← All trial deep-dives

PIONEER 6 deep-dive: oral semaglutide cardiovascular outcomes (Husain 2019 NEJM)

Last verified 2026-05-28 · Phase 3a · Completed (results published June 2019) · NCT02692716 ↗

By Eli Marsden · Founding Editor
Editorially reviewed (not clinically reviewed) · How we verify contentLast reviewed

PIONEER 6 (A Trial Investigating the Cardiovascular Safety of Oral Semaglutide in Subjects With Type 2 Diabetes) was a multicenter, randomized, double-blind, placebo-controlled, event-driven phase 3a non-inferiority trial sponsored by Novo Nordisk. The trial was designed as a pre-approval cardiovascular safety study to satisfy the FDA's 2008 guidance on diabetes drugs. Investigators randomized 3,183 adults with type 2 diabetes who were either at least 50 years old with established cardiovascular or chronic kidney disease, or at least 60 years old with cardiovascular risk factors only, 1:1 to oral semaglutide titrated to 14 mg once daily or matching placebo, on top of standard care. The primary endpoint was time to first three-point major adverse cardiovascular event — cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke — adjudicated by a blinded committee. Husain and colleagues published the results in the New England Journal of Medicine on June 11, 2019. The trial met its pre-specified non-inferiority margin and supported FDA approval of Rybelsus in September 2019 as the first orally administered GLP-1 receptor agonist.

Enrollment
3,183
Duration
Event-driven; median follow-up 15.9 months (planned minimum exposure was shorter than the multi-year duration of comparable CVOTs)
Drug
Oral semaglutide (up to 14 mg once daily)
Population
Adults with type 2 diabetes and HbA1c ≥7.0% who were either aged ≥50 years with established cardiovascular disease (myocardial infarction, ischemic stroke, peripheral arterial disease, prior coronary revascularization, or chronic heart failure NYHA II-III) or chronic kidney disease (eGFR <60 mL/min/1.73 m²), or aged ≥60 years with cardiovascular risk factors only. Mean age 66 years, 31.6% female, mean BMI 32.3 kg/m², mean HbA1c 8.2%, mean diabetes duration 14.9 years. 84.7% had established cardiovascular disease or chronic kidney disease at entry. Background therapy included metformin (77%), insulin (61%), sulfonylureas (29%), statins, antihypertensives, and antiplatelet agents per regional standard of care.

Primary endpoint

Three-point MACE: first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke (time-to-first-event, oral semaglutide vs placebo)

Treatment arm

61 of 1,591 (3.8%)

Comparator

76 of 1,592 (4.8%)

Treatment difference: Hazard ratio 0.79 (95% CI 0.57–1.11); P<0.001 for non-inferiority; P=0.17 for superiority

Non-inferior to placebo on three-point MACE — the regulatory question the trial was powered to answer. The 21% directional reduction in the point estimate did not reach statistical superiority; the upper bound of the confidence interval crossed 1.0. PIONEER 6 was designed and powered for non-inferiority on a short event-driven timeline, not for a superiority claim.

Secondary endpoints

EndpointTreatmentComparatorDifference
Cardiovascular death (component of MACE; pre-specified hierarchical secondary)

51% relative reduction in cardiovascular death. This component is the most clinically striking PIONEER 6 result, but the publication and FDA reviews caution against over-interpretation: pre-specified hierarchical testing stopped at the prior non-significant step, so cardiovascular death is reported as nominally significant rather than confirmatorily significant.

15 of 1,591 (0.9%)30 of 1,592 (1.9%)Hazard ratio 0.49 (95% CI 0.27–0.92)
Nonfatal myocardial infarction (component of MACE)

No reduction in nonfatal MI; the point estimate was numerically higher on oral semaglutide but the confidence interval was wide and crossed 1.0. This is a different component pattern than SUSTAIN-6, where nonfatal MI trended favorably.

37 of 1,591 (2.3%)31 of 1,592 (1.9%)Hazard ratio 1.18 (95% CI 0.73–1.90)
Nonfatal stroke (component of MACE)

Numerically lower nonfatal stroke on oral semaglutide, not statistically significant on its own. Event counts were small over the short median follow-up.

12 of 1,591 (0.8%)16 of 1,592 (1.0%)Hazard ratio 0.74 (95% CI 0.35–1.57)
Death from any cause (all-cause mortality; pre-specified secondary)

49% relative reduction in all-cause mortality — the most provocative secondary signal from the trial. As with cardiovascular death, hierarchical testing had already exited so this is nominally significant rather than confirmatorily significant. The signal has been one of the most-debated findings in GLP-1 cardiology and was a major driver of subsequent SOUL trial design.

23 of 1,591 (1.4%)45 of 1,592 (2.8%)Hazard ratio 0.51 (95% CI 0.31–0.84)
Expanded cardiovascular composite (CV death, nonfatal MI, nonfatal stroke, hospitalization for heart failure, or hospitalization for unstable angina)

Directionally favorable on the broader composite but not statistically significant, consistent with the primary MACE result. Heart-failure hospitalization was numerically similar between arms.

79 of 1,591 (5.0%)100 of 1,592 (6.3%)Hazard ratio 0.82 (95% CI 0.61–1.10)
HbA1c change at week 26 (mean change from baseline, on-treatment analysis)

Confirms oral semaglutide produces glycemic lowering comparable to other GLP-1 receptor agonists at the 14 mg dose, although PIONEER 6 was a cardiovascular safety trial rather than a glycemic-efficacy trial.

−1.0 percentage points−0.3 percentage pointsEstimated treatment difference −0.7 percentage points (95% CI not reported in primary paper; consistent with PIONEER glycemic-efficacy trials)
Body weight change at end of treatment (mean change from baseline)

Modest weight reduction relative to obesity-dose subcutaneous semaglutide 2.4 mg. PIONEER 6 used the T2D dose (up to 14 mg oral) and recruited a population with high baseline burden of cardiovascular disease and insulin use.

−4.2 kg−0.8 kgEstimated treatment difference −3.4 kg
Severe or blood-glucose-confirmed symptomatic hypoglycemia

Consistent with the broader GLP-1 class: hypoglycemia risk comes from concomitant insulin or sulfonylureas, not from semaglutide itself.

Rates broadly similar between arms (semaglutide did not increase hypoglycemia on its own)Baseline rate from background insulin and sulfonylureasNo clinically meaningful imbalance reported
Premature treatment discontinuation due to adverse events

Approximately double the placebo discontinuation rate, concentrated in the dose-escalation period. The premature-discontinuation rate has been a recurring real-world adherence concern for oral semaglutide.

11.6% (185/1,591)6.5% (104/1,592)Higher on oral semaglutide, driven by gastrointestinal events during titration

Adverse events

EventTreatment rateComparator rate
Any serious adverse event

Overall serious adverse events were lower on oral semaglutide, largely reflecting fewer cardiovascular events in a high-risk population.

18.9% (301/1,591)22.5% (358/1,592)
Discontinuation due to adverse events

Gastrointestinal events were the dominant reason for stopping. Discontinuation peaked during the 4–8 week titration period.

11.6% (185/1,591)6.5% (104/1,592)
Nausea

Most common adverse event; mostly mild-to-moderate and concentrated in early treatment. Per the FDA Rybelsus label, ~20% of patients on the 14 mg dose report nausea.

Higher on oral semaglutide than placebo (consistent with the PIONEER program; specific incidence not the headline endpoint in the safety publication)Lower (placebo rate similar to other GLP-1 CVOT placebo arms)
Acute pancreatitis (adjudicated)

Numerically lower on oral semaglutide; no signal of increased pancreatitis risk. Consistent with the broader semaglutide safety database.

0.1% (1/1,591)0.6% (10/1,592)
Acute kidney injury

Numerically higher on oral semaglutide but not statistically significant. Often linked to volume depletion from GI side effects during titration.

1.5% (24/1,591)1.0% (16/1,592)
Diabetic retinopathy complications

No statistically significant excess in PIONEER 6, in contrast with the SUSTAIN-6 signal for subcutaneous semaglutide. The FDA label still flags retinopathy monitoring for patients with a history of retinopathy and rapid HbA1c improvement.

7.1% (113/1,591)6.3% (101/1,592)
Malignant neoplasms

No imbalance in overall cancer incidence over the median 15.9-month follow-up. The trial was too short to address long-latency cancer risk.

Rates balanced between arms over the trial periodRates balanced between arms over the trial period

Subgroup analyses

  • Established cardiovascular disease or chronic kidney disease at baseline (~85% of cohort): MACE hazard ratio consistent with the overall result (~0.79)

    Pre-specified subgroup; the secondary-prevention majority drove the trial's enrollment and event accrual.

  • Lower-risk participants (≥60 years with CV risk factors only, no established disease; ~15% of cohort): Small subgroup with few events; not informative for cardiovascular efficacy

    The trial was not powered to draw conclusions in this lower-risk slice.

  • Background insulin use: MACE direction consistent across insulin users and non-users

    Approximately 61% of the cohort was on insulin at baseline, reflecting the advanced-disease population.

  • Region (Europe vs North America vs Asia vs other): No clear heterogeneity reported in the publication

    Multinational enrollment across 21 countries; effects directionally consistent across regions, though event counts in any single region were small.

Clinical significance

PIONEER 6 is the regulatory and scientific foundation for treating oral semaglutide (Rybelsus) as cardiovascularly safe in adults with type 2 diabetes — but it is not the trial that earned Rybelsus a cardiovascular-risk-reduction label. The trial was designed and powered for non-inferiority on a short event-driven timeline, and that is what it delivered. The eye-catching 51% reduction in cardiovascular death and 49% reduction in all-cause mortality were pre-specified secondary endpoints but fell outside the hierarchical testing chain after the primary failed superiority, so they are nominally significant rather than confirmatorily significant. The MI and stroke components individually did not reach significance, partly because the median 15.9-month follow-up accrued relatively few events. The trial achieved its regulatory purpose — pre-approval cardiovascular safety — and seeded the SOUL trial (NCT03914326), the longer, larger superiority-powered CVOT of oral semaglutide that reported in 2024. Until SOUL, PIONEER 6 is the best available cardiovascular evidence for Rybelsus, and most cardiologists treat the mortality signal as hypothesis-generating rather than practice-changing.

Frequently asked questions

Why did PIONEER 6 only show non-inferiority and not superiority?

Because that is what it was designed and powered to do. PIONEER 6 was a pre-approval cardiovascular safety study built to satisfy the FDA's 2008 diabetes-drug guidance, which requires a sponsor to rule out a hazard ratio above roughly 1.3 for major adverse cardiovascular events. To minimize time-to-approval, the trial was event-driven with a relatively low pre-specified number of MACE events, a median follow-up of just 15.9 months, and a non-inferiority margin as the primary regulatory question. The primary three-point MACE hazard ratio was 0.79 (95% CI 0.57–1.11), which comfortably cleared non-inferiority (P<0.001) but did not reach the threshold for superiority (P=0.17). The trial was simply too short and too small to confirm superiority — that question is being answered by the longer SOUL trial of oral semaglutide.

How does PIONEER 6 compare to SUSTAIN-6 (subcutaneous semaglutide)?

Both trials enrolled adults with type 2 diabetes at high cardiovascular risk and tested semaglutide against placebo, but the designs and conclusions differ. SUSTAIN-6 (Marso 2016, NEJM, PMID 27633186) enrolled 3,297 participants, used subcutaneous semaglutide 0.5 or 1.0 mg weekly over a median 104 weeks, and was pre-specified for both non-inferiority and superiority; it reported a three-point MACE hazard ratio of 0.74 (P=0.02 for superiority). PIONEER 6 enrolled 3,183 participants, used oral semaglutide up to 14 mg daily over a median 15.9 months, and was powered only for non-inferiority; it reported a MACE hazard ratio of 0.79, which cleared non-inferiority but did not reach superiority. The Ozempic cardiovascular-risk-reduction label rests on SUSTAIN-6; Rybelsus does not currently carry an analogous cardiovascular-risk-reduction indication.

Should patients on Rybelsus take it for cardiovascular protection?

Not as a primary reason in current labeling. The FDA-approved Rybelsus label is for glycemic improvement in adults with type 2 diabetes; it does not include a cardiovascular-risk-reduction indication, even though PIONEER 6 established cardiovascular safety. For patients who specifically need a GLP-1 receptor agonist with a cardiovascular-risk-reduction indication, the FDA has approved subcutaneous semaglutide (Ozempic), liraglutide (Victoza), dulaglutide (Trulicity), and others. Many endocrinologists nevertheless view the PIONEER 6 mortality signal as encouraging and use Rybelsus comfortably in patients with cardiovascular disease, but the formal regulatory claim has to wait on the SOUL trial of oral semaglutide.

Was PIONEER 6 too short to be meaningful?

It was short by CVOT standards, but the design was deliberate. With a median follow-up of 15.9 months and only 137 primary MACE events accrued, the trial accumulated about a quarter of the events of larger CVOTs like LEADER or REWIND. That gave it adequate statistical power for non-inferiority but not for superiority. Short follow-up also limits inference about endpoints that take years to declare themselves — long-latency cancers, durable kidney outcomes, late-emerging retinopathy. The trial achieved its primary regulatory purpose: pre-approval cardiovascular safety. The follow-on SOUL trial (NCT03914326) was designed specifically to address the duration limitation with a multi-year, superiority-powered design in a similar population.

Does the all-cause mortality signal mean anything?

It is one of the most-debated findings in GLP-1 cardiology. The reported 49% reduction in all-cause mortality (HR 0.51; 95% CI 0.31–0.84) and 51% reduction in cardiovascular death (HR 0.49; 95% CI 0.27–0.92) are large effects with confidence intervals that exclude 1.0. The caveat is statistical: pre-specified hierarchical testing exited after the primary failed superiority, so these are nominally significant rather than confirmatorily significant under the trial's analysis plan. Event counts were also small (23 deaths on oral semaglutide vs 45 on placebo) and the follow-up short. The FDA accepted the signal as supportive of overall benefit-risk but did not approve a mortality-related label claim. Most cardiologists treat it as hypothesis-generating, with confirmation pending the SOUL superiority trial and consistent with the broader pattern of GLP-1 cardiovascular benefit seen across the class.

How does PIONEER 6 fit alongside LEADER, REWIND, and SELECT?

PIONEER 6 sits in a family of GLP-1 cardiovascular outcomes trials with consistent direction-of-effect but different populations, drugs, and durations. LEADER (Marso 2016, NEJM, PMID 27295427) tested liraglutide in 9,340 adults with T2D over a median 3.8 years and reported a MACE hazard ratio of 0.87 (P=0.01 for superiority). REWIND (Gerstein 2019, Lancet, PMID 31189511) tested dulaglutide in 9,901 adults with T2D over a median 5.4 years and reported a MACE hazard ratio of 0.88 (P=0.026). SELECT (Lincoff 2023, NEJM, PMID 37952131) tested subcutaneous semaglutide 2.4 mg in 17,604 adults with obesity but without diabetes over a mean 39.8 months and reported a MACE hazard ratio of 0.80 (P<0.001). PIONEER 6 is the smallest and shortest of the group; its directional 0.79 hazard ratio is consistent with the class effect but was not powered to be confirmed on its own.

References

  1. 1.Husain M, Birkenfeld AL, Donsmark M, Dungan K, Eliaschewitz FG, Franco DR, Jeppesen OK, Lingvay I, Mosenzon O, Pedersen SD, Tack CJ, Thomsen M, Vilsbøll T, Warren ML, Bain SC; PIONEER 6 Investigators. Oral Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2019. PMID: 31185157.
  2. 2.Novo Nordisk A/S. A Trial Investigating the Cardiovascular Safety of Oral Semaglutide in Subjects With Type 2 Diabetes (PIONEER 6) — Study Results. ClinicalTrials.gov, NCT02692716. 2019. https://clinicaltrials.gov/study/NCT02692716
  3. 3.Marso SP, Bain SC, Consoli A, Eliaschewitz FG, Jódar E, Leiter LA, Lingvay I, Rosenstock J, Seufert J, Warren ML, Woo V, Hansen O, Holst AG, Pettersson J, Vilsbøll T; SUSTAIN-6 Investigators. Semaglutide and Cardiovascular Outcomes in Patients with Type 2 Diabetes. N Engl J Med. 2016. PMID: 27633186.
  4. 4.Marso SP, Daniels GH, Brown-Frandsen K, et al.; LEADER Steering Committee; LEADER Trial Investigators. Liraglutide and Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2016. PMID: 27295427.
  5. 5.Lincoff AM, Brown-Frandsen K, Colhoun HM, et al.; SELECT Trial Investigators. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. N Engl J Med. 2023. PMID: 37952131.
  6. 6.Gerstein HC, Colhoun HM, Dagenais GR, et al.; REWIND Investigators. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet. 2019. PMID: 31189511.
  7. 7.U.S. Food and Drug Administration. Rybelsus (semaglutide) tablets — Prescribing Information (initial approval September 2019). FDA Drug Label. 2019. https://www.accessdata.fda.gov/drugsatfda_docs/label/2019/213051s000lbl.pdf

Related trial deep-dives