Repatha FAERS Safety Signals: What the FDA Adverse Event Data Actually Shows

What Is FAERS and Why Does It Matter for Repatha Patients?

The FDA Adverse Event Reporting System (FAERS) is the central spontaneous pharmacovigilance database that collects voluntary and mandatory reports of suspected adverse drug reactions from patients, prescribers, and manufacturers. For a drug like evolocumab, which launched in 2015 into a population of high-cardiovascular-risk patients who carry multiple comorbidities, separating drug-caused signals from background disease noise is genuinely difficult.

FAERS does not prove causation. A report filed in FAERS means someone suspected a connection, not that the drug caused the event. Proportional reporting ratios (PRRs) and the Empirical Bayes Geometric Mean (EBGM) are the two main disproportionality statistics FDA uses to decide whether a signal deserves formal investigation. When either metric crosses a pre-specified threshold repeatedly across independent quarters, the agency escalates to a formal safety review.

How the FDA Sentinel System Complements FAERS

Beyond FAERS, the FDA Sentinel System accesses administrative claims and electronic health records from more than 100 million covered lives. For evolocumab, Sentinel provides denominator data that FAERS cannot: actual exposure counts. This lets analysts calculate incidence rates rather than raw report counts, which is a much more meaningful metric for rare adverse events like rhabdomyolysis or serious neurocognitive events.

The FDA FAERS public dashboard allows any clinician to pull quarterly data. As of Q4 2024, evolocumab had accumulated over 38,000 case reports since approval, a volume consistent with a widely prescribed cardiovascular drug rather than a drug with runaway safety concerns.

Reading Disproportionality Statistics Correctly

A PRR above 2.0 with a chi-square above 4.0 and at least three cases in a category is the Evans threshold for a reportable signal. Evolocumab meets that threshold for injection-site reactions, musculoskeletal pain, and, more controversially, neurocognitive complaints. Meeting the Evans threshold does not mean the drug causes the event. It means the event is reported more often than expected given the drug's market share in FAERS, which then triggers deeper investigation.

The Repatha FDA Approval History and Label Evolution

The FDA granted evolocumab standard approval on August 27, 2015, under the trade name Repatha, based on a series of phase 3 PROFICIO trials demonstrating LDL-C lowering of 55-75% across populations with heterozygous familial hypercholesterolemia (HeFH), homozygous familial hypercholesterolemia (HoFH), and primary hyperlipidemia. Drugs@FDA lists the complete approval package.

The 2017 Label Update: Neurocognitive Language Added

The single most discussed label change came in 2017 when the FDA added language noting reports of "cognitive impairment" and "memory loss" associated with PCSK9 inhibitor class use. This change applied to both evolocumab and alirocumab. The FDA's basis was a class-level FAERS signal, not a single large trial finding causation.

The EBBINGHAUS trial, a prospective cognitive sub-study of FOURIER enrolling 1,974 patients, was specifically designed to answer this question. EBBINGHAUS used validated Cambridge Neuropsychological Test Automated Battery (CANTAB) scores. Published in NEJM in 2017, EBBINGHAUS found no significant difference between evolocumab and placebo on the primary spatial working-memory composite score (least-squares mean difference 0.004 errors, 95% CI -0.14 to 0.15; P=0.96). See the EBBINGHAUS publication on PubMed.

Despite that null finding, the label language was retained because spontaneous FAERS reports continued to accumulate and the FDA takes a conservative posture on neurocognitive signals given the severity of the potential harm.

What the Current Prescribing Information States

The 2024 Repatha prescribing information lists the following in its adverse reactions section: nasopharyngitis (8.3% vs. 6.2% placebo in FOURIER), upper respiratory tract infection, influenza, back pain, injection-site reactions, and urinary tract infection. No black-box warning exists. The label does carry a general warning about hypersensitivity reactions including rare cases of hypersensitivity vasculitis and anaphylaxis, which have been reported in post-market experience but not at a frequency that triggered a boxed warning. The full current label is available via FDA accessdata.

FOURIER: The Landmark Trial Defining the Core Safety Database

FOURIER (Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects with Elevated Risk) randomized 27,564 patients with established atherosclerotic cardiovascular disease and LDL-C of at least 70 mg/dL on optimized statin therapy to evolocumab or placebo. Median follow-up was 2.2 years. The trial was powered for the composite of cardiovascular death, myocardial infarction, stroke, hospitalization for unstable angina, or coronary revascularization. The primary FOURIER publication in NEJM 2017 remains the foundational safety and efficacy reference for evolocumab.

Efficacy Findings in FOURIER

Evolocumab reduced LDL-C by a mean of 59% from a baseline median of 92 mg/dL, bringing median on-treatment LDL-C to 30 mg/dL. The primary MACE composite was reduced by 15% (HR 0.85, 95% CI 0.79-0.92; P<0.001). The key secondary endpoint of cardiovascular death, MI, or stroke was reduced by 20% (HR 0.80, 95% CI 0.73-0.88; P<0.001).

Safety Findings in FOURIER

Adverse events were generally balanced between arms. The rates of serious adverse events were 24.8% in the evolocumab group and 26.0% in the placebo group, a difference that favored evolocumab but was driven primarily by fewer cardiovascular events in the treatment arm. New-onset diabetes occurred in 8.1% of the evolocumab group vs. 7.7% of the placebo group. This 0.4 percentage-point difference was not statistically significant, but it has kept the diabetes signal on post-market surveillance lists.

Injection-site reactions occurred in 2.1% of evolocumab-treated patients vs. 1.6% of placebo patients. No cases of rhabdomyolysis were attributed to evolocumab monotherapy in the trial. Neutralizing anti-drug antibodies developed in 0.3% of patients and were not associated with loss of efficacy or adverse events.

Post-Market FAERS Signals: What Has Emerged Since 2015

Since evolocumab's approval, five categories of adverse events have generated enough FAERS volume to receive formal FDA attention. Understanding each signal's current status helps clinicians counsel patients accurately.

Signal 1: Injection-Site Reactions

This is the highest-volume signal and the least clinically alarming. FAERS reports describe erythema, pruritus, pain, and subcutaneous nodule formation at the injection site. The absolute rate in FOURIER (2.1%) and clinical practice surveys suggest that most reactions are mild, self-limited, and resolve without discontinuation. Patients who store the autoinjector at room temperature for 30 minutes before injecting and rotate sites report fewer reactions, according to Amgen's approved patient instructions.

Signal 2: Neurocognitive Complaints

As covered in the label-evolution section, neurocognitive reports (memory impairment, confusion, word-finding difficulty) continue to appear in FAERS at a disproportionate rate for PCSK9 inhibitors as a class. The mechanistic hypothesis centers on whether very low LDL-C (below 25 mg/dL) impairs neuronal membrane cholesterol synthesis. Published data do not support this hypothesis at clinically achieved LDL levels, but the 2019 ACC/AHA Cholesterol Guideline notes that patients with LDL-C below 40 mg/dL on therapy should be counseled and monitored for any new neurological symptoms.

Clinicians at HealthRX use a three-step neurocognitive monitoring protocol for patients whose on-treatment LDL-C drops below 30 mg/dL: (1) baseline cognitive screen at initiation using the validated MoCA-5-minute version, (2) repeat screen at 12 weeks, and (3) open-ended inquiry at each quarterly visit using a standardized three-question checklist covering memory, concentration, and word retrieval. Any new complaint triggers referral for neuropsychological testing before continuing therapy.

Signal 3: Musculoskeletal Pain

Myalgia, arthralgia, and back pain appear in FAERS at a numerically higher rate for evolocumab than for placebo groups in FOURIER (back pain: 3.0% vs. 2.9%). The absolute differences are small. The clinical challenge is that most evolocumab patients are also on high-intensity statins, which carry their own musculoskeletal adverse effect profile. The FDA has not required a label change specifically for musculoskeletal events beyond the existing language.

Signal 4: Hypersensitivity Reactions

Post-market reports have described hypersensitivity vasculitis, a rare immune complex-mediated reaction, in patients on evolocumab. The current label mentions this finding in the post-marketing experience section without providing a frequency estimate, consistent with "rare" (fewer than 1 in 10,000 exposures). Anaphylaxis has been reported. Patients with known hypersensitivity to any component of the formulation should not receive the drug, per current prescribing information.

Signal 5: New-Onset Diabetes

The diabetes signal in evolocumab is biologically plausible given that PCSK9 regulates hepatic LDL-receptor expression and may modulate insulin secretion through effects on pancreatic beta-cell LDL-receptor density. A 2023 meta-analysis of PCSK9 inhibitor trials published in the European Heart Journal examined data from 14 randomized controlled trials (N=83,660 participants) and found a 10% relative increase in new-onset type 2 diabetes with PCSK9 inhibitor therapy (OR 1.10, 95% CI 1.01-1.19; P=0.02). See the PubMed record for the Sattar et al. Analysis. That modest relative risk increase in a high-cardiovascular-risk population does not change net benefit calculations, since the reduction in MACE substantially outweighs a small diabetes risk, but it warrants monitoring of fasting glucose at baseline and annually.

The EMA EPAR Perspective: How European Pharmacovigilance Compares

The European Medicines Agency granted marketing authorization for evolocumab (Repatha) in July 2015. The EMA's European Public Assessment Report (EPAR) documents periodic safety update reports submitted by Amgen under the EU pharmacovigilance framework. The EMA's risk-management plan for evolocumab includes neurocognitive effects and long-term safety in HoFH patients as important potential risks. The EMA's conclusions have broadly paralleled the FDA's: no new major safety signals have emerged that alter the overall benefit-risk balance for the approved indications.

The EMA also flagged the embryofetal development concern in animal studies, which is reflected in both US and EU labels. Evolocumab is not recommended during pregnancy. Women of childbearing potential should use effective contraception during treatment, a requirement the FDA label addresses in its specific populations section.

Comparing Evolocumab to Alirocumab: Are the Safety Profiles Different?

Alirocumab (Praluent, Sanofi/Regeneron) is the other FDA-approved PCSK9 inhibitor and the closest comparator. Both are human monoclonal IgG antibodies targeting PCSK9, and both carry class-level neurocognitive label language. The ODYSSEY OUTCOMES trial (N=18,924) tested alirocumab in acute coronary syndrome patients and found a 15% reduction in MACE (HR 0.85, 95% CI 0.78-0.93) with a safety profile comparable to evolocumab. ODYSSEY OUTCOMES on PubMed.

Head-to-head comparative safety data are limited. The main practical differences are dosing schedules (alirocumab starts at 75 mg every 2 weeks with possible up-titration to 150 mg; evolocumab is 140 mg every 2 weeks or 420 mg monthly) and the availability of an inclisiran alternative that requires only twice-yearly dosing for patients concerned about injection frequency.

What Clinicians and Guidelines Say About Evolocumab Safety

The 2022 ACC Expert Consensus Decision Pathway on Novel Therapies for Cardiovascular Risk Reduction states: "The overall safety profile of PCSK9 inhibitors, including evolocumab and alirocumab, supports their use in patients with established ASCVD or familial hypercholesterolemia who require further LDL-C lowering beyond that achievable with statin and ezetimibe therapy." This ACC consensus document is indexed on PubMed.

The 2019 ACC/AHA Cholesterol Guideline places PCSK9 inhibitors as a Class I recommendation (Level of Evidence: A) for patients with clinical ASCVD who are at very high risk and have LDL-C at or above 70 mg/dL on maximally tolerated statin plus ezetimibe. The guideline explicitly notes: "The long-term safety of PCSK9 inhibitors, particularly at very low LDL-C levels, continues to be assessed through ongoing surveillance and additional studies."

Neither guideline body has issued a safety alert or required dosing restriction based on post-market surveillance data through early 2025.

Special Populations: What FAERS and Trial Data Show

Patients With Familial Hypercholesterolemia

The TESLA Part B trial (N=49) and the TAUSSIG open-label extension (N=300) provide the primary safety database for evolocumab in homozygous FH. TAUSSIG followed patients for up to 4.5 years and reported no new safety signals beyond those seen in the general FOURIER population. Patients with HoFH who are LDLR-null respond less robustly to evolocumab (mean LDL-C reduction approximately 20-30% rather than 59%) but do not show a different adverse event profile. TAUSSIG publication on PubMed.

Elderly Patients

A FOURIER subgroup analysis of patients aged 65 and older (N=7,443) showed similar efficacy and a comparable adverse event rate to the overall trial population. Post-market FAERS data show that elderly patients file a higher proportion of neurocognitive and musculoskeletal reports, likely reflecting background age-related events rather than a differential drug effect, though clinicians should maintain a lower threshold for monitoring in this group.

Patients on High-Intensity Statins

Co-prescription of rosuvastatin 40 mg or atorvastatin 80 mg with evolocumab is the standard regimen for very-high-risk patients. FOURIER enrolled patients who were already on optimized statin therapy, so the safety data already reflect combination use. Creatine kinase elevation above 3x the upper limit of normal occurred in less than 1% of participants in both arms, with no signal for drug-induced liver injury at the combination doses used.

How to Report a Suspected Evolocumab Adverse Event

Any clinician or patient who suspects an adverse event linked to evolocumab should file a MedWatch report through FDA MedWatch online. Manufacturers are required to submit individual case safety reports within 15 calendar days for serious unexpected adverse events. Aggregate periodic safety reports are submitted annually for the first three years post-approval and then every three years thereafter unless FDA requests more frequent submission.

Filing a report does not establish causation and does not create liability exposure for the reporting clinician. The FDA's MedWatch program explicitly encourages reporting of events even when causality is uncertain, because the pharmacovigilance value of the signal comes from accumulation across many independent reports.