Estimated Oral Semaglutide Exposure Has Distinct Relationships With Glycaemic Response, Weight Loss and Gastrointestinal Tolerability

Aims: Oral semaglutide absorption is subject to inter-individual variability. We investigated whether estimated individual exposure (eC avg ) provides predictive information beyond the prescribed dose in a real-world cohort of patients with type 2 diabetes (T2D). Materials and methods: We retrospectively included patients initiating oral semaglutide to calculate eC avg at each follow-up visit, using a validated population pharmacokinetic model incorporating dose, weight, sex, ethnicity and GI disorders, but without measuring plasma semaglutide concentrations. Mixed-effects models were used to assess the associations of dose and eC avg with changes in HbA1c and percent body weight, and to compare model fit. For GI side effects, time-adjusted fixed-effects logistic regression was employed. Results: We analysed 256 participants (31.6% women, mean age 65.6 years). During a median follow-up of 19 months, HbA1c declined by 0.7% and body weight by 7.5%; 41.8% of patients reported GI side effects. Both dose and eC avg were significantly associated with all outcomes when analysed individually. For ΔHbA1c, the dose model outperformed the eC avg model (p < 0.01), while eC avg showed modestly improved fit for weight loss (p < 0.001). The exposure-effect relationship was clearly right-shifted for weight loss compared to the glycaemic effect. For GI side effects, eC avg remained significantly associated with the outcome after adjustment for dose residuals, indicating that eC avg had greater predictive value for tolerability. Conclusions: Although estimated and not measured, the exposure to oral semaglutide may provide outcome-dependent information. While the prescribed dose seems sufficient to predict glycaemic response, estimated exposure provides additional value for weight loss and GI tolerability.