GLP-1 Receptor Agonist Drug Trends Analyzed via AI
Key Clinical Summary
- GLP-1 receptor agonists (GLP-1 RAs) lead in efficacy: Liraglutide, semaglutide, and tirzepatide showed 10–22% weight loss in trials.
- Multi-receptor agonists show promise: Investigational agents like retatrutide and NA-931 exceeded 24% weight loss.
- Artificial intelligence (AI) analysis confirms trade-offs: Traditional drugs offer moderate weight loss but greater safety concerns; GLP-1 RAs show stronger efficacy but higher cost and gastrointestinal (GI) side effects.
At a poster session during ObesityWeek 2025, researchers presented a machine learning–based meta-analysis of anti-obesity drug trials from the last 20 years. The findings highlighted the superiority of GLP-1 receptor agonists (GLP-1 RAs) in promoting weight loss while balancing safety and efficacy across multiple drug classes. This session provided data-driven insights into how obesity pharmacotherapy has evolved and where future treatments may be headed.
Session Highlights
In Poster-712, researchers applied proprietary AI and machine learning (ML) tools to systematically review and synthesize phase 2 and 3 trial data for obesity drugs published between 2004 and 2024. This included both US Food and Drug Administration (FDA)-approved agents and investigational compounds.
The analysis demonstrated that GLP-1 RAs—including liraglutide, semaglutide, and tirzepatide—achieved the highest weight loss outcomes (10–22%) among approved therapies. Newer investigational agents such as retatrutide (GLP-1/GIP/glucagon) and NA-931 (IGF-1/GLP-1/GIP/glucagon) were found to deliver even greater efficacy, with weight loss exceeding 24%, along with encouraging safety profiles.
Older medications such as orlistat and phentermine were associated with modest weight loss (5–15%) but limited long-term use due to tolerability and safety issues. The study also used AI-driven predictive modeling to evaluate efficacy trends by receptor targets and to simulate potential outcomes of combination therapies, such as semaglutide with NA-931.
Expert Perspectives
According to the presenting researchers, the evolving pharmacologic landscape reflects increasing receptor complexity and drug potency. The data suggest a shift from single-target therapies to multi-agonist agents that deliver enhanced weight loss while potentially improving metabolic outcomes.
The use of AI provided a scalable method for identifying efficacy and safety patterns across 2 decades of trial data.
This approach also supports drug developers and clinicians in understanding risk-benefit trade-offs and in identifying optimal patient populations for emerging therapies.
Implications for Practice
This AI-enhanced meta-analysis reinforces the growing clinical utility of GLP-1 RAs for obesity management while also pointing toward future multi-agonist compounds. Clinicians may anticipate expanded treatment options that offer improved efficacy, but must remain attentive to cost, tolerability, and long-term safety monitoring.
Conclusion
Over the past 20 years, anti-obesity pharmacotherapy has advanced significantly, with GLP-1 RAs now setting the benchmark for efficacy. AI and ML tools are helping to identify next-generation therapies and optimize treatment strategies, guiding both current practice and future drug development. Further clinical trials will be needed to validate these promising combination therapies.
Reference
Tran ZV, Willis MA, Tran LL. Poster-712 obesity drugs over the past 20 years: a meta-analysis of clinical trials using ML and AI. Presented at the ObesityWeek; November 4-7, 2025. Atlanta, Georgia.
