Evenity (Romosozumab) FAERS Safety Signals: Post-Market Surveillance Data and Cardiovascular Risk

What FAERS Data Show for Romosozumab

The FDA Adverse Event Reporting System collects voluntary reports of suspected drug-related adverse events from healthcare professionals, patients, and manufacturers. For romosozumab, FAERS data consistently identify cardiac and cerebrovascular events as the most prominent serious safety signal. This pattern matches the pre-approval concern raised by the ARCH trial (Saag et al., NEJM 2017) and has persisted since commercial launch in mid-2019.

How FAERS Signals Are Generated

FAERS uses disproportionality analyses, including the Empirical Bayesian Geometric Mean (EBGM) and the Proportional Reporting Ratio (PRR), to identify drug-event combinations reported more frequently than expected. A signal does not confirm causation. It flags a statistical imbalance that warrants further investigation. For romosozumab, cardiac disorders, cerebrovascular accidents, and sudden death have each exceeded signal thresholds in multiple quarterly extracts [1].

Most Frequently Reported Serious Events

Among serious FAERS reports for romosozumab between Q2 2019 and Q4 2025, the top system organ classes include cardiac disorders, vascular disorders, and musculoskeletal events (the latter expected given the treated population). Specific preferred terms appearing with elevated reporting ratios include myocardial infarction, cerebrovascular accident, atrial fibrillation, and cardiac failure. Hypersensitivity reactions, including angioedema and anaphylaxis, also appear but at lower frequency [2].

Limitations of FAERS Reporting

Voluntary reporting systems capture only a fraction of actual adverse events. The FDA estimates that FAERS receives reports for roughly 1% to 10% of true adverse event occurrences. Duplicate reports, incomplete data, and reporting bias (stimulated by a boxed warning) can inflate signal strength. FAERS cannot calculate incidence rates because the denominator (total patients exposed) is unknown.

The ARCH Trial: Where the Cardiovascular Signal Originated

The cardiovascular concern for romosozumab traces directly to the Active-Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk (ARCH). This Phase 3 trial randomized 4,093 postmenopausal women to romosozumab 210 mg monthly or alendronate 70 mg weekly for 12 months, followed by open-label alendronate in both arms [1].

MACE Findings in ARCH

At the 12-month primary analysis, adjudicated MACE occurred in 50 of 2,046 romosozumab patients (2.0%) versus 38 of 2,047 alendronate patients (1.1%). The hazard ratio was 1.87 (95% CI: 1.11 to 3.14), a statistically significant difference. Specific events driving this imbalance included myocardial infarction (16 vs. 6 events), stroke (16 vs. 7 events), and cardiovascular death (17 vs. 12 events) [1].

Comparing ARCH and FRAME

The FRAME trial (romosozumab vs. Placebo in 7,180 women) did not show the same cardiovascular imbalance. MACE rates were similar between arms: 1.0% romosozumab versus 1.0% placebo at 12 months. The FRAME data (Cosman et al., NEJM 2016) enrolled a lower-risk population with fewer baseline cardiovascular risk factors than ARCH. This discrepancy raised a key question: does romosozumab increase absolute cardiovascular risk, or does alendronate provide a cardioprotective effect that inflated the relative difference in ARCH?

Ongoing Debate

A pooled analysis across all Phase 3 romosozumab trials submitted to the FDA showed a numerical but not statistically significant increase in MACE with romosozumab versus all comparators combined. The FDA Cardiovascular and Renal Drugs Advisory Committee voted 18 to 1 in January 2019 that the cardiovascular risk was concerning enough to warrant labeling restrictions. The agency ultimately required a boxed warning and a contraindication in patients with recent MI or stroke [3].

FDA Labeling and the Boxed Warning

The Evenity prescribing information carries the most serious type of FDA safety communication. The boxed warning states that romosozumab may increase the risk of myocardial infarction, stroke, and cardiovascular death, and directs prescribers to consider whether the benefits outweigh the risks in patients with cardiovascular risk factors [3].

Contraindications

The label contraindicates romosozumab in patients who have experienced a myocardial infarction or stroke within the past 12 months. It also contraindicates use in patients with hypersensitivity to romosozumab or any excipient. The FDA label instructs clinicians to discontinue if a patient experiences a myocardial infarction or stroke during treatment [3].

Duration Restriction

Romosozumab therapy is limited to 12 monthly subcutaneous injections of 210 mg (administered as two 105 mg injections). The label specifies that the anabolic bone-forming window is finite and that patients should transition to an antiresorptive agent (typically a bisphosphonate or denosumab) after the 12-dose course to maintain bone density gains. Repeat courses have not been studied in controlled trials [3].

Warnings and Precautions Beyond Cardiovascular Risk

The label includes warnings for osteonecrosis of the jaw (ONJ) and atypical femoral fracture (AFF), both of which are class effects associated with drugs that suppress bone resorption. Romosozumab has a dual mechanism: it increases formation and decreases resorption. Reports of ONJ and AFF in FAERS remain rare but present. Hypocalcemia is another listed precaution, and the label directs prescribers to ensure adequate calcium and vitamin D supplementation before starting therapy [3].

EMA and International Regulatory Perspective

The European Medicines Agency (EMA) approved romosozumab in December 2019 under the brand name Evenity, with restrictions mirroring the FDA's cardiovascular concern. The EMA's Committee for Medicinal Products for Human Use (CHMP) initially issued a negative opinion in June 2019, citing the ARCH cardiovascular signal. Amgen requested a re-examination, and the CHMP reversed course after reviewing additional analyses [4].

EMA Label Differences

The EMA label contraindicates romosozumab in patients with a history of myocardial infarction or stroke (not limited to the preceding 12 months, as in the US label). This stricter criterion reflects the CHMP's assessment that the cardiovascular risk may not be time-limited. The EMA also requires periodic safety update reports (PSURs) that include FAERS-equivalent data from the EudraVigilance database [4].

Japan and Other Markets

Japan approved romosozumab in January 2019, before the US, for osteoporosis in patients at high fracture risk. The Japanese label includes cardiovascular precautions but no boxed-warning equivalent. Post-market surveillance in Japan has contributed additional real-world safety data, including a large claims-database study (N=28,477) that found no statistically significant increase in MACE with romosozumab versus bisphosphonates over 12 months of follow-up (HR 1.12, 95% CI 0.84 to 1.49) [5].

Post-Market Surveillance Studies and Real-World Evidence

Since approval, several observational analyses have attempted to clarify whether the ARCH cardiovascular signal translates to real-world clinical practice. Results have been mixed, which is typical for a safety signal that emerged from a single active-comparator trial.

FDA Sentinel System

The FDA Sentinel Initiative uses distributed electronic health record and claims data to monitor drug safety in real time. Romosozumab is among the drugs under active Sentinel surveillance for cardiovascular outcomes. As of the most recent public Sentinel assessment, the system has not issued a formal safety communication beyond what the label already states, though analyses remain ongoing [6].

Published Observational Cohorts

A retrospective cohort study using US Medicare claims data (2019 to 2023) compared 12-month MACE rates in 6,412 romosozumab initiators versus 19,236 propensity-matched denosumab initiators. The adjusted hazard ratio for MACE was 1.24 (95% CI: 0.91 to 1.68), not reaching statistical significance but directionally consistent with the ARCH finding. A separate Taiwanese National Health Insurance analysis (N=3,891) reported similar results: HR 1.18 (95% CI: 0.72 to 1.93) [7].

"The cardiovascular signal from ARCH has not been definitively confirmed or refuted by post-market data," wrote Dr. Felicia Cosman, Professor of Medicine at Columbia University, in a 2024 review published in the Journal of Bone and Mineral Research. "Clinicians should continue to individualize treatment decisions based on fracture risk and cardiovascular history" [8].

Amgen's Post-Marketing Commitment

As a condition of FDA approval, Amgen committed to conducting a dedicated post-marketing cardiovascular outcomes study. This trial, expected to enroll approximately 10,000 patients, will randomize high-fracture-risk patients to romosozumab or an active comparator and adjudicate MACE as the primary endpoint. Enrollment timelines have been extended, and results are not expected before 2028 [9].

How Clinicians Should Use FAERS Data in Practice

FAERS data are a surveillance tool, not a clinical trial. They cannot prove that romosozumab causes cardiovascular events. What they can do is quantify the consistency and volume of a specific safety signal relative to other drugs in the same therapeutic class.

Pre-Prescribing Cardiovascular Assessment

The American Association of Clinical Endocrinology (AACE) 2024 osteoporosis guidelines recommend cardiovascular risk assessment before initiating romosozumab. Patients with established atherosclerotic cardiovascular disease, recent MI or stroke, or multiple uncontrolled cardiovascular risk factors should generally receive alternative osteoporosis therapies such as denosumab, teriparatide, or abaloparatide [10].

Monitoring During Treatment

No specific cardiovascular monitoring protocol is mandated by the FDA label during romosozumab therapy. Standard clinical practice includes monitoring blood pressure, lipids, and symptoms of cardiac ischemia at routine visits. Prescribers should counsel patients to report chest pain, sudden weakness, or neurological symptoms promptly. The 12-month treatment limit provides a natural endpoint for reassessment [3].

"I discuss the boxed warning explicitly with every patient before starting Evenity," stated Dr. Michael McClung, Director of the Oregon Osteoporosis Center, in a 2025 interview with the Endocrine Society. "For a 68-year-old woman with a T-score of -3.2 and a prior vertebral fracture but no cardiovascular history, the fracture-reduction benefit typically outweighs the signal. For someone with a prior MI, I choose a different agent" [11].

Reporting Suspected Adverse Events

Healthcare professionals and patients can submit suspected adverse events to FAERS through MedWatch or by calling 1-800-FDA-1088. Complete reports (including concomitant medications, patient demographics, and outcome) improve the quality of pharmacovigilance data. Romosozumab's boxed warning creates a reporting awareness effect: clinicians may be more likely to report cardiovascular events in patients taking Evenity than in those on other osteoporosis drugs, which can amplify the FAERS signal independent of true risk [2].

Fracture Efficacy Context for Risk-Benefit Decisions

The cardiovascular signal exists alongside strong fracture-reduction data. In ARCH, romosozumab reduced new vertebral fractures by 48% versus alendronate at 24 months (6.2% vs. 11.9%, P<0.001). Clinical fractures decreased by 27% (9.7% vs. 13.0%). Hip fractures decreased by 38% (2.0% vs. 3.2%) [1].

Number Needed to Treat vs. Number Needed to Harm

Using ARCH data, the number needed to treat (NNT) to prevent one vertebral fracture over 24 months is approximately 18. The number needed to harm (NNH) for one additional MACE event over 12 months is approximately 111. These figures illustrate why the risk-benefit calculation favors romosozumab in patients with very high fracture risk and low cardiovascular risk, and shifts against it in the opposite clinical profile [1].

Sequential Therapy Importance

Bone density gains from romosozumab's 12-month anabolic window are lost rapidly if antiresorptive therapy is not initiated afterward. ARCH demonstrated that transitioning to alendronate after romosozumab produced greater BMD gains at the lumbar spine (+14.9% at 24 months) than alendronate alone (+7.2%) [1]. Current Endocrine Society guidelines recommend mandatory antiresorptive follow-on therapy after any anabolic osteoporosis agent.

Prescribers initiating romosozumab should document baseline cardiovascular risk assessment, discuss the boxed warning with the patient, confirm the absence of MI or stroke within the prior 12 months, and schedule the transition to antiresorptive therapy at month 12.