In brief
A prespecified mortality analysis of the randomized VESALIUS-CV trial, published online on 31 August 2026, found lower all-cause mortality with evolocumab (Repatha) than with placebo in selected high-risk adults who had not previously experienced a heart attack or stroke. The five-year Kaplan–Meier mortality estimates were 7.9% versus 9.7% (hazard ratio 0.80; 95% CI 0.70–0.91).
This is evidence in a carefully selected high-risk population, not proof that Repatha extends life for everybody with elevated LDL-C. Many participants already had qualifying atherosclerosis; others had high-risk diabetes.
Who was studied, and why is that important?
VESALIUS-CV randomized 12,257 participants with qualifying atherosclerosis or high-risk diabetes, without previous myocardial infarction or stroke. Eligibility required elevated LDL-C (at least 90 mg/dL), non-HDL-C (at least 120 mg/dL) or ApoB (at least 80 mg/dL) at baseline. Participants received evolocumab or placebo on top of background care, and median follow-up was 4.6 years.
"No previous heart attack or stroke" does not mean low cardiovascular risk. Some participants had documented atherosclerosis and others had diabetes with additional risk characteristics. It would be wrong to apply this result to healthy adults with borderline cholesterol.
What happened to all-cause mortality?
Clinical evidence · Absolute event rates
12,257 high-risk participants without previous MI or stroke; 5-year Kaplan–Meier estimates · Bars share a 0–12% scale.
Five-year estimated mortality, not raw proportion of all randomized patients. Population had qualifying atherosclerosis or high-risk diabetes; not generalizable to everyone with high LDL-C. Prespecified analysis in Circulation ↗
In the prespecified analysis, there were 434 deaths with evolocumab and 539 with placebo. The reported five-year Kaplan–Meier estimates were 7.9% and 9.7%. The hazard ratio was 0.80 (95% CI 0.70–0.91; p=0.0005), consistent with an approximately 20% lower relative hazard of death over follow-up.
The estimated absolute five-year difference was 1.8 percentage points. This number is an estimate at the specified time point rather than the direct proportion of deaths among all randomized participants; individual risk and benefit will vary.
Did deaths from every cause decline?
Clinical evidence · Absolute event rates
Prespecified cause-of-death analysis; five-year Kaplan–Meier estimates · Bars share a 0–12% scale.
Estimated cardiovascular death, a cause-specific secondary analysis. Do not compare this risk difference directly with all-cause mortality or treat it as an independent trial. Original prespecified mortality analysis ↗
The analysis reported cardiovascular deaths of 156 versus 195 (five-year estimates 2.8% vs 3.6%; HR 0.79, 95% CI 0.64–0.98). Noncardiovascular deaths were 229 versus 268 (4.2% vs 5.0%; HR 0.85, 95% CI 0.71–1.01); that latter confidence interval includes no effect. Therefore it would overstate the findings to claim separately proven reductions in every cause of death.
The authors explored whether preventing nonfatal cardiovascular events could indirectly influence later mortality. These modeling analyses are informative but should not be mistaken for an independent randomized comparison of those mechanisms.
How does this fit with VESALIUS-CV cardiovascular outcomes?
The parent randomized study had already shown fewer first major cardiovascular events in this high-risk population. The new work addresses a different question: all-cause and cause-specific mortality. A prespecified mortality analysis is stronger than an unplanned post hoc claim, but should still be interpreted with its analysis plan and statistical context in mind.
PCSK9 monoclonal antibodies such as evolocumab have extensive randomized outcomes evidence, but this result does not mean that every PCSK9-targeting medicine, including oral compounds and siRNA approaches, has demonstrated the same survival effect.
Who might discuss treatment with a clinician?
Patients with documented atherosclerosis, high-risk diabetes, severe hypercholesterolemia or familial hypercholesterolemia who remain above an individualized LDL-C/ApoB goal despite tolerated therapy may discuss additional options, including PCSK9 inhibitors, with their clinician. Treatment depends on baseline risk, local approvals, affordability and potential benefit, not one headline.
Do not stop or replace existing treatment based on this article. Lifestyle remains useful alongside medication but cannot be assumed to reproduce the survival result seen in the randomized trial.
What are the limitations?
Important constraints include selection of high-risk participants, a median follow-up of 4.6 years, differences between time-to-event estimates and crude risks, uncertainty around cause-specific mortality and the fact that this was a prespecified secondary mortality analysis rather than a trial randomized around the expectation of a fixed all-cause mortality difference. Additional longer-term follow-up could refine estimates.
Our conclusions refer specifically to evolocumab in the VESALIUS-CV population. They do not establish a class-wide mortality effect for every LDL-lowering drug.
Primary sources and further reading
PCSK9 inhibitors: evidence and treatment options · 2026 LDL guideline · How low should LDL-C and ApoB go?