← All trial deep-dives

ELIXA trial deep-dive: lixisenatide CV outcomes after ACS (Pfeffer 2015)

Last verified 2026-05-28 · Phase 3 · Completed; primary results reported December 2015 · NCT01147250 ↗

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

ELIXA (Evaluation of Lixisenatide in Acute Coronary Syndrome) was a multicenter, randomized, double-blind, placebo-controlled phase 3 trial sponsored by Sanofi to satisfy the U.S. Food and Drug Administration 2008 guidance requiring every new diabetes drug to demonstrate cardiovascular safety. Investigators randomized 6,068 adults with type 2 diabetes who had been hospitalized for an acute coronary syndrome event — ST-elevation myocardial infarction, non-ST-elevation myocardial infarction, or unstable angina with biomarker elevation — within the previous 180 days, in a 1:1 ratio to once-daily subcutaneous lixisenatide titrated to 20 mcg or matching placebo, on top of locally determined standards of care. The primary endpoint was a four-point major adverse cardiovascular event composite: cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina. Median follow-up was 25 months. Pfeffer and colleagues reported in the New England Journal of Medicine on December 3, 2015 a hazard ratio of 1.02 (95% CI 0.89 to 1.17) — noninferior to placebo (P<0.001) but not superior (P=0.81). ELIXA was the first completed GLP-1 cardiovascular outcomes trial and remains the historical baseline for the class.

Enrollment
6,068
Duration
Median follow-up 25 months
Drug
Lixisenatide 20 mcg (Adlyxin / Lyxumia)
Population
Adults aged 30 or older with type 2 diabetes who had been hospitalized for an acute coronary syndrome event (STEMI, NSTEMI, or unstable angina with biomarker elevation) within the previous 180 days. HbA1c at screening was required to be between 5.5% and 11%. Patients were excluded for type 1 diabetes, coronary artery bypass graft after the qualifying ACS event, planned revascularization within 90 days, history of pancreatitis or pancreatectomy, personal or family history of medullary thyroid cancer, or required use of other incretin-based therapy. The enrolled population had a mean age of 60 years, was 69% male, with a mean HbA1c of 7.7%, mean BMI 30.2 kg/m², and mean diabetes duration of 9.3 years. The qualifying ACS event was an MI in 83% of participants and unstable angina in 17%.

Primary endpoint

Four-point composite MACE: cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for unstable angina (time-to-first-event)

Treatment arm

406 of 3,034 (13.4%)

Comparator

399 of 3,034 (13.2%)

Treatment difference: Hazard ratio 1.02 (95% CI 0.89 to 1.17); P<0.001 for noninferiority, P=0.81 for superiority

Lixisenatide met the FDA noninferiority threshold (upper 95% CI bound below 1.3) but did not demonstrate superiority. ELIXA was the first completed GLP-1 cardiovascular outcomes trial and established the regulatory baseline that the class is cardiovascular-safe in a post-ACS population.

Secondary endpoints

EndpointTreatmentComparatorDifference
Expanded composite: cardiovascular death, nonfatal MI, nonfatal stroke, hospitalization for unstable angina, or hospitalization for heart failure

No significant difference when heart-failure hospitalization was added to the primary composite.

456 of 3,034 (15.0%)469 of 3,034 (15.5%)Hazard ratio 0.97 (95% CI 0.85 to 1.10)
Six-point composite: cardiovascular death, nonfatal MI, nonfatal stroke, hospitalization for unstable angina, hospitalization for heart failure, or coronary revascularization

No signal across the broadest cardiovascular composite reported.

661 of 3,034 (21.8%)659 of 3,034 (21.7%)Hazard ratio 1.00 (95% CI 0.90 to 1.11)
Cardiovascular death (individual MACE component)

No effect on cardiovascular mortality, in contrast to the 22% reduction reported a year later in LEADER for liraglutide.

156 of 3,034 (5.1%)158 of 3,034 (5.2%)Hazard ratio 0.98 (95% CI 0.78 to 1.22)
Nonfatal myocardial infarction (individual MACE component)

No effect on nonfatal MI; numerical rates marginally higher with lixisenatide.

8.6% (event rate)8.1% (event rate)Hazard ratio 1.03 (95% CI 0.87 to 1.22)
Nonfatal stroke (individual MACE component)

No effect on stroke. Contrast with SUSTAIN-6, where semaglutide reduced nonfatal stroke by 39% — one of the earliest signals that within-class GLP-1 effects vary.

1.9% (event rate)1.8% (event rate)Hazard ratio 1.12 (95% CI 0.79 to 1.58)
Hospitalization for unstable angina (individual MACE component)

Few events; the unstable-angina component contributed minimally to the overall composite.

1.0% (event rate)1.3% (event rate)Hazard ratio 1.11 (95% CI 0.47 to 2.62)
Hospitalization for heart failure

No effect on heart-failure hospitalization in a population already enriched for left-ventricular dysfunction. Notable because the contemporaneous SAVOR-TIMI 53 trial of saxagliptin had raised a heart-failure signal for the DPP-4 class.

4.0% (event rate)4.2% (event rate)Hazard ratio 0.96 (95% CI 0.75 to 1.23)
All-cause mortality

No effect on total mortality over 25 months of follow-up.

211 of 3,034 (7.0%)223 of 3,034 (7.4%)Hazard ratio 0.94 (95% CI 0.78 to 1.13)
Change in HbA1c at end of treatment

Modest glycemic separation; trial protocol permitted open-label glycemic rescue in both arms, narrowing the observed difference.

Mean reduction approximately 0.27 percentage points greater than placeboReference (background-glycemic-therapy adjusted)Least-squares mean difference -0.27 percentage points (95% CI -0.31 to -0.22); P<0.001
Change in body weight at end of treatment

Small weight separation, consistent with the short-acting prandial GLP-1 pharmacology of lixisenatide and the older, sicker post-ACS population.

-0.7 kg vs placeboReferenceLeast-squares mean difference -0.7 kg (95% CI -0.9 to -0.5); P<0.001
Percent change from baseline in urinary albumin-to-creatinine ratio (UACR) at week 108 (prespecified secondary, reported in detail by Muskiet 2018)

Lixisenatide reduced progression of urinary albumin excretion in participants with baseline macroalbuminuria; no significant effect in normo- or microalbuminuria subgroups. Reported in the ELIXA renal substudy (Muskiet et al., Lancet Diabetes Endocrinol 2018, PMID 30292589).

+24% from baseline+34% from baselineAdjusted treatment difference -1.69% (95% CI -11.69 to 8.31) overall; -39.18% (95% CI -68.5 to -9.84; P=0.007) in the macroalbuminuria subgroup

Adverse events

EventTreatment rateComparator rate
Any serious adverse event

Total serious-adverse-event rate was numerically lower with lixisenatide than placebo.

20.6%22.1%
Discontinuation due to adverse event

Gastrointestinal events were the most common reason for discontinuing lixisenatide, consistent with the GLP-1 class profile.

11.4%7.2%
Nausea

Most common AE with lixisenatide; typically prandial and concentrated in the first weeks of titration.

24.9%6.2%
Vomiting12.4%2.7%
Diarrhea8.4%5.6%
Symptomatic hypoglycemia

Similar overall hypoglycemia rates; the glucose-dependent insulin-release mechanism limits hypoglycemia even with added glycemic effect.

16.3%15.0%
Severe hypoglycemia (adjudicated)

No excess severe hypoglycemia with lixisenatide.

1.0% (30 events)1.0% (31 events)
Acute pancreatitis (adjudicated)

No increased pancreatitis risk; rate numerically lower with lixisenatide. Pre-specified safety endpoint after class-wide regulatory concern.

0.2% (5 events)0.3% (8 events)
Pancreatic neoplasms (adjudicated)

Fewer pancreatic neoplasms with lixisenatide; numerical imbalance favored the active arm.

0.1% (3 events)0.3% (9 events)
Allergic reactions (adjudicated)

No excess allergic reactions despite the peptide structure of the molecule.

0.4% (11 events)0.6% (17 events)

Subgroup analyses

  • Index event: STEMI vs NSTEMI vs unstable angina: Hazard ratios consistent across qualifying-event types (interaction P=0.18)

    No subgroup showed a superiority signal; the neutral primary result was reproduced across MI severity strata.

  • Time from qualifying ACS event to randomization (within 60 days vs 60-180 days): Hazard ratios consistent across recency strata (interaction P>0.20)

    No effect of how recently the qualifying ACS event occurred on the treatment effect.

  • Baseline estimated glomerular filtration rate (<60 vs ≥60 mL/min/1.73m²): Hazard ratios consistent across renal-function strata

    No meaningful interaction with baseline kidney function on the primary outcome.

  • Baseline HbA1c (<7.5% vs ≥7.5%): Hazard ratios consistent across glycemic strata

    Glycemic control at baseline did not modify the cardiovascular treatment effect.

  • Age (<65 vs ≥65 years): Hazard ratios consistent across age strata

    No meaningful age effect on the neutral primary outcome.

  • Baseline albuminuria status (normo- vs micro- vs macroalbuminuria): UACR progression reduced 39% in the macroalbuminuria subgroup (P=0.007); no effect in lower-albuminuria strata

    From the Muskiet 2018 renal substudy. The macroalbuminuria signal was hypothesis-generating and consistent with the broader GLP-1 class kidney pattern later confirmed in LEADER, REWIND, and FLOW.

Clinical significance

ELIXA is the historical baseline for the entire GLP-1 cardiovascular outcomes literature. It was the first such trial to complete and report, beating LEADER to publication by seven months and SUSTAIN-6 by nine months. The neutral primary result — hazard ratio 1.02 for the four-point MACE composite — satisfied the FDA cardiovascular-safety requirement but did not establish superiority, leaving open the question of whether GLP-1 receptor agonists offered cardioprotection in addition to glycemic control. That question was answered in 2016 when LEADER and SUSTAIN-6 reported positive results for liraglutide and semaglutide respectively. The contrast between ELIXA's neutral result and the later positive trials is generally attributed to three factors: the short-acting prandial pharmacology of lixisenatide compared with the longer-acting human GLP-1 analogues, the unusually high-risk post-ACS population in ELIXA that limited the ability to detect long-term atherosclerotic-event reductions over only 25 months of follow-up, and the smaller-than-typical separation in glycemic control between arms. Lixisenatide received FDA approval as Adlyxin in July 2016 on the basis of its glycemic-efficacy program but never achieved meaningful U.S. commercial uptake; Sanofi voluntarily discontinued Adlyxin in the United States in January 2023. The Lyxumia brand persists in select international markets. ELIXA's lasting contribution to the field is methodological: it set the design template — event-driven, FDA noninferiority bar, broad MACE composite — that every subsequent GLP-1 outcomes trial has followed.

Frequently asked questions

Why was ELIXA's effect smaller than LEADER or SUSTAIN-6?

Three factors are routinely cited. First, pharmacology: lixisenatide is a short-acting prandial GLP-1 receptor agonist with a half-life of roughly three hours, dosed once daily before breakfast and producing a transient postprandial signal. Liraglutide (LEADER) is a once-daily human GLP-1 analogue with a half-life of about 13 hours, and semaglutide (SUSTAIN-6) is once-weekly with a half-life of about a week — both produce continuous receptor activation that translates into greater glycemic, weight, and likely vascular effects. Second, follow-up duration: ELIXA ran a median of 25 months versus 3.8 years in LEADER and 2.1 years in SUSTAIN-6, limiting the time available to detect atherosclerotic-event reductions. Third, the post-ACS population in ELIXA had unusually high event rates and aggressive background care, which compresses the effect window. The neutral result is therefore consistent with — rather than contradicting — the later positive trials in the GLP-1 class.

Was lixisenatide commercially successful?

No. Lixisenatide was developed by Zealand Pharma and licensed to Sanofi, who launched it as Lyxumia in the European Union in 2013 and as Adlyxin in the United States in July 2016. U.S. uptake was limited because by the time of FDA approval the class had moved to once-weekly dosing — exenatide weekly (Bydureon) and dulaglutide (Trulicity) were already established, and once-weekly semaglutide (Ozempic) launched in late 2017 with substantially greater efficacy. Sanofi voluntarily discontinued Adlyxin in the United States in January 2023, citing low commercial demand. Lyxumia remains marketed in select international territories. The fixed-ratio combination iGlarLixi (Soliqua in the U.S., Suliqua in the EU), which pairs insulin glargine with lixisenatide in a single daily pen, continues to be sold and is the most active commercial channel for the molecule today.

Does ELIXA inform current GLP-1 prescribing decisions?

Indirectly. ELIXA is not used to drive prescribing for any specific drug, because lixisenatide is no longer marketed in the United States and the field has moved to once-weekly semaglutide, dulaglutide, and tirzepatide. Its remaining relevance is twofold. First, ELIXA established that GLP-1 receptor agonists are cardiovascular-safe in a high-risk post-acute-coronary-syndrome population — a useful baseline for prescribers managing patients on a GLP-1 who experience an MI or unstable-angina event. Second, ELIXA's neutral result clarified that not every GLP-1 produces cardiovascular benefit; the positive signals in LEADER, SUSTAIN-6, REWIND, AMPLITUDE-O, and SELECT are pharmacology- and exposure-specific, not a universal class property. Modern guideline language reflects this nuance: the American Diabetes Association lists liraglutide, semaglutide, and dulaglutide as having demonstrated cardiovascular benefit, but does not extend that recommendation to lixisenatide.

Why was ELIXA designed specifically for the post-acute-coronary-syndrome population?

Two reasons. First, regulatory efficiency: the FDA 2008 cardiovascular-safety guidance required new diabetes drugs to rule out a hazard ratio above 1.3 for MACE, and that ratio can be statistically confirmed faster in a high-event-rate population. Enrolling patients within 180 days of an acute coronary syndrome event ensured a high background MACE rate, allowing Sanofi to reach the prespecified number of adjudicated events in roughly two years rather than four. Second, scientific rationale: at the time ELIXA was designed (2010), the dominant concern was whether GLP-1 receptor agonists were safe to use in the acutely vulnerable period after an MI or unstable angina hospitalization, given some preclinical signals around heart rate and arrhythmia. ELIXA answered both questions simultaneously — meeting the FDA bar and demonstrating safety in the post-ACS window — but the design also limited the trial's ability to detect long-term atherosclerotic-event reductions of the kind observed in the more cardiovascular-stable populations enrolled in LEADER and SUSTAIN-6.

How does ELIXA inform the GLP-1 receptor agonist class cardiovascular effect?

ELIXA is the data point that prevents the field from treating GLP-1 cardiovascular benefit as a uniform class effect. The neutral result for lixisenatide stands alongside the neutral result for once-weekly exenatide (EXSCEL, 2017) and the neutral result for oral semaglutide (PIONEER 6, 2019, which was powered for noninferiority only), and on the other side positive trials for liraglutide (LEADER), injectable semaglutide (SUSTAIN-6 and SELECT), dulaglutide (REWIND), albiglutide (HARMONY OUTCOMES), and efpeglenatide (AMPLITUDE-O). The dividing line tracks roughly with structural-similarity to human GLP-1 and duration of receptor exposure rather than with the simple distinction of being a GLP-1. Meta-analyses pooling all GLP-1 trials report a roughly 12% reduction in three-point MACE on average, but the heterogeneity between individual agents is real, and ELIXA is the foundational evidence for that heterogeneity.

Did ELIXA reveal anything about kidney outcomes?

Yes, in the dedicated renal substudy by Muskiet and colleagues published in Lancet Diabetes Endocrinology in 2018 (PMID 30292589). The prespecified analysis of percent change in urinary albumin-to-creatinine ratio from baseline to week 108 showed no overall treatment effect, but a 39% relative reduction in UACR progression in the subgroup of participants who entered the trial with macroalbuminuria (UACR above 300 mg per gram). There was no effect in the normoalbuminuria or microalbuminuria subgroups, and no effect on harder kidney endpoints such as doubling of serum creatinine or end-stage kidney disease — those events were too rare in 25 months of follow-up to detect a signal. The macroalbuminuria finding was hypothesis-generating rather than confirmatory and was consistent with the broader kidney-protection pattern that the GLP-1 class later established in the LEADER renal substudy (Mann 2017), the REWIND exploratory kidney analysis, and ultimately the FLOW trial in 2024.

References

  1. 1.Pfeffer MA, Claggett B, Diaz R, Dickstein K, Gerstein HC, Køber LV, Lawson FC, Ping L, Wei X, Lewis EF, Maggioni AP, McMurray JJ, Probstfield JL, Riddle MC, Solomon SD, Tardif JC; ELIXA Investigators. Lixisenatide in Patients with Type 2 Diabetes and Acute Coronary Syndrome. N Engl J Med. 2015. PMID: 26630143.
  2. 2.Muskiet MHA, Tonneijck L, Huang Y, Liu M, Saremi A, Heerspink HJL, van Raalte DH. Lixisenatide and renal outcomes in patients with type 2 diabetes and acute coronary syndrome: an exploratory analysis of the ELIXA randomised, placebo-controlled trial. Lancet Diabetes Endocrinol. 2018. PMID: 30292589.
  3. 3.Sanofi. Evaluation of Cardiovascular Outcomes in Patients With Type 2 Diabetes After Acute Coronary Syndrome During Treatment With AVE0010 (Lixisenatide) — ELIXA. ClinicalTrials.gov, NCT01147250. 2015. https://clinicaltrials.gov/study/NCT01147250
  4. 4.U.S. Food and Drug Administration. Adlyxin (lixisenatide) New Drug Application 208471 — Approval Package. FDA Drugs@FDA, July 2016. 2016. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2016/208471Orig1s000TOC.cfm

Related trial deep-dives