Integrative Biomedical Research

1. Introduction

There is a certain vertigo in trying to write about obesity care right now, because the ground keeps moving. A decade ago, a clinician counseling a patient with severe obesity had, realistically, diet, exercise, and perhaps a referral for bariatric surgery to offer; today, the same conversation might include a once-weekly injection capable of producing weight loss that rivals surgical outcomes. It is tempting to call this a revolution, and in many respects it is one. But revolutions are rarely tidy, and this one has left behind a trail of open questions that this review tries, however incompletely, to work through.

Obesity itself has been reframed. It is no longer adequate, clinically or scientifically, to describe it as simple caloric excess or a failure of willpower; it is now understood as a chronic, progressive, and relapsing multisystem disease that disrupts homeostatic regulation across nearly every organ system (Lempesis & Dalamaga, 2026; Abdallah et al., 2026). The numbers involved are, frankly, difficult to hold in one's head. The Global Burden of Disease Study 2021 estimated that 2.11 billion adults worldwide were living with overweight or obesity that year, and the trajectory projected forward suggests that close to two-thirds of adults over 25 will be affected by 2050 (Lempesis & Dalamaga, 2026). This is not a distant or abstract epidemiological curiosity-it is a burden that spans childhood through old age, and one that accelerates premature death through chronic cardiorenal disease, type 2 diabetes mellitus (T2DM), metabolic dysfunction-associated steatotic liver disease (MASLD), and a widening list of obesity-related malignancies (Salama et al., 2025; Podder et al., 2026; Abdallah et al., 2026). Behavioral and lifestyle interventions remain, at least in principle, the foundation of obesity management, yet their real-world performance is sobering; adherence erodes over time, and attrition from structured lifestyle programs is the rule rather than the exception (Salama et al., 2025; de Arriba Munoz et al., 2025). It was against this backdrop of limited, hard-won, and often disappointing behavioral efficacy that pharmacological innovation became less of an option and more of an imperative (Podder et al., 2026; Abdallah et al., 2026).

What has followed can reasonably be described as a pharmacological revolution, one centered on the therapeutic exploitation of incretin biology (Garcia-Gorrita et al., 2025; Abdallah et al., 2026). Glucagon-like peptide-1 receptor agonists (GLP-1RAs) and, more recently, multi-receptor co-agonists have pushed the ceiling of non-surgical weight loss to a place few clinicians would have predicted even five years ago (Sancho-Haro et al., 2026; Abdallah et al., 2026). The underlying biology is elegant, if not entirely simple: endogenous GLP-1 and glucose-dependent insulinotropic polypeptide (GIP) are released from the gut in response to a meal, and together they orchestrate a pleiotropic set of metabolic responses that extend well beyond glycemic control (La Vignera et al., 2026). Synthetic mimics of this system, such as once-weekly semaglutide 2.4 mg, act on hypothalamic satiety circuitry while also slowing gastric emptying, and in doing so achieve mean weight reductions of approximately 15% (Celebi et al., 2026; Pardali et al., 2026). Tirzepatide, which co-activates both the GIP and GLP-1 receptors, pushes further still, with trial-level weight reductions between 20.9% and 22.5% attributable to the way it recruits complementary neuroendocrine and adipose-tissue pathways simultaneously (Pardali et al., 2026; Abdallah et al., 2026). For the first time, pharmacotherapy is closing-not just narrowing, but genuinely closing-the historic efficacy gap with bariatric surgery, and doing so without the perioperative risk that surgery necessarily carries (La Vignera et al., 2026).

And yet. It would be a disservice to the field, and arguably a disservice to patients, to stop the story there. Underneath the striking topline efficacy numbers sit several biological and physiological limitations that are only now coming into full view (Lempesis & Dalamaga, 2026). Chief among these is the fact that obesity appears to be biologically defended with a tenacity that pharmacotherapy has not fully overcome; weight loss tends to plateau around 18 months, and stopping the drug does not simply freeze progress in place-it reverses it (Lempesis & Dalamaga, 2026; Abdallah et al., 2026). Within roughly a year of discontinuation, patients regain about two-thirds of the weight they had lost, and the cardiometabolic gains that had accompanied that loss tend to erode alongside it (Lasik & Ukleja-Sokolowska, 2026; Abdallah et al., 2026). Compounding this, the composition of the weight lost during treatment is not uniformly favorable: rapid pharmacologically induced energy restriction disproportionately strips lean tissue alongside fat, with skeletal muscle and lean mass accounting for somewhere between 25% and 40% of total weight reduction under GLP-1RA and dual co-agonist therapy (Lempesis & Dalamaga, 2026; Sancho-Haro et al., 2026). This raises legitimate concern about sarcopenic obesity, declining resting metabolic rate, and physical frailty-concerns that become sharper still in older adults and in patients who already carry metabolic comorbidities (Santic et al., 2026; Sancho-Haro et al., 2026).

Tolerability presents its own set of complications. Gastrointestinal adverse events-nausea, vomiting, diarrhea, constipation, in varying combinations-affect up to 80% of treated individuals and represent, by most accounts, the single most common reason patients stop taking these drugs early (Pardali et al., 2026). These symptoms are usually transient and cluster around dose titration, but "usually transient" is cold comfort to a patient living through them, and real-world cohort data suggest that 46% to 65% of patients discontinue therapy within the first year regardless (Podder et al., 2026). Beyond the everyday tolerability burden, there are less common but more serious signals-modestly elevated risks of cholelithiasis and cholecystitis, and questions about interaction with pre-existing diabetic retinopathy-that warrant ongoing surveillance (Podder et al., 2026). And for the newer dual- and triple-receptor agonists now entering the pipeline, the honest answer is that we simply do not yet know what decades of sustained receptor activation will mean for thyroid C-cell biology, pancreatic health, or other outcomes that only reveal themselves over long follow-up (Lempesis & Dalamaga, 2026; Podder et al., 2026).

None of this occurs on a level playing field. Perhaps the starkest limitation of the incretin era is not biological at all, but economic. List prices for second-generation anti-obesity medications in the United States sit between $1,000 and $1,600 per month, a figure that effectively prices most eligible patients out of the market absent substantial insurance support (Podder et al., 2026). Coverage, when it exists, is inconsistent and often restrictive-Medicare, for instance, is statutorily barred from covering GLP-1RAs for obesity indications alone (Podder et al., 2026). The consequence is predictable, if no less troubling for being so: prescribing and fill rates track closely with socioeconomic status, insurance type, and geography, leaving Black, Hispanic, and socially vulnerable populations-groups that already carry a disproportionate share of the obesity burden-with disproportionately lower access (Podder et al., 2026). Absent deliberate policy intervention, it is difficult to see how the rapid diffusion of these high-cost therapies does anything but widen existing health disparities (Podder et al., 2026).

Finally, and perhaps most uncomfortably, many of these unknowns are amplified rather than resolved when the population in question is children. Childhood obesity has itself become epidemic in scope (Salama et al., 2025), and while semaglutide and liraglutide now carry FDA approval for adolescents aged 12 and older, younger children remain, for the most part, an evidence vacuum (Salama et al., 2025). What sustained pharmacological appetite suppression means for pubertal timing, physical development, and psychiatric wellbeing over years-not months-of exposure is genuinely unknown (Batra et al., 2026; Abdallah et al., 2026). Regulators have already flagged concerns about slowed growth velocity and delayed bone mineral density accrual with prolonged exposure, concerns serious enough to justify routine pediatric surveillance rather than reassurance (Abdallah et al., 2026).

It is this tension-extraordinary efficacy set against biological limitation, tolerability burden, inequitable access, and pediatric uncertainty-that motivates the present review. Rather than treating GLP-1-era pharmacotherapy as a solved problem, we approach it as a rapidly maturing but still incomplete story, and we structure the remainder of this review around three research questions and five objectives that, together, attempt to map both what is now well established and what still needs to be worked out.

Several open questions frame the analysis that follows. The first concerns phenotypic variation: how might different clinical presentations—what some clinicians informally describe as "slow burn" or "hungry gut" phenotypes—shape not just the rate of weight loss but its composition, particularly the ratio of fat to lean mass lost, in adults treated with dual GIP/GLP-1 agonists (Garcia-Gorrita et al., 2025; Santic et al., 2026)? A second, more structural question asks to what extent socioeconomic status and geographic variation in insurance-coverage policy independently predict real-world 12-month persistence with GLP-1RA therapy among racially diverse populations living with severe obesity (Podder et al., 2026). A third question turns toward the pediatric population and toward time horizons the current literature has barely begun to address: what is the long-term safety profile—specifically with respect to renal function and bone mineral density—and the developmental impact of sustained GLP-1 receptor activation in children aged 6 to 12, followed over five years (Salama et al., 2025; Abdallah et al., 2026)?

Building from these questions, this review pursues five interconnected objectives. The first is mechanistic: to elucidate the molecular and physiological basis of GLP-1, GIP, and glucagon receptor co-agonism, and to evaluate how synergistic pathway activation improves metabolic parameters relative to single-receptor agonism (Lempesis & Dalamaga, 2026; Podder et al., 2026). The second turns to real-world performance rather than trial efficacy, aiming to quantify the global persistence and adherence gap across modern obesity pharmacotherapies in diverse healthcare systems, and to identify the clinical, tolerability, and financial factors that most consistently drive early discontinuation (Podder et al., 2026; Dziewierz & Siudak, 2026). Third, the review examines the impact of rapid pharmacological weight loss on body composition, with particular attention to skeletal muscle quality, neuromuscular function, and the extent to which precision protein nutrition and structured resistance exercise can preserve lean mass during treatment (Santic et al., 2026; Sancho-Haro et al., 2026). Fourth, it appraises the clinical, ethical, and policy dimensions of this therapeutic era, specifically how high acquisition costs and restrictive insurance coverage contribute to racial, ethnic, and geographic inequity in access to evidence-based obesity care (Podder et al., 2026). Finally, the review works toward something more constructive than critique alone: a multidisciplinary clinical management framework that integrates structured dietetic support, pediatric monitoring protocols, and behavioral intervention, with the goal of helping patients move from pharmacologically supported appetite suppression toward durable, longer-term metabolic rehabilitation (Lasik & Ukleja-Sokolowska, 2026; Balasubaramaniam et al., 2026).

2. Re-Evaluating Pediatric and Adult Obesity in the Incretin Era

2.1 Obesity Reconceived: From Body Mass Index to a Multisystem, Staged Disease

It has become almost a cliche to say that obesity is "not just about willpower," and yet the clinical infrastructure of medicine has been slow to catch up with that idea. Obesity is now understood-properly, we would argue-as a complex, progressive, and relapsing multisystem disease, one that impairs nearly every physiological axis it touches (Abdallah et al., 2026; Lempesis & Dalamaga, 2026). The scale involved is difficult to overstate: excess adiposity, combined with underweight, now affects more than 800 million adults and roughly 160 million children and adolescents worldwide (Garcia-Gorrita et al., 2025; Abdallah et al., 2026). Much of the downstream damage traces back to visceral fat, which drives chronic low-grade systemic inflammation, oxidative stress, and ectopic lipid deposition in organs that were never built to store fat (Sado et al., 2026; Abdallah et al., 2026). The clinical fallout is familiar to anyone working in metabolic medicine-type 2 diabetes mellitus, metabolic dysfunction-associated steatotic liver disease, and the increasingly used umbrella term cardiovascular-kidney-metabolic (CKM) syndrome (Bharaj et al., 2025; Sado et al., 2026).

Recognizing the limits of body mass index as a sole diagnostic anchor, the 2025 Lancet Commission proposed something of a paradigm shift (Abdallah et al., 2026). It separates "preclinical obesity"-excess adiposity without demonstrable organ dysfunction, but carrying high future cardiometabolic risk, often flagged by a waist-to-height ratio above 0.5-from "clinical obesity," in which measurable organ harm is already present across metabolic, cardiovascular, or physical domains (Abdallah et al., 2026). Not everyone is convinced this framing helps; the European Society for the Study of Obesity has voiced concern that a "preclinical" label might, paradoxically, breed clinical inertia rather than prompt earlier action (Abdallah et al., 2026). Still, whatever one makes of the semantics, the underlying logic-that risk stratification should occur before irreversible organ damage sets in-seems difficult to argue against (Salama et al., 2025; Abdallah et al., 2026).

The roots of this crisis, moreover, reach back further than adulthood. Pediatric obesity has itself reached epidemic proportions, and it does not sit quietly in childhood-it drags early clustering of cardiometabolic risk factors (insulin resistance, dysglycemia, dyslipidemia) along with it, seeding subclinical atherosclerosis well before adulthood begins (Salama et al., 2025; Abdallah et al., 2026). Given that close to 80% of adolescents with severe obesity remain obese as adults, the case for early, scalable public health intervention writes itself (Abdallah et al., 2026). Chile's 2016 Food Labeling and Advertising Law offers one of the few structural success stories here: mandatory warning labels on ultra-processed foods, paired with restrictions on child-targeted marketing, measurably reduced the probability of early childhood excess weight at a population level (Abdallah et al., 2026).

2.2 The Incretin Revolution: Mechanistic Basis of GLP-1 and Multi-Receptor Agonism

The mechanistic story underlying modern obesity pharmacotherapy begins, somewhat elegantly, in the gut. Endogenous GLP-1 and GIP are secreted postprandially by intestinal L-cells and K-cells respectively, coordinating glucose-dependent insulin secretion, glucagon suppression, and broader nutrient handling (La Vignera et al., 2026; Pardali et al., 2026). In obesity and T2DM, this endogenous signal is blunted-both secretion and receptor sensitivity decline (Podder et al., 2026). Synthetic GLP-1 receptor agonists sidestep this deficit entirely by achieving supraphysiological receptor occupancy, which stimulates hypothalamic and hindbrain satiety centers, slows gastric emptying, and appears to quiet the intrusive "food noise" that drives much of overeating behavior (La Vignera et al., 2026; Dziewierz & Siudak, 2026; Podder et al., 2026). (Figure 1 summarizes this receptor-level cascade schematically.)

What has changed the field most in the last two to three years, though, is the move from single-peptide analogs toward multi-receptor co-agonism (Lempesis & Dalamaga, 2026; Podder et al., 2026). Tirzepatide, a single acylated peptide, activates both GLP-1 and GIP receptors simultaneously (Vasile et al., 2026; Podder et al., 2026). The GLP-1 arm handles the now-familiar anorexigenic and gastric-slowing effects, while the GIP arm contributes something distinct-enhanced white adipose tissue lipid buffering and improved insulin sensitivity-that appears to act synergistically rather than redundantly (Vasile et al., 2026; Podder et al., 2026). There is also preclinical evidence, worth flagging even if still early, that GIP receptor signaling within the area postrema may blunt the nausea and vomiting pathways typically triggered by strong GLP-1 stimulation, which would help explain the somewhat improved tolerability profile seen with dual agonism relative to what one might predict from GLP-1 potency alone (Vasile et al., 2026). Clinically, this "twincretin" effect translates into weight reductions up to 20.9%, compared with roughly 15% for selective GLP-1RAs (Pardali et al., 2026; Podder et al., 2026).

The pipeline does not stop at dual agonism. Retatrutide adds a third receptor to the mix, co-activating GLP-1, GIP, and glucagon (GCG) receptors (Lempesis & Dalamaga, 2026; Podder et al., 2026). The glucagon component is mechanistically interesting in its own right: it stimulates hepatic mitochondrial oxidation and raises resting energy expenditure, which helps counteract the metabolic-rate decline that typically accompanies rapid weight loss-a compensatory adaptation that has, historically, undermined the durability of weight-loss interventions (Lempesis & Dalamaga, 2026; Podder et al., 2026). In phase 2 data, retatrutide produced weight reductions as high as 24.2% at 48 weeks and normalized liver fat content in more than 85% of treated participants, a striking result for a disease (MASLD) that has otherwise proven stubbornly difficult to treat pharmacologically (Lempesis & Dalamaga, 2026; Abdallah et al., 2026; Podder et al., 2026). Combination approaches are advancing in parallel-CagriSema, pairing semaglutide with the amylin analog cagrilintide, is now in phase 3 evaluation and appears to recruit distinct neuroendocrine satiety circuitry, which may offer a further efficacy increment without necessarily sacrificing tolerability (Lempesis & Dalamaga, 2026; Pardali et al., 2026).

2.3 Pediatric Obesity Care and the Emerging Role of Digital Health Interventions

Pediatric obesity poses a genuinely different clinical problem than its adult counterpart, not least because premature cardiometabolic injury can begin before a child has finished growing (de Arriba Munoz et al., 2025; Salama et al., 2025). Although semaglutide and liraglutide now carry FDA approval from age 12 upward, real-world pediatric pharmacotherapy is dogged by attrition, inconsistent persistence, and legitimate developmental concern (Salama et al., 2025; Batra et al., 2026). Unlike adults, children cannot really be treated in isolation from their family system-durable behavior change in this age group typically requires a family-centered ecosystem, not just a prescription (de Arriba Munoz et al., 2025; Abdallah et al., 2026).

One promising, if still early, response to this gap comes from digital health interventions. de Arriba Munoz et al. (2025) evaluated a mobile precision-health platform-the Adhera Caring Digital Program-layered on top of standard pediatric pharmacotherapy (GLP-1RAs, metformin, or orlistat) in 40 families. The platform combines AI-driven personalization with cognitive-behavioral educational content and real-time family progress tracking. At a 150-day interim analysis, the program showed high usability and strong behavioral engagement, and appeared to meaningfully stabilize medication adherence, improve dietary quality, and lower caregiver-reported treatment burden (de Arriba Munoz et al., 2025). It is a small, interim dataset, and we should be cautious about over-extrapolating from it, but the underlying logic-that digital tools can bridge pharmacological therapy and durable family-centered lifestyle change-seems sound and worth pursuing at scale (de Arriba Munoz et al., 2025; Dziewierz & Siudak, 2026).

2.4 Systemic and Cardiometabolic Impact: CKM Syndrome and Surrogate Markers of Insulin Resistance

Obesity and insulin resistance sit at the center of cardiovascular-kidney-metabolic (CKM) syndrome, a construct that captures cardiorenal injury, myocardial remodeling, and endothelial dysfunction as a single interconnected process rather than three separate diseases (Bharaj et al., 2025; Voziki et al., 2026). The cardioprotective potential of GLP-1 receptor activation is no longer speculative; it has been demonstrated in large outcomes trials. The SELECT trial, which enrolled more than 17,000 adults with overweight or obesity and established cardiovascular disease but without diabetes, found that semaglutide 2.4 mg reduced major adverse cardiovascular events by 20% relative to placebo (Abel et al., 2024; Lala & Gulati, 2025; Podder et al., 2026). What stands out about this result is that the benefit emerged early and appeared largely independent of the absolute magnitude of weight loss-suggesting a direct anti-inflammatory or anti-atherogenic mechanism rather than a purely weight-mediated one (Lala & Gulati, 2025; Podder et al., 2026). On the renal side, the FLOW trial was stopped early for efficacy after semaglutide reduced chronic kidney disease progression and cardiorenal mortality by 24% in patients with T2DM and established renal impairment (Abel et al., 2024; Podder et al., 2026). In heart failure with preserved ejection fraction, STEP-HFpEF and SUMMIT showed that semaglutide and tirzepatide improved functional capacity, reduced heart-failure exacerbations, and lowered inflammatory markers such as C-reactive protein (Lala & Gulati, 2025; Grudniewska et al., 2025; Podder et al., 2026).

SGLT2 inhibitors occupy an adjacent, mechanistically distinct niche in the CKM space (Bharaj et al., 2025; Podder et al., 2026). By blocking glucose and sodium reabsorption in the proximal tubule, agents such as canagliflozin, dapagliflozin, and empagliflozin restore tubuloglomerular feedback, reduce intraglomerular pressure, and blunt renal hyperfiltration injury-all through an insulin-independent mechanism (Bharaj et al., 2025; Podder et al., 2026). Across the pivotal trial programs (DAPA-HF, EMPEROR-Reduced, EMPEROR-Preserved, DELIVER, CREDENCE, DAPA-CKD, EMPA-KIDNEY), this class consistently reduced heart-failure hospitalization and slowed CKD progression, regardless of diabetes status (Bharaj et al., 2025). There is even preliminary neuroimaging evidence suggesting SGLT2 inhibitors may improve central hypothalamic insulin sensitivity, hinting at metabolic reach beyond the kidney (Abel et al., 2024).

Because gold-standard measures of insulin resistance-namely the hyperinsulinemic-euglycemic clamp-are simply too resource-intensive for routine practice, the field has leaned on validated surrogate indices instead (Voziki et al., 2026). The Metabolic Score for Insulin Resistance (METS-IR) and the Single Point Insulin Sensitivity Estimator (SPISE) combine routine anthropometric and lipid data with fasting glycemic measures to approximate what the clamp would otherwise capture directly (Voziki et al., 2026). In a real-world comparative study of 100 patients, Voziki et al. (2026) found that both GLP-1RAs and SGLT2 inhibitors improved METS-IR and SPISE over six months-but interestingly, the improvement tracked with BMI and HbA1c reduction only in the GLP-1RA group, whereas SGLT2i-associated improvement appeared largely independent of weight change, hinting at genuinely distinct underlying mechanisms rather than a shared final common pathway (Voziki et al., 2026).

2.5 Nutritional Priorities, Muscle Preservation, and the Practical Limits of Long-Term Adherence

For all their efficacy, modern anti-obesity medications remain constrained-perhaps more than anything else-by tolerability. Up to 80% of patients starting GLP-1-based therapy experience gastrointestinal adverse events, chiefly nausea, vomiting, diarrhea, and constipation, clustering predictably around dose titration (Pardali et al., 2026). In real-world practice, these symptoms are the dominant driver of early discontinuation, with 46% to 65% of patients stopping therapy within the first year (Podder et al., 2026). Less commonly discussed, but clinically relevant, is oral health: repeated vomiting and delayed gastric transit introduce real risk of enamel erosion, xerostomia, and dental caries (Bijoch, 2026). Psychiatric safety also warrants ongoing attention-recent reviews are broadly reassuring about the psychological safety of GLP-1RAs as a class, but regulators still recommend careful screening for worsening depression or suicidal ideation in vulnerable patients, and that caution seems reasonable given how novel these exposures are at a population scale (Batra et al., 2026).

Table 4 (placed after the conclusion, alongside the other study tables) summarizes the principal clinical and nutritional challenges associated with GLP-1-era pharmacotherapy-sarcopenic obesity and muscle loss, rapid regain after discontinuation, gastrointestinal intolerance, and pediatric developmental risk-together with the countermeasures currently recommended for each.

Because these medications work chiefly by inducing a severe, sustained caloric deficit, they inevitably reshape eating behavior and nutrient intake more broadly (Zambrano-Villacres et al., 2026; Balasubaramaniam et al., 2026). That reshaping carries risk: progressive malnutrition, micronutrient deficiency (vitamin B12 in particular), and physical frailty are all plausible, and increasingly documented, consequences of profound appetite suppression (Santic et al., 2026; Zambrano-Villacres et al., 2026). The practical response-though not yet universally implemented-is to fold registered dietitian nutritionists into obesity pharmacotherapy pathways as a matter of routine, mirroring the structured medical nutrition therapy models that bariatric surgery programs adopted years ago (Dziewierz & Siudak, 2026; Balasubaramaniam et al., 2026). Nutritional care, in other words, cannot be an afterthought bolted onto a prescription; it needs to be personalized to the patient's metabolic phenotype, symptom profile, and rate of weight loss, always paired with structured dietary and physical countermeasures aimed at preserving muscle quality, functional capacity, and durable cardiorenal health (Santic et al., 2026; Sancho-Haro et al., 2026; Balasubaramaniam et al., 2026).

2.6 Synthesis and Future Directions

Taken together, the literature reviewed here traces a genuine historic shift-from a body mass index-centric, arguably reductive, view of obesity toward a chronic, progressive, multisystem disease framework (Abdallah et al., 2026). The 2025 Lancet Commission criteria, by separating preclinical from clinical obesity, allow for a degree of risk stratification that simply was not available before, directing early lifestyle and digital-monitoring support toward preclinical risk while reserving potent multi-receptor pharmacotherapy or metabolic-bariatric surgery for clinical obesity (de Arriba Munoz et al., 2025; Abdallah et al., 2026).

Looking ahead, the therapeutic pipeline shows no sign of slowing-once-monthly injectables, oral small-molecule GLP-1RAs such as orforglipron, and muscle-sparing co-therapies are all advancing through clinical development, and each will likely demand a more personalized approach to prescribing than the field has historically practiced (Lempesis & Dalamaga, 2026; Podder et al., 2026). A particularly promising research direction involves fasting incretin and pharmacogenomic profiling-variants in genes such as GLP1R and ARRB1-with the goal of shifting obesity pharmacotherapy away from trial-and-error dosing toward genuinely biomarker-guided drug selection (La Vignera et al., 2026; Podder et al., 2026). Ultimately, though, no amount of biological precision will resolve this epidemic on its own; durable progress will require pairing individualized, digitally supported precision medicine with structural pricing reform and health policy that actively works against, rather than passively tolerates, inequitable access (de Arriba Munoz et al., 2025; Abel et al., 2024; Abdallah et al., 2026).

3. Methods

3.1 Review Design and Reporting Framework

This paper is best described as a structured narrative review rather than a systematic review in the strictest sense, though we have tried, wherever feasible, to borrow the transparency conventions of the latter. We did not prospectively register a protocol-a limitation we acknowledge openly-but the search, screening, and synthesis steps described below were followed consistently and are reported here in enough detail that another reviewer, working from the same source list, should be able to reproduce our search logic and arrive at a substantially overlapping evidence base. Reporting was informed by the general structure of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, adapted for a narrative rather than a quantitative synthesis, and by conventions typical of narrative reviews indexed on PubMed.

3.2 Information Sources and Search Strategy

We searched PubMed/MEDLINE as the primary bibliographic database, supplemented by manual cross-referencing of citation lists within the retrieved articles to capture closely related work that indexing terms alone might have missed. The search window was restricted to publications dated 2024 through 2026, reflecting our intent to characterize the current, rapidly evolving incretin era rather than to provide a historical account of obesity pharmacotherapy.

Search terms combined Medical Subject Headings with free-text keywords across four conceptual clusters, joined using standard Boolean operators: (1) disease terms-"obesity," "overweight," "childhood obesity," "cardiovascular-kidney-metabolic syndrome"; (2) pharmacological terms-"GLP-1 receptor agonist," "semaglutide," "tirzepatide," "retatrutide," "cagrilintide," "SGLT2 inhibitor," "incretin"; (3) outcome terms-"weight loss," "body composition," "lean mass," "treatment persistence," "medication adherence," "cardiovascular outcomes," "chronic kidney disease"; and (4) population/equity terms-"pediatric," "adolescent," "health disparities," "insurance coverage," "drug pricing." An illustrative PubMed query took the form: ("GLP-1 receptor agonist" OR semaglutide OR tirzepatide AND (obesity OR "weight loss" AND ("body composition" OR "treatment persistence" OR "health equity" AND ("2024": "2026").

3.3 Eligibility Criteria

We included peer-reviewed original research articles, randomized controlled trials, prospective and retrospective cohort studies, registry-based real-world evidence studies, and narrative or systematic reviews published in English between 2024 and 2026 that addressed at least one of the following: the pharmacology or clinical efficacy of GLP-1, GIP, or glucagon receptor-targeted agents in adults or children with overweight or obesity; body-composition outcomes of pharmacological weight loss; real-world adherence or persistence with incretin-based therapy; cardiorenal or metabolic outcomes trials involving GLP-1RAs or SGLT2 inhibitors; or health-equity, access, and pricing dimensions of obesity pharmacotherapy. We excluded conference abstracts without full-text availability, case reports involving fewer than five patients, animal-only studies without a translational discussion relevant to human dosing or safety, and articles for which the full text could not be retrieved.

3.4 Study Selection and Data Extraction

Titles and abstracts identified through the search strategy were screened for topical relevance against the eligibility criteria above. Full texts of potentially eligible articles were then retrieved and reviewed in full to confirm inclusion. From each included source, we extracted, where reported: study design and population characteristics; sample size; drug, dose, and duration; weight-loss and HbA1c outcomes; body-composition measures (fat mass, lean mass, and the technique used to derive them, such as dual-energy X-ray absorptiometry, magnetic resonance imaging, or bioelectrical impedance analysis); cardiorenal outcomes; adherence and persistence metrics, including the specific discontinuation-gap definition used by each source; and reported adverse events. Extracted data were organized into four evidence tables (Tables 1-4) spanning adult pharmacology, pediatric and adolescent pharmacotherapy, global adherence and persistence epidemiology, and body-composition remodeling, respectively; these tables are presented after the conclusion of this manuscript, together with two summary figures synthesizing the mechanistic (Figure 1) and adherence-related (Figure 2) findings from the literature review, and two additional figures synthesizing comparative efficacy (Figure 3) and body-composition (Figure 4) findings from the results section.

3.5 Synthesis Approach

Given the heterogeneity of study designs, populations, and outcome definitions across the included literature-ranging from randomized phase 2/3 trials to national administrative-claims registries-a formal meta-analysis was neither feasible nor, in our judgment, appropriate. We instead conducted a structured narrative and thematic synthesis, grouping findings under five recurring themes that emerged consistently across the reviewed literature: (1) receptor pharmacology and comparative efficacy; (2) pediatric and adolescent-specific efficacy and safety; (3) real-world adherence and persistence; (4) body-composition remodeling and musculoskeletal preservation; and (5) equity, access, and policy considerations. Where multiple sources reported quantitatively similar findings, we report the range across sources rather than a single pooled estimate, and we have tried, throughout, to flag where evidence rests on a single study versus a convergent body of literature, since the two should not be read with equal confidence.

3.6 Risk of Bias and Limitations of the Method

We did not apply a formal, standardized risk-of-bias instrument (such as the Cochrane RoB 2 tool or ROBINS-I) uniformly across every included study, given the mixed-methods nature of this narrative synthesis; this represents an acknowledged limitation relative to a full systematic review. Where a source's design carried an obvious threat to internal validity-most commonly, unmeasured confounding in retrospective claims-based persistence studies, or short follow-up in phase 2 trials-we have tried to flag this explicitly within the results and discussion rather than presenting findings as uniformly definitive. Real-world persistence figures, in particular, are highly sensitive to how each registry defines a "discontinuation gap" (ranging from 60 to more than 90 days across the sources reviewed), and direct cross-country comparisons should therefore be interpreted with appropriate caution.

4. Results

Figure 1. Receptor-level synergy underlying incretin-based obesity pharmacotherapy. Endogenous nutrient-stimulated release of GLP-1, GIP, and (for triple agonists) recruitment of glucagon receptor signaling act on distinct but complementary tissues-central satiety circuitry, adipose lipid handling, and hepatic mitochondrial oxidation, respectively-to produce a graded increase in weight-loss efficacy from selective GLP-1 receptor agonism through dual and triple receptor co-agonism, as synthesized from the literature reviewed in Section 2.2.

Figure 2. Global attrition funnel for GLP-1 receptor agonist therapy, from prescription to long-term persistence. The funnel traces the proportion of patients lost at each stage of the treatment pathway-pharmacy abandonment, discontinuation after the first fill, 12-month persistence, and 36-month persistence-illustrating the wide cross-country variation documented in Table 3 and discussed in Section 2.5 and Section 4.3.

4.1 Comparative Efficacy and Glycemic Profiles of Adult Incretin-Based and Cardiorenal Therapies

Synthesizing across the included pharmacological literature, a fairly clear efficacy hierarchy emerges, one that scales closely with the number of receptors engaged (Podder et al., 2026; Figure 3). Selective GLP-1RAs, represented most prominently by semaglutide, sit at the foundation of this hierarchy. In the STEP 1 trial, semaglutide 2.4 mg weekly produced a mean total body weight reduction of approximately 14.9% over 68 weeks (Sancho-Haro et al., 2026; Santic et al., 2026; Table 1), alongside HbA1c reductions of 1.5% to 2.0% in patients with T2DM (Abel et al., 2024; Podder et al., 2026).

Tirzepatide extends this ceiling considerably. By recruiting GIP-mediated adipose lipid handling alongside GLP-1-mediated central satiety signaling, it achieved a dose-dependent mean weight reduction of 20.9% to 22.5% at 72 weeks in SURMOUNT-1 (Sancho-Haro et al., 2026; Santic et al., 2026; Table 1), and the head-to-head SURMOUNT-5 trial confirmed its superiority over high-dose semaglutide, with 20.2% weight loss versus 13.7% at 72 weeks (Podder et al., 2026). Retatrutide, which adds glucagon receptor activation to the dual-agonist backbone, produced the largest effect observed across the reviewed literature-24.2% mean weight loss at 48 weeks in phase 2 trials (Lempesis & Dalamaga, 2026; Podder et al., 2026; Table 1; Figure 3)-an effect plausibly related to glucagon-driven increases in hepatic mitochondrial oxidation and resting energy expenditure, which counter the metabolic-rate decline typically seen with rapid energy restriction (Lempesis & Dalamaga, 2026; Podder et al., 2026).

These metabolic gains appear to translate into measurable cardiorenal protection rather than remaining confined to the scale (Abel et al., 2024; Podder et al., 2026). The SELECT trial found a 20% relative risk reduction in major adverse cardiovascular events with semaglutide 2.4 mg among adults with cardiovascular disease but without diabetes, an effect that emerged early in follow-up and appeared to operate at least partly independent of the magnitude of weight loss achieved (Abel et al., 2024; Podder et al., 2026). The FLOW trial similarly demonstrated a 24% reduction in major kidney disease events with once-weekly semaglutide in patients with T2DM and established renal impairment (Abel et al., 2024; Podder et al., 2026). SGLT2 inhibitors, by comparison, produced far more modest weight reduction-2.0 to 3.0 kg-yet delivered substantial, mechanistically distinct organ protection, reducing heart-failure hospitalization and slowing renal decline through an insulin-independent pathway even in advanced chronic kidney disease (Bharaj et al., 2025; Voziki et al., 2026; Table 1).

4.2 Pediatric and Adolescent Pharmacotherapy: Efficacy and Specialized Safety Signals

Pediatric outcomes, drawn together in Table 2, tell a more heterogeneous story than the adult literature. Weekly semaglutide 2.4 mg, approved for adolescents 12 years and older, achieved a 16.1% mean BMI reduction at week 68 in the pivotal STEP TEENS trial, compared with a 0.6% increase in the placebo arm (Salama et al., 2025). Real-world data complicate this picture, however; a UK pediatric clinic cohort found a more modest 8.9% mean weight loss at 12 months, with a meaningful subgroup of patients showing no measurable reduction at all-an efficacy-effectiveness gap that mirrors, and arguably exceeds, what is seen in adult populations (Salama et al., 2025).

For children younger than 12, the evidence base thins considerably. Liraglutide 3.0 mg, evaluated off-label in the SCALE Kids trial for children aged 6 to under 12, achieved a 5.8% mean BMI reduction at 56 weeks versus a 1.6% increase with placebo, but at the cost of substantial gastrointestinal burden-80% of children reported nausea, vomiting, or diarrhea (Salama et al., 2025; Table 2). Older, centrally acting agents show more modest and phenotype-specific effects: topiramate combined with lifestyle modification produced a 3.4% BMI reduction over 12 months in a retrospective cohort of youth aged 12-18 (Salama et al., 2025), while lisdexamfetamine, approved for pediatric ADHD but used off-label in severe pediatric obesity, reduced BMI z-scores by approximately 0.41 (Salama et al., 2025). Setmelanotide remains a notable exception to the general pattern of modest pediatric efficacy, but only within its narrow approved population of monogenic obesity syndromes (POMC, PCSK1, or LEPR deficiency, and Bardet-Biedl syndrome), where it achieved an 18% mean BMI reduction in children as young as two to five years old (Salama et al., 2025; Table 2). Across all pediatric agents reviewed, regulatory advisories consistently flag the need for longitudinal monitoring of growth velocity and bone mineral density accrual with sustained exposure (Salama et al., 2025).

4.3 The Efficacy-Effectiveness Gap: Global Adherence and Real-World Persistence

Figure 3. Comparative weight-loss efficacy across incretin-based agents in pivotal phase 2/3 trials. Mean placebo-uncorrected body weight reduction is plotted for liraglutide, semaglutide, tirzepatide, retatrutide, and CagriSema, illustrating the graded increase in efficacy associated with progressive receptor co-agonism described in Table 1 and Section 4.1.

Figure 4. Composition of pharmacologically induced weight loss across representative body-composition substudies. Stacked bars show the proportion of total weight loss attributable to fat mass versus lean mass across the STEP 1 DXA substudy, SURMOUNT-1 DXA substudy, a real-world bioelectrical impedance analysis cohort, and the COURAGE trevogrumab combination trial, illustrating the modifiability of lean-mass loss discussed in Table 4 and Section 4.4.

Table 1. Pharmacological characteristics and trial efficacy of anti-obesity and glycemic agents in adults. The table compares selective GLP-1 receptor agonists, dual GIP/GLP-1 and triple-receptor co-agonists, and SGLT2 inhibitors across receptor targets, dosing regimens, weight-loss and HbA1c efficacy, cardiorenal trial outcomes, and adverse-event profiles, drawing on pivotal phase 2/3 and outcomes trials cited in the source references.

Table 2. Pediatric and adolescent anti-obesity pharmacotherapy. Agents are presented with their FDA-approved or off-label indication and age group, dosing and titration schedule, reported clinical efficacy outcomes, mechanism of action, and principal safety concerns relevant to growing children and adolescents.

If the efficacy data described above represent the best case for these medications, the persistence data represent something closer to the honest case-and the gap between the two is, frankly, one of the more sobering findings of this review (Dziewierz & Siudak, 2026). Figure 2 traces this attrition across the treatment cascade, from initial prescription through long-term persistence, and Table 3 presents the underlying country-level data in full.

Large-scale US pharmacy benefit and claims data indicate that roughly 15% of approved GLP-1RA prescriptions are never filled at all, and a further 36% of patients discontinue after only a single fill (Dziewierz & Siudak, 2026; Figure 2). Among patients who do initiate therapy, matched electronic health record cohorts show 12-month discontinuation rates of 52.5% for semaglutide and 55.9% for tirzepatide, with long-term commercial persistence collapsing to just 8.1% by three years (Dziewierz & Siudak, 2026; Table 3; Figure 2).

Persistence, however, is not a fixed biological property of these drugs-it appears to be substantially a function of health-system design. In Sweden and Denmark, where GLP-1RA therapy for T2DM sits within a public reimbursement framework, 12-month persistence reached approximately 76% and 79% respectively, though even within Denmark's universal, copayment-capped system, patients in the lowest household-income tertile still carried a 36% higher relative risk of 12-month discontinuation than those in the highest tertile (Dziewierz & Siudak, 2026; Table 3). By contrast, in healthcare settings where patients bear the full cost of therapy out of pocket, persistence falls off a cliff: only 6.6% of patients remained on therapy at 12 months in Poland's private LUX MED cohort, and in Colombia, persistence dropped from an already-alarming 13.8% at six months to just 0.2% at twelve months (Dziewierz & Siudak, 2026; Table 3). Across nearly every setting reviewed, the two dominant, recurring drivers of discontinuation were gastrointestinal intolerance and direct or indirect financial burden-with US annual out-of-pocket costs for these agents estimated between $7,000 and $16,000 (Dziewierz & Siudak, 2026; Podder et al., 2026).

4.4 Body Composition Remodeling: Lean Mass Attrition and Musculoskeletal Preservation Strategies

Because these medications achieve such large absolute weight reductions, the composition of what is actually being lost has become a central question in its own right-arguably as important, clinically, as the topline percentage (Sancho-Haro et al., 2026; Santic et al., 2026). Across the body-composition literature summarized in Table 4 and visualized in Figure 4, lean mass consistently accounts for a substantial share of total weight lost, typically in the range of 25% to 40% absent targeted nutritional or exercise intervention (Dziewierz & Siudak, 2026; Sancho-Haro et al., 2026).

In the STEP 1 DXA substudy, semaglutide was associated with a 9.7% reduction in lean body mass, representing roughly 40% of total weight lost (Wilding et al., 2021; Table 4; Figure 4). The SURMOUNT-1 DXA substudy found a somewhat more favorable ratio for tirzepatide-a mean loss of 5.6 kg of lean soft tissue over 72 weeks, or approximately 26% of total weight loss (Look et al., 2025; Table 4; Figure 4)-while a real-world bioelectrical impedance analysis cohort found lean mass accounting for closer to 25% to 30% of weight loss over 12 months (Dziewierz & Siudak, 2026; Table 4).

Encouragingly, this proportion appears highly modifiable rather than fixed. In the trial reported by Lundgren et al. (2021), adding supervised, structured exercise to liraglutide therapy essentially eliminated lean mass and bone mineral density loss altogether, whereas patients on liraglutide alone experienced significant declines in both (Lundgren et al., 2021; Table 4). Consensus recommendations converging across the reviewed literature call for dietary protein intake of at least 1.2 to 1.5 g/kg body weight per day, distributed across meals, combined with progressive resistance training at least two to three sessions weekly (Sancho-Haro et al., 2026; Santic et al., 2026). Investigational pharmacological approaches are moving in the same direction-Phase 2 data combining semaglutide with myostatin/activin inhibitors such as trevogrumab or garetosmab preserved 50% to 80% of lean mass during active treatment, shifting the composition of weight loss so that up to 93% of total mass lost derived exclusively from adipose tissue (Regeneron Pharmaceuticals, Inc., 2025; Table 4; Figure 4)-a result that, if replicated in larger and longer trials, could meaningfully change how clinicians think about "high-quality" pharmacological weight loss.

5. Discussion

5.1 Reconciling Extraordinary Efficacy with Biological Limitation

It would be easy, reading only the efficacy tables, to

Table 3. Global epidemiology of treatment adherence and persistence with GLP-1 and dual incretin-based therapies. Country- and registry-level findings are compared across discontinuation-gap definitions, early attrition rates, long-term persistence, and the principal structural barriers reported in each healthcare setting.

Table 4. Evidence map of body composition remodeling and lean-mass loss mitigation strategies. Trials and cohorts are compared by measurement technique, relative fat- versus lean-mass change, the proportion of total weight loss attributable to lean tissue, and nutritional or pharmacological strategies shown to mitigate musculoskeletal loss.

conclude that obesity pharmacotherapy has essentially arrived at a solved state. We do not think that conclusion is warranted, and the evidence synthesized in this review suggests something more nuanced: efficacy has genuinely arrived, but durability has not followed at the same pace. The plateau in weight loss around 18 months, and the rapid rebound-roughly two-thirds of lost weight within a year of discontinuation-observed across the literature (Lasik & Ukleja-Sokolowska, 2026; Abdallah et al., 2026) suggests that these agents function less like a cure and more like a chronic, ongoing life-support system for altered energy homeostasis. That reframing has real clinical consequences: if obesity pharmacotherapy is understood as lifelong therapy rather than a finite course, then questions of affordability, tolerability, and long-term surveillance stop being secondary concerns and become, arguably, the central clinical question (Table 1; Table 3).

5.2 The Efficacy-Effectiveness Gap Is Not Primarily a Biological Problem

One of the more striking patterns to emerge from this synthesis is how much of the "efficacy-effectiveness gap" appears to be structural rather than pharmacological. The nearly sevenfold difference in 12-month persistence between reimbursed Scandinavian systems and out-of-pocket markets such as Poland and Colombia (Dziewierz & Siudak, 2026; Table 3; Figure 2) is difficult to explain through biology alone-patients are not fundamentally different across these settings in how their gastrointestinal tracts respond to GLP-1 agonism. What differs is who can afford to stay on therapy long enough to get past dose titration, and who cannot. This has an uncomfortable implication for how the field talks about "real-world effectiveness": persistence data collected predominantly from insured, high-income cohorts likely overstate what effectiveness will look like once these drugs diffuse more broadly, unless accompanied by deliberate pricing and coverage reform (Podder et al., 2026).

5.3 Weight Loss Quality: Rethinking Success Beyond the Number on the Scale

We would also argue, based on the body-composition evidence reviewed here (Table 4; Figure 4), that the field's near-exclusive focus on percentage weight loss as the primary efficacy metric deserves reconsideration. A patient who loses 20% of body weight with 40% of that loss coming from lean tissue is not equivalent, clinically, to a patient who loses the same percentage with only 10% from lean tissue-yet current trial reporting and, frankly, most marketing materials treat these as interchangeable outcomes. The fact that structured resistance exercise (Lundgren et al., 2021) and, more speculatively, myostatin/activin-pathway co-therapies (Regeneron Pharmaceuticals, Inc., 2025) can shift this ratio so substantially suggests that "quality-adjusted" weight loss is an achievable and measurable target, not merely an aspirational one. We think body-composition endpoints deserve a more central place in future trial design and in routine clinical monitoring, particularly for older adults and patients with pre-existing sarcopenia risk.

5.4 Pediatric Uncertainty Warrants Humility, Not Just Caution

The pediatric evidence base (Table 2) leaves us, honestly, somewhat uneasy-not because the available trials show alarming signals, but because so much of the relevant developmental window (growth velocity, pubertal timing, long-term bone mineral density accrual, psychiatric outcomes over years rather than months) simply has not been studied yet, particularly in children under 12 (Salama et al., 2025; Batra et al., 2026; Abdallah et al., 2026). It is worth being explicit that "no evidence of harm" and "evidence of no harm" are different statements, and pediatric obesity pharmacotherapy currently sits closer to the former. We would argue this supports continued off-label use only within structured surveillance protocols-incorporating routine DXA monitoring and growth tracking-rather than either blanket restriction or unmonitored expansion.

5.5 Toward an Integrated, Equity-Conscious Clinical Framework

Pulling these threads together, we would propose that durable success in the GLP-1 era depends on treating pharmacotherapy as one component of an integrated system rather than a standalone intervention. That system should include structured dietetic support modeled on bariatric surgery pathways (Balasubaramaniam et al., 2026), routine resistance training prescription, pediatric-specific surveillance protocols, digital family-engagement tools where feasible (de Arriba Munoz et al., 2025), and-perhaps most importantly, given the persistence data reviewed here-deliberate policy engagement around drug pricing and insurance coverage. Absent that last piece in particular, we suspect the equity gaps documented in this review will not close on their own, no matter how much further pharmacological efficacy improves.

5.6 Limitations

This review carries limitations that should temper how its conclusions are read. It is a narrative rather than systematic synthesis, drawn from a relatively small set of recent sources rather than an exhaustive database search, and we did not apply a formal risk-of-bias tool uniformly across included studies. Persistence and adherence data in particular are drawn from heterogeneous registries using inconsistent discontinuation-gap definitions, which limits precise cross-country comparison even as the broad direction of the findings appears robust. Finally, because much of the underlying evidence base is itself recent-many of the cited trials and cohorts were published within the last one to two years-longer-term outcomes, particularly regarding pediatric development and rare safety signals with dual- and triple-receptor agonists, remain genuinely unknown rather than merely under-reviewed here.

6. Conclusion

Obesity pharmacotherapy has moved, within a remarkably short span of time, from a modest adjunct to lifestyle intervention into something capable of bariatric-level weight loss. That is a genuine and important achievement. But the evidence reviewed here suggests the field is not yet finished-durability of response, disproportionate lean-mass loss, gastrointestinal tolerability, pediatric developmental uncertainty, and profound global inequities in affordability and access all remain substantially unresolved. Selective GLP-1 receptor agonists, dual GIP/GLP-1 co-agonists, and emerging triple agonists each extend the efficacy ceiling further, yet real-world persistence data show that trial efficacy translates unevenly, and often poorly, into sustained population-level benefit once cost and tolerability enter the picture. Going forward, we would argue that success in this field should be measured less by weight-loss percentage alone and more by a composite of durability, tissue-quality of weight lost, safety across the lifespan, and equitable accessibility-an admittedly harder standard to meet, but the one that ultimately matters to patients.

Author Contributions

N.S.A. contributed to the conception and design of the review, literature search, analysis and synthesis of the relevant evidence, and drafting of the manuscript. H.N.H. contributed to the literature search, interpreted the evidence regarding efficacy, tolerability, treatment persistence, and long-term outcomes, and critically revised the manuscript. N.N.A. contributed to the analysis and interpretation of evidence related to equity, pediatric populations, body composition, and clinical translation and critically revised the manuscript for important intellectual content. All authors reviewed and approved the final version of the manuscript and agreed to be accountable for all aspects of the work.

Acknowledgements

The authors would like to acknowledge the Ministry of National Guard Health Affairs, Riyadh, Saudi Arabia, and the Department of Environmental & Public Health, Eastern Kentucky University, United States, for their academic and institutional support. The authors also acknowledge the researchers whose published studies contributed to the scientific foundation of this review.

References