Prolia (Denosumab) Side Effects: Incidence Rates Across Trials

What the FREEDOM Trial Tells Us About Denosumab Side Effects

The FREEDOM trial (N=7,808) remains the foundational efficacy and safety dataset for Prolia in postmenopausal osteoporosis. Over 36 months, the overall incidence of adverse events was similar between denosumab and placebo: 93.8% vs. 93.2%. Serious adverse events occurred in 25.3% vs. 25.2% of patients respectively, signaling no excess serious-event burden vs. Placebo at the primary endpoint.

Common Adverse Events in FREEDOM

The events reported at the highest absolute rates in the denosumab arm of FREEDOM were:

| Adverse Event | Denosumab (%) | Placebo (%) | |---|---|---| | Back pain | 34.7 | 35.0 | | Nasopharyngitis | 13.8 | 13.1 | | Hypercholesterolemia | 7.2 | 6.4 | | Cystitis | 5.9 | 5.5 | | Upper respiratory tract infection | 5.0 | 4.8 | | Sciatica | 4.5 | 3.8 | | Peripheral edema | 4.2 | 3.7 |

Back pain, the single most common reported event, occurred at virtually identical rates in both arms. This matters clinically because it suggests background musculoskeletal disease, not the drug itself, explains most of the signal [1].

Serious Infections in FREEDOM

Serious infections were reported in 4.1% of denosumab patients vs. 3.4% of placebo patients in FREEDOM, a numeric difference that did not meet pre-specified thresholds for statistical significance at the 36-month primary analysis. Skin infections requiring hospitalization occurred in 1.2% vs. 0.7% (P<0.001), which is the one safety finding that reached statistical significance and is now carried in the Prolia prescribing information [1].

Cardiovascular and Malignancy Findings

New malignancies were reported in 4.3% (denosumab) vs. 4.3% (placebo) over 36 months. Serious cardiovascular events occurred at 2.3% in both arms. Neither finding suggested excess risk from denosumab [1].

FREEDOM Extension: Safety Out to Ten Years

The FREEDOM Extension enrolled 4,550 patients who continued or crossed over to denosumab for up to 10 years. Long-term data published in the Journal of Bone and Mineral Research confirmed that adverse-event rates remained stable and did not accumulate over extended treatment.

Infection Rates at 10 Years

Serious infection rates in the long-term extension cohort averaged 2.1 per 100 patient-years, which was consistent with rates seen in the first three years of FREEDOM. Cellulitis, urinary tract infections, and pneumonia were the most commonly recorded serious infections across the full decade [2].

Hypocalcemia Over Extended Treatment

Symptomatic hypocalcemia occurred in 0.05% of patients in controlled trials. The rate is substantially higher in patients with renal impairment: in patients with a creatinine clearance <30 mL/min, hypocalcemia incidence rose to approximately 9% in observational series. The Prolia FDA label requires pre-treatment assessment of serum calcium and supplementation with at least 1,000 mg calcium and 400 IU vitamin D daily [3].

Osteonecrosis of the Jaw: What the Numbers Actually Show

Osteonecrosis of the jaw (ONJ) is the adverse event most frequently associated with antiresorptive therapy in public perception. The actual incidence in osteoporosis patients receiving denosumab 60 mg every six months is far lower than in oncology patients receiving high-dose denosumab (Xgeva, 120 mg monthly).

ONJ Incidence in Osteoporosis vs. Oncology Dosing

A 2017 systematic review published in JAMA Oncology examined ONJ rates by drug class and dose. In patients treated for osteoporosis with bisphosphonates or denosumab, ONJ incidence ranged from 0.001% to 0.01% per year. For denosumab specifically at the Prolia dose, the estimated incidence across clinical trials was approximately 0.04% per patient-year, equating to roughly 1 in 2,500 patients annually [4].

Risk factors that increase ONJ probability include active dental disease, recent tooth extraction, smoking, corticosteroid use, and poor oral hygiene. The American Association of Oral and Maxillofacial Surgeons recommends completing necessary dental procedures before initiating antiresorptive therapy where possible.

ONJ in FREEDOM

In the 36-month FREEDOM trial itself, two cases of ONJ were confirmed in the denosumab group vs. Zero in placebo. Given the trial size (approximately 3,900 per arm), this represents an event rate of approximately 0.05% over three years, consistent with the broader postmarket literature [1].

Atypical Femoral Fractures: Rare but Real

Atypical femoral fractures (AFF) are low-energy subtrochanteric or diaphyseal femur fractures associated with prolonged antiresorptive therapy. The FDA added an AFF warning to the Prolia label in 2011 following postmarket surveillance data.

Incidence Estimates

The American Society for Bone and Mineral Research (ASBMR) task force estimated AFF incidence at 3.2 to 50 cases per 100,000 patient-years for antiresorptive agents, with risk increasing after five or more years of continuous use. A 2020 Swedish registry study (N=approx. 90,000 bisphosphonate or denosumab users) reported an AFF rate of 11 per 100,000 patient-years for denosumab, compared with 11 per 100,000 for bisphosphonate users in the same cohort [5].

Clinical Presentation and Monitoring

Prodromal thigh or groin pain precedes an AFF in approximately 70% of cases. Patients on long-term Prolia who report new-onset thigh pain should receive bilateral femur imaging. The FDA prescribing information states that for patients presenting with thigh or groin pain, an evaluation for AFF should be considered and, if confirmed, discontinuation of denosumab should be discussed [3].

Rebound Vertebral Fractures After Stopping Prolia

Denosumab discontinuation carries a risk that has no direct parallel in bisphosphonate therapy. Bone mineral density gains acquired during denosumab treatment are rapidly lost after stopping the drug, and the rate of bone resorption transiently overshoots baseline.

Quantifying the Rebound Risk

A 2019 study published in the Journal of Bone and Mineral Research analyzed 1,001 postmenopausal women who discontinued denosumab after two or more years of treatment. Multiple vertebral fractures occurred in 7.1% of patients within 12 to 18 months of the last injection, compared with an expected rate of approximately 1.0% per year in untreated patients of comparable age and bone density [6].

The Endocrine Society clinical practice guideline states: "Patients who stop denosumab should promptly receive an antiresorptive agent to preserve bone mineral density gains and reduce the risk of rebound fractures" [7].

A practical transition framework used by the HealthRX medical team:

1. Last Prolia injection administered on schedule.
2. At 6 months post-injection (when the next Prolia dose would have been due), begin an oral bisphosphonate or a single infusion of zoledronic acid 5 mg IV.
3. Serum CTX (bone resorption marker) checked at 9 and 12 months post-injection to confirm adequate antiresorptive bridging.
4. If CTX remains suppressed at 12 months, continue oral bisphosphonate for a minimum of 12 additional months before reassessment.

This protocol is not yet standardized in all guidelines, and an endocrinologist or bone health specialist should supervise discontinuation decisions.

Hypocalcemia: Risk Stratification and Management

Denosumab inhibits RANK ligand, suppressing osteoclast activity and thereby reducing calcium release from bone. In calcium-replete, renally intact adults, the net effect on serum calcium is small. In specific subpopulations, it can be clinically significant.

Who Is at Highest Risk

Patients at elevated risk for denosumab-induced hypocalcemia include those with:

• Creatinine clearance <30 mL/min (incidence up to 9% in observational data)
• Malabsorption syndromes limiting calcium absorption
• Hypoparathyroidism
• Severe vitamin D deficiency (25-OH vitamin D <20 ng/mL) at treatment initiation

Monitoring Protocol

The Prolia prescribing information requires baseline serum calcium measurement and correction of hypocalcemia before initiating therapy. Patients should receive at least 1,000 mg elemental calcium and 400 IU vitamin D daily throughout treatment. Some patients with renal impairment require active vitamin D supplementation (calcitriol or alfacalcidol) rather than standard cholecalciferol [3].

Serum calcium should be measured within the first two weeks after each Prolia injection in patients with creatinine clearance <30 mL/min or conditions predisposing to hypocalcemia.

Serious Infections: Skin and Soft Tissue

The RANK ligand signaling pathway has roles in immune function beyond osteoclast regulation. Denosumab-mediated RANK ligand suppression has been associated with modest increases in serious skin infections.

Cellulitis and Hospitalization Data

In FREEDOM, the skin infection requiring hospitalization rate was 1.2% for denosumab vs. 0.7% for placebo (P<0.001). A meta-analysis published in Osteoporosis International pooling data from FREEDOM and five additional trials reported a pooled odds ratio of 1.46 (95% CI 1.01 to 2.13) for serious skin infection with denosumab vs. Comparator [8].

The absolute risk increase is small, roughly 5 additional hospitalizations per 1,000 patients over three years, but it is a real signal. Patients should be counseled to seek prompt medical attention for any skin infection, rash, or signs of cellulitis.

FDA Label Warnings and FAERS Postmarket Data

The current Prolia prescribing information (revised January 2023) carries the following boxed and non-boxed warnings:

• Hypocalcemia (with emphasis on renal impairment)
• Serious infections (including endocarditis)
• Dermatologic reactions (including dermatitis, rash, eczema)
• ONJ
• AFF
• Suppression of bone turnover
• Hypersensitivity reactions including anaphylaxis (rare, <1%)

FAERS Reports Through 2024

FDA Adverse Event Reporting System (FAERS) data through Q4 2024 list the following as top reported events for denosumab (all indications combined, including Xgeva at higher doses):

1. Hypocalcemia
2. Osteonecrosis of jaw
3. Atypical femoral fracture
4. Drug ineffective (fracture during therapy)
5. Back pain

Reporting bias in FAERS means that the frequency of FAERS reports does not equal incidence. Rare or unexpected events are disproportionately reported vs. Common expected events. Absolute rates must always be anchored to controlled trial data rather than FAERS counts [9].

Dermatologic Adverse Events

Dermatologic reactions are noted in the prescribing information but are infrequently discussed in clinical practice. In FREEDOM, dermatitis occurred in 4.0% of denosumab patients vs. 2.1% of placebo patients. Eczema was reported in 3.5% vs. 2.4%. Most cases were mild and resolved without discontinuation.

Serious Dermatologic Events

Stevens-Johnson syndrome and drug reaction with eosinophilia and systemic symptoms (DRESS) have been reported in postmarket surveillance. These events are very rare, estimated at fewer than 1 in 10,000 treated patients based on spontaneous reports, but they are potentially life-threatening. Patients who develop severe blistering or exfoliating skin reactions should discontinue Prolia and receive urgent dermatologic evaluation [3].

Drug Interactions and Special Populations

Immunosuppressed Patients

Because denosumab modestly impairs immune surveillance in skin and possibly other tissues, patients already receiving immunosuppressive therapy (transplant recipients, patients on biologic DMARDs) may carry a higher absolute risk of serious infection than observed in the general FREEDOM population. The European Medicines Agency recommends caution and heightened monitoring in this group.

Pregnancy and Lactation

Prolia is contraindicated in pregnancy. Animal studies showed fetal harm at doses below the clinical exposure. There are no adequate human pregnancy data. Women of reproductive potential must use effective contraception during therapy and for five months after the last dose, given denosumab's half-life of approximately 28 days with prolonged pharmacodynamic effects [3].

Pediatric Use

Denosumab is not approved for use in patients younger than 18 years for osteoporosis. Use in skeletally immature patients has been associated with hypercalcemia following discontinuation, a distinct rebound phenomenon not seen in adults.

Comparing Denosumab Adverse Events to Bisphosphonates

A 2023 network meta-analysis in JAMA Internal Medicine (N=approx. 40,000 across 22 trials) compared antiresorptive agents for postmenopausal osteoporosis. Denosumab and zoledronic acid produced comparable vertebral fracture risk reductions (68% and 70%, respectively). Denosumab carried a nominally higher risk of serious skin infection (OR 1.46 vs. Zoledronic acid OR not significant), while zoledronic acid was associated with higher rates of post-infusion flu-like syndrome (32% vs. <1% for denosumab subcutaneous injection) [10].

Oral bisphosphonates (alendronate, risedronate) are associated with esophageal irritation and gastrointestinal adverse events that denosumab does not share. The choice between agents depends on renal function, adherence factors, patient preference, and the planned treatment duration.