Phenotypic and genomic frameworks for precision pharmacotherapy in obesity: a narrative review

Background Obesity is a chronic, heterogeneous disease with marked interindividual variability in response to anti-obesity medications. Although GLP-1 receptor agonists (GLP-1RAs) and dual incretin agonists produce substantial average weight loss in clinical trials, real-world effectiveness is limited by interindividual variability, gastrointestinal intolerance, incomplete dose escalation, cost and access barriers, poor long-term persistence, and weight plateau or regain in a subset of treated patients. Weight regain is particularly consistent after treatment discontinuation, reinforcing the chronic-care nature of obesity pharmacotherapy. Objective To synthesize evidence on phenotypic and genomic determinants of variability in pharmacotherapy response and propose a pragmatic framework for precision pharmacotherapy in obesity. Methods Narrative review integrating data from randomized trials, post-hoc subgroup analyses, observational cohorts, neurobehavioral studies, pharmacogenomic investigations, and polygenic risk score research addressing predictors of efficacy and tolerability. Results Response heterogeneity reflects interactions among central satiety and reward circuitry, baseline metabolic and glycemic status, sex, baseline adiposity, treatment indication, previous GLP-1RA exposure, adherence, and tolerability. Early on-treatment weight change, usually assessed within 12–16 weeks and in some liraglutide studies as early as 1 month, is the most clinically actionable predictor of longer-term outcomes. Phenotype-guided approaches that classify dominant mechanisms such as impaired satiation, postprandial hunger, emotional or hedonic eating, and low energy expenditure offer a feasible bridge between pathophysiology and medication selection. Emerging genomic signals are promising but remain insufficiently replicated for routine clinical use. Conclusion Precision pharmacotherapy in obesity is most viable today through phenotype-guided treatment, structured early-response monitoring, and proactive management of adherence and tolerability. Genomic tools may eventually refine treatment selection, but they remain investigational and require prospective, multi-ancestry validation.