Molecular obesity therapies advance, but precision prescribing remains largely unproven

A 2026 narrative review finds that obesity pharmacotherapy is moving beyond single-pathway appetite suppression toward drugs that combine molecular targets, with multi-agonist candidates producing some of the largest weight-loss results reported in recent trials. However, the evidence supports a narrower form of personalization than the term “precision obesity medicine” can suggest: receptor-targeted therapies are clinically actionable, genotype-directed treatment is established for selected forms of genetic obesity, and most biomarkers, polygenic scores and multi-omics profiles remain investigational. The review synthesizes clinical trials, mechanistic studies and translational research published between 2016 and 2026, with foundational studies included where needed to explain established pathways.

The biological rationale for this shift is that obesity is not governed by a single defect. Hypothalamic and brainstem circuits integrate hunger, satiety, nutrient availability and energy expenditure, while gastrointestinal hormones communicate with the brain through blood signals, vagal pathways and neural circuits. In the arcuate nucleus of the hypothalamus, NPY/AgRP neurons promote food intake, whereas POMC neurons release melanocortin peptides that activate MC4R and suppress feeding. Signals including leptin, insulin, GLP-1, GIP, PYY, CCK and ghrelin interact with these systems. Adipose tissue adds another layer: expansion of white adipose tissue can accompany inflammation, impaired insulin signalling and altered adipokine secretion, whereas brown adipose tissue supports energy expenditure through mitochondrial metabolism and thermogenesis.

Established GLP-1 receptor agonists exploit these pathways by prolonging satiety signalling, reducing food intake, slowing gastric emptying and improving glucose regulation. In the STEP 1 randomized trial, once-weekly semaglutide at 2.4 mg produced a mean weight reduction of 14.9% at 68 weeks in adults with overweight or obesity. Meta-analysis summarized in the review also found significant weight loss and improvements in several cardiometabolic outcomes among people without diabetes. The SELECT trial separately reported that semaglutide reduced the risk of major adverse cardiovascular events in adults with overweight or obesity who had cardiovascular disease but not diabetes. These outcomes support the clinical value of GLP-1 receptor activation, although the review emphasizes that efficacy, adverse effects, durability, comorbidities, contraindications and patient preferences must all inform drug selection.

Combining receptors has produced larger trial results without yet establishing molecular profiling as a routine selection tool. In SURMOUNT-1, tirzepatide, which activates both GIP and GLP-1 receptors, achieved mean weight reductions of 15.0% to 20.9% at 72 weeks, depending on dose. Retatrutide, an investigational triple agonist targeting GLP-1, GIP and glucagon receptors, reached up to 24.2% mean weight reduction at 48 weeks in a phase 2 randomized trial. Survodutide, a GLP-1/glucagon co-agonist under phase 3 development, produced dose-dependent weight loss and a 14.9% reduction at 46 weeks with the 4.8-mg dose in phase 2 testing. The mechanisms are complementary: GLP-1 signalling suppresses appetite, GIP signalling affects insulin action and metabolism, and glucagon agonism may increase lipid utilization and energy expenditure. Greater receptor complexity, however, does not automatically mean fewer adverse effects or less treatment resistance.

Other experimental combinations extend beyond incretin signalling. Cagrilintide, a long-acting amylin analogue, produced mean weight reductions of 6.0% to 10.8% at 26 weeks in a phase 2 randomized trial, with greater loss at higher doses. Amycretin, which targets GLP-1 and amylin receptors, produced up to 24.3% mean weight reduction at week 36 in a phase 1b/2a randomized study and is being developed in subcutaneous and oral forms. These results remain investigational, with longer-term efficacy, durability and safety unresolved. The gastrointestinal adverse effects frequently reported with obesity drugs are particularly important because they can lead to dose reduction or discontinuation. In a cross-sectional survey summarized by the review, nausea and vomiting were the most common patient complaints, while inadequate glycaemic control, nausea or vomiting and other gastrointestinal events were leading reasons physicians gave for stopping GLP-1 receptor agonists.

Molecular defects can guide treatment much more securely in a small but clinically important subset of patients with obesity. Setmelanotide activates MC4R and is approved for specified genetic or syndromic forms of obesity, including Bardet–Biedl syndrome and biallelic POMC, PCSK1 or LEPR deficiency, as well as acquired hypothalamic obesity under stated age criteria in the United States and European Union. The position of the defect matters. In POMC or LEPR deficiency, the molecular fault lies upstream of an intact MC4R pathway, so agonism can bypass it; in MC4R deficiency, the drug’s target itself is defective, and MC4R deficiency is not an approved indication. Phase 3 evidence summarized in the review found that about 80% of participants with POMC deficiency and approximately 45% with LEPR deficiency lost at least 10% of body weight after about one year.

That genotype-directed paradigm should not be confused with pharmacogenomic prediction in common polygenic obesity. Genome-wide association studies have identified many variants associated with obesity susceptibility, but most individual effects are small and the number of reported loci varies with population, phenotype definition, sample size and analytical method. Polygenic risk scores depend on how they are calibrated and therefore are not generally ready to determine an individual patient’s drug. Pharmacogenetic associations with response to GLP-1 receptor agonists have likewise been small, inconsistent and poorly reproduced across populations. Multi-ancestry evidence suggests that known genetic factors explain only a minor share of variability in weight loss with these agents. Preliminary weight loss with tirzepatide in people with pathogenic MC4R mutations also shows that pharmacological efficacy can persist despite disruption of a physiological pathway, complicating simple predictions based on the affected gene.

Disease-associated biomarkers should likewise not be confused with predictive markers. Inflammatory measures such as tumor necrosis factor, interleukin-6 and C-reactive protein, circulating microRNAs, adipocytokines, blood-cell profiles, oxidative-stress markers and features of the gut microbiota can characterize obesity, metabolic dysfunction or possible prognosis. A biomarker becomes clinically actionable only when its assay is reliable, its association with response to a defined treatment is prospectively and independently validated, and using it demonstrably improves decisions or patient outcomes compared with standard care. Most proposed markers meet only part of this chain. Preliminary phenotype-guided treatment is similarly unresolved: one study reported about 75% greater average weight loss in a phenotype-guided group than with standard pharmacological treatment, but the categories were hypothesis-based and require multicentre replication, external validation and comparison with current incretin therapies.

More sophisticated molecular profiling could eventually identify patients with distinct routes to response or resistance. Proteomic and metabolomic signatures, adipose-cell composition, gene-expression patterns and epigenetic changes may reveal biological subgroups, while epigenetic persistence after weight loss could help explain why obesity sometimes recurs. These approaches remain investigational because prospective validation, reproducibility across populations and evidence of improved outcomes after molecular stratification are still lacking. Early weight response, baseline metabolic characteristics, comorbidity profile and tolerability are clinically actionable now, but they represent conventional individualized care rather than validated molecular precision medicine. Exercise may also complement medication: one cited study found greater loss with exercise plus liraglutide than with either alone, and 45% maintained at least 10% weight loss one year after discontinuation, compared with 29% for exercise alone, 16% for liraglutide alone and 10% for placebo.

The review’s central conclusion is that molecularly guided obesity treatment is a spectrum of evidence rather than a single precision paradigm. GLP-1 receptor agonists and GIP–GLP-1 co-agonism are established clinical therapies, and setmelanotide provides a validated example of treatment selected by the site of a causal genetic defect. By contrast, glucagon-containing multi-agonists, amylin combinations, GDF15–GFRAL modulation and adipose-specific targets remain under development. The next decisive step is not merely finding more correlates of response, but demonstrating that a reproducible molecular test can prospectively select a treatment and produce better patient outcomes than existing clinical assessment. Until that evidence arrives, molecular understanding is most powerful for explaining mechanisms and identifying narrow genotype-directed opportunities, not for routinely matching most people with obesity to a molecularly optimized drug.

Subject of Research: Molecular targets, clinical efficacy, and precision approaches in obesity pharmacotherapy

Article Title: Molecularly guided pharmacotherapy for obesity: emerging targets, clinical efficacy, and personalized therapeutic approaches

Article References: Mao, J., & Hu, R. (2026). Molecularly guided pharmacotherapy for obesity: emerging targets, clinical efficacy, and personalized therapeutic approaches. Frontiers in Pharmacology, 17, 1988828. https://doi.org/10.3389/fphar.2026.1988828

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1988828

Keywords: obesity pharmacotherapy, GLP-1 receptor agonists, incretin therapy, multi-agonists, genotype-directed treatment, pharmacogenomics, biomarkers, multi-omics, tirzepatide, precision medicine, Molecularly, guided