Granhall · Clinical pharmacokinetics 2019 · Randomized, double-blind, placebo-controlled trial · n=242

Safety and Pharmacokinetics of Single and Multiple Ascending Doses of the Novel Oral Human GLP-1 Analogue, Oral Semaglutide, in Healthy Subjects and Subjects with Type 2 Diabetes.

Cited 183 times in the scientific literature.

Level 2 - randomized trial

Two randomized, double-blind, placebo-controlled phase 1 trials.

PubMed 30565096 · doi:10.1007/s40262-018-0728-4 · record verified 2026-08-29

What was done

Researchers evaluated the safety, tolerability, and pharmacokinetics of oral semaglutide co-formulated with the absorption enhancer sodium N-(8-[2-hydroxybenzoyl] amino) caprylate (SNAC) across two randomized, double-blind, placebo-controlled trials. The first was a single-dose, first-in-human trial in 135 healthy males testing 2 to 20 mg semaglutide with 150 to 600 mg SNAC versus placebo with SNAC. The second was a 10-week, once-daily trial evaluating 20 or 40 mg semaglutide (with 300 mg SNAC) versus placebo (with or without SNAC) in 84 healthy males and 40 mg semaglutide (with 300 mg SNAC) versus placebo (with or without SNAC) in 23 males with type 2 diabetes.

What was found

Oral semaglutide was safe and well-tolerated, with predominantly mild gastrointestinal adverse events. In the single-dose study, exposure was highest with the 300 mg SNAC formulation. In the multiple-dose study, semaglutide exposure was approximately twofold higher with 40 mg compared with 20 mg in healthy males (dose-proportional) and was comparable between healthy males and males with type 2 diabetes. The elimination half-life was approximately 1 week across all groups. Specific numerical pharmacokinetic metrics (such as AUC or Cmax) were not reported in the abstract.

Why it matters

These phase 1 trials established that co-formulating semaglutide with 300 mg SNAC enables oral absorption with a 1-week half-life, providing the foundational pharmacokinetic and safety rationale for once-daily oral dosing.

Limits

The study included only male participants, limiting generalizability to females. Sample sizes in the diabetes subgroup were small (n = 23), and specific numerical pharmacokinetic parameters and confidence intervals were omitted from the abstract.

Cited by