PIONEER 3 published in JAMA: oral semaglutide beats sitagliptin

Results from PIONEER 3, a phase 3a trial of once-daily oral semaglutide, were published in JAMA. Among 1,864 adults with type 2 diabetes that was not controlled by metformin alone or metformin plus a sulfonylurea, the 7 mg and 14 mg daily tablets lowered blood sugar and body weight significantly more than sitagliptin 100 mg after 26 weeks. The lowest dose tested, 3 mg daily, did not show non-inferiority to sitagliptin on HbA1c ^[1].

What the trial tested

PIONEER 3 was a randomized, double-blind, double-dummy trial run at 206 sites in 14 countries over 78 weeks, from February 2016 to March 2018. Of 2,463 people screened, 1,864 were randomized: 466 to oral semaglutide 3 mg, 466 to 7 mg, 465 to 14 mg, and 467 to sitagliptin 100 mg. Everyone assigned to semaglutide started at 3 mg per day and stepped up every four weeks — first to 7 mg, then to 14 mg — until they reached their assigned dose ^[1].

Participants averaged 58 years old (SD 10), with a mean starting HbA1c of 8.3% (SD 0.9%) and a mean body mass index of 32.5 (SD 6.4). Women made up 47.2% of the group (n=879) ^[1].

The main outcome was change in HbA1c from the start of the study to week 26. The key secondary outcome was change in body weight over the same period. Both were also measured at weeks 52 and 78. The statistical plan required each semaglutide dose to first clear a non-inferiority margin of 0.3% on HbA1c before it could be tested for superiority on HbA1c or weight ^[1].

The numbers

Compared with sitagliptin at week 26, oral semaglutide 7 mg cut HbA1c by an additional 0.3 percentage points (95% CI, −0.4% to −0.1%) and 14 mg by an additional 0.5 points (95% CI, −0.6% to −0.4%). Both differences had P values below .001 ^[1].

For body weight at week 26, the 7 mg dose produced 1.6 kg more weight loss than sitagliptin (95% CI, −2.0 to −1.1 kg) and the 14 mg dose 2.5 kg more (95% CI, −3.0 to −2.0 kg), again with P < .001 for both ^[1].

The 3 mg dose is the key exception. The authors report that non-inferiority of semaglutide 3 mg on HbA1c "was not demonstrated" — meaning this trial did not show it kept pace with sitagliptin on blood sugar ^[1].

At week 78, reductions in both HbA1c and body weight were still statistically significantly greater with the 14 mg dose than with sitagliptin ^[1]. The abstract does not give the week-78 numbers for the 7 mg dose, so those are not available here.

A total of 1,758 participants (94.3%) completed the trial, but 298 stopped treatment early. Early discontinuation rates were 16.7% for semaglutide 3 mg, 15.0% for 7 mg, 19.1% for 14 mg, and 13.1% for sitagliptin ^[1]. The abstract does not break down the specific side effects behind those discontinuations.

Why it matters for patients

Most GLP-1 medicines at the time of this trial were injections. PIONEER 3 was the first phase 3 comparison of an oral GLP-1 receptor agonist against a different class of glucose-lowering pill — sitagliptin, a DPP-4 inhibitor widely used as an add-on to metformin ^[1]. That makes it a direct head-to-head between two tablets taken by mouth, rather than a pill-versus-shot comparison.

The dose findings matter in practice. The 3 mg dose in this trial functioned as a starting step that was escalated every four weeks in the higher-dose groups ^[1]. The fact that 3 mg did not demonstrate non-inferiority on HbA1c suggests the benefit seen in this trial was tied to reaching the higher doses, not the starter dose.

The size of the difference also matters. A 0.3 to 0.5 percentage point advantage on HbA1c and a 1.6 to 2.5 kg weight advantage over 26 weeks are real but modest differences between two active drugs ^[1] — not a comparison against placebo.

Finally, the authors themselves note a limit: "Further research is needed to assess effectiveness in a clinical setting" ^[1]. Trial conditions, with scheduled visits and structured dose escalation, differ from everyday care.

What happens next

The trial ran from February 2016 to March 2018 and is registered as NCT02607865 ^[1]. JAMA published an accompanying commentary, "The Future of the GLP-1 Receptor Agonists," by I.B. Hirsch ^[1]. Regulatory status for oral semaglutide in the United States is not addressed in the source material provided here.