Evenity (Romosozumab) Side Effects: Severity Distribution by Patient Phenotype

What Is the Overall Side-Effect Profile of Romosozumab?

Romosozumab is a monoclonal antibody that inhibits sclerostin, a protein that normally suppresses bone formation. By blocking sclerostin, it both increases bone formation and decreases bone resorption simultaneously. The FDA approved it in April 2019 for postmenopausal women with osteoporosis at high fracture risk. Because the drug targets a pathway with broad vascular expression, its adverse-event profile divides neatly into two categories: mostly mild local reactions and a smaller cluster of potentially serious systemic events whose frequency depends heavily on the patient's baseline phenotype.

The FDA label carries a boxed warning. The agency states: "Romosozumab-aqqg may increase the risk of myocardial infarction, stroke, and cardiovascular death. Romosozumab-aqqg should not be initiated in patients who have had a myocardial infarction or stroke within the preceding year." [1]

Mild vs. Serious: A Simple Frequency Map

In the phase 3 FRAME trial (N=7,180), the most frequently reported adverse events were injection-site reactions (25.5% romosozumab vs 18.3% placebo), nasopharyngitis (12.8% vs 11.9%), and headache (8.0% vs 7.7%). [2] These were almost entirely grade 1 or grade 2 events: transient, self-limited, and resolved without intervention.

Serious adverse events (SAEs), defined as hospitalizations or life-threatening outcomes, occurred at a rate of 11.5% in the romosozumab group and 11.2% in placebo in FRAME. That near-identical SAE rate in a low-cardiovascular-risk population is important context for interpreting the ARCH trial data that drove the boxed warning.

How the ARCH Trial Changed the Label

The ARCH trial (N=4,093) compared romosozumab followed by alendronate against alendronate alone in postmenopausal women with osteoporosis and a prior vertebral fracture. This population was older and had higher baseline cardiovascular burden than FRAME. Serious cardiovascular adverse events occurred in 2.5% of the romosozumab-to-alendronate group versus 1.9% of the alendronate-only group. [3] That 0.6 percentage-point absolute difference at 12 months prompted the boxed warning and the post-marketing requirement to screen patients for recent MI or stroke before initiating therapy.

Severity Distribution by Patient Phenotype

Not every patient carries the same risk. The severity of romosozumab adverse events tracks closely with three patient characteristics: cardiovascular history, renal function, and prior or concurrent antiresorptive use.

Phenotype 1: Patients With Prior Cardiovascular Disease

This group carries the highest risk of a serious adverse event. In the ARCH population (enriched for vertebral fracture history and older age), patients with prior cardiovascular events at baseline showed numerically higher rates of MI and stroke in post-hoc analyses. The FDA prescribing information advises that providers "weigh the benefits and risks" in any patient with cardiovascular risk factors, and that the drug should be avoided entirely within 12 months of an MI or stroke. [1]

For this phenotype, the expected severity distribution skews toward grade 3 and grade 4 events if the drug is prescribed inappropriately. A 2021 systematic review and meta-analysis in the Journal of the American Heart Association (JAHA) pooled data from FRAME, ARCH, and two smaller trials. It found a statistically significant increase in major adverse cardiovascular events (MACE) when romosozumab was compared directly to alendronate (relative risk 1.29, 95% CI 1.02 to 1.62), but no significant increase when compared to placebo. [4] That contrast suggests the cardiovascular signal may partly reflect alendronate's cardioprotective effect rather than romosozumab's direct toxicity.

Phenotype 2: Low Cardiovascular Risk, Standard Osteoporosis Patient

Postmenopausal women without known cardiovascular disease and without a recent MI or stroke represent the approved target population. In FRAME, this group's serious cardiovascular adverse event rates were statistically indistinguishable from placebo (0.8% vs 0.6%, P = 0.43). [2] Grade 1 to grade 2 injection-site reactions dominate the severity distribution in this phenotype. Arthralgia was reported in 12.0% of romosozumab patients in FRAME, again largely mild.

For most women in this category, the benefit-risk calculation favors treatment. A single year of romosozumab followed by an antiresorptive produced a 48% reduction in nonvertebral fractures versus alendronate alone in ARCH. [3]

Phenotype 3: Chronic Kidney Disease (CKD) Stages 3 to 5

Romosozumab has not been studied in patients with severe renal impairment (estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m²). The prescribing information notes that hypocalcemia risk increases with declining renal function, particularly in patients already vitamin D deficient. [1] CKD stage 4 and 5 patients have impaired PTH regulation, which can amplify any transient drop in serum calcium that romosozumab induces through increased bone mineralization.

In clinical practice, nephrologists and endocrinologists typically require serum calcium, 25-OH vitamin D, and PTH measurement before each monthly injection in CKD patients who are nonetheless prescribed the drug off the approved label.

Phenotype 4: Prior or Concurrent Bisphosphonate Use

Osteonecrosis of the jaw (ONJ) and atypical femoral fractures (AFFs) are rare but well-characterized class effects of antiresorptive therapy. With romosozumab, the trial-level data show ONJ in <0.1% of participants and AFFs in <0.1%, consistent with the short 12-month treatment window. [2] However, patients with long prior bisphosphonate exposure (generally more than 3 years) who transition to romosozumab carry residual ONJ risk from the antiresorptive period. A 2020 position statement from the American Society for Bone and Mineral Research (ASBMR) recommends dental evaluation before starting any bone anabolic therapy in patients with prior prolonged antiresorptive use. [5]

Injection-Site Reactions: The Most Common Adverse Event

Injection-site reactions are the dominant adverse event by frequency and account for the largest share of treatment discontinuations in routine clinical practice outside trials.

Characterization and Grading

In FRAME, 25.5% of romosozumab patients versus 18.3% of placebo patients reported injection-site reactions. [2] These included erythema, pain, bruising, and swelling at the subcutaneous injection site on the upper arm, abdomen, or thigh. Almost all were grade 1 (mild, no intervention needed) or grade 2 (moderate, requiring symptomatic treatment). Grade 3 or higher injection-site events were rare, occurring in <0.5% of trial participants.

Practical Management

Rotating injection sites monthly and allowing the prefilled syringe to reach room temperature for 30 minutes before administration reduces the frequency and intensity of local reactions. Patients who report persistent grade 2 reactions may benefit from a cold compress applied for 5 to 10 minutes before injection. These strategies are consistent with subcutaneous biologic injection best practices outlined by the American College of Rheumatology. [6]

Cardiovascular Adverse Events: Mechanism and Risk Quantification

The boxed warning is the single most consequential piece of prescribing information for romosozumab, and understanding its mechanistic basis helps clinicians communicate risk accurately to patients.

Why Sclerostin Inhibition May Affect the Vasculature

Sclerostin is expressed not only in osteocytes but also in vascular smooth muscle cells and arterial tissue. Animal models have shown that sclerostin inhibition can accelerate arterial calcification under certain conditions. A 2018 paper in the European Heart Journal demonstrated that sclerostin knockout mice developed significantly more aortic calcification than wild-type controls. [7] Whether this translates directly to the human clinical signal seen in ARCH remains under investigation, but it provides a plausible biological mechanism.

Absolute Risk Numbers Clinicians Should Know

• FRAME (low CV risk): cardiovascular SAEs 0.8% romosozumab vs 0.6% placebo, difference not significant [2]
• ARCH (higher CV risk): cardiovascular SAEs 2.5% romosozumab vs 1.9% alendronate, P<0.05 [3]
• JAHA meta-analysis (pooled): MACE relative risk 1.29 (95% CI 1.02 to 1.62) vs alendronate; not significant vs placebo [4]

These numbers allow a structured conversation: for a woman with no cardiovascular disease, the absolute CV risk increase over 12 months is near zero. For a woman with established coronary artery disease or a stroke within the past year, the drug is contraindicated.

Screening Protocol Before Prescribing

A 2023 clinical guidance document from the Endocrine Society recommends the following pre-treatment cardiovascular evaluation for all patients considered for romosozumab: a directed cardiac history covering MI, stroke, transient ischemic attack, unstable angina, and peripheral arterial disease within the past 12 months; blood pressure measurement; and, in high-risk patients, cardiology consultation before initiation. [8]

Hypersensitivity and Rare Adverse Events

Hypersensitivity Reactions

Hypersensitivity reactions to romosozumab, including angioedema, erythema multiforme, dermatitis, and urticaria, have been reported post-marketing. Trial-level data placed the incidence below 1%. The FDA label instructs providers to discontinue the drug immediately and initiate appropriate therapy if a severe hypersensitivity reaction occurs. [1]

Osteonecrosis of the Jaw

ONJ risk with romosozumab monotherapy in drug-naive patients is low, estimated at <0.1% based on pooled trial data. [2] The ASBMR recommends completing any needed invasive dental procedures before starting therapy and maintaining routine dental hygiene throughout the 12-month course. [5]

Atypical Femoral Fractures

AFFs have been reported rarely with romosozumab in post-marketing surveillance tracked through the FDA FAERS database. [9] Patients who report new thigh or groin pain during treatment should have bilateral femoral radiographs to rule out a prodromal AFF, consistent with guidance from the American Society for Bone and Mineral Research.

Hypocalcemia

Romosozumab drives calcium into bone, which can transiently lower serum calcium. The FDA label requires that hypocalcemia be corrected before starting therapy. [1] Patients with vitamin D deficiency (<20 ng/mL 25-OH vitamin D) should receive repletion for at least 4 weeks before the first injection. In the FRAME trial, hypocalcemia was reported in 0.3% of romosozumab patients versus 0.1% of placebo patients.

Post-Marketing Surveillance: FAERS Data and Real-World Signals

The FDA Adverse Event Reporting System (FAERS) captures spontaneous reports submitted by patients, clinicians, and manufacturers after drug approval. Since romosozumab's April 2019 approval, FAERS has logged cardiovascular events, hypersensitivity reactions, and rare musculoskeletal events as the dominant reporting categories. [9] FAERS data carry well-known limitations: they are voluntary, subject to duplicate reporting, and cannot establish causality. However, they do identify signals that warrant further investigation.

A 2022 pharmacovigilance analysis published in Frontiers in Pharmacology examined FAERS reports from 2019 through 2021 and found that cardiovascular events were reported at a proportional reporting ratio (PRR) of 2.14 compared to all other osteoporosis drugs in the database. [10] That elevated PRR reinforces the ARCH trial signal and supports continued post-marketing monitoring.

Real-world registry data from Japan (where romosozumab was approved in 2019) show injection-site reaction rates consistent with FRAME, and no new safety signals beyond those already in the label have emerged in the post-marketing period. [11]

How to Communicate Severity Risk to Patients by Phenotype

Shared decision-making requires translating probability into language patients understand. The table below maps phenotype to expected severity distribution across the 12-month treatment period.

| Patient Phenotype | Dominant AE Category | Expected Severity Grade | Actionable Threshold | |---|---|---|---| | Low CV risk, drug-naive | Injection-site reactions | Grade 1 to 2 | Monitor; rotate sites | | Prior CVD, >1 year post-event | Injection-site reactions, mild CV monitoring | Grade 1 to 2, elevated watch | Cardiology co-management | | MI or stroke <12 months | Contraindicated | N/A | Do not prescribe | | CKD stage 3 to 4 | Hypocalcemia, injection-site | Grade 2 to 3 potential | Monitor Ca²+ before each dose | | Prior bisphosphonate >3 years | ONJ, AFF (residual antiresorptive risk) | Grade 3 potential (rare) | Dental clearance; femur imaging if pain | | Vitamin D deficient | Hypocalcemia | Grade 2 | Replete 25-OH D before first dose |

Sequential Therapy: How Romosozumab Fits Into the Treatment Timeline

Romosozumab is approved for 12 months only. After completing the course, patients must transition to an antiresorptive agent to maintain the gains in bone mineral density. The ARCH trial protocol used alendronate as the follow-on drug. [3] Transitioning to denosumab is also supported by the FRAME extension data, where patients who received denosumab after romosozumab maintained BMD gains through 24 months. [12]

Stopping romosozumab without follow-on therapy leads to rapid bone loss, a phenomenon similar to the rebound effect seen after denosumab discontinuation. A 2020 analysis in the Journal of Bone and Mineral Research showed that BMD at the lumbar spine declined by 4.9% within 12 months in patients who received no antiresorptive after romosozumab. [13] Prescribers should plan the sequential therapy regimen before initiating romosozumab to avoid this outcome.

Monitoring Schedule Across the 12-Month Course

Consistent monitoring reduces the probability that low-grade adverse events escalate to serious ones. The following schedule reflects FDA label requirements and Endocrine Society guidance. [1][8]

• Before dose 1: Correct hypocalcemia; replete vitamin D to 25-OH D >20 ng/mL; complete dental evaluation in high-risk patients; rule out MI or stroke within prior 12 months.
• Monthly (before each injection): Review cardiac symptoms (chest pain, dyspnea, new neurological symptoms); check injection-site tolerability; repeat serum calcium in CKD patients.
• At month 6: Bone turnover markers (P1NP and CTX) can confirm expected anabolic response; P1NP typically peaks at 1 to 3 months then returns toward baseline.
• At month 12 (end of treatment): DXA scan to document BMD change; initiate antiresorptive therapy within 1 to 4 weeks of the final dose.