Semaglutide from Bench to Bedside: The Experimental Journey Towards a Transformative Therapy for Diabetes, Obesity and Metabolic Liver Disorders

Type 2 diabetes (T2D) is characterized by chronically elevated glycemic values due to inadequate synthesis and release of insulin from the pancreatic β-cells to meet the demands of the body [1]. T2D is typically associated with risk factors such as a family history of the disease, being overweight, having a sedentary lifestyle, and being over forty-five years old [2]. Currently T2D affects more than 400 million people worldwide, with its future prevalence projected to increase. It is a major cause of blindness, non-traumatic lower limb amputations, chronic kidney disease (CKD), chronic liver disease, certain types of cancers and cardiovascular morbidity, leading to an overall expenditure of $11,500 per patient [3,4]. Driven by persistent hyperglycemia, target organ damage in T2D results from cellular stress, overproduction of reactive oxygen species (ROS), advanced glycation end products (AGEs), and protein kinase C (PKC) activation, ultimately leading to the development of microvascular and macrovascular diseases. These mechanisms collectively trigger and sustain inflammation, atherosclerosis, and organ-specific complications such as retinopathy, nephropathy, neuropathy, metabolic-dysfunction associated steatotic liver disease (MASLD), cardiopathy, and brain disease [5]. T2D is often linked with obesity, hence the term "diabesity" used to describe this common and dangerous combination [6].

Obesity is a chronic-relapsing disease characterized by an excess of adipose tissue resulting from prolonged mismatched energy intake exceeding energy expenditure [7]. Hypercaloric diets and sedentary lifestyles lead to the accumulation of lipids in subcutaneous (sc) fat depots, visceral adipose tissue, liver, muscle, and other ectopic sites [8]. These accumulated lipids release free fatty acids and pro-inflammatory cytokines, leading to systemic, low-grade, subclinical chronic inflammation associated with insulin resistance, exacerbating ectopic fat deposition and metabolic dysfunction, ultimately affecting renal, hepatic, cardiovascular, and endocrine health [7,9,10,11,12]. While not perfect, body mass index (BMI) remains the most commonly used tool for assessing obesity in epidemiological studies [7].

Assessment of the global burden of disease (GBD) 2021 database attributable to high BMI indicates a significant and consistent increase in global rates of obesity in adults over the past 30 years, with major variations among countries and region-specific trends. From 1990 to 2021, high BMI-attributable deaths and Disability Adjusted Life Years (DALYs) have risen significantly for both males and females. Females have seen a substantial increase in the absolute numbers of deaths and DALYs, while males have shown a more significant increase in age-standardized rates of these metrics. Age-standardized rates for women have remained relatively stable, whereas men have experienced a 15.0% increase in death rates and a 31.2% increase in DALYs [13]. Due to various interactions among genetic, economic, psycho-socio-cultural, behavioral, and pharmacological factors, the prevalence of obesity has risen globally, especially among certain racial/ethnic groups and individuals with lower educational attainment and income [14]. Individuals living with obesity face a higher annual financial burden in terms of healthcare expenses and medications compared to those of average weight [15]. The burden of both T2D and obesity has increased at a rapid pace among metabolic disorders [16], highlighting the urgent need to address these two common and closely linked metabolic disorders as a public health priority.

Glucagon-like peptide 1 (GLP-1) is an incretin hormone synthesized and released by L cells of the small and large bowel post-prandially, particularly after carbohydrate-containing meals [17]. Physiologically GLP-1 stimulates insulin secretion, inhibits glucagon secretion to lower blood glucose levels, slows gastric emptying, increases feelings of satiety, and reduces appetite. This leads to limited food intake and generally results in body weight loss (Figure 1) [18].

Various cell types express GLP-1 receptor (GLP-1R) including pancreatic alpha, beta and delta cells, neurons of the central nervous system (particularly hypothalamus and brain stem), endothelial cells, smooth muscle cells in blood vessels, cardiomyocytes and various regions of the gastrointestinal (GI) tract [19,20]. This diversity of target cells explains why a variety of physiological effects may result from the activity of the GLP-1 hormone, leading to GLP-1R agonists (GLP-1RAs) exerting a range of biological activities.

The discovery of GLP-1 by Joel Habener and Svetlana Mojsov in the 1980s, and Lotte Knudsen's development sustained-acting versions of this hormone for obesity in 2000, has been summarized by Friedman [21]. GLP-1RAs, by increasing insulin, reducing appetite, delaying stomach emptying, and decreasing liver glucose production are a pharmacological class of drugs (the most prescribed drugs include semaglutide, liraglutide, and dulaglutide) that mimic the natural GLP-1 hormone to help manage T2D and obesity by lowering blood sugar and promoting weight loss [22]. Table 1 summarizes the names, approved indications, route and timing of administration of the GLP-1RAs (Modified from: [23]).

The most common side effects of GLP-1RAs include nausea, vomiting, diarrhea, and constipation, which are often of limited severity and tend to improve over time [24]. Moreover, abdominal pain, anorexia, and headaches may occur and, more rarely, gallstones, pancreatitis, and severe allergic reactions are observed [25].

Pharmacological treatments for obesity provide a valuable adjunct to lifestyle intervention, which often achieves limited weight loss and is difficult to maintain [26]. In this context, GLP-1 RAs are positioned as a viable tool alongside lifestyle changes, often used as a step-up therapy when lifestyle modifications alone are insufficient. They can be used before bariatric surgery to aid in weight loss and improve safety, to complement the results of surgery or to prevent weight regain after surgery [27,28].

Liraglutide played a crucial role in the development of semaglutide, showing the effectiveness of a modified, long-acting GLP-1 analogue for managing diabetes and obesity [29]. Building on the success of liraglutide, semaglutide included a fatty acid moiety to enhance albumin binding and reduce clearance rates. Structural improvements increased albumin affinity, allowing for once-weekly dosing [29,30]. Here, we will comprehensively examine the key steps leading to the discovery and development of semaglutide that was initially developed for T2D, with subsequent recognition of its weight loss benefits [31]. To achieve this, we will incorporate evidence from closely interconnected fields such as medicinal chemistry and animal pharmacology. This will help us chart the human and non-human studies used to characterize cellular targets and pharmacological effects. Additionally, we will explore ongoing investigations into new applications and formulations of semaglutide in humans. Finally, we will address real-world data with a specific emphasis on post-marketing drug monitoring and safety concerns. Table 2 [32,33,34,35,36,37] provides an at-glance chronology of the key pre-clinical discoveries, clinical trials and regulatory decisions that have shaped the journey of semaglutide from its early laboratory design to its current benchmark status in metabolic medicine.

Building on the overview provided in Table 2, the following sections will guide the reader chronologically through the pivotal milestones that have shaped semaglutide's journey from molecular design to widespread clinical application.

The native incretin hormone GLP-1 is a 30-amino-acid peptide rapidly secreted after meals, capable of stimulating glucose-dependent insulin release. Unfortunately, its pharmacological utility is limited by a plasma half-life of only a few minutes, primarily due to (i) rapid N-terminal degradation by the exopeptidase dipeptidyl-peptidase-4 (DPP-4) and (ii) fast renal clearance of the small peptide [32]. The chemical architecture of semaglutide was therefore designed to address these two liabilities while retaining full agonistic activity at the GLP-1 receptor.

The enzymatic attack by DPP-4 occurs at the Ala8–Glu9 bond of endogenous GLP-1. In semaglutide, alanine at position 8 is substituted with the unnatural residue 2-aminoisobutyric acid (Aib, also termed 2-methyl-alanine). This single change sterically hinders DPP-4, rendering the analogue essentially resistant to proteolysis and preserving its biological integrity for many hours after administration [29,32].

Even when DPP-4-resistant, short linear peptides are removed quickly by glomerular filtration. To slow renal elimination, a C18 di-acid fatty chain is covalently attached to Lys26 through a spacer composed of two 8-amino-3,6-dioxaoctanoic acid units and a glutamic acid moiety. This hydrophobic appendage promotes non-covalent, high-affinity binding to circulating albumin, effectively turning the plasma protein into a reversible depot and raising the terminal half-life to roughly one week, compatible with once-weekly (OW) dosing in humans, which enhances accessibility compared to daily regimens [38].

A second Lys to Arg substitution at position 34 eliminates a potential competing acylation site, ensuring homogeneous conjugation exclusively at Lys26 and simplifying downstream purification. Apart from these deliberate modifications, the remaining residues map precisely onto GLP-1(7-37), preserving the receptor-interacting pharmacophore and maintaining picomolar potency. The final free-acid salt has the empirical formula C₁₈₇H₂₉₁N₄₅O₅₉ and a molecular weight of 4113 Da, which is still small enough for solid-phase peptide synthesis yet large enough, after acylation, to facilitate binding to serum albumin efficiently [32,37].

The three structural interventions (DPP-4 resistance, albumin tethering, single-site acylation) provide several pharmacological advantages: Extended exposure: The mean elimination half-life of approximately 160 h allowing for convenient OW sc or once-daily oral regimens. Stable receptor stimulation: The prolonged, yet still glucose-dependent insulinotropic action minimizes the risk of hypoglycemia. Improved patient adherence: The less frequent injections result in higher compliance compared to daily GLP-1 analogues. Manufacturing scalability: The site-specific chemical conjugation after yeast fermentation ensures high product purity at commercial scale.

Therefore, semaglutide represents a rationally engineered peptide whose molecular modifications directly align with the desired clinical profile: sustained glycemic control, significant weight reduction, and OW convenience without compromising safety or effectiveness.

Using homogeneous binding competition assays on plasma membranes prepared from BHK cells and BHK cells expressing a reporter construct containing human GLP-1R and the CRE firefly luciferase, it was demonstrated that semaglutide has high affinity for GLP-1R (0.38 ± 0.06 nM) [32]. Human embryonic kidney (HEK-293) cells that over-expressed the GLP-1 receptor were utilized to compare semaglutide with native GLP-1(7-37) [39,40]. Competitive radioligand displacement demonstrated sub-nanomolar affinity, which was comparable to liraglutide and only slightly weaker than the endogenous peptide [32]. In cyclic-AMP accumulation assays, semaglutide displayed full agonism with an EC₅₀ in the low-nanomolar range, and β-arrestin recruitment indicated balanced downstream signaling. The incorporation of 2-aminoisobutyric acid at position 8 made the molecule completely resistant to DPP-4 cleavage when incubated with recombinant enzyme for 24 h, while native GLP-1 lost 50% activity within minutes [32,38].

After sc dosing in rats (0.1–1.0 mg/kg) and cynomolgus monkeys (0.03–0.3 mg/kg), peak plasma concentrations were reached within 8–12 h and modulated food preference, reduced food intake, and caused weight loss without decreasing energy expenditure [41,42,43]. The albumin-binding C18 di-acid sidechain prolonged the elimination half-life to approximately 30 h in rats and 60 h in monkeys, which is roughly 10-fold longer than native GLP-1, while absolute bioavailability exceeded 85% [42]. No sex differences emerged in clearance or steady-state volume of distribution. Metabolite mapping revealed limited biotransformation: the major species in plasma remained intact parent peptide, with minor degradation products formed by endopeptidases rather than DPP-4, corroborating the in vitro stability data [44].

The efficacy of semaglutide was first demonstrated in diet-induced obese (DIO) mice [41]. Once-daily administration of semaglutide at a dose of 30 nmol/kg) normalized fed plasma glucose levels within 48 h and reduced HbA1c by approximately 1.5 percentage points over a four-week period. In db/db mice, weekly dosing of semaglutide at a dose of 100 nmol/kg led to a 30–40% decrease in non-fasting glucose levels, preserved pancreatic β-cell area, significantly improved liver damage, and improved gut microbiome compared to the control group that received only the vehicle [45]. Additionally, monitoring of food intake revealed a 15–20% reduction in mice treated with semaglutide, resulting in a cumulative weight loss of 8–10% compared to weight gain in the control group. When compared head-to-head with liraglutide at equivalent molar doses, semaglutide showed superior durability, which was attributed to its extended systemic exposure.

In order to meet regulatory expectations for first-in-human trials, a thorough safety pharmacology and toxicology program was carried out. This program included genetic toxicology screenings, assessment of organ-system functions, and long-term repeat-dose studies in two different species.

Safety and pharmacokinetics of oral semaglutide were investigated by Granhall et al. [33] in two double-blind, randomized, placebo-controlled trials (RCTs). In the single-dose trial, semaglutide exposure was highest when co-formulated with 300 mg of the absorption enhancer sodium N-(8-[2-hydroxybenzoyl] amino) caprylate (SNAC). In the multiple-dose trial, semaglutide exposure showed dose proportionality with 40 mg being twofold higher than 20 mg of oral semaglutide in healthy males, without any differences between healthy males compared to T2D male subjects. Based on this study, oral semaglutide co-formulated with 300 mg SNAC was chosen for further clinical development. Moreover, pharmacokinetics supported once-daily dosing of oral semaglutide.

The pharmacological properties of oral semaglutide were investigated by Xie et al. [49], who found that all doses of oral semaglutide exhibited dose-dependent increases in semaglutide exposure in 32 healthy Chinese adults at steady state. They also noted significantly reduced body weight and fasting plasma glucose levels (p = 0.0001 and p = 0.0011, respectively) compared to placebo. The most common adverse effects (AEs) reported were GI symptoms. However, there were no severe or confirmed symptomatic hypoglycemic events, serious AEs or deaths reported.

A systematic review of 17 published studies [50] found that semaglutide has a predictable pharmacokinetic profile, with a long half-life supporting OW sc dosing. Both oral and sc forms showed increased area under the curve (AUC) and C_max with higher doses. Factors such as food, water volume, and dosing schedules were found to influence drug the absorption of orally administered semaglutide. Drug–drug interactions with semaglutide were limited, requiring no dose adjustment fir subjects with upper GI disorders, impaired renal function, or impaired hepatic function. The study also suggested that exposure to semaglutide may be influenced by body weight, but further investigation is needed to confirm this.

Semaglutide bioavailability showed minor disparities between healthy subjects and those with T2D in terms of clearance and absorption rate, as well as between injection sites [51]. A trial involving 36 healthy Chinese individuals randomly assigned to receive OW sc semaglutide 0.5 mg (n = 12), 1.0 mg (n = 12), or placebo (n = 12) found no dose adjustment was necessary. The primary endpoint of the study was steady-state semaglutide exposure, defined as the area under the curve over a dosing interval at steady state (AUC_{0-168 h,SS}) [52].

Wharton et al. [53] assessed GI (AEs) occurring after sc 2.4 mg semaglutide OW in adults with overweight or obesity using pooled data from the Semaglutide Treatment Effect in People With Obesity (STEP) 1–3 trials for participants randomized to 68 weeks of semaglutide 2.4 mg (n = 2117) or placebo (n = 1262). Nausea, diarrhea, vomiting and constipation, the majority of which were non-serious, mild-to-moderate, transient, and occurring more often during or immediately after dose escalation, were more common with semaglutide 2.4 mg than placebo. However, these issues rarely resulted in treatment discontinuation. Mediation analysis showed that semaglutide-induced weight loss was largely independent of GI AEs, and maintenance treatment with semaglutide at the dose of 2.4 mg was well tolerated.

The effects of semaglutide on metabolic outcomes and gastric emptying were investigated by Hjerpsted et al. [54] in a double-blind RCT. These authors found that compared to placebo, semaglutide improved fasting and post-prandial glucose levels (as indicated by reduced glucose, insulin, and C-peptide serum concentrations) and lipid metabolism. It also delayed first-hour post meal gastric emptying, which may potentially contribute to a slower entry of glucose into the bloodstream, without affecting overall gastric emptying.

The dose–response relationship of semaglutide versus placebo on glycemic control among individuals with T2D was investigated in a 12-week, double-blind phase 2 RCT [30]. A total of 415 individuals were randomly assigned to receive either sc semaglutide OW with or without dose escalation, open-label liraglutide once daily, or placebo. The data showed that after 12 weeks, semaglutide reduced HbA1c in a dose-dependent manner and body weight among individuals with T2D. Mild-to-moderate GI AEs typical of GLP-1 RAs were mitigated by dose escalation. Based on this study semaglutide doses of 0.5 and 1.0 mg OW, with a 4-week dose escalation were selected for phase 3 trials.

O'Neil et al. [55] conducted a double-blind, multicenter phase 2 RCT with 957 non-diabetic adults living with obesity who were randomized to either the active treatment group or a matching placebo. The data showed that all doses of semaglutide induced significant reductions in bodyweight compared to placebo and liraglutide. The treatment was generally well tolerated, with no new safety issues identified. The most commonly observed AE being were dose-related GI symptoms, as expected with GLP-1RAs.

In their meta-analysis of 16 studies totaling 5997 patients, Karimi et al. [56] compared the effectiveness of semaglutide versus liraglutide, dulaglutide, or tirzepatide. The data indicated that semaglutide reduced HbA1c values compared to liraglutide. However, no significant differences were observed regarding fasting glucose, BMI, and weight change between semaglutide and liraglutide. Semaglutide was more effective than dulaglutide in reducing HbA1c levels and fasting glycemia although no significant differences were found in body weight and BMI changes. When compared to semaglutide tirzepatide was more effective at reducing HbA1c levels without any clear superiority in changes in body weight and fasting blood glucose. Regarding drug transition, moving from liraglutide to semaglutide did not significantly impact HbA1c levels but did induce weight loss and reduce fasting blood glucose. Conversely, transitioning from dulaglutide to semaglutide did not modify HbA1c and weight changes. The study concludes that analyses consistently demonstrate superior efficacy of semaglutide compared to liraglutide in reducing both HbA1c levels and weight. Moreover, it is more effective than dulaglutide in diminishing fasting glycemia. On the other hand, tirzepatide reduces HbA1c values more effectively than semaglutide.

To investigate changes in body weight and cardiometabolic risk factors after treatment, Wilding et al. conducted an extended analysis of 327 participants from the STEP 1 trial [57]. One year after discontinuing treatment with OW sc 2.4 mg semaglutide in combination with lifestyle intervention, participants regained approximately two-thirds of their initial weight loss, along with similar reversals in cardiometabolic variables. These findings highlight the persistent nature of obesity and suggest that continuous treatment is crucial for maintaining long-term improvements in weight and overall health.

A systematic review conducted by Feier et al. [48] analyzed 10 RCTs totaling 14,550 participants (7830 assigned to semaglutide). Additionally, 18 other studies reporting data from the same RCTs were separately discussed and assessed for the incidence of thyroid cancer and the spectrum of semaglutide-associated AE, emphasizing the importance for patient management. The incidence of thyroid cancer among subjects treated with semaglutide indicated no significant risk (<1%). AEs were mostly GI in nature, with nausea (2.05% to 19.95%) and diarrhea (1.4% to 13%) being the most common. Nasopharyngitis and vomiting were also notable, with prevalence rates of 8.23% and 5.97%, respectively. Other AEs included hyperlipasemia (6.5%), headaches (7.92%), anorexia (7%), influenza-like symptoms (5.23%), dyspepsia (5.18%), and constipation (6.91%). Serious AEs ranged from 7% to 25.2%, underscoring the importance of careful monitoring of patients. Importantly, the GI AEs associated with semaglutide are frequent but should not overshadow its treatment benefits, especially considering the low incidence of thyroid cancer.

The SUSTAIN trials investigated the efficacy of semaglutide in over 12,000 adults with T2D through a large-scale, multi-center, double-blind, RCT. Participants had a mean age of 57.4 years (25.8% were aged 65 or above), with 54.2% identifying as male and 78.6% as White [58]. The average duration of T2D was 7.4 years, with a baseline HbA1c of 8.5% and a mean BMI of 35.7 kg/m². The majority presented with hypertension and dyslipidemia and were receiving diabetes polypharmacy. Trial durations were 40 weeks or longer to evaluate long-term outcomes, with participants assigned in a 1:1 ratio to receive OW semaglutide or placebo [58].

To evaluate the effectiveness of semaglutide in achieving glycated HbA1c targets in individuals with T2D while minimizing adverse outcomes such as weight gain, hypoglycemia, and GI side effects, DeVries et al. [59] conducted an analysis of data from phase IIIa SUSTAIN 1 to 5 clinical trials. In total, 3,918 T2D participants were randomized to receive OW sc semaglutide at doses of 0.5 mg or 1.0 mg, or comparators including placebo, sitagliptin 100 mg, exenatide extended release 2.0 mg, or insulin glargine. The proportion of participants achieving HbA1c < 53 mmol/mol (7.0%) without weight gain or severe/BG-confirmed symptomatic hypoglycemia ranged from 47% to 66% with semaglutide 0.5 mg and 57% to 74% with semaglutide 1.0 mg, compared to 7% to 19% for placebo and 16% to 29% for active comparators (all p < 0.0001). These findings demonstrate that semaglutide is significantly more effective than comparators in helping patients meet HbA1c targets while avoiding undesirable outcomes.

In head-to-head comparisons, the 1.0 mg dose of semaglutide showed a greater reduction in HbA1c, and body weight compared to the 0.5 mg dose [60].

In the Semaglutide Effects on Cardiovascular Outcomes in People with Overweight or Obesity (SELECT) trial [61], semaglutide reduced major adverse cardiovascular events by 20% in adults with cardiovascular disease, overweight or obesity, and no diabetes. Over 208 weeks, semaglutide led to sustained weight loss (−10.2%), reduced waist circumference (−7.7 cm), and lower waist-to-height ratio (−6.9%) compared to placebo, with meaningful results across all subgroups. Serious AEs were less frequent with semaglutide, though discontinuation rates increased as baseline BMI decreased. Weight loss and anthropometric improvements persisted for up to four years [61].

The SUSTAIN trials have demonstrated that semaglutide significantly reduced body weight, BMI, and waist circumference compared to placebo. Additionally, this drug led to improvements in cardiometabolic markers, such as blood pressure and cholesterol levels. A substantial proportion of patients treated with semaglutide achieved clinically significant weight loss thresholds (≥5%, ≥10%, ≥15%), with these benefits appearing to be maintained over time, as indicated by data from the SUSTAIN 5 trial [62], which employed a different drug (PHEN/TPM CR). The studies further assessed adiposity indicators, including waist circumference and BMI, revealing that semaglutide contributed to reductions in these metrics and enhancements in health outcomes related to cardiovascular risk and glycemic control.

The SUSTAIN-6 trial [63] showed that OW semaglutide significantly reduced the risk of major adverse cardiovascular events (MACE) by 26% compared to placebo in patients with T2D. This reduction in MACE was consistent across various subgroups based on sex, age, and baseline cardiovascular risk levels. The primary components of MACE (cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke) were also reduced, although the reduction in cardiovascular death alone was not statistically significant in the primary analysis [63].

Building on these findings, Lincoff et al. [35] reported that in their multicenter, double-blind, event-driven superiority RCT of 17,604 individuals (8803 receiving semaglutide and 8801 receiving placebo) with a mean (± SD) duration of exposure of 34.2 ± 13.7 months, weekly sc semaglutide at a dose of 2.4 mg was superior to placebo in reducing the incidence of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke at a mean follow-up of 39.8 months. (Funded by Novo Nordisk; SELECT ClinicalTrials.gov number NCT03574597).

Aroda et al. [64] assessed the safety and tolerability of semaglutide in the SUSTAIN (subcutaneous) and PIONEER (oral) phase IIIa trials, which included 11,159 patients with type 2 diabetes. GI AEs were reported in 41.9% of those receiving sc semaglutide and 39.1% of those receiving oral semaglutide, compared to 22.0% and 24.8%, respectively, in comparator groups, primarily at the initiation of treatment. The incidence of fatal AEs, renal disorders, acute pancreatitis, malignant neoplasms, hypoglycemia, diabetic retinopathy, heart failure, and other cardiovascular events was comparable across all cohorts. Gallstone formation was more common among patients treated with semaglutide, and an increased rate of diabetic retinopathy was observed with sc administration in the SUSTAIN 6 trial. Both formulations resulted in a modest elevation in pulse rate without an increase in arrhythmia risk. Sc semaglutide demonstrated a reduction in major adverse cardiovascular event (MACE) risk, while oral semaglutide was shown to be non-inferior.

In their analysis of the US Komodo Health data, Ruseva et al. [65] identified adults with a BMI ≥ 30 kg/m² or those with a BMI between 25 and 29.9 kg/m² accompanied by at least one weight-related comorbidity. Following two years of semaglutide therapy (1.7 or 2.4 mg), results from 179 patients revealed a mean weight reduction of −17.5 kg (−16.2%), while 369 patients experienced a mean decrease in BMI of −6.0 kg/m² (p < 0.0001). Notably, the proportions of patients achieving weight reductions of ≥5%, ≥10%, ≥15%, and ≥20% were 90.5%, 69.8%, 46.9%, and 29.6%, respectively. These outcomes demonstrate clinically significant and sustained decreases in both weight and BMI.

The STEP trials played a crucial role in gathering evidence to establish semaglutide's formal indication for weight management in adults with obesity or overweight, ultimately leading to its approval in the US, Europe, UK, and Canada. These trials collectively showed that a weekly 2.4 mg sc dose, when combined with lifestyle intervention, significantly reduces body weight and improves associated health conditions. Semaglutide is now approved for patients with a BMI of ≥30 kg/m² or a BMI of ≥27 kg/m² with at least one weight-related comorbidity [66].

The safety of OW sc 2.4 mg semaglutide versus placebo, beyond the reduction in MACEs among individuals living with overweight or obesity and exhibiting overt CVD has been evaluated by Kushner et al. [84]. Data has shown that the fraction of individuals experiencing severe AEs (SAEs) was significantly lower with semaglutide than placebo, attributed to a decrease in heart disease. Moreover, semaglutide led to significantly more discontinuations due to AEs than placebo, primarily from GI AEs, while the rates of discontinuations due to serious AEss and suicide/self-injury were not statistically different. Gallbladder disorders occurred significantly more often with semaglutide due to gallstones; conversely, the incidence of cholecystitis was identical. Therefore, this study confirmed that the SELECT study reaffirms the long-term safety profile observed in previous studies, without identifying any novel safety concerns for OW 2.4 mg semaglutide.

Chiappini et al. [85] conducted the first descriptive analysis of the characteristics of Adverse Event Reports (AER) documenting the misuse and abuse potential of semaglutide compared to other GLP-1RAs (e.g., albiglutide, dulaglutide, exenatide, liraglutide, lixisenatide, and tirzepatide) by analyzing the FDA Adverse Events Reporting System (FAERS) pharmacovigilance data set. In the study period (January 2018–December 2022), the most represented reports involved dulaglutide (37.6%), semaglutide (26.1%), and liraglutide (25.0%), totaling 31,439 AERs. Semaglutide exhibited an increased number of reported AERs compared to other GLP-1RAs, with the majority of reports from the US involving female adults. Reports of drug misuse, abuse, and withdrawal were most commonly associated with semaglutide compared to the other selected GLP-1 analogues. Specifically, 'drug abuse', 'drug withdrawal syndrome', and 'prescription drug used without a prescription' were reported more than 3.5 times as frequently, and 'intentional product use issue' was reported almost two times as frequently indicating that further studies are essential to determine the extent and type of semaglutide misuse/abuse among the general public and vulnerable sub-groups of individuals.

Finally, the European Medicines Agency (EMA) safety committee Pharmacovigilance Risk Assessment Committee (PRAC) has completed its review of medicines containing semaglutide. This review was prompted by concerns about a potential increased risk of developing non-arteritic anterior ischemic optic neuropathy (NAION), an eye condition that can lead to vision loss. PRAC has determined that NAION is a very rare side effect of semaglutide medicines including Ozempic, Rybelsus and Wegovy [86]. As a result, PRAC has recommended that NAION be included in the product information for semaglutide medicines as a side effect with a frequency of 'very rare' (affecting up to 1 in 10,000 people). If patients using semaglutide experience a sudden loss of vision or a rapid deterioration in eyesight, they should promptly contact their doctor. If NAION is confirmed, treatment with semaglutide should be discontinued.

By reviewing the mechanisms through which semaglutide may act centrally and peripherally in regulating appetite and energy homeostasis, Moiz et al. [18] pinpoint that semaglutide exerts an anorexigenic effect by mimicking the hormone GLP-1 in the Central Nervous System. This is achieved through the activation of appetite-suppressing neurons and inhibition of appetite-stimulating neurons in the arcuate nucleus of the hypothalamus, serving as the primary mechanism of action, as well as in other brain areas. Furthermore, semaglutide also contributes to appetite control by impacting dopamine reward signaling, gastric emptying, and influencing other cerebral pathways. Additionally, semaglutide may act synergistically with the hormone leptin to reduce food intake and increase energy expenditure.

A monocentric double-blind, placebo-controlled, parallel-group trial was conducted by Kapitza et al. [87]. These authors enrolled 75 adults with T2D who were being treated with diet and exercise and/or metformin monotherapy. The participants were randomly assigned to OW s.c. semaglutide 1.0 mg (n = 37) or a placebo (n = 38). The data showed significant improvement in β-cell function and glycemic control, as assessed by fasting, postprandial, and overall (AUC_0–24h) glucose and glucagon responses, as well as insulin secretion rate, among those with T2D who received semaglutide.

Semaglutide exerts anti-inflammatory and cardio-renal protective mechanisms through direct and indirect effects. It down-regulates systemic inflammation by modulating immune responses in adipose tissue and inhibiting pro-inflammatory pathways such as interleukin-6, tumor necrosis factor-α, and S100A8/A9 [88].

The anti-inflammatory and anti-atherogenic properties of semaglutide, along with reductions in blood pressure, body weight, and improvements in lipid profile, contribute to cardiovascular protection [89]. Direct renal protection is achieved through decreased inflammation, fibrosis, and albuminuria, while renal protection is achieved through improved glycemic homeostasis, blood pressure, and body weight [90]. Together, these mechanisms significantly reduce the risk of CKD progression, kidney failure, or cardiovascular death in patients with T2D and CKD [35].

Despite the extensive SUSTAIN and STEP programs, several knowledge gaps remain. First, most trials have only followed participants for up to 2 years. Longer surveillance is necessary to understand the durability of glycemic control, cardiovascular protection and the extent of weight regain after drug discontinuation, as seen in extension studies. Second, the mechanisms behind inter-individual variability in efficacy and GI tolerability are not well understood. Multi-omics profiling (genomics, metabolomics, microbiomics) could help identify predictive biomarkers to guide dose selection and reduce dropouts. Third, there is still limited pediatric data, especially regarding long-term effects on bone accrual, pubertal development and neuro-cognitive endpoints in children and adolescents. Lastly, while semaglutide is being explored beyond diabetes and obesity, phase II and III programs must confirm benefits seen in surrogate markers translate to actual clinical outcomes.

Chemical innovation surrounding the GLP-1 pharmacophore is progressing rapidly. Oral semaglutide tablets are already on the market, and next-generation analogues with dual or triple receptor activity (e.g., GLP-1/GIP or GLP-1/glucagon co-agonists) show promise for greater weight reduction [91]. Artificial intelligence-driven approaches that integrate deep learning-based protein design with functional screening for isolation of long-acting GLP-1 RA are on the way to help in designing novel compound with improved stability and efficacy [92]. Combination therapies are also being explored, such as a phase II study combining semaglutide with the FXR agonist cilofexor and the ACC inhibitor firsocostat, which showed greater improvements in hepatic steatosis and metabolic biomarkers compared to semaglutide alone. In addition, digital therapeutics, mobile applications that offer dietary coaching, physical-activity tracking and adherence prompts, could help reinforce lifestyle changes and prevent post-withdrawal weight regain. Lastly, the production of biosimilars after patent exclusivity ends, or the use of compulsory-licensing statutes by governments, may increase access and control costs in high-prevalence settings.

An emerging strategy to amplify the clinical impact of semaglutide is to administer it alongside peptide partners that engage complementary metabolic pathways [26,91]. The leading example is its co-development with cagrilintide (formerly AM833), a long-acting analogue of amylin, the hormone co-secreted with insulin that slows gastric emptying, suppresses glucagon after meals and enhances meal-related satiety. Whereas semaglutide acts predominantly through GLP-1 receptors to boost glucose-dependent insulin secretion and centrally dampen appetite, cagrilintide stimulates amylin receptors located in distinct hypothalamic and brain-stem nuclei [93]. Pre-clinical studies have shown that the two signals converge additively, producing larger body-weight reductions than either peptide alone and without reciprocal attenuation of glucoregulatory effects.

This concept has already advanced to the clinic. In a multinational phase-II trial, adults with obesity received OW semaglutide 2.4 mg combined with escalating doses of cagrilintide. After 32 weeks, mean weight loss in the highest-dose cohort reached approximately 17–18% of baseline, outstripping the ≈14% loss observed with semaglutide monotherapy; glycemic, blood-pressure and lipid endpoints improved in parallel. The GI profile (chiefly nausea and vomiting) was slightly more pronounced in the combination group, but most events were transient and of mild-to-moderate severity, leading to discontinuation rates comparable with the single-agent arm [26]. These data have triggered a global phase-III program (REDEFINE) that will test long-term cardiovascular and renal outcomes as well as durability of weight control [94].

Pharmacologically, the peptide pair is well-matched: both molecules have been engineered for OW sc injection via albumin-binding fatty-acid acylation, allowing co-formulation without complex titration schedules. No relevant pharmacokinetic interaction has been detected to date. Nevertheless, careful dose-escalation remains essential because the overlapping satiety signals can potentiate early GI. Of note, modelling work suggests that combined GLP-1 and amylin receptor agonism may attenuate the post-treatment weight regain sometimes observed after semaglutide withdrawal, an effect already hinted at in extension cohorts.

The cagrilintide partnership is not the only avenue for combination therapy [39]. A phase II proof-of-concept study in non-alcoholic steatohepatitis combined semaglutide with the FXR agonist cilofexor, and the acetyl-CoA carboxylase (ACC) inhibitor firsocostat [95]. The triple regimen produced larger reductions in hepatic fat content and broader metabolic benefits than semaglutide alone while maintaining acceptable tolerability. Parallel investigations are pairing semaglutide with SGLT2 inhibitors, GIP co-agonists and centrally acting appetite suppressants, all aiming to transcend the ceiling of efficacy achievable through single-pathway modulation.

From a regulatory perspective, the use of two peptides that already possess extensive safety dossiers may permit streamlined approval pathways, provided that post-marketing pharmacovigilance confirms the absence of novel safety liabilities. Economically, the dual-peptide approach will face scrutiny because of its higher acquisition cost, yet modeling analyses suggest that deeper and more durable weight loss could shorten time-to-goal and lower downstream cardiovascular and renal expenditures, partially offsetting upfront spending.

Ultimately, biomarker-guided precision medicine, leveraging baseline amylin levels, eating-behavior phenotypes or genetic polymorphisms in amylin-receptor signaling, may identify the sub-populations most likely to derive incremental benefit, thereby maximizing the therapeutic and economic value of GLP-1/amylin co-therapy. As such, the semaglutide-cagrilintide paradigm exemplifies how layering complementary hormonal cues can push pharmacological weight loss toward the territory historically reserved for bariatric surgery, while preserving the established cardiometabolic advantages of GLP-1 receptor agonism. This blueprint is rapidly being emulated by other combination strategies that seek to deliver multidimensional, disease-modifying benefits for complex metabolic disorders.

A striking feature of the SUSTAIN and STEP programs is the broad dispersion of individual outcomes: some participants lose more than 20% of their initial body weight, whereas others achieve only modest reductions despite identical dosing and adherence. Understanding this heterogeneity is pivotal for maximizing therapeutic benefit and for containing the substantial drug-acquisition costs that currently limit access.

The search for predictive biomarkers begins with pharmacogenetics. Single-nucleotide polymorphisms in GLP-1R or in genes governing peptide trafficking (e.g., DPP4, albumin) could modulate receptor signaling, degradation kinetics or tissue distribution, thereby influencing both efficacy and GI tolerability. Early sequencing studies have already linked certain GLP-1R variants to differential HbA1c responses with older incretin analogues; prospective genotyping in forthcoming semaglutide trials should clarify clinical relevance. Beyond fixed DNA markers, dynamic "omics" layers offer additional discriminatory power. Baseline metabolomic signatures that reflect hepatic de novo lipogenesis, branched-chain amino-acid catabolism, or bile-acid pools may stratify patients by their propensity to mobilize adipose stores under GLP-1 agonism. Likewise, gut-microbiome profiles enriched in Prevotella or Bacteroides species—taxa implicated in short-chain fatty-acid production and energy harvest—might predict the degree of weight loss achieved when appetite is pharmacologically suppressed.

Clinical phenotyping remains equally informative. Patients with pronounced visceral adiposity, high baseline fasting insulin, or rapid gastric emptying have shown larger weight-loss increments and fewer hypoglycemic events in post hoc analyses. Continuous glucose-monitoring data further reveal that responders display an early, sharp attenuation of post-prandial excursions within the first two weeks, whereas non-responders exhibit a flatter trajectory. Incorporating those early digital biomarkers into adaptive algorithms could allow dose intensification or timely switch to alternative therapies.

Safety prediction is another frontier. Polymorphisms in serotonin-receptor genes, variants in the calcitonin pathway, or pre-existing dysbiosis characterized by methane-producing archaea may forecast susceptibility to nausea, gall-bladder events, or constipation, respectively. Pre-emptive identification of such risk factors would enable slower titration schedules, prophylactic anti-emetics, or selection of alternative weight-management strategies.

Implementation science will ultimately determine the clinical utility of these candidate markers. Pragmatic trials that embed genotyping, metabolomics, and microbiome sequencing into routine care can generate real-world evidence on cost-effectiveness. Machine-learning models fed with multi-modal data—demographics, genetics, laboratory values, CGM streams, wearable-derived activity metrics—can then produce individualized response probabilities delivered seamlessly to electronic health-record systems. Such precision triage is especially pertinent given recent estimates that 4.5 million U.S. adults meet eligibility criteria for semaglutide therapy: even modest improvements in response rates or discontinuation avoidance would translate into substantial absolute gains in population health and budgetary sustainability.

In short, biomarker-guided precision medicine offers a pragmatic path to amplify the impact of semaglutide while safeguarding health-system resources. By identifying likely responders, forecasting adverse-event risk, and informing dynamic dose adjustments, such an approach can convert a "one-size-fits-all" injectable into a tailored intervention fully aligned with the principles of value-based metabolic care.

The precise use of semaglutide is likely to improve both effectiveness and cost-efficiency. Algorithms that consider baseline BMI category, visceral-fat distribution, renal function, gut-hormone profiles and behavioral traits could help predict responders versus non-responders. This information could also guide decisions on whether monotherapy, combination therapy or bariatric surgery is the most appropriate course of action. Continuous glucose monitoring and smart-pen data streams can be used to adjust doses in real-time using machine learning models and shape diagnostics [96]. Similarly, analyzing patient reviews of semaglutide, with the goal of better understanding its real-world effectiveness and safety for weight management might help with surveillance and improve individualized regimens [97]. This approach can help maintain glycemic targets while minimizing GI AEs. Furthermore, pharmacogenetic screening for variants that affect peptide clearance or GLP-1 receptor signaling may eventually support genotype-directed titration schedules. With up to 4.5 million U.S. adults meeting eligibility criteria for semaglutide therapy, it will be crucial to deploy this treatment in a stratified manner to maximize its public health impact within limited budgets.