Approximately 10% of participants in controlled obesity trials fail to achieve even 5% total body weight loss on GLP-1 or dual GLP-1/GIP receptor agonists — with real-world non-response rates likely exceeding this benchmark. A genome-wide association study identified variants in GLP1R linked to differential weight-loss efficacy, and GIPR variants associated with nausea and vomiting specifically on tirzepatide, though these findings remain unreplicated and hypothesis-generating. Across glycemic, weight-loss, and dual-metabolic contexts, early on-treatment response remains the only clinically actionable predictor currently available.
The variability problem with GLP-1 receptor agonists has been hiding in plain sight behind headline-grabbing average outcomes. Semaglutide and tirzepatide trials report extraordinary mean weight loss, but averages obscure a meaningful minority who receive expensive, potentially side-effect-laden therapy with minimal benefit. This review crystallizes what the field has resisted acknowledging: no baseline biomarker — hormonal, genetic, microbiome-derived, or behavioral — yet reliably stratifies responders before treatment begins. The GLP1R and GIPR genetic findings are tantalizing but single-cohort and self-reported, making clinical genotyping premature. Practically, this means clinicians should establish structured 4–12 week response checkpoints rather than waiting months before pivoting therapy. For longevity-focused adults, the implication is clear: non-response is real, not rare, and early reassessment is both medically sound and economically rational. This is a confirmatory synthesis rather than a paradigm shift, but its clinical utility lies in legitimizing structured discontinuation protocols.