SAT-140 A Drug-Drug Interaction Trial of Oral Semaglutide with Levothyroxine and Multiple Coadministered Tablets

Introduction: Oral semaglutide contains the GLP-1 analog semaglutide co-formulated with the absorption enhancer, sodium N-(8-[2-hydroxybenzoyl] amino) caprylate (SNAC). Aims: This trial investigated (A) the effect of oral semaglutide and SNAC alone on the pharmacokinetics (PK) of levothyroxine (thyroxine), a frequently used oral medication with similar dosing instructions to oral semaglutide, and (B) the effect of multiple co-administered tablets on oral semaglutide PK. Methods: This was a 2-part, open-label, one-sequence crossover trial in 45 healthy subjects. Part A comprised three periods during which subjects received single-dose thyroxine 600 μg alone, with SNAC 300 mg, and with oral semaglutide 14 mg at steady-state. Primary endpoint was area under the total thyroxine (T4) serum concentration-time curve (0-48 h) after thyroxine, baseline-corrected for endogenous total T4 ((bc)AUC(0-48h,T4)). No effect of oral semaglutide or SNAC alone was declared if the 90% CI for the treatment ratio (thyroxine ± oral semaglutide or SNAC) was entirely within the pre-defined interval (0.80-1.25). In part B, oral semaglutide 14 mg at steady-state was co-administered with 5 placebo tablets (used as model tablets) once-daily for 5 weeks. Primary endpoint was area under the semaglutide plasma concentration-time curve during a 0-24 h dosing interval at steady state (AUC(0-24h)); no effect was confirmed if the 90% CI for the treatment ratio (oral semaglutide ± placebo) was entirely within the pre-defined interval (0.7000-1.4286). Safety and tolerability were also assessed. Results: In part A, co-administration of steady-state oral semaglutide with thyroxine resulted in a 33% increase in (bc)AUC(0-48h,T4); the 90% CI for the treatment ratio was not within the no effect interval. Oral semaglutide had no effect on baseline-corrected maximum total T4 concentration ((bc)C(max,T4)). Co-administration of SNAC alone did not increase exposure of total T4. In part B, co-administration of multiple tablets with steady-state oral semaglutide resulted in decreased semaglutide AUC(0-24) (34%) and C(max) (32%); part of the treatment ratio 90% CIs were not within the no effect interval for both endpoints. The safety profile was as expected for GLP-1 receptor agonists with gastrointestinal adverse events the most frequent. Conclusion: PK of thyroxine were influenced by co-administration of steady-state oral semaglutide. However, no change in clinical practice is required as close monitoring of co-administered thyroxine is already part of medical guidance. No obvious effect was seen with SNAC alone so increased thyroxine exposure may be due to the known gastric emptying delaying effect of the GLP-1 component. Absorption of oral semaglutide is affected by co-administration with multiple tablets (5), which is addressed in clinical dosing guidance.