Bisphosphonate Complications Including Osteonecrosis of the Jaw

Bisphosphonates have proven benefit in the treatment of bony disease secondary to multiple myeloma and solid tumors. Intravenous bisphosphonates are known to prolong time to development of skeletal-related complications such as fractures, vertebral compressions, cord compressions requiring radiation or surgery; improve pain control, improve hypercalcemia and without significant deterioration in quality of life and performance status. No consistent benefit in survival has been demonstrated with the use of these agents in patients with lytic or blastic bony disease.^1,2 

Myeloma-related lytic bony disease is now understood to be secondary to increased osteoclastic activity and impaired osteoblastic activity. Myeloma cells are known to secrete both stimulators of osteoclast activation such as receptor activator of nuclear factor-κB ligand (RANKL) and soluble molecules such as dickkopf 1 (DKK) that inhibit osteoblastic activity.³ Bisphosphonates inhibit osteoclast function and therefore block the formation of “punched out” lytic bony lesions and consequent manifestations of lytic bony disease. The American Society of Clinical Oncology published clinical practice guidelines⁴ on the role of bisphosphonates in the management of myeloma. Oral bisphosphonates such as clodronate (not available in the US) and intravenous bisphosphonates pamidronate and zoledronic acid demonstrated reduction in skeletal complications; however, time to first skeletal-related event was more consistently examined in studies evaluating intravenous bisphosphonates. This panel also raised several questions that remain to be answered, including the timing to initiate and stop therapy, role in asymptomatic patients, integration of these agents with other treatments for bony disease and a cost-benefit analysis.

Since 2003 several reports have been published highlighting the adverse effect profile of this class of agents including the development of osteonecrosis of the jaw and renal insufficiency in patients treated with bisphos-phonates.^5,–7 The purpose of this discussion is to focus on the complications of supportive therapy with bisphosphonates in the management of multiple myeloma.

The mechanism of bisphosphonates-induced osteonecrosis is unclear. However, several predisposing factors have been identified. Periodontal disease, dentoalveolar surgery, prior trauma, corticosteroid therapy, immune-compromised state predisposing to increased risk of infection, possible vascular insufficiency and an underlying hypercoagulable state secondary to underlying malignancy have been implicated as possible etiologic or precipitating factors. Pathologically, the findings are indistinguishable from radiation-induced osteonecrosis. The unique predisposition for bony necrosis in the jaw may be related to the microenvironment in the maxilla and mandible secondary to compromised vascular supply and presence of oral microflora that may super-infect a poorly healing wound in the jaw bone. In addition, the constant microtrauma from jaw movement and lack of adequate bone remodeling in the presence of bisphosphonates may contribute to the genesis of osteonecrosis of the jaw (ONJ). Although spontaneous osteonecrosis is reported in a minority of cases, the majority of patients give a history of some prior dental or oral surgical manipulation. Several investigators have identified increasing duration of exposure, type of bisphosphonate, older age and prior history of a dental procedure as risk factors for the development of ONJ,^8,9 especially in patients with multiple myeloma. Median time to development of this complication varies with the type of bisphosphonate used for lytic bony disease. Pamidronate is 10 times more potent than clodronate, while zoledronate is 10,000 times more potent in its osteoclastic inhibitory activity than the oral bisphosphonate. It is therefore expected that the time to development of OJN is relatively shorter with the more potent agents.

Typical presentation is in the form of a nonhealing extraction socket, presence of exposed bone, gingival swelling or purulent discharge, when local debridement and antibiotics are ineffective. Often, a nonhealing ulcer or exposed bone may be detected on routine oral care that may remain asymptomatic, until superinfection sets in when swelling, pain, loosening of teeth and discharge may develop. Occasionally, pain in the jaw bone may be the only symptom without any evidence of radiological abnormalities. Eighty percent of patients report an antecedent dental procedure prior to presentation. Median age in one series was 68 years, mandible as the site of involvement is seen in two-thirds of the patients and maxillary involvement is in a third.

The true incidence of ONJ is hard to determine as the use of oral and intravenous bisphosphonates is widespread and osteonecrosis is likely an underrecognized entity. However, there has clearly been an increase in the observation and reporting of this entity over the last 3 years coinciding with the prolonged exposure to potent bisphosphonates for the management of symptomatic malignant bony disease. Durie and colleagues¹⁰ conducted a web-based survey. Of the 904 respondents with the diagnosis of myeloma, 62 patients had the diagnosis of ONJ and an additional 54 patients had suspicious findings. Of the patients with myeloma, 71% had received zoledronic acid, and 29% had received only pamidronate. The 36-month estimate of osteonecrosis or suspicious findings did not differ in this series between patients with myeloma and breast cancer. Additionally, no other concomitant treatment with thalidomide or corticosteroids conferred an additional risk. At 36 months censure of data, ONJ was noted to develop in 10% of patients receiving zoledronic acid, compared with 4% of patients receiving pamidronate. In this series, the mean time to the onset of ONJ among patients receiving zoledronic acid was 18 months, as compared with 6 years for patients receiving pamidronate. Other investigators have also reported retrospective series, suggesting an incidence of 1.8–12.8% in patients with multiple myeloma (Table 1 ). Durie’s web-based voluntary reporting of data and the other studies shown in Table 1  may have inherent limitations associated with such retrospective analyses. Some of these studies have analyzed records of patients treated with bisphosphonates prior to this complication of ONJ being widely recognized and therefore may contribute to underreporting of the incidence. Bannias et al report an incidence of ONJ increasing from 1.5% among patients treated for 4–12 months to 7.7% for treatment of 37–48 months.⁸ Additionally, the cumulative hazard of developing ONJ was significantly higher with exposure to zoledronic acid compared with pamidronate alone or sequential pamidronate and zoledronic acid (P < 0.001) In our updated series of 143 patients with documented ONJ,¹⁵ where data on duration of treatment were available by history in 56 patients, the median time to development of ONJ with the use of oral bisphosphonates, pamidronate and zoledronate was 54, 34 and 16 months respectively (Figure 1; see Color Figures, page 515). Therefore the risk of development of ONJ varies with the type of bisphosphonate used and duration of exposure, with more potent agents increasing the risk with shorter durations of exposure.

Oral surgical colleagues have reported a staging system from I to III based on the presence of symptoms such as pain, degree of swelling, ulceration, exposure of bone, response to topical antibiotic rinses, requirement for debridement and requirement of intravenous antibiotics. This staging system helps direct appropriate local treatment (Table 2, Figure 2a-c; see Color Figures, page 515 515 515 ).

There is no clear consensus on a standard treatment algorithm for this entity. The American Academy of Oral Medicine has published a position paper¹⁶ outlining the management of patients with ONJ. In addition, Novartis pharmaceuticals, the manufacturers of both the commercially available intravenous bisphosphonates, established an expert panel in June 2004 and updated in May 2006 that provided recommendations for the prevention, diagnosis and treatment of ONJ and has used this ‘white paper’ as an educational resource for all health care providers.¹⁷ If ONJ is suspected, panoramic and cross-sectional imaging is recommended to determine the extent of necrosis and the presence of a sequestrum or osteomyelitis, and to exclude other possible etiologies for these manifestations such as impacted teeth, bony cysts or metastatic disease. In the presence of soft tissue swelling or purulent discharge, microbial cultures may help identify concomitant super-infection and direct appropriate antimicrobial therapy. Microbial cultures may identify infection with actinomyces or mixed aerobic and anaerobic bacteria. Tissue biopsies are only recommended if metastatic disease is suspected. Any additional dental trauma is to be avoided as it may further precipitate and delay wound healing in this relatively avascular tissue. When biopsies are performed, tissue microbial cultures should be obtained.

Staging ONJ appropriately based on clinical and radiographic findings may be used to direct specific local and systemic therapy. Stage I disease as characterized by asymptomatic detection of exposed bone without soft tissue infection, may be managed conservatively with a non-surgical conservative approach to avoid further osseous injury. In addition, daily irrigation and oral antimicrobial rinses (0.12% chlorhexidine gluconate) are recommended. Clinical follow-up with an oral surgeon or dentist is recommended at least every 3 months. Dentures may be worn but should be adjusted to avoid further trauma to bone and soft tissues and should be removed at night. Stage II disease characterized by presence of symptoms around the area of exposed bone secondary to soft tissue swelling and/or bone infection may require culture-directed long-term and maintenance antimicrobial therapy, analgesic management, in addition to conservative measures outlined for stage I disease. Occasionally, minor bony debridement may be necessary to reduce sharp edges for reducing trauma to surrounding tissues. Stage III disease is characterized by the presence of a pathological fracture (not related to metastatic disease), exposed bone associated with soft tissue infection, which is not manageable with antibiotics alone due to the volume of necrotic bone. This degree of necrosis usually requires surgical debridement/resection to reduce the volume of necrotic bone in addition to conservative measures of analgesics, culture directed oral/intravenous antibiotics and oral antimicrobial rinses. Additional information and treatment recommendations from the International Myeloma Foundation for patients with ONJ are available on line at http://meloma.org/main.jsp?type=article&tab_id=0&id=1223.

The keystone in management of ONJ is its prevention. As risk factors and precipitating factors are now better understood, prevention of this entity with specific precautions would be ideal. Prior to initiating bisphosphonate therapy, except when indicated emergently for the management of hypercalcemia, it is recommended that all patients undergo a routine dental clinical examination and an appropriate radiographic study. All patients should be educated about this possible complication and instructed to avoid elective invasive dental procedures that may not heal completely prior to starting therapy. Once started on maintenance bisphosphonate therapy, patients should have routinely scheduled oral assessments at a frequency determined by dental and hematology/oncology care givers depending on general oral health and concomitant risk factors. Hematologists and oncologists are advised to perform a brief oral examination on routine follow-up visits looking for open ulcers, swelling of soft tissues, drainage, exposed or necrotic bone, or non-healing of an extraction socket. Dental surveillance includes a review of oral care, examination of dentures, if any, and adjustments as needed to avoid tissue injury, and routine dental cleanings without soft tissue injury. The expert panel also recommended avoiding tooth extractions and instead recommended endodontic (root canal) therapy where appropriate. When invasive dental procedures are to be performed electively, some investigators have recommended withholding intravenous bisphosphonates for 1 to 3 months before the procedure^18,19 and resuming treatments after oral healing is complete. Although this short period of interrupting the exposure to bisphosphonates is unlikely to change bone osteoclastic and remodeling environment, it may abrogate the anti-angiogenic properties of bisphosphonates and allow for soft tissue healing.

There are currently no data available to recommend discontinuing, maintaining or temporarily interrupting bisphosphonate therapy once ONJ has developed. As bisphosphonates are not metabolized and continue to maintain high concentrations within bone for long periods of time, it is unclear if altering the dose, schedule or type of bisphosphonate affects the symptomatic complication rate of ONJ. Continuing dialogue between hematologists and dental and oral surgeon colleagues and careful evaluation of risk/benefit assessment should guide further therapy on an individualized basis. When bisphosphonate therapy is to be resumed post development of ONJ, it may be prudent to use pamidronate in lieu of zoledronate and also to administer it at less frequent intervals than the standard monthly infusions.

Intravenous bisphosphonates are not metabolized and they are eliminated almost exclusively by the kidneys. Zoledronic acid and pamidronate clearance is dependent on the patient’s creatinine clearance. Bisphosphonates can cause deterioration in renal function and often are used in patients with baseline renal insufficiency. In clinical trials, the incidence of renal toxicity with administration of 4 mg of zoledronic acid as an intravenous infusion over 15 minutes is described from 9% to 15%. In these trials, renal function deterioration was defined as follows: for patients with normal baseline creatinine, increase of 0.5 mg/dL; for patients with abnormal baseline creatinine, increase of 1.0 mg/dL. Postmarketing publications report the possibility of development of significant renal dysfunction requiring dialysis.^20,21 The true incidence of this severe complication is unknown. Chang et al²⁰ reported 72 cases in the FDA Adverse Event Reporting System from August 2001 to March 2003. Forty-two patients were being treated for myeloma bony disease. Of the 72 patients, 27 required dialysis and 18 died. It is unclear if progressive disease, bisphosphonate use or concomitant usage of nonsteroidal drugs contributed to this toxicity. In this series, renal failure developed after an average of 56 days after administration of zoledronic acid. Twenty-five percent of patients received only one dose of zoledronate and developed renal failure after an average of 11 days. Bisphosphonate infusions were only resumed in clinical trials when serum creatinine returned to within 10% of baseline value. It is therefore recommended that serum creatinine be monitored prior to each dose of intravenous administration.

Renal toxicity is both dose and infusion time dependent. In a large phase III trial comparing zoledronate with pamidronate in over 1600 patients²² with myeloma and breast cancer, it was determined that decreasing the dose of zoledronic acid from 8 mg to 4 mg was associated with decreased risk of renal toxicity; in addition, prolonging the infusion time from 5 to 15 minutes was associated with decreased risk of renal toxicity. In this trial, at the recommended doses and infusion durations, the incidence of renal deterioration was 10.7% with zoledronate, similar to that with pamidronate (9.3%)

The mechanism of renal toxicity is thought to be secondary to acute tubular injury. Markowitz et al described the renal biopsy findings²³ in 6 patients who developed renal failure following zoledronate. The predominant finding was marked tubular degenerative changes. In addition, varying degrees of tubular atrophy, interstitial fibrosis, interstitial inflammation and mild to moderate vascular disease was noted. Pamidronate-induced toxic tubular necrosis has also been described.^24,25 Collapsing focal segmental glomerulosclerosis has been reported with use of high doses of pamidronate.²⁶ Unexplained albuminuria has also been reported with the administration of pamidronate. An expert panel recommends monitoring urine proteinuria, particularly for albuminuria periodically, for monitoring this toxicity in patients receiving pamidronate.

In view of the now published experience of renal toxicity with widespread use of potent bisphosphonates, it is important to serially monitor renal function, especially prior to each dose, maintain hydration status, adjust dosage for reduced creatinine clearance at baseline during initiation of therapy and only resume therapy when renal function returns to near baseline when an acute deterioration has occurred. As progressive myeloma can also cause renal deterioration, and the standard duration of bisphosphonate usage is indefinite, a constant risk/benefit assessment is warranted by treating clinicians.

The disease course of patients with multiple myeloma is often complicated by the presence of renal insufficiency, which is sometimes progressive to end-stage renal disease requiring dialysis. There are limited data available to provide clear recommendations on the use of bisphosphonates in this setting. Several case series^27,–30 describe the safe use of pamidronate in renal failure in doses of 60 to 90 mg administered every 1 to 2 months. Most of the safety data emerge from the management of hypercalcemia in patients with renal insufficiency and from series evaluating the usefulness of pamidronate in severe secondary hyperparathyroidism in patients undergoing hemodialysis.²⁹ Given the renal toxicity profile of bisphosphonates as described above it would be prudent to use a reduced dose of the bisphosphonate as an infusion over a more prolonged period of time. However, clinical trials are lacking in providing clear dose, schedule, and duration of infusion recommendations.

The initial trials that evaluated the efficacy of bisphosphonate in multiple myeloma compared 9 months of intravenous therapy to placebo. No Level I evidence is available to make recommendations at this time regarding the choice of agent, duration of therapy and any modifications to the dosing schedule. However, recently the Mayo Clinic consensus statement for the use of bisphosphonates in multiple myeloma¹⁸ recommended monthly infusions of bisphosphonates (preferably pamidronate in view of the lower risk of ONJ with this agent and lack of proven inferiority to zoledronate in preventing bone disease) until 2 years and withholding further treatment if the disease state has achieved a remission or a stable plateau phase off treatment for myeloma. If the patient still requires active therapy, this panel recommends reducing the frequency of bisphosphonate administration to every 3 months.

Bisphosphonates are an important adjunct to systemic biological and chemotherapeutic management of multiple myeloma. They clearly provide an important enhancement in the quality of life by reducing skeletal related events and reducing pain; however, now with widespread use, post-marketing experience suggests caution toward monitoring for toxicities. Osteonecrosis of the jaw and renal toxicities can lead to significant morbidity and impairment in quality of life. Future trials must evaluate the optimal dosing, duration of treatment, potency of agent used, frequency of administration and quality of life end points to enable clinicians to make evidence-based recommendations in the management of their patients with symptomatic malignant bony disease.