Reclast (Zoledronic Acid) vs Prolia (Denosumab): Side-Effect Profile Head-to-Head

Do Reclast and Prolia share any side effects?

Both are antiresorptive drugs, and both labels warn about osteonecrosis of the jaw, atypical femoral fractures and severe bone, joint and muscle pain [2][1]. Both labels also address low blood calcium: the Prolia label has a boxed warning for severe hypocalcemia in advanced kidney disease, and both labels say pre-existing hypocalcemia must be corrected before treatment [2][1]. Beyond these shared warnings the profiles differ. Reclast's most frequent complaints cluster in the first three days after an infusion, while Prolia's most consequential problems are calcium handling and what happens once the injections stop [1][2]. The percentages below come from separate placebo-controlled trials, not a direct Reclast-versus-Prolia trial.

Reclast is a bisphosphonate given as a 5 mg intravenous infusion over at least 15 minutes, once a year for postmenopausal osteoporosis treatment and once every 2 years for prevention [1]. Prolia is a RANK ligand inhibitor given as 60 mg subcutaneously every 6 months by a healthcare provider, with pregnancy ruled out first [2]. The supplement targets differ too: the Reclast label recommends that osteoporosis patients average at least 1,200 mg calcium and 800 to 1,000 IU vitamin D daily, and the Prolia label says patients should take calcium 1,000 mg and at least 400 IU vitamin D daily [1][2].

Reclast: the reaction after the infusion

The Reclast label describes an acute-phase reaction: fever in 18%, myalgia in 9%, flu-like symptoms in 8%, headache in 7% and arthralgia in 7%. Most of these symptoms begin within the first 3 days and usually resolve within 3 days, though they can last 7 to 14 days, and they become less frequent with later doses [1].

The 3-year osteoporosis trial of 7,736 women aged 65 to 89 gives the comparison with placebo: pyrexia 17.9% vs 4.6%, myalgia 11.7% vs 3.7%, influenza-like illness 8.8% vs 2.7%, headache 12.4% vs 8.1%, arthralgia 23.8% vs 20.4%, and nausea 8.5% vs 5.2% [1].

Acetaminophen given after the infusion may reduce the incidence of these symptoms, according to the label [1]. In the hip-fracture trial, where patients received acetaminophen at the infusion and for 72 hours afterwards, fever was 7% and arthralgia 3% [1].

Eye inflammation (iritis, uveitis and episcleritis) occurred in up to 0.2% of Reclast patients across trials [1].

Prolia: is there a reaction after the injection?

The Prolia label's table of reactions reported in at least 2% of postmenopausal women and more often than placebo lists myalgia at 2.9% versus 2.4% on placebo; fever and flu-like illness do not appear in that table [2]. The rest of the table is close to placebo: back pain 34.7% vs 34.6%, pain in extremity 11.7% vs 11.1%, musculoskeletal pain 7.6% vs 7.5%, hypercholesterolemia 7.2% vs 6.1%, and cystitis 5.9% vs 5.8% [2]. Withdrawal because of adverse events was 2.4% on Prolia and 2.1% on placebo [2].

Hypocalcemia and kidney disease: Prolia's boxed warning

The Prolia label carries a boxed warning for severe hypocalcemia in patients with advanced chronic kidney disease (eGFR below 30 mL/min/1.73 m2), including dialysis-dependent patients, with hospitalization, life-threatening events and fatal cases reported [2]. The FDA added this warning on 19 January 2024 [4]. In female dialysis-dependent Medicare patients, the 12-week cumulative incidence of severe hypocalcemia was 41.1% with denosumab (n=1,523) versus 2.0% with oral bisphosphonates (n=1,281), and most cases begin 2 to 10 weeks after each injection, with the greatest risk in weeks 2 to 5 [4]. The label directs prescribers to evaluate for chronic kidney disease-mineral and bone disorder before starting and to monitor serum calcium weekly for the first month and monthly thereafter in these patients [2].

In the 3-year trial of postmenopausal women, serum calcium fell below 8.5 mg/dL at some visit in 1.7% taking Prolia versus 0.4% taking placebo, all of whom were taking calcium and vitamin D. Among participants with normal kidney function, calcium reached its lowest level about 10 days after dosing [2]. In a 55-person study across kidney function levels without calcium or vitamin D supplementation, calcium below 7.5 mg/dL or symptomatic hypocalcemia occurred in 0%, 10%, 29% and 29% of the normal function, creatinine clearance 50 to 80, creatinine clearance below 30 and hemodialysis groups [2].

For Reclast, hypocalcemia is a contraindication and the label says it must be corrected before treatment. The Reclast label has no boxed warning [1].

What happens when each drug is stopped

With Prolia, bone turnover rises above pretreatment levels about 9 months after the last dose, and new vertebral fractures occurred as early as 7 months after the last dose (on average 19 months), with a prior vertebral fracture predicting multiple fractures. If Prolia is discontinued, the label directs prescribers to transition the patient to another antiresorptive therapy [2]. In a post hoc analysis of 1,001 participants who stopped denosumab, the vertebral fracture rate rose from 1.2 per 100 participant-years on treatment to 7.1 off treatment, close to the 8.5 seen in people who stopped placebo [3]. Among those with at least one off-treatment vertebral fracture, 60.7% of denosumab stoppers had more than one, versus 38.7% of placebo stoppers (p=0.049) [3]. A later post hoc analysis reported annualized multiple-vertebral-fracture rates of 3.6 per 100 patient-years after stopping placebo, 2.9 after 3 years or less of denosumab, and 7.5 after more than 3 years, with an odds ratio of 3.0 for longer treatment; the authors suggested closer bone-protective follow-up, such as zoledronic acid, for patients stopping after 3 years [5].

With Reclast, the label states that the safety and effectiveness data cover three years, that the optimal duration has not been determined, and that patients at low risk of fracture should be considered for discontinuation after 3 to 5 years, with periodic re-evaluation of fracture risk [1]. In the extension of the 3-year trial, 1,233 women were randomized to 3 more years of zoledronic acid or to placebo: femoral neck bone density stayed level with continued treatment and dropped slightly after stopping — a 1.04% between-group difference — but remained above pretreatment levels. New morphometric vertebral fractures numbered 14 with continued zoledronic acid versus 30 with placebo [6].

Serious infections

The Prolia label notes that RANKL is expressed on activated T and B lymphocytes and in lymph nodes, so a RANKL inhibitor may increase infection risk [2]. In the 3-year trial, nonfatal serious infections occurred in 3.3% of placebo patients and 4.0% of Prolia patients, while infections resulting in death were 0.2% in both groups [2]. Serious infections, including skin infections such as cellulitis, are listed among the label's warnings, and patients are advised to seek prompt medical attention for signs of infection [2].

Infection is not among the Reclast label's warnings and precautions. The Reclast label lists hypocalcemia, renal impairment, osteonecrosis of the jaw, atypical fractures, severe bone, joint, and muscle pain, and asthma among its warnings and precautions [1].

Osteonecrosis of the jaw

An international consensus review estimated ONJ incidence in osteoporosis patients at 0.001% to 0.01%, against 1% to 15% in oncology patients receiving high doses at frequent intervals [7]. In the Reclast 3-year osteoporosis trial of 7,736 women, one placebo patient and one Reclast patient had symptoms consistent with ONJ, and there were none in the 2,127-patient hip-fracture trial [1]. No ONJ cases were reported in the 3-year FREEDOM trial [8]; across the extension, 7 cases occurred in the long-term group and 6 in the crossover group [9]. Both labels call for a routine oral exam before treatment, and both note that risk may increase with longer exposure [2][1].

Atypical femoral fractures

Both labels warn about atypical low-trauma fractures of the femoral shaft and direct prescribers to evaluate thigh or groin pain [2][1]. The Reclast label directs prescribers to re-evaluate the risk and benefit of continuing and to consider interrupting bisphosphonate therapy if an atypical fracture is found [1]. In the 10-year denosumab extension, one atypical femoral fracture occurred in each group [9].

Atrial fibrillation

In the Reclast 3-year trial, adjudicated serious atrial fibrillation events occurred in 1.3% (50 patients) on zoledronic acid versus 0.4% (17) on placebo; the NEJM report gives 50 versus 20. All atrial fibrillation events were 2.5% versus 1.9%. More than 90% occurred more than a month after the infusion, and an ECG substudy of 559 patients found no difference 9 to 11 days after the infusion [1]. The hip-fracture trial did not show the same pattern, with serious atrial fibrillation at 1.0% versus 1.2% [1]. The FREEDOM investigators reported no increase in cardiovascular disease with denosumab [8].

Bone, joint and muscle pain

Severe bone, joint and muscle pain is a labeled warning for both drugs. The Reclast label says to withhold future doses if severe symptoms occur [1]; the Prolia label says to discontinue if severe symptoms develop [2]. The ordinary aches around a Reclast infusion belong to the acute-phase reaction and usually settle within days [1]. For Prolia, the trial rates for back pain, extremity pain and musculoskeletal pain were essentially identical to placebo [2].

Bottom line

Reclast's reactions are concentrated in the first three days after an infusion, and the label reports that acute-phase symptoms decrease with later doses [1]. Prolia's key concerns include calcium control, particularly in advanced kidney disease, and increased fracture risk after stopping, skipping, or delaying an injection [2][4][3]. Both carry the rare labeled risks of osteonecrosis of the jaw and atypical femoral fracture [2][1]. For context on why either is used, vertebral fracture reductions were 70% with yearly zoledronic acid and 68% with denosumab in their separate pivotal trials [10][8].