing on diet, physical activity and social support	Ad libitum diet (according with Dietary Guidelines for Diabetes in China (2017 Edition)	Best-practice routine care
Age, years (mean, SD)	52.9 ± 7.6	58.6 ± 6.7	53.4 ± 11.72	41.9 ± 5.4
Duration of diabetes, years (mean, SD)	3 ± 3	5 ± 8	6.65 ± 3.25	1.8 ± 0.96
BMI, kg/m2 (mean, SD)	36 ± 9	35.8 ± 5.9	24.23 ± 2.58	35 ± 5.2
Definition of remission	HbA1c < 6.5% (48 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications	HbA1c < 6.5% (48 mmol/mol) after discontinuing all antidiabetic medications at a single point in time	HbA1c < 6.5% (48 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications	HbA1c < 6.5% (48 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications*
Remission (%) at 1 year	49 vs 4 (P < 0.0001)	11.5 vs 2 (P = NA)	44 vs 0 (P < 0.0001)	61 vs 12 (P < 0.0001)
Remission (%) at 2 years I vs C	40 vs 4 (P < 0.0001)	9.2 vs 2 (P < 0.01)	NA	NA
Remission (%) at 4 years I vs C	NA	7.3 vs 3.5 (P < 0.01)	NA	NA
Remission (%) at 5 years I vs C	10 vs 5 (P = 0.31)	NA	NA	NA
Remission (%) at 12 years I vs C	NA	3.7 vs 1.95 (P = NA)	NA	NA

In particular, the percentages of remission at 1, 2, 4, 5 and 12 years of follow-up are summarised

Data regarding HbA1c and changes in BMI or body weight were not presented in the table because of substantial heterogeneity among the included trials in terms of study design, type and duration of the intervention, baseline characteristics, and outcome reporting. A direct comparison or synthesis of these variables could lead to misinterpretation

HbA1c glycated haemoglobin, BMI body mass index, NA not acquired

*This was a secondary endpoint

In the DIADEM-I trial, a total of 147 patients with T2DM under 50 years of age (mean age 42 years) with a diabetes duration of less than 3 years were enrolled. The intervention group underwent a 12-week total diet replacement phase using a low-energy (800–820 kcal/day) formula meal replacement, followed by a 12-week structured food reintroduction phase. Subsequently, participants managed their own energy-restricted diets. Physical activity support began with a walking target of 10,000 steps per day and progressed to unsupervised activity for at least 150 min per week. At the 1-year follow-up, diabetes remission was achieved in 61% of the intervention group, with an average total weight loss (TWL) of 12% [37]. In the DiRECT trial and its extension study, a total of 298 patients with T2DM (149 participants in the extension study) were enrolled. Participants were aged 55.9 ± 7.3 years in the control group and 52.9 ± 7.6 years in the intervention group, with a mean diabetes duration of 3 years (less than 6 years total). Remission was achieved using a very-low-calorie diet intervention, which included a total diet replacement with a nutritionally complete formula (825–853 kcal/day) for 12 weeks, followed by stepped food reintroduction and structured weight maintenance support. Diabetes remission rates at 1, 2, and 5 years were 46%, 40%, and 10%, respectively [38]. Notably, a higher percentage of participants who achieved 10–15% TWL and > 15% TWL attained remission at 1 year (57% and 86%, respectively) compared to those with 5–10% TWL (34%) [39]. The Look AHEAD (Action for Health for Diabetes) study is the largest multicentric RCT of an intensive lifestyle intervention among adults with T2DM. An observational analysis of the Look AHEAD cohort was conducted to investigate the association of an intensive lifestyle intervention with frequency of partial and complete remission of T2DM. A total of 5145 T2DM overweight adults (BMI > 25 kg/m², HbA1c 7.37 ± 1.2) aged 45 to 76 years, with a mean diabetes duration of 5 years were randomised to either an intensive lifestyle-based weight loss intervention or a diabetes support and education intervention. The intervention group underwent a strict counselling program, with total caloric intake of 1200 to 1800 kcal/day, through reductions in total and saturated fat intake and by increasing physical activity levels to 175 min/week. The intervention group was significantly more likely to have sustained remission both at year 2 of follow-up, 9.2% [95% CI 7.9% to 10.4%] vs 1.7% [95% CI 1.2% to 2.3%] (P < 0.001), and year 4 of follow-up, 3.5% [95% CI 2.7% to 4.3%] vs 0.5% [95% CI 0.2% to 0.8%] (P = 0.02). Remarkably, rates of remission were higher (15–21%) among persons with substantial weight loss [40]. Another recent RCT investigated the effect of intermittent calorie restriction on T2DM remission. The study enrolled 72 patients with T2DM, a mean age of 53.4 ± 11.72 years and a mean diabetes duration of up to 11 years (6.65 ± 3.25 years in the intervention group). The intervention consisted of an intermittent calorie-restricted diet (three meals per day) with 840 kcal/day for 15 days, alternating with 10 days of an ad libitum diet. Both groups followed the Dietary Guidelines for Diabetes in China (2017 Edition). At the end of the trial, remission was achieved in 44.4% of patients in the intervention group, with an average TWL of 9% [41]. These findings underscore the importance of early, intensive dietary interventions in patients with newly diagnosed T2DM, as longer disease duration is associated with more pronounced beta-cell dysfunction and reduced likelihood of remission.

Pharmacological Interventions Inducing T2DM Remission

Current options for inducing diabetes remission without MS remain limited. Novel antidiabetic medications, such as SGLT2-Is and glucagon-like peptide 1 receptor agonists (GLP1-RA), have demonstrated significant glycaemic efficacy and safety profiles, as supported by several RCTs [42–46]. Furthermore, emerging evidence suggests that newer medications may have glycaemic independent efficacy on weight loss [47]. Nevertheless, weight loss remains the dominant factor in achieving remission. Several trials have assessed the impact of different pharmacological interventions on remission; a comparative summary of key studies is reported in Table 2.

Table 2.

RCTs on pharmacological intervention-induced type 2 diabetes remission: a comparative summary of key studies

	Weng et al. (2008) [49]	McInnes et al. (2017) [5]	McInnes et al. (2020) [48]	McInnes et al. (2023) [7]
Study design	RCT	RCT	RCT	RCT
Patients enrolled (n)	382	83	154	160
Intervention (I)	Transient intensive insulin therapy (via continuous subcutaneous insulin infusion (I1) or multiple daily injections (I2)	Caloric restriction (500–750 kcal/day), moderate-intensity physical activity ≥ 150 min/week treatment with insulin glargine, metformin, and acarbose	Lifestyle modifications combined with insulin glargine, metformin, and dapagliflozin	Lifestyle modifications combined with insulin glargine/lixisenatide and metformin
Comparator (C)	Oral hypoglycaemic agents (metformin and/or gliclazide)	Best-practice routine care	Best-practice routine care	Best-practice routine care
Age, years (mean, SD)	50 ± 11	57.1 ± 10.3	56.8 ± 10	56.2 ± 11.2
Baseline HbA1c, % (mean, SD)	9.8 ± 2.3	6.6 ± 0.6	6.7 ± 0.6	6.8 ± 0.7
Duration of diabetes, years (mean, SD)	0 (newly diagnosed)	1 ± 1	3 ± 2	NA (< 3)
BMI, kg/m2 (mean, SD)	25.1 ± 3.3	33.2 ± 5.8	33.3 ± 5.8	31.9 ± 5.7
Definition of remission	HbA1c < 6.3% (45 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications	Partial: HbA1c < 6.5% (48 mmol/mol)/complete: HbA1c < 6.0% (42 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications*	Partial: HbA1c < 6.5% (48 mmol/mol)/complete: HbA1c < 6.0% (42 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications	Partial: HbA1c < 6.5% (48 mmol/mol)/complete: HbA1c < 6.0% (42 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications*
Remission (%) at 4 months I vs C	NA	22.2 vs 10.7 (P = NA)	14.3 vs 7.8 (P = 0.21)	NA
Remission (%) at 6 months I vs C	NA	NA	NA	38.0 vs 19.8 (P = 0.01)
Remission (%) at 12 months I vs C	(I1) 51.1/(I2) 44.9 vs 26.7 (P = 0.0012)	NA	NA	27.8 vs 14.8 (P = 0.05)
Remission (%) at 18 months I vs C	NA	NA	NA	22.8 vs 11.1 (P = 0.06)

In particular, the percentages of remission at 4, 6, 12, 18 months of follow-up are summarised

Data regarding HbA1c and changes in BMI or body weight were not presented in the table because of substantial heterogeneity among the included trials in terms of study design, type and duration of the intervention, baseline characteristics, and outcome reporting. A direct comparison or synthesis of these variables could lead to misinterpretation

HbA1c glycated haemoglobin, BMI body mass index, RCT randomised controlled trial, NA not acquired

*This was a secondary endpoint

In a randomised, parallel, open-label pilot trial with 83 participants (mean age 57.1 ± 10.3 years; HbA1c 6.6 ± 0.6%), intensive lifestyle and pharmacological interventions were compared to standard care over 52 weeks. Participants in the intervention group underwent caloric restriction (reducing daily intake by 500–750 kcal/day) and engaged in ≥ 150 min of moderate-intensity physical activity per week, alongside treatment with insulin glargine, metformin, and acarbose. At the end of the trial, 22.2% of participants in the intervention group achieved remission compared to 10.7% in the standard care group [RR (95% CI) 2.07 (0.58–7.47)] [5]. In a follow-up study by the same group, 154 participants (mean age 56.8 ± 10 years; HbA1c 6.7 ± 0.6%) were followed for 64 weeks. The intervention group received lifestyle modifications combined with insulin glargine, metformin, and dapagliflozin. Diabetes remission was achieved in 14.3% of the intervention group compared to 7.8% of the standard care group [RR (95% CI) 1.8 (0.7–4.7); (P = 0.21)] [48]. A recent RCT by McInnes et al. investigated the efficacy of combining intensive lifestyle modifications with insulin glargine/lixisenatide and metformin in 160 patients with T2DM (mean age 56.2 ± 11.2 years; diabetes duration < 3 years; BMI 31.9 ± 5.7 kg/m²). At 24 weeks, 38.0% of the intervention group achieved remission compared to 19.8% of controls [RR (95% CI) 1.92 (1.14–3.24), (P = 0.01)]. However, at 64 weeks, remission rates dropped to 22.8% versus 11.1%, respectively [RR (95% CI) 2.05 (0.98–4.29; (P = 0.06)]. These findings suggest that while intensive interventions are effective in the short term, their effects on remission may not be sustained [7].

Early intensive insulin therapy has also been shown to extend diabetes remission. A 2008 multicentric RCT enrolled 382 participants (mean age 50 ± 11 years; HbA1c 9.8 ± 2.3%; BMI 25.1 ± 3.3 kg/m²) and randomised them to either transient intensive insulin therapy (via continuous subcutaneous insulin infusion [CSII] or multiple daily injections [MDI]) or oral glucose-lowering agents (metformin and/or gliclazide). At 1 year, remission rates were significantly higher in the insulin groups (51.1% in CSII and 44.9% in MDI) compared to the oral agent group (26.7%), (P = 0.0012) [49]. However, the association between insulin and weight gain highlights the importance of careful patient selection. Recent advancements include tirzepatide, a dual glucose-dependent insulinotropic peptide (GIP) and GLP-1 receptor agonist. Tirzepatide has shown significant reductions in HbA1c levels and considerable weight loss among both patients with and without diabetes and obesity [47, 50, 51]. Pooled data from nine RCTs (total of 10,121 participants randomised either to tirzepatide (5, 10 or 15 mg, once-weekly) or control) demonstrated that treatment with tirzepatide versus control resulted in a significant increase in the odds for achievement of normoglycaemia [OR (95% CI) 16.81 (7.83–36.09)], (P < 0.001) [52]. In the SURPASS-2 trial, tirzepatide at doses of 5, 10, and 15 mg once weekly were superior to semaglutide (1 mg) in terms of HbA1c reduction (− 2.01%, − 2.24%, and − 2.30% vs − 1.86%, respectively) and weight loss (− 7.6 kg, − 9.3 kg, and − 11.2 kg vs − 5.7 kg) over 40 weeks [53]. Sustained remission may require ongoing treatment; however, some authors consider this state as “pharmacological remission” [9, 52]. The STEP1 trial extension reported that participants regained two-thirds of their prior weight loss within 1 year after discontinuing semaglutide (2.4 mg weekly), suggesting that maintenance of cardiometabolic benefits requires continued therapy [54]. Similarly, the SURMOUNT-4 trial found that participants who discontinued tirzepatide regained 14% of their lost weight, whereas those who continued treatment achieved further weight reductions (− 5.5%) [55].

Emerging Therapies and Future Directions

Recent advances in diabetes research have highlighted the potential of novel therapeutic strategies beyond traditional pharmacological and lifestyle approaches. Promising new treatments, including dual and triple agonists, are currently under investigation and may offer additional therapeutic options, though extensive preclinical and clinical studies are still needed [56]. Mazdutide (GLP-1/GIP dual agonist) [57], cotadutide (GLP-1/glucagon dual agonist) [58] and retatrutide (GLP-1/GIP/glucagon triple agonist) [59] have shown the capacity to enhance glycaemic control and induce significant weight loss through synergistic mechanisms. Early-phase trials indicate that these agents may offer superior efficacy in reducing HbA1c and promoting diabetes remission compared to current GLP-1 receptor agonists [60, 61]. Emerging therapies targeting chronic low-grade inflammation, a key driver of IR, such as interleukin-1 inhibitors and colchicine have demonstrated interesting effects on glycaemic control and reducing cardiovascular risk in patients with T2DM [62, 63]. Additionally, encouraging genetic biomarkers such as specific microRNA profiles appear able to predict the likelihood to achieve remission [64]. As these intriguing fields evolve, combining this innovative prognostic approach with emerging individualised treatments may be the key for achieving durable diabetes remission and improved long-term outcomes.

Metabolic Surgery Inducing T2DM Remission

Bariatric surgery, more appropriately referred to as MS, encompasses procedures that alter the stomach or intestines to reduce food intake or absorption, primarily to achieve weight loss and mitigate associated health risks. According to the National Institutes of Health guidelines, patients with T2DM with a BMI above 35 kg/m² are eligible for bariatric surgery. Evidence strongly supports that intensive lifestyle management and MS can induce T2DM remission by significantly reducing body weight [65]. The six most frequently performed MS procedures include jejunoileal bypass (JIB), Roux-en-Y gastric bypass (RYGB), vertical banded gastroplasty (VBG), biliopancreatic diversion (BPD) with or without duodenal switch (DS), adjustable gastric banding (AGB), and sleeve gastrectomy (SG). These procedures vary in their mechanisms, creating new pathways for nutrient metabolism and reducing the stomach’s capacity [65–67]. Between 2013 and 2015, RYGB and SG were the most commonly performed MS procedures globally, with approximately 64.7% of patients achieving diabetes remission postoperatively. In addition to diabetes remission, MS has been associated with improvements in depression, hypertension, musculoskeletal pain, sleep apnoea, and gastroesophageal reflux disease [68]. Some studies report even higher remission rates, with up to 72% of patients with T2DM achieving remission 2 years postoperatively [69], solidifying MS as the most effective treatment for T2DM remission; a comparative summary of key studies is reported in Table 3.

Table 3.

Meta-analyses and RCTs on metabolic surgery-induced type 2 diabetes remission: a comparative summary of key studies

	Courcoulas et al. ARMMS-T2D (2024) [109]	Kirwan et al. ARMMS-T2D (2022) [110]	Mingrone et al. (2021) [70]	Courcoulas et al. (2020) [111]	Ding et al. (2015) [112]	Foschi et al. (2019) [113]	Halperin et al. (2014) [114]	Parikh et al. (2014) [115]	Sjoholm et al. (2022) [116]	Dixon et al. (2013) [117]
Study design	Pooled analysis from 4 RCTs (STAMPEDE, TRIABETES, SLIMM-T2D, CROSSROADS)	Pooled analysis from 4 RCTs (STAMPEDE, TRIABETES, SLIMM-T2D, CROSSROADS)	RCT	RCT	RCT	RCT	RCT	RCT	Prospective controlled intervention study	RCT
Patients enrolled (n)	262	316	60	61	40	60	38	57	701	154
Intervention (I)	RYGB, SG, AGB	RYGB, SG, AGB	I1: RYGB, I2: BPD	I1: RYGB, I2: LAGB	LAGB	II-DD-SG	RYGB	RYGB	AGB, GB	MGB
Comparator (C)	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	Medical therapy and lifestyle management	RYGB
Age, years (mean, SD)	49.9 ± 8.3	49.9 ± 8.2	43.67 ± 7.57†	47.3 ± 6.6	51 ± 10	52.5 ± 3.04†	51.64 ± 6.16†	50.4 ± 8.92†	49.3 ± 6.19†	39.5 ± 10.7
Baseline HbA1c, % (mean, SD)	8.5 ± 1.5	8.7 ± 1.7	8.67 ± 1.43†	7.8 ± 1.9	8.2 ± 1.2	7.6 ± 0.19†	8.53 ± 1.25†	NA	7.75 ± 1.46†	9.1 ± 1.7
Duration of diabetes, years (mean, SD)	8.5 ± 5.4	8.4 ± 5.4	NA	NA	9 ± 5	4.4 ± 0.65	10.28 ± 6.24†	NA	NA	NA (< 5)
BMI, kg/m2 (mean, SD)	36.4 ± 3.5	36.3 ± 3.4	NA	35.7 ± 3.1	36.5 ± 3.7	42.5 ± 1.35	36.25 ± 3.42†	32.6 ± 1.75†	41.5 ± 4.8†	37.2 ± 8.8
Definition of remission	HbA1c < 6.5% (48 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications*	HbA1c < 6.5% (48 mmol/mol) for at least 3 months after discontinuing all antidiabetic medications	HbA1c < 6.5% ((48 mmol/mol) and fasting glycaemia < 5.55 mmol/L without ongoing medication for at least 1 year	Partial: HbA1c < 6.5% (48 mmol/mol)/complete: HbA1c < 5.7% (37 mmol/mol) after discontinuing all antidiabetic medications at follow-up	Partial: HbA1c < 7% (48 mmol/mol)/complete: HbA1c < 6.5% (37 mmol/mol) at 1 year (whether or not on pharmacological treatment)	Complete: HbA1c < 6.0% (42 mmol/mol) for at least 1 year after discontinuing all antidiabetic medications	Attaining glycaemic control (fasting plasma glucose levels below 126 mg/dL and HbA1c below 6.5% at 1 year whether or not on pharmacological treatment	Not meeting the American Diabetes Association criteria for T2DM, without the use of diabetes medications (fasting glucose ≥ 126 mg/dL or glucose, ≥ 200 at 120 min after 75 g oral glucose loader HbA1c ≥ 6.5%)	HbA1c < 7% (48 mmol/mol) or fasting blood glucose < 6.1 mmol/L (plasma glucose < 7.0 mmol/L) without receipt of diabetes medication	HbA1c < 6.0% (42 mmol/mol), whether or not on pharmacological treatment
Remission (%) at 6 month I vs C	NA	NA	NA	NA	NA	76.6 vs 3.3 (P < 0.0001)	NA	65 vs 0 (P < 0.0001)	NA	NA
Remission (%) at 1 year I vs C	NA	NA	NA	(I1) 60 vs (I2) 29 vs (C) 0	33 vs 23 (P = 0.457)	76.7 vs 0 (P < 0.0001)	58 vs 16 (P = 0.03)	NA	NA	85.5 vs 48.5 (P < 0.001)
Remission (%) at 2 years I vs C	NA	NA	NA	(I1) 45 vs (I2) 29 vs (C) 0	NA	82.8 vs 4 (P < 0.0001)	NA	NA	69.8 vs 15.9 (P < 0.001)	NA
Remission (%) at 3 years I vs C	NA	37.5 vs 2.6 (P < 0.001)	NA	(I1) 40 vs (I2) 29 vs (C) 0	NA	89.7 vs 4 (P < 0.0001)	NA	NA	NA	NA
Remission (%) at 5 years I vs C	NA	NA	NA	(I1) 30 vs (I2) 19 vs (C) 0	NA	68 vs 8 (P < 0.0001)	NA	NA	NA	NA
Remission (%) at 7 years I vs C	37.7 vs 17.3 (P = NA)	NA	NA	NA	NA	NA	NA	NA	NA	NA
Remission (%) at 10 years I vs C	NA	NA	(I1) 25 vs (I2) 50 vs (C) 0	NA	NA	NA	NA	NA	34.1 vs 6.5 (P < 0.001)	NA
Remission (%) at 12 years I vs C	40 vs 19.6 (P = 0.03)	NA	NA	NA	NA	NA	NA	NA	NA	NA

In particular, the percentages of remission at 6 months, 1, 2, 3, 5, 7, 10, 12 years of follow-up are summarised (RCTs reported in the table are not included in meta-analyses)

Data regarding HbA1c and changes in BMI or body weight were not presented in the table because of substantial heterogeneity among the included trials in terms of study design, type and duration of the intervention, baseline characteristics, and outcome reporting. A direct comparison or synthesis of these variables could lead to misinterpretation

HbA1c glycated haemoglobin, BMI body mass index, RYGB Roux-en-Y gastric bypass, SG sleeve gastrectomy, AGB adjustable gastric banding, GB gastric bypass, BPD biliopancreatic diversion, LAGB laparoscopic adjustable gastric banding, II-DD-SG ileal interposition with duodenal diversion sleeve gastrectomy, MGB mini gastric bypass, RCT randomised controlled trial, NA not acquired

*This was a secondary endpoint, †The weighted average and combined standard deviation were derived by the authors using the sample sizes, means, and standard deviations of the study groups

In a 10-year, open-label, single-centre RCT involving 60 participants (mean age 43.9 ± 7.6 years; HbA1c 8.9 ± 1.7%), participants were assigned to RYGB, BPD, or medical therapy with lifestyle interventions (20:20:20 ratio). At the 10-year mark, the medical therapy group showed improved glycaemic control compared to baseline, but surgical treatment led to significantly greater HbA1c reductions [− 0.8 ± 1.0% for medical therapy vs − 2.4 ± 1.6% for BPD and − 1.9 ± 1.6% for RYGB; (P < 0.0097)]. While RYGB achieved T2DM remission in 75% of patients at 2 years, remission rates declined to 25% at 10 years [70]. Over the past two decades, several endoscopic bariatric procedures have emerged as less invasive alternatives for primary and secondary obesity treatment. Restrictive gastric procedures, such as intragastric balloons and endoscopic gastroplasty, have shown effectiveness in inducing weight loss compared to diet modifications alone [71]. Advanced endoscopic techniques, including argon plasma coagulation with endoscopic full-thickness suturing systems (APC-TORe) and Re-EndoSleeve, have demonstrated efficacy in managing weight regain and metabolic alterations after RYGB or SG. The continuous evolution of endoscopic bariatric procedures holds promise for further improvements in metabolic outcomes [65].

Emerging Molecular Mechanisms and Future Directions

Recent studies have identified additional molecular pathways influenced by MS, beyond traditional weight loss mechanisms. While weight loss is the most intuitive outcome, MS also induces profound endocrine changes. These changes arise from altered nutrient absorption due to gastrointestinal restructuring, particularly the rapid delivery of nutrients to the ileum, which increases haematic GLP-1 concentrations [72]. Elevated GLP-1 levels improve beta-cell function and insulin sensitivity, especially after SG or RYGB [73, 74]. GLP-1 enhances satiety, stimulates insulin secretion, and suppresses glucagon release in response to glucose ingestion. It also reduces lipogenesis, FFA concentrations and lipotoxicity [75]. Additionally, fibroblast growth factors (FGF-19 and FGF-21) have emerged as potential biomarkers for weight loss after MS. These hormones regulate bile acid biosynthesis, glucose metabolism, and lipid metabolism in pancreatic islets, the liver, and adipose tissue [76]. Microbiome changes are another significant effect of MS. Studies have shown that SG and RYGB lead to decreased Firmicutes and increased Bacteroidetes and Proteobacteria populations, alterations linked to improvements in metabolic health; gut microbiota modifications enhance short-chain fatty acid production, improving gut barrier integrity and systemic insulin sensitivity [77]. Alterations in enterohepatic circulation, induced by procedures such as RYGB and SG, play a critical role in enhancing insulin sensitivity and glycaemic control. Increased circulating bile acids activate the farnesoid X-receptor and Takeda G-protein receptor 5, which regulate glycolipidaemic profile, improve energetic efficiency, and modulate inflammation [78]. Additionally, the epigenetic modifications related to dietary intervention and MS seem to affect gene expression and the pathogenesis of obesity and T2DM. The reversible nature of epigenetic modifications observed in genes involved in insulin signalling and energy metabolism may offer future therapeutic strategies and provide novel biomarkers for predicting long-term metabolic benefits of MS [79]. Looking forward, integrating MS with emerging pharmacological agents, such as dual GLP-1/GIP receptor agonists (e.g. tirzepatide), the development of non-invasive endoscopic techniques and individualised therapies based on genetic and microbiome profiles are expected to ensure safer and more durable metabolic outcomes [80].

Long-Term Outcomes After T2DM Remission

T2DM remission is increasingly recognised as an achievable target for patients with diabetes, offering remarkable potential benefits in terms of reducing long-term morbidity, improving quality of life, and preserving beta-cell function. The Look AHEAD study, a multicentric RCT, evaluated the effects of a 12-year intensive lifestyle intervention. Among the intervention group, the prevalence of remission was 11.2% at year 1, decreasing by approximately 0.7 percentage points annually, reaching 3.7% by year 12. Participants who achieved any remission during the study had significant benefits, including a 33% lower rate of CKD [HR (95% CI) 0.67 (0.52–0.87)] and a 40% reduction in the composite measure of CVD [HR (95% CI) 0.60 (0.47–0.79)] [81]. The broader metabolic benefits of weight loss in achieving diabetes remission are well documented. These include improvements in arterial hypertension, dyslipidaemia, and reduced medication requirements [39]. Recent scientific consensus highlights the intricate relationship between metabolic dysfunction and hepatic steatosis, leading to the redefinition of metabolic dysfunction-associated steatotic liver disease (MASLD) [82]. Achieving the necessary weight loss for diabetes remission also confers benefits for MASLD and metabolic associated steatohepatitis (MASH). In an RCT involving 288 individuals with biopsy-proven MASH, MS resulted in histological resolution of MASH without worsening fibrosis at the 1-year follow-up in 56% of patients who underwent RYGB and 57% of those who underwent SG, compared to 16% in the control group (medical care and lifestyle modifications) [83, 84].

Role of Pharmacological and Lifestyle Synergies

Emerging evidence suggests that combining intensive lifestyle interventions with novel pharmacological therapies enhances the likelihood of achieving diabetes remission. GLP1-RAs, SGLT2-Is and GIP/GLP1-RA have already demonstrated efficacy in improving glycaemic control, reducing body weight, and enhancing beta-cell function. For instance, the SURPASS trials revealed that tirzepatide achieved superior HbA1c and weight reductions compared to standard treatments [47]. When combined with lifestyle modifications, these therapies extend the metabolic benefits by targeting multiple pathways, including insulin sensitivity, contrasting lipotoxicity and regulating satiety [38, 39]. Early achievement of diabetes remission provides the greatest advantage in terms of cardiovascular prevention. Long-term follow-up of the UK Prospective Diabetes Study (UKPDS) demonstrated that early glycaemic control has durable benefits. Reductions in microvascular events persisted, and reductions in cardiovascular events and mortality were observed 10 years after the trial concluded [85]. Emerging evidence suggests that hyperglycaemia imprints early changes at the cellular level, predisposing individuals to future diabetic complications. Studies indicate that failure to achieve an HbA1c below 6.5% within the first year of diagnosis establishes a long-term risk for microvascular and macrovascular complications. This phenomenon, where the effects of early hyperglycaemia persist even after achieving euglycaemia, is referred to as “metabolic memory” or the “legacy effect” [86–88].

Predictors of T2DM Remission

The identification of predictive factors for diabetes remission has garnered significant scientific and clinical interest over the past decade. Prediction of diabetes remission can be helpful in clinical practice and preoperative consultation before MS for decision-making, and several indices have been purposed. Still and colleagues developed a novel scoring system, the DiaRem score, to predict the likelihood of T2DM remission after RYGB [65]. This scoring system incorporates four preoperative clinical variables: insulin use, age, HbA1c levels, and the type of antidiabetic drugs used. The DiaRem score has proven to be a valid tool for predicting diabetes remission following RYGB [89]. Other scores such as Ad-DiaRem [90], DiaBetter [91], the individualised metabolic surgery (IMS) score [92], and the ABCD score [93] have been used for predicting diabetes remission after bariatric surgery, with different results [90]. Barthold et al. [94], in a multiethnic cohort study of 5982 patients, demonstrated that the probability of T2DM remission after MS increases significantly with early-stage T2DM and TWL of 10–15%, plateauing at 20–25% TWL. For patients with advanced T2DM, including those on insulin therapy, remission was observed only with 20–25% TWL. Similar findings were reported in a large population-based study involving 72,920 patients who underwent MS [67], indicating that in long-standing diabetes, substantial TWL is required to achieve remission. The DiRECT study provided additional insights, associating diabetes remission with the reversal of ectopic lipid accumulation in the liver (improving insulin sensitivity) and the pancreas (restoring insulin secretion) [38]. A retrospective cohort study analysing 815 patients (68.9% female, mean age 52.1 ± 11.5 years; BMI 45.1 ± 7.7 kg/m²) with a follow-up of 7.3 ± 3.8 years identified several predictors of remission, including preoperative diabetes duration, baseline HbA1c, pre-surgical insulin use, the number of antidiabetic medications, and %TWL (all P < 0.01). Remission rates were proportionally associated with TWL quartiles: Q1 (40.9%), Q2 (52.7%), Q3 (53.1%), and Q4 (56.1%) (P = 0.02) [95]. Although moderate TWL may predict initial T2DM remission, even lower TWL after MS can lead to significant improvements in glycaemic control, hypertension, dyslipidaemia, obstructive sleep apnoea, and functional capacity [66]. Although numerous predictive factors have been investigated, only a limited number have been found to be statistically significant among different studies; positive and negative predictive factors of T2DM remission are summarised in Figs. 1 and 2.

Fig. 1.

Positive predictors of T2DM remission. Only predictors with statistically significant odds ratio on T2DM remission reported on key studies have been included. *When metabolic surgery is indicated as the treatment of choice. GLP1-RA glucagon-like peptide 1 receptor agonists, RYGB Roux-en-Y gastric bypass, TWL total weight loss, T2DM Type 2 Diabetes Mellitus

Fig. 2.

Negative predictors of T2DM remission. Only predictors with statistically significant odds ratio on T2DM remission reported on key studies have been included. HbA1c glycated haemoglobin, BMI body mass index, T2DM Type 2 Diabetes Mellitus

Emerging Approaches in Predicting Diabetes Remission

Recent advances in predictive modelling have highlighted the potential of integrating molecular, genetic, and computational tools to refine the prediction of diabetes remission. Biomarkers such as adiponectin, C-peptide levels, and inflammatory markers (e.g. C-reactive protein (CRP), IL-6) are gaining attention for their ability to predict metabolic improvements post surgery [14, 96]. Additionally, genetic polymorphisms, including variants in the TCF7L2 and FTO genes, have been associated with differential responses to weight loss and glycaemic control, offering the potential for individualised risk stratification [97]. Artificial intelligence (AI) and machine learning (ML) algorithms have revolutionised predictive analytics. By analysing large datasets, AI-driven models can identify complex interactions among clinical, genetic, and metabolic variables, enabling more accurate predictions of remission outcomes [98]. Cao et al. proposed a prediction model for T2DM remission using a deep learning AI algorithm (convolutional neural network, CNN). They compared its predictive capability with the four most commonly used scores, demonstrating a better predictive ability than traditional indices [99]. Another study employed artificial neural networks to identify nonlinear correlations among clinical and genomic factors to differentiate between patients with and without surgery-induced diabetes remission. This approach highlighted insulin treatment, baseline HbA1c, the use of insulin-sensitising agents, baseline serum insulin levels, and eight single nucleotide polymorphisms (ABCA1, ARHGEF12, CTNNBL1, GLI3, PROK2, RYBP, SMUG1, and STXBP5) as the most informative variables, achieving relevant internal validation performance (84% accuracy, AUC 0.92) [100]. These innovations promise to enhance patient selection, optimise surgical outcomes, and pave the way for truly personalised diabetes management.

Discussion

T2DM remission is increasingly recognised as an attainable therapeutic goal, particularly among individuals with early-stage disease and central obesity phenotypes. Among the available interventions, MS remains the most effective strategy, with remission rates exceeding 60% at 2 years, primarily owing to profound changes in gastrointestinal hormones, bile acid metabolism, and weight loss. Intensive lifestyle interventions, especially those achieving ≥ 10% TWL, also demonstrate significant efficacy, with remission rates ranging from 34% to 61% in structured dietary programs as demonstrated in DiRECT [38], Look AHEAD [40] and DIADEM-I [37]. Pharmacological approaches, particularly those targeting glucose homeostasis and adipose tissue dysfunction, have emerged as promising strategies for remission. Among pharmacological interventions, tirzepatide, a dual GLP-1/GIP receptor agonist, represents a paradigm shift in T2DM management, as it directly modulates visceral and ectopic fat deposition, key drivers of IR and metabolic dysfunction [50]. A recent meta-analysis [52] demonstrated that tirzepatide achieves higher rates of diabetes pharmacological remission compared to placebo [OR (95% CI) 68.2 (9.41–493.98) in SURPASS-1] [47], semaglutide [OR (95% CI) 2.87 (2.23–3.69) in SURPASS-2] [53], or glargine [OR (95% CI) 115.20 (7.14–1858.92) in SURPASS-AP-Combo] [101], with superior reductions in HbA1c and body weight. Tirzepatide’s ability to preferentially target visceral adiposity and its effects on pancreatic beta-cell function [102] suggest it may play a significant role in inducing and maintaining diabetes remission. However, no tirzipatide effectiveness data are available according to the current definition of remission.

Achieving T2DM remission should be a critical theme for present and future research, as it holds the potential to redefine the therapeutic landscape of T2DM. This requires a proactive approach to identify and intensively treat patients in the early stages of their disease. Higher probabilities of maintaining long-standing remission are associated with specific factors, including younger age, lower baseline HbA1c, fewer diabetes medications, absence of insulin use, shorter disease duration, greater TWL%, and the reversal of ectopic fat in key metabolic organs. In light of the evidence from cardiovascular outcome trials and the effects of glucose-lowering therapies with pleiotropic mechanisms on hard outcomes [46, 103–108], the suspension of drug therapy once glycaemic target is reached—an essential criterion for defining “diabetes remission”—is clearly controversial. On the basis of our experience, and considering the chronic and relapsing nature of T2DM, in patients with positive predictive factors for long-term remission (Fig. 1), after effective adaptation to appropriate lifestyle modifications and the achievement of remission-specific glycaemic targets, alongside a significant reduction in weight and visceral fat, it is possible to reconsider drug therapy. This would involve a tailored maintenance treatment, strict monitoring of metabolic balance, with additional intensive treatment cycles introduced in response to early signs of metabolic decompensation or the new accumulation of visceral fat deposits. However, further studies are required to methodologically refine and identify personalised treatment regimens. A multimodal approach that integrates MS, intensive lifestyle interventions, and pharmacotherapy—tailored to the patient’s phenotype—may optimise long-term outcomes. Furthermore, advancing our understanding of adipose tissue dynamics and emerging tools, like the DiaRem score, machine learning models, and biomarkers such as FGF-21 and specific microRNAs hold promise for customising personalised treatment plans. These advancements will be crucial in shaping the next era of precision medicine for diabetes remission, allowing for more personalised and durable therapeutic solutions. This underscores the importance of a multidisciplinary and patient-centred approach to enhance long-term diabetes management, remission, and sustained metabolic benefits.

This manuscript provides evidence from RCTs investigating behavioural and dietary strategies, pharmacological interventions, and MS for the remission of T2DM. RCTs remain the gold standard for evaluating the efficacy of therapeutic interventions. This is particularly relevant in the context of T2DM remission, where the precise effect size, temporal relationship, and causality between interventions and outcomes can only be ascertained through adequately controlled experimental designs. Registry-based studies, while important, are typically subject to confounding, bias, and selection effects and thus serve a complementary—rather than primary—role in establishing therapeutic efficacy. Our review specifically aimed to summarise the mechanistic rationale and clinical efficacy of interventions associated with remission. For this purpose, we prioritised high-quality evidence from RCTs. Nevertheless, we acknowledge the relevance of long-term real-world follow-up data and their potential contribution to understanding sustainability and recurrence. Future research incorporating both RCTs and real-world evidence is warranted to optimise integrated care pathways for individuals with T2DM.

Conclusion

Remission of T2DM is no longer just a dream—it is an increasingly tangible therapeutic goal. Thanks to advances in MS, intensive lifestyle interventions, and novel pharmacological agents such as dual incretin receptor agonists, remission has become a realistic and achievable outcome, especially when interventions are timely and tailored to the patient’s phenotype. However, the durability of remission and the definition itself—particularly given the ongoing need for some pharmacologic support—remain areas of clinical and ethical debate. The current evidence emphasises the need for a paradigm shift: from managing chronic hyperglycaemia to proactively inducing and maintaining remission through a patient-centred, multidisciplinary, and precision medicine approach. Thus, while the dream of remission has begun to materialise, its long-term reality depends on our ability to personalise and sustain integrated care strategies.

Author Contributions

Salvatore Corrao was responsible for the study concept and planning of the work described, co-responsible for draft writing and responsible for critical revision of the manuscript and for the final reviewing and editing. Fabio Falcone was co-responsible for planning and writing the original draft and strongly contributed to the final reviewing and editing, hence he should be considered a co-first author. Luigi Calvo, Luigi Mirarchi and Simona Amodeo contributed to the final reviewing and editing. All authors agreed and approved the final submitted version of the manuscript.

Funding

No funding or sponsorship was received for this study or publication of this article.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.

Declarations

Conflict of Interest

All the authors declare that they have no competing interests. Salvatore Corrao, Fabio Falcone, Luigi Mirarchi, Simona Amodeo and Luigi Calvo have nothing to disclose.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.