Preprints.org · Oct 27, 2026

Defining and Predicting Inadequate Weight Response to Oral Semaglutide

By Labate AM, Villari P, Cimino E et al.

Background: Oral semaglutide produces substantial average weight loss, but the term non-responder combines clinically different situations. Failure to cross a numerical threshold may reflect limited treatment exposure, intolerance, a slow trajectory or persistently inadequate efficacy. Identifying these situations is particularly important when evidence from diabetes doses is applied to oral obesity treatment. Objective: To critically examine definitions of inadequate weight response to oral semaglutide and assess whether baseline characteristics or early treatment patterns identify patients likely to derive less weight benefit. Methods: A narrative synthesis drew on searches of PubMed, Embase, Web of Science and CENTRAL, supplemented by primary trial appendices and conference materials. Evidence was interpreted separately by oral dose, diabetes status, follow-up, analysis population and publication type. Observed responder counts were distinguished from model-based estimates. No quantitative pooling was performed. Results: The apparent frequency of weight loss below 5% varied substantially with population, dose and denominator. At the end of OASIS 4, 40/192 participants with observed in-trial measurements and 20/157 with on-treatment measurements remained below 5% loss. These proportions do not establish biological resistance. Baseline sex and body weight showed exploratory signals in OASIS 1, whereas age, BMI, diabetes duration and baseline HbA1c did not consistently distinguish weight response across studies. An OASIS 4 post hoc analysis associated early weight loss with subsequent trajectories, but participants below 5% loss at week 16 still achieved a mean 7.0% reduction at week 64. Behavioral and exposure models remain exploratory. Conclusions: Current evidence supports a time-dependent, treatment-specific assessment of inadequate weight response rather than a fixed baseline non-responder phenotype. Early trajectories inform expectations but do not validate an automatic stopping rule. Persistent low weight response should be described alongside achieved dose, administration, persistence, tolerability and other treatment outcomes.

doi.org/10.20944/preprints202610.2609.v1Open accessInterventions