Repatha Long-Term Side Effects: Eight Years of Safety Data Reviewed

What the Long-Term Safety Data Actually Shows

Repatha's long-term safety profile rests on three big datasets: the FOURIER trial (27,564 patients, median 2.2 years), the FOURIER-OLE extension (6,635 patients, median 5 additional years, maximum total exposure 8.4 years), and the EBBINGHAUS cognitive substudy (1,204 patients in the original trial plus 306 in the long extension).

Here's the safety summary that holds up across these studies:

The pattern is consistent: the things people worry about with very low LDL haven't materialized, and the things that do occur are predictable from earlier trials.

The Diabetes Question

Statins increase the risk of new-onset diabetes by a small but real amount, especially in people who already have prediabetes. Since Repatha lowers LDL even further than statins, a reasonable worry is whether it amplifies that effect.

The data says it doesn't. In FOURIER, evolocumab did not increase new-onset diabetes, including in patients with prediabetes at baseline, and HbA1c plus fasting glucose stayed similar to placebo over the trial.

Broader meta-analyses of PCSK9 inhibitor trials reach the same conclusion across thousands of patients. FOURIER-OLE's longer follow-up confirmed the same result over up to 8 years.

There's one wrinkle worth knowing. Real-world cohorts and post-acute coronary syndrome data have reported modest increases in fasting glucose, especially early in treatment and especially in people with baseline prediabetes. The signal is small, transient, and concentrated in patients who were already heading toward diabetes.

If you have prediabetes and start Repatha, expect to monitor your fasting glucose and HbA1c closely in the first year. After that, the long-term data is reassuring.

What About Cognitive Side Effects

PCSK9 inhibitors push LDL to levels rarely seen outside of newborns. Since LDL helps build cell membranes, including in the brain, the obvious concern is whether very low LDL impairs cognition over years.

This question got tested directly. EBBINGHAUS was a cognitive substudy nested inside FOURIER, with patients put through formal neuropsychological testing on executive function, working memory, episodic memory, and psychomotor speed. After a median of 19 months, there was no difference between evolocumab and placebo on any cognitive domain, and no association between LDL level and cognitive change.

Patient-reported everyday cognition in the broader FOURIER cohort was similar between evolocumab and placebo, including in patients whose LDL dropped below 20 mg/dL.

Pharmacovigilance databases do show disproportionate reports of neurocognitive complaints with PCSK9 inhibitors as a class, but spontaneous-report data can't establish causality and tends to capture the loudest patient experiences rather than the typical one. The controlled-trial evidence is consistent: no measurable cognitive harm.

The Very Low LDL Question

About 63% of patients in FOURIER-OLE achieved LDL below 40 mg/dL at 12 weeks. Many got below 20 mg/dL. A specific FOURIER-OLE analysis looked at safety outcomes by achieved LDL across the full follow-up of up to 8.6 years and found no significant association between lower LDL (down to under 20 mg/dL) and increased risk of cancer, cataracts, hemorrhagic stroke, diabetes, neurocognitive events, muscle events, or non-cardiovascular death.

This matters because it directly tests the "lower is too low" hypothesis. If pushing LDL to extreme lows caused subtle harm, you'd expect to see it cluster in the patients who got lowest. It didn't.

How the Safety Profile Holds Up in Specific Groups

Three subgroup analyses are worth flagging because they cover situations where you might reasonably expect more side effects.

• Older patients (≥75 years). Annualized safety event rates were similar between evolocumab and placebo over a median 7.1 years of total follow-up, with no major safety concerns in this group.
• Patients with chronic kidney disease. Across CKD stages including stage 3 or worse, adverse events including kidney function decline were infrequent and similar across stages. eGFR did not deteriorate faster on evolocumab.
• Patients with familial hypercholesterolemia. Long-term studies of evolocumab in patients with familial hypercholesterolemia and broader hypercholesterolemia found a safety profile consistent with the FOURIER trials, even though these patients tend to have higher cumulative drug exposure starting at younger ages.

The Methodological Critique Worth Knowing About

One paper has pushed back against the prevailing "Repatha is safe long-term" narrative on methodological grounds. The argument: FOURIER was terminated early at a median of 2.2 years instead of the planned 3.6, which may have reduced statistical power to detect long-term mortality differences or rare safety signals. The authors note a numerically higher all-cause mortality in the evolocumab group after 2 years (4.8% vs. 4.3%) and wide confidence intervals for several safety endpoints.

This critique deserves to be taken seriously, but it's also worth context. FOURIER-OLE's open-label follow-up has now produced 5+ additional years of safety data on top of the original 2.2 years, with no excess mortality emerging and no safety signals climbing.

The methodological concern was about whether short blinded follow-up could miss something. Eight years of unblinded follow-up has now answered most of what the critique asked.

The honest position is that we have very strong long-term safety data through about 8 years, and beyond that, we're extrapolating. For a drug that's been on the market since 2015, that's a more complete safety dataset than most cardiovascular medications can claim.

What This Means If You're Taking Repatha Long-Term

The clinical translation is fairly direct.

• You're unlikely to develop diabetes from Repatha alone. If you have prediabetes, monitor more carefully in the first year.
• You're unlikely to lose cognitive function. EBBINGHAUS tested this with formal neuropsychological batteries and found nothing, even at very low LDL.
• Very low LDL appears to be safe. Pushing your LDL below 40 or even below 20 mg/dL has not shown harm in 8 years of follow-up.
• Injection site reactions are the most common nuisance. They're typically mild and local.
• The 10+ year question is genuinely open. We have excellent data through year 8. Patients starting at age 50 will eventually answer the year-15 and year-20 questions.

If you're considering whether Repatha is right for you, the conversation worth having is about cardiovascular benefit relative to your individual risk, not about whether the drug is safe long-term. The safety data, including in older patients, in CKD, and at very low LDL, is among the strongest we have for any non-statin lipid-lowering therapy.

Instalab's Repatha Program pairs you with a licensed physician who prescribes Repatha, monitors your lipid panel and safety markers over time, and adjusts your treatment based on your numbers and how you're tolerating the medication.