Question 1

What did ELIXA actually find?

Accepted Answer

On its primary endpoint — Cardiac death, a nonfatal heart attack, a nonfatal stroke, or admission for unstable angina — whichever came first — the treatment arm recorded 406 of 3,034 (13.4%) against 399 of 3,034 (13.2%) for the comparator. Hazard ratio 1.02 (95% CI 0.89 to 1.17); P<0.001 for noninferiority, P=0.81 for superiority. ★ Neutral, and deliberately so. The trial was built to rule out harm and it did — but it found no benefit either. This was the first completed cardiovascular trial in the class and it established the baseline that these drugs are safe to give shortly after a heart attack.

Question 2

How large was ELIXA, and how long did it run?

Accepted Answer

6,068 participants, over A median 25 months of follow-up. It was a Phase 3 trial, finished; headline results reported December 2015. Registered as NCT01147250, which you can look up yourself rather than taking anyone's summary for it.

Question 3

Who was studied in ELIXA?

Accepted Answer

Adults 30 and over whose type 2 diabetes had put them in hospital in the previous 180 days with an acute coronary event — a heart attack of either type, or unstable angina with raised cardiac markers. HbA1c had to sit between 5.5% and 11%. ⚠ The list of what ruled people out ran long: type 1 diabetes, bypass surgery after the qualifying event, a revascularization already scheduled inside 90 days, pancreatitis or pancreas surgery at any point, medullary thyroid cancer in the family or the person, and any other drug in this class. Average age 60, 69% men, average HbA1c 7.7%, average BMI 30.2, about 9 years since diagnosis. For 83%, the qualifying event was a heart attack. ⚠ A trial result describes the population it was measured in. If you differ from it materially — in age, in baseline weight, in whether you have diabetes — the number is context rather than a prediction.

Question 4

What side effects were reported in ELIXA?

Accepted Answer

The most commonly reported were any serious adverse event (20.6% against 22.1% on comparator), stopped the drug over a side effect (11.4% against 7.2% on comparator), nausea (24.9% against 6.2% on comparator). Slightly lower on the drug than on placebo. Rates come from this trial's population under its own dose-escalation schedule; a faster or slower titration changes what people experience.

Question 5

What else did ELIXA measure?

Accepted Answer

Secondary endpoints included the same count with heart-failure admissions added (456 of 3,034 (15.0%) against 469 of 3,034 (15.5%)); the widest count, adding stent and bypass procedures as well (661 of 3,034 (21.8%) against 659 of 3,034 (21.7%)); death from a cardiac cause (156 of 3,034 (5.1%) against 158 of 3,034 (5.2%)). ⚠ Secondary endpoints are weaker evidence than the primary one — a trial is powered for its primary outcome, and the rest are read as supporting rather than as findings in their own right.

Question 6

Does ELIXA mean this drug will work for me?

Accepted Answer

It cannot tell you that. A trial reports an average across a defined population under a fixed protocol, with monitoring and adherence support most people do not get. It is the best evidence available and it is still a distribution rather than a promise — some participants lost far more than the mean and some lost nothing. What it is genuinely good for is comparing options against each other, which is what the rest of this section is for.

Endpoint	Treatment	Comparator	Difference
The same count with heart-failure admissions added No difference. Widening the definition did not surface a hidden benefit.	456 of 3,034 (15.0%)	469 of 3,034 (15.5%)	Hazard ratio 0.97 (95% CI 0.85 to 1.10)
The widest count, adding stent and bypass procedures as well Still nothing. ★ A neutral result that stays neutral however the endpoint is drawn is far more convincing than one that appears only under a particular definition.	661 of 3,034 (21.8%)	659 of 3,034 (21.7%)	Hazard ratio 1.00 (95% CI 0.90 to 1.11)
Death from a cardiac cause ⚠ Identical in both arms — against the 22% reduction the liraglutide trial reported a year later. Same drug class, opposite result.	156 of 3,034 (5.1%)	158 of 3,034 (5.2%)	Hazard ratio 0.98 (95% CI 0.78 to 1.22)
Heart attacks that were not fatal No effect, and numerically slightly worse on the drug.	8.6% (event rate)	8.1% (event rate)	Hazard ratio 1.03 (95% CI 0.87 to 1.22)
Strokes that were not fatal ⚠ No effect at all, against the 39% reduction semaglutide produced. This pair is the earliest hard evidence that drugs in this class do not behave alike.	1.9% (event rate)	1.8% (event rate)	Hazard ratio 1.12 (95% CI 0.79 to 1.58)
Admitted with unstable chest pain Too few events either way to contribute much to the overall count.	1.0% (event rate)	1.3% (event rate)	Hazard ratio 1.11 (95% CI 0.47 to 2.62)
Admitted to hospital for heart failure ★ No effect — and worth registering as a genuine reassurance. A diabetes drug tested shortly before this had raised a heart-failure signal, which is part of why the FDA required these trials at all.	4.0% (event rate)	4.2% (event rate)	Hazard ratio 0.96 (95% CI 0.75 to 1.23)
Death from any cause at all No difference across two years.	211 of 3,034 (7.0%)	223 of 3,034 (7.4%)	Hazard ratio 0.94 (95% CI 0.78 to 1.13)
Three-month average blood sugar ⚠ Only about 0.27 points better than placebo, and that understates the drug — both arms could have other diabetes medication added when glucose ran high.	Mean reduction approximately 0.27 percentage points greater than placebo	Reference (background-glycemic-therapy adjusted)	Least-squares mean difference -0.27 percentage points (95% CI -0.31 to -0.22); P<0.001
Weight change ⛔ Under a kilogram. This is a short-acting drug taken before a meal, in an older post-heart-attack population. Nothing here bears on the weight drugs sold today.	-0.7 kg vs placebo	Reference	Least-squares mean difference -0.7 kg (95% CI -0.9 to -0.5); P<0.001
Protein leaking into the urine, at two years ⚠ Protein rose in both groups and rose less on the drug — and only in those who already had heavy protein loss. No effect in anyone else. Read the FAQ below before treating this as kidney protection.	+24% from baseline	+34% from baseline	Adjusted 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

Event	Treatment rate	Comparator rate
Any serious adverse event Slightly lower on the drug than on placebo.	20.6%	22.1%
Stopped the drug over a side effect 11.4% against 7.2%, mostly stomach and bowel effects — the same pattern as every drug in this class.	11.4%	7.2%
Nausea ★ Four times the placebo rate. Because this drug is taken before a meal and acts briefly, the nausea is tied to eating rather than spread through the week.	24.9%	6.2%
Vomiting Over four times the placebo rate.	12.4%	2.7%
Diarrhea Modestly more common than on placebo.	8.4%	5.6%
Low blood sugar with symptoms Barely different from placebo. This class only pushes insulin out when glucose is already elevated, which is why better control does not come with more lows.	16.3%	15.0%
Severe low blood sugar needing help from someone else Identical in both arms.	1.0% (30 events)	1.0% (31 events)
Pancreatitis, confirmed by independent review ★ No increase — numerically fewer on the drug. This was written in as a safety endpoint precisely because of the class-wide concern, and it found nothing.	0.2% (5 events)	0.3% (8 events)
Pancreatic tumors Fewer on the drug than on placebo, on very small numbers.	0.1% (3 events)	0.3% (9 events)
Allergic reactions No excess, despite this being a peptide derived from a lizard protein rather than a copy of the human hormone.	0.4% (11 events)	0.6% (17 events)

ELIXA: The First One, and It Was Neutral

Primary endpoint

Secondary endpoints

Adverse events

Subgroup analyses

Clinical significance

Frequently Asked Questions

References

Related trial deep-dives