Risks of therapy with bone antiresorptive agents in patients with advanced malignancy

INTRODUCTION — Parenteral administration of antiresorptive agents (bisphosphonates and denosumab, also referred to as osteoclast inhibitors, bone modifying agents, or bone targeting agents) reduces the frequency of skeletal-related events among patients with multiple myeloma and in those with bone metastases from a variety of solid tumors, including breast, lung, and prostate cancer. Prolonged therapy with these antiresorptive agents in patients with advanced malignancy is generally well tolerated, but some side effects are potentially serious and require periodic monitoring [1-3].

This topic will provide an overview of the risks of therapy with antiresorptive agents in patients with skeletal involvement from advanced malignancy. The therapeutic use of bisphosphonates and denosumab in malignant disease, and their use and side effects in other conditions, such as osteoporosis and cancer treatment-related bone loss, are discussed separately, as is a more thorough discussion of medication-related osteonecrosis of the jaw. (See "Osteoclast inhibitors for patients with bone metastases from breast, prostate, and other solid tumors" and "Multiple myeloma: The use of osteoclast inhibitors" and "Bisphosphonate therapy for the treatment of osteoporosis" and "Denosumab for osteoporosis" and "Evaluation and management of aromatase inhibitor-induced bone loss" and "Use of osteoclast inhibitors in early breast cancer" and "Side effects of androgen deprivation therapy", section on 'Preventive strategies' and "Bone metastases in advanced prostate cancer: Management" and "Medication-related osteonecrosis of the jaw in patients with cancer".)

CLASSES OF ANTIRESORPTIVE AGENTS — Two classes of antiresorptive agents are used for prevention of skeletal events in patients with advanced malignancy:

●Bisphosphonates pamidronate, zoledronic acid, and ibandronate are structural analogs of inorganic pyrophosphate. Pamidronate and zoledronic acid are approved in the United States, while ibandronate is approved in many European countries. Bisphosphonates inhibit osteoclastic bone resorption by attaching to hydroxyapatite binding sites within bone. The bisphosphonate that is released during bone resorption impairs the ability of the osteoclasts to adhere to the bony surface and resorb bone. In addition, bisphosphonates also reduce osteoclast activity by decreasing osteoclast progenitor development and recruitment, and by promoting osteoclast apoptosis. (See "Pharmacology of bisphosphonates".)

●Denosumab is a fully human monoclonal antibody that inhibits osteoclastic bone resorption by specifically binding and inhibiting the receptor activator of nuclear factor-KB ligand, a key regulator of osteoclast formation, function, and survival. (See "Normal skeletal development and regulation of bone formation and resorption".)

Both classes of agents are approved for use in patients with bone metastases from a variety of solid tumors and multiple myeloma for prevention of skeletal-related events. (See "Osteoclast inhibitors for patients with bone metastases from breast, prostate, and other solid tumors" and "Multiple myeloma: The use of osteoclast inhibitors" and "Bone metastases in advanced prostate cancer: Management" and "Use of osteoclast inhibitors in early breast cancer".)

Denosumab shares several toxicities with bisphosphonates, including osteonecrosis of the jaw and hypocalcemia. However, there are differences as well. As a fully human monoclonal antibody, denosumab is administered as a subcutaneous, rather than intravenous, injection, and its toxicity profile differs in several respects from that of bisphosphonates (table 1). Knowledge about the comparative side effect profiles of denosumab and bisphosphonates is primarily derived from the three large identically designed registration trials performed in patients with bone metastases from solid tumors as well as a phase III trial conducted exclusively in patients with multiple myeloma [4-7].

RISKS SHARED BY BOTH CLASSES OF DRUGS

Osteonecrosis of the jaw — Medication-related osteonecrosis of the jaw (MRONJ), a form of avascular necrosis first described in 2002 [8], is a relatively uncommon but potentially serious side effect of treatment with either intravenous (IV) high-potency bisphosphonates (image 1) or denosumab.

MRONJ typically presents as pain, infection, and necrotic bone in the mandible or maxilla in a patient receiving long-term treatment with high-potency antiresorptive agents. The overall risk of this complication is <2 percent in patients receiving antiresorptive agents in the first year of therapy and rises with prolonged therapy.

The incidence is slightly higher with denosumab than with bisphosphonates. Dentoalveolar surgery is a major risk factor. In view of the difficulty in treating and the morbidity associated with established MRONJ, prevention is emphasized. Treatment objectives for patients with an established diagnosis of MRONJ are to eliminate pain, control infection of the soft tissue and bone, and minimize the progression or occurrence of bone necrosis. Treatment has generally shifted away from aggressive surgical interventions toward conservative therapy with limited debridement, antibiotics, and oral rinses with chlorhexidine or hydrogen peroxide resulting in resolution in a high proportion of patients. This subject is discussed in detail elsewhere. (See "Medication-related osteonecrosis of the jaw in patients with cancer".)

Hypocalcemia and other electrolyte abnormalities — Low levels of vitamin D are especially common among patients requiring antiresorptive therapy. All patients receiving parenteral therapy with a bone-modifying agent for advanced cancer should have their calcium and vitamin D levels assessed at baseline, and low levels should be supplemented prior to initiating therapy. This is important both to minimize the risk of hypocalcemia, which can occur from antiresorptive agents, and to optimize the efficacy of antiresorptive therapy.

Daily supplementation with calcium and vitamin D is indicated for all patients unless otherwise contraindicated (eg, in patients with hypercalcemia or a history of recurrent renal stones). Most patients should receive a total of 1000 mg of calcium (diet plus supplement) and 800 to 1200 international units of vitamin D daily. However, vitamin D dosing should be based on serum levels, and recommendations regarding calcium supplementation should be individualized based on the patient's underlying malignancy (and the risk for developing hypercalcemia) and renal function. (See "Vitamin D deficiency in adults: Definition, clinical manifestations, and treatment", section on 'Dosing' and "Calcium and vitamin D supplementation in osteoporosis", section on 'Optimal intake'.)

Periodic monitoring of serum calcium levels (eg, every three months) is advised during therapy with either bisphosphonates or denosumab. More frequent monitoring is needed in symptomatic patients and those with baseline renal insufficiency, especially shortly after initiating antiresorptive therapy. The United States Prescribing Information for denosumab recommends that patients predisposed to hypocalcemia and disturbances of mineral metabolism (eg, history of hyperparathyroidism, thyroid or parathyroid surgery, malabsorption syndromes, history of small bowel resection, severe renal impairment, or receiving dialysis) have monitoring of calcium, magnesium, and phosphate levels within 14 days of initiating denosumab. If patients become hypercalcemic, calcium supplements and vitamin D should be discontinued and hypercalcemia treatment initiated depending on the level. (See "Treatment of hypercalcemia".)

Calcium homeostasis is one of the major functions of bone [9]. Since osteoclasts resorb bone and release calcium as a result effective osteoclast inhibition can result in hypocalcemia and hypophosphatemia. Most patients do not become hypocalcemic because of compensatory mechanisms, most importantly increased secretion of parathyroid hormone. In some cases, these compensatory mechanisms may be blocked (eg, prior parathyroidectomy, low vitamin D levels, hypomagnesemia hypoparathyroidism, and renal failure), resulting in hypocalcemia [3,10]. Patients are at higher risk of electrolyte imbalance if they have renal insufficiency or decreased compensatory mechanisms.

The incidence, severity, and time course of hypocalcemia and hypophosphatemia can be illustrated by the following studies:

●In a report of 120 patients who received a total of 546 infusions of zoledronic acid for multiple myeloma or another malignancy [11], hypocalcemia developed following 55 infusions (10 percent) in 42 of the patients (35 percent). Symptomatic hypocalcemia requiring IV calcium supplementation occurred in 10 patients (8 percent) despite dose adjustment for creatinine clearance (CrCl) and prophylactic supplementation with oral calcium and vitamin D. Impairment of CrCl was seen more often among subjects developing hypocalcemia than in those remaining normocalcemic. Hypomagnesemia was found in all patients who developed hypocalcemia who had measurement of serum magnesium.

In the phase III myeloma trial comparing zoledronic acid versus denosumab, all patients were supplemented with calcium >500 mg and vitamin D 400 international units daily, and all had a CrCl of at least 30 mL/min [7]. In the zoledronic acid group, hypocalcemia of any grade developed in 12 percent of the 852 patients, with only 2.5 percent (22 patients) having grade 3 or 4 events.

●Denosumab is a more potent inhibitor of bone resorption than zoledronic acid as measured by changes in bone turnover markers [4-6,12]. Thus, it is not unexpected that the incidence rates of hypocalcemia and hypophosphatemia were greater with denosumab than with zoledronic acid in all three registration trials and the phase III myeloma trial [4-6,13], and in other reports as well.

In the latest integrated analysis of patient-level data from all three conducted solid tumor trials that differed only as to tumor type, hypocalcemia ≥grade 2 (table 2) occurred in 12.4 percent of patients treated with denosumab versus 5.3 percent of those treated with zoledronic acid [14]. The median time to first occurrence of hypocalcemia was also shorter with denosumab as compared with zoledronic acid (median 3.8 versus 6.5 months). Most patients experienced only a single hypocalcemic episode; however, 43 percent of denosumab-treated patients and 32 percent of zoledronic acid-treated patients had recurrent events. Multivariate analysis suggested a greater risk of hypocalcemia in those with prostate and small cell lung cancer, and in those with reduced CrCl and higher baseline levels of bone turnover markers. In an integrated analysis of all three solid tumor trials, the incidence of hypophosphatemia of any grade was 32 percent with denosumab versus 20 percent with zoledronic acid [13].

In the phase III myeloma trial, hypocalcemia was observed in 17 percent of the 850 patients treated with denosumab, and 4 percent had grade ≥3 events [7].

●In general, with both drugs, most hypocalcemic episodes have been mild to moderate in severity, and the majority of patients remain asymptomatic. Hypocalcemia occurs within the first six months of therapy in the majority of patients (median time to first occurrence of hypocalcemia was 2.8 months and of grade >3 was 4.6 months [14]).

●While renal impairment does not affect denosumab pharmacokinetics and dose adjustments are not required for patients with acute or chronic kidney disease (in contrast to zoledronic acid), patients with renal insufficiency or end-stage kidney disease need to be followed more closely for hypocalcemia while receiving denosumab therapy as these patients are likely more dependent on parathyroid hormone-induced bone resorption:

•In one report, 4 of 17 patients with severe chronic kidney disease (CrCl ≤30 mL/min) developed albumin-adjusted serum calcium levels less than 7.5 mg/dL compared with only 1 of 38 patients with CrCl >30 mL/min (24 versus 3 percent) [15]. (See 'Proteinuria and kidney injury' below and 'Renal and electrolyte effects' below.)

•In an analysis of data from the three phase III solid tumor trials comparing monthly dosing with denosumab versus zoledronic acid, 15.5 percent of patients receiving denosumab who had CrCl 30 to <60 mL/min developed grade ≥2 hypocalcemia compared with 5.8 percent of patients receiving zoledronic acid [14].

Treatment of hypocalcemia is discussed elsewhere. (See "Treatment of hypocalcemia".)

Atypical fractures — Although they are uncommon events overall, atypical subtrochanteric and diaphyseal femoral fractures have been reported in patients receiving bisphosphonates and denosumab, which may occur after minimal or no trauma. Patients with otherwise unexplained thigh, hip, or groin pain should be evaluated to rule out a femoral fracture. We discontinue the antiresorptive agent, at least temporarily, when one of these fractures occurs.

Since 2005 several reports have been published describing unusual femoral shaft fractures primarily in postmenopausal females treated for prolonged periods with a bisphosphonate for osteoporosis. This complication has also been reported with denosumab use in both the osteoporosis and cancer settings [16]. Atypical fractures, which are often subtrochanteric or in the proximal diaphyseal region, are thought to be related to oversuppression of bone remodeling, resulting in increased skeletal fragility. They can be bilateral in up to 40 to 50 percent of cases, and circumferential cortical thickening and cortical stress lesions are often seen on radiographs preceding the fracture. They are slow or difficult to heal consistent with the patient's low bone turnover markers. (See "Denosumab for osteoporosis", section on 'Oversuppression of bone remodeling'.)

Among patients receiving long-term bisphosphonates for treatment of metastatic bone disease and multiple myeloma there are a few case reports describing atypical fractures of the femur and even fewer of metatarsal bone fractures [17-21], but the frequency with which this occurs, as well as the risk factors and the best way to manage osteoclast inhibitor therapy in such patients (ie, drug holiday or permanent discontinuation), remains unclear.

Atypical femoral fractures have also occurred after discontinuation of osteoclast inhibitor therapy. Incidence rates of 0.4 to 1.8 percent have been reported with long term denosumab use. Risk factors include prior zoledronic acid therapy and long-term therapy for longer than 3.5 years [22].

The Musculoskeletal Tumor Society guidelines provide advice on imaging and surgical management of an impending or completed atypical femur fracture.

RISKS SPECIFIC TO BISPHOSPHONATES — Side effects among patients receiving oral or parenteral bisphosphonates for non-malignant conditions are discussed separately.

The following are the most important potential side effects in patients receiving parenteral bisphosphonates for malignant disease.

Proteinuria and kidney injury — Intravenous (IV) bisphosphonates are associated with proteinuria and acute kidney injury which are both dose and infusion time dependent, and most often seen in patients with multiple myeloma. Kidney injury may be minimized by observing recommended infusion times and creatinine clearance (CrCl) adjusted dosing, optimizing hydration prior to bisphosphonate administration (particularly for patients with marked Bence Jones proteinuria), and avoiding concurrent nephrotoxic medications or contrast reagents.

A lower initial dose of zoledronic acid is indicated for patients with pre-existing kidney injury (CrCl <60 but ≥30 mL/min). Another alternative is use of denosumab. However, among patients with chronic kidney disease who are permanently on dialysis, zoledronic acid can be safely administered on a monthly basis without concern regarding the induction of further kidney injury.

Renal function must be checked before each dose in cancer patients being treated with IV bisphosphonates. Unexplained azotemia or an absolute creatinine level >1.4 mg/dL [124 micromol/L] in patients with normal baseline values should prompt temporary discontinuation of the bisphosphonate. For patients with multiple myeloma receiving pamidronate or zoledronic acid, we evaluate for albuminuria every three to six months. For patients experiencing significant albuminuria (>500 mg per 24 hours), discontinuation of the drug is advised until resolution, unless other potential causes of albuminuria are uncovered. Switching to denosumab is an alternative for these patients.

Mechanisms

●Pamidronate – Pamidronate has been associated with the development of nephrotic-range proteinuria and acute kidney injury, and a number of mechanisms have been proposed, including collapsing focal segmental glomerulosclerosis [23-25]. (See "Collapsing focal segmental glomerulosclerosis (collapsing glomerulopathy)", section on 'Bisphosphonates and other drugs' and "Overview of heavy proteinuria and the nephrotic syndrome".)

Nephrotoxicity is both dose and infusion time dependent, but it is mostly related to infusion time; the risk appears to be higher among patients who receive the drug over less than two hours [26]. Although most cases are reported in patients with multiple myeloma, albuminuria followed by azotemia has also been seen with long-term administration of pamidronate in patients with other malignancies.

●Zoledronic acid – Acute kidney injury has also been observed with zoledronic acid, although most of the renal toxicity has been reported with higher doses (8 mg) than are US Food and Drug Administration (FDA)-approved (at most 4 mg) and with infusion durations of less than 15 minutes [27-32]. The mechanism of acute kidney injury with zoledronic acid may be different from that in patients receiving pamidronate. In one case series of six patients with acute kidney injury associated with zoledronic acid, renal biopsies showed toxic acute tubular necrosis without evidence of glomerular damage [32]. These patients develop rises in creatinine often without proteinuria. However, with long-term treatment, patients may develop albuminuria that improves upon discontinuation of the drug. Others describe Fanconi syndrome (proximal tubular dysfunction causing renal tubular acidosis with glycosuria, aminoaciduria, hypokalemia, and phosphate wasting) in patients receiving zoledronic acid for metastatic cancer [33]. (See "Clinical manifestations and diagnosis of Fanconi anemia" and "Etiology and diagnosis of distal (type 1) and proximal (type 2) renal tubular acidosis".)

Acute kidney injury from either drug can progress to renal failure and the need for dialysis [25,30,34].

Incidence and risk factors

Multiple myeloma — Most reports of acute kidney injury during IV bisphosphonate therapy have involved patients with multiple myeloma [30,35,36]. As an example, of 72 cases of renal failure in association with zoledronic acid use in the FDA Adverse Event Reporting System from August 2001 to March 2003, 42 were in patients with myeloma, 22 with solid tumors, 2 with benign conditions, and 6 were unknown [30]. Underlying risk factors (advanced cancer, previous bisphosphonate exposure, and use of NSAIDs) may have contributed to the progression of renal failure; information on hydration status was not uniformly reported. Eighteen patients developed renal failure after only one dose. Of the patients who developed IV bisphosphonate-related acute kidney injury, 27 required dialysis and 18 died.

Among patients with multiple myeloma, elevated baseline serum creatinine is a risk factor for acute kidney injury. In a retrospective trial involving 300 patients with multiple myeloma treated with zoledronic acid at 15 oncology sites, renal function deteriorated in 34 patients (11 percent), while the rate was 24 percent among patients who had an elevated baseline creatinine [35]. Notably, one-half of the patients who had acute kidney injury while being treated with zoledronic acid restarted zoledronic acid after a brief delay, and only five (2.7 percent) had ongoing renal function deterioration.

However, renal disease is a common problem in multiple myeloma. The pathology is very heterogeneous and may involve a variety of different mechanisms. This can make it difficult to ascertain whether bisphosphonate use is the culprit. (See "Kidney disease in multiple myeloma and other monoclonal gammopathies: Etiology and evaluation".)

As an example, one large randomized study comparing zoledronic acid (4 mg every 21 to 28 days) with oral clodronate (1600 mg/day) in previously untreated patients with multiple myeloma showed no differences in the development of renal failure between the two arms [37]. Given that oral clodronate has not been associated with kidney problems, this study suggests that much of the deterioration in renal function among patients receiving monthly zoledronic acid is likely due to other causes, such as progressive myeloma, other comorbid conditions (eg, diabetes mellitus, hypertension), and/or other potentially nephrotoxic medications (eg, cisplatin [36]).

Other malignancies — Compared with patients who have multiple myeloma, the incidence of acute kidney injury during treatment with IV bisphosphonates is lower in patients with other types of cancer (approximately 10 percent), especially when doses are limited to 4 mg and infusion times are 15 rather than 5 minutes [4,38,39]. Incidence and severity are higher in patients with baseline kidney disease.

Management and prevention

●Choice of agent, dose, and dosing interval – Bisphosphonates are eliminated in the urine. Consistent with the FDA-approved prescribing information, a lower initial dose of zoledronic acid (ranging from 3 to 3.5 mg) is appropriate for patients with pre-existing kidney disease (CrCl <60 but ≥30 mL/min). However, there are no clinical studies to support either the efficacy or safety of these lower doses. The FDA-approved prescribing information for pamidronate does not contain similar guidance. Pamidronate can be used safely in patients with a CrCl <30 mL/min [40].

Perhaps the best alternative for patients with significant kidney disease who are not on dialysis is denosumab, which is not renally cleared and is FDA approved for preventing skeletal-related events in patients with multiple myeloma and in those with bone metastases from solid tumors. However, among patients with chronic kidney disease who are permanently on dialysis, zoledronic acid can be safely administered on a monthly basis without concern regarding the induction of further kidney injury from this potentially nephrotoxic drug. In addition, initial renal impairment among patients with multiple myeloma often returns to normal after effective anti-multiple-myeloma therapies, so patients initially treated with denosumab may be changed to zoledronic acid once their renal function improves.

For patients receiving IV bisphosphonates, acute kidney injury may be minimized by observing recommended doses and infusion times, optimizing hydration prior to bisphosphonate administration (particularly for patients with marked Bence Jones proteinuria), and avoiding concurrent nephrotoxic medications or contrast reagents:

•If pamidronate is chosen, it should be administered at a dose no higher than 90 mg at a rate no faster than over two hours.

•If zoledronic acid is chosen, it should be administered over no less than 15 minutes every three to four weeks.

•Clinicians should not attempt to shorten the infusion time, increase the dose, or reduce the dosing interval specifically to improve renal tolerance. There is no evidence that lengthening the infusion duration from 15 to 30 minutes reduces the rate of renal toxicity [41], although larger studies are needed to further define this issue. Preclinical studies show that negative renal effects of bisphosphonates are related to the maximal serum concentration (Cmax), rather than the Area Under the Curve of concentration x time (AUC), so that longer infusion times are likely to reduce kidney problems without compromising efficacy, which is related to AUC and not Cmax.

Increasingly, longer dosing intervals (ie, every 12 as compared with every 3 to 4 weeks) are being used for bisphosphonate therapy of patients with bone metastases, at least from solid tumors. (See "Osteoclast inhibitors for patients with bone metastases from breast, prostate, and other solid tumors", section on 'Dosing interval'.)

Whether or not less frequent administration of zoledronic acid reduces the incidence of kidney injury is unclear. Randomized trials, conducted mainly in breast and prostate cancer, have not reported a significantly lower rate of kidney injury with the longer dosing interval, but they may be underpowered to demonstrate this. As an example, the CALGB (Alliance) trial 70604 randomly assigned 1822 patients with bone metastases (855 with breast cancer, 689 with prostate cancer, and 278 with myeloma) to the same dose of zoledronic acid (adjusted based on calculated CrCl) every 4 weeks or every 12 weeks for two years, starting with the first dose [42]. The two-year cumulative incidence of grade 3 or 4 creatinine increase overall was lower in the every-12-week treatment group (0.5 versus 1.2 percent); while potentially clinically meaningful, this difference was not statistically significant (p = 0.10).

●Periodic assessment of kidney function and albuminuria – Renal function must be assessed serially in cancer patients who are being treated with bisphosphonates, and appropriate dose adjustments and/or temporary cessation of treatment may be required. Guidelines from the American Society of Clinical Oncology (ASCO) for use of bisphosphonates among patients with breast cancer or multiple myeloma recommend that creatinine be monitored prior to each dose of pamidronate or zoledronic acid [43,44]. Unexplained azotemia (an increase of ≥0.5 mg/dL for patients with normal baseline creatinine, or ≥1 mg/dL for patients with an abnormal baseline value) in a patient receiving IV bisphosphonates for any malignancy should prompt at least temporary discontinuation of the bisphosphonate until recovery of kidney function. If recovery does not occur or recurs with restarting of IV bisphosphonates, a switch to denosumab is advisable.

For patients with multiple myeloma receiving pamidronate or zoledronic acid, ASCO guidelines also recommend intermittent evaluation (every three to six months) for the presence of albuminuria [44]. For patients experiencing significant albuminuria (defined as >500 mg per 24 hours), discontinuation of the drug is advised until resolution of albuminuria. Importantly, there may be other causes for the albuminuria such as comorbid conditions (eg, diabetes mellitus) or the development of primary amyloidosis in this at-risk patient population. In contrast to patients with multiple myeloma, ASCO guidelines do not support routine assessment for albuminuria in patients with other malignancies such as breast cancer [43]. (See "Multiple myeloma: The use of osteoclast inhibitors", section on 'Monitoring during therapy'.)

If albuminuria does not resolve, or recurs, a switch to denosumab is advisable.

●Restarting therapy in the event of acute kidney injury – Patients who develop acute kidney injury and temporarily discontinue IV bisphosphonates should be reassessed every three to four weeks with a serum creatinine (and, in the setting of multiple myeloma, with a 24-hour urine collection for total protein and urine protein electrophoresis). If kidney function returns to within 10 percent of baseline, therapy can be reinstituted cautiously, as follows:

•Zoledronic acid should be administered over at least 15 minutes. ASCO guidelines suggest prolonging the infusion time (to 30 to 60 minutes) once the drug is restarted following a return of kidney function to within 10 percent of baseline [44], although the benefit of this approach is unclear [41]. Consistent with the FDA-approved United States Prescribing Information, reinitiation of zoledronic acid at the same dose as that used prior to treatment interruption is reasonable.

•Pamidronate should be administered over four hours, at doses no greater than 90 mg every four weeks [44]. The FDA-approved United States Prescribing Information does not provide guidelines for dosing pamidronate in patients whose abnormal kidney function has returned to within 10 percent of baseline.

If kidney function does not return to baseline, there are no data on which to base further management decisions. Switching to denosumab is a reasonable option as it has no known renal toxicity and provides comparable efficacy. (See "Osteoclast inhibitors for patients with bone metastases from breast, prostate, and other solid tumors".)

Acute phase response — In 15 to 30 percent of patients who have not previously received bisphosphonates, IV zoledronic acid and pamidronate can cause transient fever and an influenza-like syndrome [4,5,13,45-47]. The syndrome is typical of an acute phase response characterized by fever, chills, bone pain, headache, nausea, myalgias, and arthralgias and is postulated to be due to transiently increased cytokine production. These symptoms are usually mild and self-limited, resolving after several days, and most often do not occur with subsequent dosing of these drugs.

Acetaminophen or NSAIDs are both effective treatments, but acetaminophen is preferred given the potential for kidney damage with NSAIDs. If patients are receiving glucocorticoids to treat their underlying malignancy, it may be beneficial to administer these drugs on the same day as pamidronate or zoledronic acid, especially during the first several infusions as this may reduce the incidence and severity of the influenza-like syndrome. For the rare patient with ongoing symptoms following each infusion of IV bisphosphonates, switching to denosumab is an option since this drug has not been associated with acute phase reactions.

Ocular toxicities — Both oral and IV bisphosphonate use can be associated with conjunctivitis, uveitis, scleritis, and orbital inflammation. Ocular inflammation requires a prompt ophthalmologic evaluation; further treatment with the offending bisphosphonate is not recommended [48,49].

Bone, joint, or muscle pain — In a syndrome distinct from the acute phase response, rarely severe and sometimes incapacitating bone, joint, or muscle pain can occur within days, months, or years after starting a bisphosphonate and does not always resolve completely with discontinuation of therapy. This syndrome is uncommon and may be overlooked by clinicians.

Atrial fibrillation and stroke — Data on the potential increased risk of atrial fibrillation and stroke with oral bisphosphonate use in patients with osteoporosis are conflicting.

However, a significantly increased risk of atrial fibrillation/flutter and stroke has been reported in patients receiving IV bisphosphonates in the setting of a malignancy [50,51]. Such patients may be at particular risk because of the higher doses used, advanced age, and/or the prior use of chemotherapeutic agents with cardiac toxicities.

This issue was explored using Surveillance, Epidemiology, and End Results-Medicare-linked data in 6857 older (age ≥65) patients with cancer who were treated with IV bisphosphonates and 13,714 matched bisphosphonate nonusers. Results included [50]:

●Infusion of IV bisphosphonate was associated with a modestly increased risk for atrial fibrillation (hazard ratio [HR] 1.30, 95% CI 1.18-1.43), all supraventricular tachycardias (SVTs; HR 1.28, 95% CI 1.19-1.38), and stroke (HR 1.30, 95% CI 1.09-1.54).

●The risk for all SVTs increased 7 percent for every five bisphosphonate doses (HR 1.07, 95% CI 1.02-1.12).

Of note, an increased risk of atrial fibrillation/stroke was not reported in the bisphosphonate arms of any of the three registration trials of denosumab versus zoledronic acid. (See "Medication-related osteonecrosis of the jaw in patients with cancer", section on 'Dose, duration, and type of therapy'.)

RISKS SPECIFIC TO DENOSUMAB

Renal and electrolyte effects — Denosumab is a monoclonal antibody that is cleared predominantly through the liver via proteolytic degradation and nonspecific endocytosis. Elimination through the kidney is insignificant as the molecular weight is higher than the glomerular filtration threshold [52]. Consequently, monitoring of renal function and dose adjustments for pre-existing kidney injury are neither required nor recommended. However, patients with severe kidney disease (creatinine clearance [CrCl] <30 mL/min) are more prone to hypocalcemia and should be monitored more closely for this toxicity when initiating therapy. (See 'Hypocalcemia and other electrolyte abnormalities' above.)

In the four registration trials comparing monthly denosumab with zoledronic acid therapy, renal toxicity and renal adverse events were seen significantly more frequently in the zoledronic acid arm [4-7]. This was true despite obligatory renal function testing prior to each dose and only zoledronic acid being dose adjusted for decreases in renal function as recommended by the manufacturer's US Food and Drug Administration (FDA)-approved package insert. (See 'Proteinuria and kidney injury' above.)

Few data are available on the use of denosumab in patients with severe kidney injury:

●A small single trial involving 55 patients with varying degrees of kidney dysfunction included 17 with severe chronic renal disease (CrCl ≤30 mL/min or requiring hemodialysis) who received a single 60 mg dose of subcutaneous denosumab. Denosumab pharmacokinetics and induced changes in biomarkers of bone resorption were not affected by kidney disease of any severity [15].

●Another small retrospective study included 22 patients with severe chronic kidney disease (CrCl <30 mL/min) receiving monthly denosumab (120 mg subcutaneously) for the prevention of skeletal-related events from bone metastases [53]. Patients received a median of 3.5 doses of denosumab (range 1 to 21 doses). Forty-five percent of patients developed hypocalcemia, with 14 percent developing grade 3 hypocalcemia, and 32 percent developed hypophosphatemia of any grade (two patients, 9 percent, developed grade 3 events). No patient developed symptoms secondary to electrolyte disturbances. Intravenous calcium replacement was administered to 3 of the 10 patients who developed hypocalcemia, while oral calcium supplements were added or the dose increased in three others. The two patients with grade 3 hypophosphatemia were treated with supplemental phosphate, while none of the grade 2 hypophosphatemic events were treated and all resolved without intervention. In this retrospective analysis, 36 percent of patients were noted to be receiving oral calcium and vitamin D supplementation, but baseline vitamin D levels were not available for the majority of patients.

●Denosumab was also not dose adjusted for baseline kidney function in any of the phase III registration trials in metastatic solid tumors or multiple myeloma [4-7]. Renal adverse events in these trials were comparable with those seen in the placebo control arms from older bisphosphonate trials consistent with no additional renal toxicity associated with denosumab therapy [39].

We agree with consensus-based guidelines from the Kidney Disease: Improving Global Outcomes group, which recommend measuring serum calcium, phosphate, parathyroid hormone (PTH), alkaline phosphatase, and 25 hydroxyvitamin D prior to starting denosumab in patients with chronic kidney disease [54,55]. Lower pretreatment albumin-adjusted serum calcium levels or vitamin D levels may be predictive of hypocalcemia with denosumab therapy. Severe hyperparathyroidism may also be a risk factor for hypocalcemia as patients with chronic kidney disease may be dependent on PTH-induced osteoclast activity to preserve calcium levels and when administered an antiresorptive agent may not have sufficient PTH reserve to avoid hypocalcemia.

As the half-life of denosumab is 25 to 30 days, patients on hemodialysis should be monitored at each hemodialysis session for the first two weeks, followed by weekly monitoring for at least 30 days following the first dose of denosumab [54].

Rebound vertebral fractures — It is common practice to administer a single dose of a potent bisphosphonate (ie, zoledronic acid at 4 or 5 mg for a single dose) after stopping denosumab therapy to prevent rebound bone loss and fractures.

Markers of bone resorption rebound and increase rapidly after denosumab discontinuation, and this can lead to increased loss of bone mineral density and the development of vertebral fractures, particularly in patients with baseline osteoporosis, with a history of prior fracture, or on continued aromatase inhibitor therapy [56,57]. In a large phase III trial (ABCSG-18) comparing denosumab therapy with placebo in early stage breast cancer patients receiving an aromatase inhibitor, patients who stopped denosumab and continued on aromatase inhibitor therapy for at least six months had a 2.4-fold increase in clinical vertebral fractures and a 3.5-fold higher risk for multiple vertebral fractures compared with those who stopped placebo treatment [57].

Specific recommendations to prevent rebound bone loss and fractures after discontinuation of denosumab are provided separately. (See "Denosumab for osteoporosis", section on 'Discontinuation or delay of denosumab'.)

Hypersensitivity — Allergic reactions are very rare, but they can include anaphylaxis, hypotension, dyspnea, lip swelling, rash, pruritus, and urticaria and should be treated with appropriate supportive measures.

Neutralizing antibodies — Denosumab is a fully human monoclonal antibody, and thus, the incidence of antibody formation against denosumab is expected to be quite low. In clinical trials involving several thousands of patients (metastatic cancer and osteoporosis trials combined), antibodies directed against denosumab have been found rarely (<1 percent) in treated patients and none have been neutralizing. Testing for anti-denosumab antibodies is not recommended or available outside the research setting.

Infections — Receptor activator of nuclear factor-KB ligand (RANKL) is expressed on subsets of activated T and B cells as well as mature dendritic cells [58]. In knockout mice depleted of the RANKL or RANK genes, there are changes, such as loss of lymph nodes and architectural changes in the thymus, suggesting that inhibition of the RANK/RANKL pathway may affect immune function [59].

In preclinical and clinical models, an immunologic effect of denosumab has been difficult to prove [60]. However, an increased risk of infections in patients treated with denosumab has been seen in some trials:

●In the pivotal phase III FREEDOM trial demonstrating the efficacy of denosumab at a dose of 60 mg subcutaneously every six months for preventing fractures in postmenopausal females with osteoporosis, an increased number of serious adverse events related to skin infections was observed (15 subjects or 0.4 percent in the denosumab arm versus 3 or <0.1 percent in the placebo arm; p <0.05) [61]. No other infectious complication other than cellulitis was observed statistically more frequently in denosumab-treated females. All patients continued on therapy after adverse events from skin infection, and only one patient who had a long-standing history of varicose ulceration experienced a second serious skin infection on trial. (See "Denosumab for osteoporosis", section on 'Effects on immune system'.)

●In the placebo-controlled HALT trial of denosumab in males with bone loss secondary to androgen deprivation therapy for nonmetastatic prostate cancer, there was no statistical difference in the number of serious adverse events or deaths [62]. However, there were a numerically higher number of serious adverse events related to infection (5.9 versus 4.6 percent) and urinary tract infections (5.1 versus 4.4 percent) in denosumab versus placebo treated males. (See "Side effects of androgen deprivation therapy", section on 'Denosumab'.)

In the four registration trials conducted in patients with skeletal involvement from advanced malignancy, zoledronic acid was the comparator arm and, thus, a true placebo arm is not available for comparison in any of the trials [4-7]. Nevertheless, in none of these trials was the incidence of infectious adverse events or serious adverse events statistically different between the treatment groups despite the large number of patients randomized. An integrated analysis of patient-level data from all three solid tumor trials concluded that there was no difference in the risk of infectious adverse events (43.4 versus 42.9 percent) or infectious serious adverse events (11.6 versus 10.9 percent) with denosumab as compared with zoledronic acid [13].

Hypercalcemia — Severe and life-threatening hypercalcemia requiring immediate therapy has been observed in patients with giant cell tumor of bone and in juvenile patients with growing skeletons after discontinuation of treatment with denosumab [63]. This rare side effect is thought to be secondary to rapid recovery of osteoclastic activity triggering unregulated bone resorption and hypercalcemia.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Medication-related osteonecrosis of the jaw".)

SUMMARY AND RECOMMENDATIONS

●Shared risks

•MRONJ – Medication-related osteonecrosis of the jaw (MRONJ) typically presents as pain, infection, and necrotic bone in the mandible or maxilla in a patient receiving long-term treatment with high-potency antiresorptive agents. Specific recommendations for prevention and management are provided separately. (See "Medication-related osteonecrosis of the jaw in patients with cancer".)

•Electrolyte abnormalities

-All patients receiving a parenteral antiresorptive agent for advanced cancer should have calcium and vitamin D levels assessed at baseline, and low levels supplemented prior to initiating therapy. Daily supplementation with calcium and vitamin D is recommended for all patients unless otherwise contraindicated. (See 'Hypocalcemia and other electrolyte abnormalities' above.)

-We monitor serum magnesium, calcium, and phosphate at least quarterly during therapy with an antiresorptive agent. Monthly monitoring is needed for symptomatic patients and those with baseline kidney injury. Hypercalcemia should prompt discontinuation of calcium supplements and vitamin D, and minimizing ingestion of calcium-containing foods.

•Atypical fractures

-Atypical subtrochanteric and diaphyseal femoral fractures may occur after minimal or no trauma. Patients with otherwise unexplained thigh, hip, or groin pain should be evaluated to rule out a femoral fracture. Pain in the foot may suggest a metatarsal foot fracture.

-We discontinue the antiresorptive agent, at least temporarily, when one of these fractures occurs. (See 'Atypical fractures' above.)

●Risks specific to bisphosphonates

•Proteinuria and kidney injury

-Intravenous (IV) bisphosphonates are associated with proteinuria and acute kidney injury, which are both dose and infusion time dependent, and most often seen in patients with multiple myeloma. (See 'Proteinuria and kidney injury' above.)

-Kidney injury may be minimized by observing recommended infusion times and creatinine clearance (CrCl) adjusted dosing, optimizing pretreatment hydration, and avoiding concurrent nephrotoxic medications or iodinated contrast. (See 'Management and prevention' above.)

-A lower initial dose of zoledronic acid is indicated for patients with pre-existing chronic kidney disease (CrCl <60 but ≥30 mL/min). Another alternative is use of denosumab. However, for patients with chronic kidney disease who are permanently on dialysis, zoledronic acid can be safely administered every four weeks without concern for further kidney injury.

-For cancer patients being treated with IV bisphosphonates, kidney function must be checked prior to each dose. Unexplained azotemia or an absolute creatinine level >1.4 mg/dL (124 micromol/L) in patients with normal baseline values should prompt temporary discontinuation of the bisphosphonate.

-For patients with multiple myeloma, we evaluate for albuminuria every three to six months. If significant albuminuria develops (>500 mg per 24 hours), discontinuation of the drug is advised until resolution, unless other potential causes of albuminuria are uncovered. If albuminuria does not resolve, or recurs, a switch to denosumab is advisable.

•Other side effects

-A commonly observed side effect of IV bisphosphonates is a complex of flu-like symptoms. (See 'Acute phase response' above.)

Rarely, patients may experience unremitting, incapacitating bone, joint, or muscle pain, or possibly atrial fibrillation/stroke. (See 'Bone, joint, or muscle pain' above and 'Atrial fibrillation and stroke' above.)

-Oral and IV bisphosphonates can also be associated with conjunctivitis, uveitis, scleritis, and orbital inflammation. Ocular inflammation requires a prompt ophthalmologic evaluation and drug discontinuation. (See 'Ocular toxicities' above.)

●Risks specific to denosumab

•Denosumab is not cleared by the kidneys, and dose adjustments or monitoring of renal function are not required during therapy. However, patients with severe chronic kidney disease are at greater risk for hypocalcemia and hypophosphatemia, and they should be monitored more closely. For those on hemodialysis it is reasonable to check calcium and phosphate levels at each session for the first two weeks, followed by weekly monitoring for an additional two weeks and then monthly. (See 'Renal and electrolyte effects' above.)

•It is common practice to administer a single dose of a potent bisphosphonate (ie, zoledronic acid at 4 or 5 mg for a single dose) after stopping denosumab therapy to prevent rebound bone loss and fractures. Specific recommendations are provided separately. (See "Denosumab for osteoporosis", section on 'Discontinuation or delay of denosumab'.)

•Rare allergic reactions can include anaphylaxis, hypotension, dyspnea, lip swelling, rash, pruritus, and urticaria. (See 'Hypersensitivity' above.)

•A slightly increased risk of infections in patients treated with denosumab has been seen in some trials, but immunologic impairment has been difficult to prove. (See 'Infections' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Robert A Kyle, MD, who contributed to an earlier version of this topic review.