Journal of the National Cancer Institute Advance Access originally published online on June 27, 2007
JNCI Journal of the National Cancer Institute 2007 99(13):1016-1024; doi:10.1093/jnci/djm025
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ARTICLES |
Intravenous Bisphosphonate Therapy and Inflammatory Conditions or Surgery of the Jaw: A Population-Based Analysis
Affiliations of authors: Departments of Preventive Medicine and Community Health (GSW, YFK, JLF, JSG) and Internal Medicine (YFK, JLF, JSG), and Sealy Center on Aging (GSW, YFK, JLF, JSG), University of Texas Medical Branch, Galveston, TX
Correspondence to: Gregg Wilkinson, PhD, Department of Preventive Medicine and Community Health, University of Texas Medical Branch, 301 University Blvd, Galveston, TX 77555-1153 (e-mail: gswilkin{at}utmb.edu).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Background: Recent reports have identified an association between osteonecrosis of the jaw or facial bones and treatment with nitrogen-containing intravenous bisphosphonates. We investigated this association by use of data from the Surveillance, Epidemiology, and End Results (SEER) program linked to Medicare claims.
Methods: We identified 16073 cancer patients who were diagnosed between January 1, 1986, and December 31, 2002, and were treated intravenously with the bisphosphonates pamidronate and/or zoledronic acid between January 1, 1995, and December 31, 2003. We matched 28698 bisphosphonate nonusers, at a 2:1 ratio, to 14349 bisphosphonate users on month and year of the first bisphosphonate administration received by users, cancer type, age, sex, risk factors for osteonecrosis (diabetes, alcoholism, cigarette smoking, obesity, hyperlipemia, pancreatitis, or chemotherapy with L-asparaginase), bone metastasis, and SEER program geographic region. Patients were followed until the study's end on December 31, 2003; loss of coverage from Medicare Parts A and B; or one of the following outcomes: a diagnosis of inflammatory conditions or osteomyelitis of the jaw, surgery on the facial bones, or death, whichever occurred first.
Results: Use of intravenous bisphosphonates was associated with an increased risk of jaw or facial bone surgery (hazard ratio [HR] = 3.15, 95% confidence interval [CI] = 1.86 to 5.32) and an increased risk of being diagnosed with inflammatory conditions or osteomyelitis of the jaw (HR = 11.48, 95% CI = 6.49 to 20.33), compared with nonuse. The absolute risk at 6 years for any jaw toxicity was 5.48 events per 100 patients using intravenous bisphosphonates and 0.30 events per 100 patients not using such drugs. The risk of each outcome increased as cumulative dose increased (e.g., for 4–8 infusions, HR for operations on the jaw and facial bones = 3.63, 95% CI = 0.77 to 17.08; for more than 21 infusions, HR = 9.18, 95% CI = 1.74 to 48.53).
Conclusion: Users of intravenous bisphosphonates had an increased risk of inflammatory conditions, osteomyelitis, and surgical procedures of the jaw and facial bones. The increased risk may reflect an increased risk for osteonecrosis of the jaw.
Prior knowledge Associations between osteonecrosis of the jaw or facial bones and treatment with intravenous bisphosphonates have been reported from small studies. Study design Population-based cohort study of data from the Surveillance, Epidemiology, and End Results program linked to Medicare claims. Contribution In a large study of 14 349 bisphosphonate users and 28 698 nonusers, use of bisphosphonates, compared with its nonuse, was associated with substantially increased risks of jaw or facial surgery or of being diagnosed with inflammatory conditions or osteomyelitis of the jaw. Implications Increased risks of jaw complications among bisphosphonate users may reflect an increased risk for osteonecrosis of the jaw. Increasing oral hygiene and avoiding tooth extractions may lead to a decrease in the incidence of jaw complications. Given the increasing use of intravenous bisphosphonate therapy for treating osteoporosis, these patients should be followed carefully for adverse bone events involving facial bones. Limitations Because there is no disease code for facial or jaw osteonecrosis or aseptic necrosis of the jaw, the authors used surrogates, which could introduce some misclassification. Reliance on Medicare claims data required that patients and physicians be aware of their condition or that patients had undergone treatment by a clinician who billed Medicare; underreporting could have occurred if patients had unreported disease or were treated by a dentist who could not charge Medicare.
|
Bisphosphonates have been used in the United States to treat osteoporosis since 1977 (1). Highly potent intravenously administered nitrogen-containing bisphosphonates were introduced for the treatment of cancer-related bone lesions and hypercalcemia in 1991 (i.e., introduction of pamidronate) and in 2001 (i.e., introduction of zoledronic acid) (2,3). Intravenous bisphosphonates are also effective in preventing osteopenia in patients with prostate cancer undergoing androgen deprivation therapy (4,5). Starting in 2003, a growing number of case reports and case series have been published that link bisphosphonate administration, particularly bisphosphonates administered intravenously, with the previously rare condition of osteonecrosis of the jaw and facial bones (6–11). A causal link between bisphosphonates and jaw osteonecrosis was originally challenged by the pharmaceutical manufacturer Novartis Pharmaceuticals (12) and others (13,14), but the association was striking, and reports proliferated. In September 2004, Novartis Pharmaceuticals added osteonecrosis of the jaw to the list of possible toxicities in product inserts for pamidronate and zoledronic acid.
Until recently, most reports of jaw osteonecrosis appeared in oral surgery journals. A few cohort studies have recently been reported that involve small patient populations (15–18), and one Web-based survey of 1203 users (19) has also been reported. Recent press reports (20,21) and a systematic review of data from 368 case patients reported in 10 case series studies (1) have raised the visibility of this association in the medical community and general public.
The mechanism that underlies the association between intravenous bisphosphonate treatment and jaw osteonecrosis is not understood. Current explanations include an infectious etiology, the loss of blood supply, or the suppression of bone turnover (1,7,22,23). There is also a lack of information on the epidemiology of the toxicity. There is no information from prospective trials and only limited information from observational cohorts on incidence, time course, or other risk factors (15–19).
Recent studies have used tumor registry data from the Surveillance, Epidemiology, and End Results (SEER) program linked to Medicare claims data to examine toxicities of parenterally administered drugs (24,25). A challenge in using this approach to investigate toxicity from intravenous bisphosphonates is that there is no specific code in the ninth revision of the International Classification of Diseases (ICD-9) for osteonecrosis of the jaw. Therefore, we chose two outcomes with ICD-9 codes that should be closely related to jaw osteonecrosis: 1) a diagnosis of inflammatory conditions or osteomyelitis of the jaw and 2) an operation on the jaw or facial bones. In this study, we investigated the association between intravenous treatment with pamidronate or zoledronic acid and a subsequent diagnosis of inflammatory conditions or osteomyelitis of the jaw or an operation on the jaw or facial bones.
|
Patients and Methods |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Data Sources
The SEER–Medicare linked database contains tumor characteristics and demographic information for Medicare beneficiaries newly diagnosed with cancer in geographic regions covered by the SEER program (26). In 2002, the program included registries that covered approximately 26% of the US population. Medicare data were linked to these patients' SEER records through a collaborative program between the National Cancer Institute and the Center for Medicare and Medicaid Services. Approximately 94% of the SEER records for beneficiaries aged 65 years and older were matched to the patients' claims for medical services.
Study Subjects
We first selected individuals diagnosed with any cancer during the period from January 1, 1986, through December 31, 2002. Using the Health Care Procedure Coding System (HCPCS) drug administration codes J2430 for pamidronate and J3487 for zoledronic acid, we then identified subjects (bisphosphonate recipients) among those who had received at least one injection of the bisphosphonate pamidronate disodium (Aredia) or zoledronic acid (Zometa) from January 1, 1995, through December 31, 2003. Patients who were not enrolled in both Medicare part A and part B for the 12 months before the first bisphosphonate injection, who were members of a health maintenance organization for 12 months before the first injection, or whose cancer was first diagnosed by autopsy or indicated on a death certificate were excluded from this study. Individuals who experienced a study outcome in the 12 months before study entry were also excluded. Overall, data were available for 16073 bisphosphonate users for dose–response analyses (see Table 3).
For analyses comparing site of jaw toxicity associated with bisphosphonate use (see Fig. 1 and Tables 1 and 2), we matched each bisphosphonate user to two nonusers. Nonusers were selected from among cancer patients who had not received any bisphosphonate therapy from January 1, 1995, through December 31, 2003. Nonusers who were not enrolled in both Medicare part A and part B for the 12 months before study entry, whose cancer was first diagnosed by autopsy or indicated on a death certificate, or who experienced a study outcome in the 12 months before study entry were excluded from the study. Study entry for a nonuser was the month and year of the first bisphosphonate injection received by the user to which the nonuser was matched. We used a two-step sequential matching process. In the first step, patients who had not received bisphosphonates were matched with those who had by type of cancer (breast, lung, prostate, multiple myeloma, or all other cancers), age at bisphosphonate administration (<65, 65–69, 70–74, 75–79, or 
80 years), month and year of the first bisphosphonate injection, sex (male or female), number of risk factors (0, 1, or 2) for osteonecrosis (diabetes, alcoholism, cigarette smoking, obesity, hyperlipemia, pancreatitis, or chemotherapy with L-asparaginase) (27), bone metastasis (yes or no), and SEER region (Connecticut, Detroit, Hawaii, Iowa, New Mexico, Seattle, Utah, Atlanta, Rural Georgia, Kentucky, Louisiana, New Jersey, or California). We were able to match 12484 users to 24968 nonusers with this method. The second step was to match the remaining users and nonusers with somewhat less stringent criteria: type of cancer (same as above), age (<65, 65–74, or 75 years), sex, and number of risk factors (0 or 1). This step yielded an additional 1865 users matched to 3730 nonusers. All users who could not be matched with two nonusers in either the first or second step had a diagnosis of multiple myeloma (n = 1724). All matched users (n = 14349) and nonusers (n = 28698) had the same selection criteria of Medicare enrollment, reporting sources, and exclusion criteria.
|
|
|
Risk Factor Scores
As stated above, the HCPCS drug administration codes J2430 and J3487 were used to identify patients who received intravenous bisphosphonate therapy. Patient demographic characteristics (such as age, sex, and ethnicity), type of cancer, and tumor characteristics were derived from the SEER Patient Entitlement and Diagnosis Summary File. A risk factor index for osteonecrosis was developed by summarizing the occurrence among study subjects of the following conditions: diabetes (ICD-9 code 250.x), alcoholism (codes 291.x and 303.x), cigarette smoking (code 305.1), obesity (code 278.0), hyperlipidemia (codes 272.0–272.4), pancreatitis (codes 577.0 and 577.1), and chemotherapy with L-asparaginase (code J9020) (27). In addition, a comorbidity score was calculated by use of Klabunde's adaptation of the Charlson comorbidity index in the SAS macro provided by the National Cancer Institute (28). For both the risk factor index and comorbidity score, we searched the inpatient, outpatient, and physician files for any of those diagnoses in the 12 months before study entry, as previously described (27,28). To avoid double counting the effect of diabetes, we removed diabetes from the Klabunde comorbidity score above. Receipt of parenteral corticosteroids and diagnosis of bone metastasis in the 12 months before to the 1 month after date of study entry (for bisphosphonate users, month and year of first bisphosphonate injection; for nonusers, month and year of first bisphosphonate injection of the user to which a nonuser is matched) were assessed with HCPCS codes J1094, J1100, J1700, J1710, J1720, and J2650 and with ICD-9 code 198.5.
Outcomes
The study outcomes were diagnoses of inflammatory conditions or osteomyelitis of the jaw (ICD-9 526.4) and operations on the facial bones and joints (ICD-9 procedure codes 76.0–76.6 and 76.9 or Current Procedure Terminology codes 21025, 21026, 21030, 21032–21034, 21040, 21045–21047, 21081, 21127, 21193–21196, 21198, 21199, 21206, 21210, 21215, and 21244–21249) (28). In preliminary analyses, we also assessed the utility of ICD-9 code 773.40 (aseptic necrosis—site unspecified). However, that diagnosis appeared to be used for aseptic necrosis at sites other than the jaw and facial bones. For example, more than 60% of patients with the code 733.40 also had a site-specific code for aseptic necrosis (e.g., aseptic necrosis of the hip or the shoulder). Therefore, we did not include 733.40 as an outcome in our analyses. However, in the analyses that did include it, the direction and statistical significance of the results were unaltered.
Dose Estimation
Medicare claims forms document the number of units administered in each bisphosphonate injection. We estimated cumulative dose by calculating the number of units recorded during the period of follow-up for each bisphosphonate user. Because the units for pamidronate and zoledronic acid differ, we used 4 mg of zoledronic acid and 90 mg of pamidronate as equivalent doses. Both the number of units and the equivalent dose were analyzed as continuous and as categorical variables (number of injections: 
3, 4–8, 9–13, 14–21, or >21). We considered the number of injections to be an indicator of dose and compared the number of injections with the number of units and dose-equivalent measures.
Statistical Analysis
Unadjusted event-free survival was estimated by the Kaplan–Meier method (29). Multivariable survival analyses (see Tables 2 and 3) were performed by use of Cox proportional hazards regression with the dependent variable being time to first occurrence of a study outcome (i.e., inflammatory conditions or osteomyelitis of the jaw or surgery on the jaw or facial bones). We tested the assumption of proportionality in the Cox model by determining that the logarithm of the baseline cumulative hazard rates and the Schoenfeld residuals were proportional with follow-up time (29,30). Patients were censored at death, at loss of Medicare part A and part B coverage, or at the end of the study (i.e., December 31, 2003). Patient and cancer characteristics and treatment with intravenous bisphosphonate therapy were treated as independent variables.
|
We assessed the dose–response relationship among intravenous bisphosphonate recipients in two ways. First, cumulative dose over time was modeled in Cox proportional hazard models as a time-dependent covariate among all study patients who received bisphosphonate therapy (n = 16073). Second, after restricting eligibility to recipients with more than 12 months of follow-up time (n = 6343), we estimated the hazard ratio (HR) for any study outcome by cumulative dose category for injections received during the first 12 months. All analyses were performed with the SAS package of computer programs version 9.1 (SAS Institute, Cary, NC). We did not estimate power before beginning the study. All statistical tests were two-sided.
|
Results |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Characteristics of all intravenous bisphosphonate users and of the matched intravenous bisphosphonate users and nonusers are presented in Table 1. Among all intravenous bisphosphonate users, 82.3% were diagnosed with cancer of the lung, breast, or prostate or with multiple myeloma. The use of intravenous bisphosphonates to treat patients in the study increased from 170 patients in 1995 to 5348 patients in 2003. Although zoledronic acid was introduced in 2001, a separate drug administration code for it did not exist until late 2002. Therefore, we had no reliable information on type of intravenous bisphosphonate used in 2001 and 2002. In 2003, 15.5% of new bisphosphonate users received pamidronate, 80.5% received zoledronic acid, and 4.0% received both. Among patients who received intravenous bisphosphonate therapy, 28.0% also received parenteral corticosteroids compared with 12.3% in matched nonusers. In addition, 60.1% of users had bone metastases. The two-step matching procedure that we used appears to have been successful because the distribution of age, sex, type of cancer, risk factors, and year of drug administration were not statistically significantly different between users and matched nonusers. The differences in distribution of SEER regions between users and matched nonusers were also small. However, bone metastasis (60.4% versus 52.6%) and use of intravenous steroids (29.8% versus 12.3%) were still higher among intravenous bisphosphonate users than nonusers, respectively.
We investigated event-free survival among the matched intravenous bisphosphonate users and nonusers by use of Kaplan–Meier curves. Results for three outcomes are shown in Fig. 1: surgery on facial bones or jaw, diagnosis of inflammatory conditions or osteomyelitis of the jaw, and either of these outcomes. As noted above, we used a diagnosis of inflammatory conditions or osteomyelitis of the jaw or operations on the jaw and facial bones as potential indicators for jaw osteonecrosis because there is no specific ICD-9 code for the diagnosis of jaw osteonecrosis. By 18 months, the survival curves for the bisphosphonate users differed statistically significantly from those of the nonusers for each outcome (P<.001) by log-rank test. The curves for users and nonusers continued to diverge over time. At 6 years, 5.48% (95% confidence interval [CI] = 3.63% to 7.33%) of users had been diagnosed with inflammatory conditions or osteomyelitis of the jaw or underwent jaw or facial surgery, compared with 0.30% (95% CI = 0.19% to 0.42%) of nonusers, resulting in an absolute risk of 5.48 events per 100 bisphosphonate users and an absolute risk of 0.30 events per 100 nonusers at 6 years.
We next investigated whether the association between intravenous bisphosphonate therapy and jaw toxicity was independent of other factors that might contribute to jaw toxicity (Table 2). The risk index is composed of specific comorbidities associated with increased risk of aseptic necrosis, including diabetes, alcoholism, cigarette smoking, obesity, hyperlipemia, pancreatitis, and chemotherapy with L-asparaginese (27). After adjustment for these potential confounders, intravenous bisphosphonate use, compared with nonuse, was strongly associated with an elevated risk of a diagnosis of inflammatory conditions or osteomyelitis of the jaw (HR = 11.48, 95% CI = 6.49 to 20.33), operations on the facial bones (HR = 3.15, 95% CI = 1.86 to 5.32), or either outcome (HR = 4.94, 95% CI = 3.33 to 7.34) (Table 2).
We also conducted multivariable analyses in which we examined dose–response relationships among all bisphosphonate users (Table 3). Increases in the risk of inflammatory conditions or osteomyelitis of the jaw were observed with increases in the cumulative intravenous bisphosphonate dose (for 4–8 dose-equivalent units of intravenous bisphosphonate compared with 3 units or fewer, adjusted HR = 1.56, 95% CI = 0.67 to 3.60; for at least 21 units, HR = 3.57, 95% CI = 1.46 to 8.75). Similar patterns were observed for the risk of operations on the jaw and facial bones (for 4–8 dose-equivalent units compared with 3 units or fewer, HR = 3.63, 95% CI = 0.77 to 17.08; for at least 21 units, HR = 9.18, 95% CI = 1.74 to 48.53) and for either operations on the jaw or facial bones or inflammatory conditions or osteomyelitis of the jaw (for 4–8 dose-equivalent units compared with 3 units or fewer, HR = 1.61, 95% CI = 0.76 to 3.44; for at least 21 units, HR = 3.67, 95% CI = 1.58 to 8.51). For all three outcomes investigated, hazard ratios increase as cumulative dose increases. Other factors associated with any adverse outcome were year of drug administration, SEER registry geographic region, and a diagnosis of multiple myeloma.
The dose–response analyses in Table 3 used the cumulative bisphosphonate dose for the entire study period as a time-dependent covariate. In a separate analysis, we assessed the total bisphosphonate dose received for the 12 months after the first dose and assessed jaw toxicity starting at 12 months (Table 4). These analyses were restricted to the 6343 patients with more than 12 months of follow-up after receipt of a bisphosphonate. Each one-dose increase in bisphosphonates received in the first 12 months was associated with an 8% increase in the risk of any adverse jaw outcome (HR = 1.08, 95% CI = 1.02 to 1.14).
|
In the multivariable models shown in Tables 2 and 3, we also tested for interactions between bisphosphonate use and all other covariates (such as age, type of cancer, or sex of patient). There were no statistically significant interactions. Thus, the association of bisphosphonate use with jaw toxicity did not vary by age, sex, type of cancer, or other characteristics analyzed.
|
Discussion |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
We found that use of intravenously administered nitrogen-containing bisphosphonates, compared with nonuse, was associated, in a dose–response manner, with an increase in Medicare claims related to jaw and facial bone disease. This study is, to our knowledge, the first of a large, representative population-based cohort of bisphosphonate users and matched nonusers, and our results are consistent with those of several previous studies (15,16,19). In a study of 252 patients using bisphosphonates, Bamias et al. (15) reported cumulative hazards of 3% (95% CI = 1% to 5%), 7% (95% CI = 1% to 13%), and 11% (95% CI = 3% to 19%) for osteonecrosis of the jaw at 24, 36, and 48 months after first use, respectively. Dimopoulos et al. (16) reported that, for a 10-year period, 15 patients (7.4%) of the 202 multiple myeloma patients studied who received intravenous bisphosphonates were diagnosed with osteonecrosis of the jaw.
In this study, we found not only a strong association between intravenous bisphosphonate therapy and jaw and facial bone disease but also a pattern in which risk estimates for this disease increased as the cumulative dose of intravenous bisphosphonate increased. This finding is consistent with the observation that bisphosphonates have a long half-life and remain in the bone for an extensive period of time (15). It has been speculated that oversuppression of bone turnover may be related to the length and amount of exposure and the long half-life that characterize bisphosphonates (1,15).
Several previous studies have found that osteonecrosis of the jaw is more likely to occur among patients treated with zoledronic acid than with other bisphosphonate formulations. The largest of these studies, a study of 1203 patients who had received either pamidronate or zoledronic acid and who responded to an online survey (19), reported that 10% of the patients using zoledronic acid and 4% of those using pamidronate developed osteonecrosis of the jaw or had suspicious symptoms after 36 months of follow-up. In a study of 303 multiple myeloma patients, Zervas et al. (17) reported that a higher risk for osteonecrosis of the jaw was associated with use of zoledronic acid than use of pamidronate (relative risk = 9.5, 95% CI = 1.0 to 83.2). Bamias et al. (15) found that all cases of jaw osteonecrosis occurred among patients treated with zoledronic acid (n = 17), either alone (n = 7) or in combination with pamidronate (n = 9) or ibandronate (n = 1). They also reported that the cumulative risk associated with duration of treatment increased from 1% after 1 year to 21% after 3 years of treatment with zoledronic acid compared with a cumulative risk of 0% after 2 years to 7% after 4 years of treatment with pamidronate alone or in combination with zoledronic acid (19). Thus, the risk of osteonecrosis of the jaw appears to be more strongly associated with zoledronic acid use than with pamidronate use. We were unable to address the issue of relative toxicity of zoledronic acid versus pamidronate in our analyses because, although zoledronic acid was introduced in 2001, there was a delay in issuing a separate billing code for zoledronic acid until 2002.
No clear associations were found by us between adverse outcomes and sex, ethnicity, type of cancer, comorbidity, or with known risk factors for jaw and facial bone disease, including diabetes, alcoholism, cigarette smoking, obesity, hyperlipidemia, pancreatitis, chemotherapy with L-asparaginase, or receipt of parenteral corticosteroids (27). That is, the risks for inflammatory disease or osteomyelitis of the jaw and for jaw surgery after exposure to intravenously administered bisphosphonate were not affected by the presence of other risk factors.
This study has several limitations. First, because there is no ICD-9 code for facial or jaw osteonecrosis or aseptic necrosis of the jaw, we relied on indirect measures, including operations on the jaw or facial bones or a diagnosis of inflammatory conditions or osteomyelitis of the jaw (31). Some misclassification was likely present because it was not possible to separate patients with osteonecrosis from those with the other conditions in these categories. Second, our reliance on Medicare claims required that patients and physicians be aware of their condition or had undergone treatment by a clinician who billed Medicare. Many patients may have had unreported disease that would only be recognized by screening patients who were receiving or had received bisphosphonate therapy (32), and many patients may have been treated by dentists, rather than by an oral surgeon, without a Medicare charge being generated. If such underreporting existed, however, it would have caused us to underestimate the number of osteonecrosis cases among both bisphosphonate users and nonusers, which would have caused us to underestimate the absolute risk of jaw complications associated with use of intravenous bisphosphonates. Third, Medicare claims data provide no information on oral bisphosphonate use. In several case series of jaw osteonecrosis, fewer than 10% of such patients had received oral bisphosphonates (1,6–12). A recent case–control study using health insurance claims data (32) reported a nonstatistically significant 15% increase in jaw procedures associated with oral bisphosphonate use. Fourth, as noted above, we are unable to assess the relative toxicities of pamidronate versus zoledronic acid because of the delay in introduction of a specific Medicare billing code for zoledronic acid. In 2006, another nitrogen-containing bisphosphonate, ibandronate, was approved by the Food and Drug Administration for intravenous administration to treat osteoporosis (33), but our study contains no information on use of that agent. Also, as previously mentioned, there were too few cases in the SEER–Medicare data to allow a comparison between pamidronate and zoledronic acid on jaw toxicity. Fifth, a limitation common to all observational studies is that an association does not prove a causal link. Others (12–14) have suggested, for example, that the underlying cancer might be responsible for the osteonecrosis and also that patients using intravenous bisphosphonates are likely to have used other therapies, such as corticosteroids, that might be responsible for the toxicity.
Our findings, however, would argue against all of these limitations. First, the presence of bone metastases is not strongly associated with the outcomes in our study. Among bisphosphonate users, the risk of osteomyelitis or jaw surgery was similar in cancer patients with bone metastases and in those without metastases (Table 3). Second, in multivariable analyses (Tables 2, 3, and 4), none of the risk factors, such as multiple comorbidities or use of parenteral corticosteroids, had much impact on the relationship between intravenous bisphosphonate use and adverse bone outcome. Third, the dose–response relationship between intravenous bisphosphonate use and jaw disease supports the possibility of a causal association.
Even if the association is ultimately shown to be causal, the toxicity of any therapy must be balanced against its efficacy. Intravenous bisphosphonate treatment is associated with a substantial benefit in the form of decreased bone loss and fractures experienced by cancer patients who have bone metastases (34). For example, in clinical trials, intravenous pamidronate was statistically significantly associated with a 19% absolute reduction in skeletal events (P<.001) (i.e., pathologic fracture or need for radiation or surgery) in patients with metastatic breast cancer involving bone and a 17% absolute reduction of skeletal events (P<.001) in patients with multiple myeloma (35). Moreover, increasing attention is being given to promoting oral hygiene and avoiding tooth extractions in patients who have received intravenous bisphosphonates (2,3,7,23). Such attention may lead to a decrease in incidence of jaw complications.
Future research should include large clinical trials of intravenous bisphosphonate users who are followed long enough for the development of adverse events such as osteonecrosis of the jaw. Given the increasing use of intravenous bisphosphonate therapy for patients with severe osteoporosis, it is important that patients receiving therapy with both established and new formulations of intravenous bisphosphonates be followed carefully for adverse bone events involving the facial bones that may be indicative of osteonecrosis of the jaw. Large population-based observational studies that rely on Medicare and commercial health insurance claims would also be informative, as would studies of data contained in electronic medical records databases, such as the General Practice Research Database in Great Britain.
|
Funding |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
Supported in part by grants (P50CA105631, R24HS011618) from the Public Health Service. The funding bodies had no role in data extraction and analyses, in the writing of the manuscript, or in the decision to submit the manuscript for publication.
Funding to pay the Open Access publication charges for this article was provided by US Public Health Service Grant No. P50CA105631.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
This study used the linked SEER–Medicare database. The interpretation and reporting of these data are the sole responsibilities of the authors. The authors acknowledge the efforts of the Applied Research Program, National Cancer Institute; the Office of Research, Development and Information, Centers for Medicare and Medicaid Services; Information Management Services, Inc; and SEER Program tumor registries in the creation of the SEER–Medicare database.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsPatients and MethodsResultsDiscussionFundingReferencesNotes |
|---|
(1) Woo SB, Hellstein JW, Kalmar JR. Systematic review: bisphosphonates and osteonecrosis of the jaws. Ann Intern Med (2006) 144:753–61.
(2) U.S. Food and Drug Administration. Label and approval history, AREDIA, NDA no. 020035. Available at: http://www.accessdata.fda.gov/. [Last accessed: June 9, 2006.].
(3) U.S. Food and Drug Administration. Label and approval history, ZOMETA, NDA no. 021223. Available at: http://www.accessdata.fda.gov/. [Last accessed: April 26,2006.].
(4) Smith MR, McGovern FJ, Zietman AL, Fallon MA, Hayden DL, Schoenfeld DA, et al. Pamidronate to prevent bone loss during androgen deprivation therapy for prostate cancer. N Engl J Med (2001) 345:948–55.
(5) Smith MR, Eastham J, Gleason DM, Shasha D, Tchekmedyian S, Zinner N. Randomized controlled trial of zoledronic acid to prevent bone loss in men receiving androgen deprivation therapy for nonmetastatic prostate cancer. J Urol (2003) 169:2008–12.[CrossRef][ISI][Medline]
(6) Marx RE. Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic. J Oral Maxillofac Surg (2003) 61:1115.[CrossRef][ISI][Medline]
(7) Marx RE, Sawatari Y, Fortin M, Broumand V. Bisphosphonate-induced exposed bone (osteonecrosis/osteoporosis) of the jaws: risk factors, recognition, prevention, and treatment. J Oral Maxillofac Surg (2005) 63:1567–75.[CrossRef][ISI][Medline]
(8) Ruggiero SL, Mebrotra B, Rosenberg TJ, Engroff SL. Osteonecrosis of the jaws associated with the use of bisphosphonates: a review of 63 cases. J Oral Maxillofac Surg (2004) 62:527–34.[CrossRef][ISI][Medline]
(9) Bagan JV, Murillo J, Jimenez Y, Poveda R, Milian MA, Sanchis JM, et al. Avascular jaw osteonecrosis in association with cancer chemotherapy: series of 10 cases. J Oral Pathol Med (2005) 34:120–3.[CrossRef][ISI][Medline]
(10) Migliorati CA, Schubert MM, Peterson DE, Seneda LM. Bisphosphonate associated osteonecrosis of mandibular and maxillary bone: an emerging oral complication of supportive cancer therapy. Cancer (2005) 104:83–93.[CrossRef][ISI][Medline]
(11) Farrugia MC, Summerlin DJ, Krowiak E, Huntley T, Freeman S, Borrowdale R, et al. Osteonecrosis of the mandible or maxilla associated with the use of new generation bisphosphonates. Laryngoscope (2006) 116:115–20.[CrossRef][ISI][Medline]
(12) Tarassoff P, Csermak K. Avascular necrosis of the jaws: risk factors in metastatic cancer patients [letter]. J Oral Maxillofac Surg (2003) 61:1238–9.[ISI][Medline]
(13) Bone HG, Hosking D, Devogelaer JP, Tucci JR, Emkey RD, Tonino RP, et al. Ten years’ experience with alendronate for osteoporosis in postmenopausal women. N Engl J Med (2004) 350:1189–99.
(14) Bone HG, Santora AC. Ten years of alendronate treatment for osteoporosis in postmenopausal women [letter]. N Engl J Med (2004) 351:191–2. author reply.
(15) Bamias A, Kastritis E, Bamia C, Moulopoulos LA, Melakopoulos I, Bozas G, et al. Osteonecrosis of the jaw in cancer after treatment with bisphosphonates: incidence and risk factors. J Clin Oncol (2005) 23:8580–7.
(16) Dimopoulos MA, Kastritis E, Anagnostopoulos A, Melakopoulos I, Gika D, Moulopoulos LA, et al. Osteonecrosis of the jaw in patients with multiple myeloma treated with bisphosphonates: evidence of increased risk after treatment with zoledronic acid. Haematologica (2006) 91:968–71.
(17) Zervas K, Verrou E, Teleioudis E, Vahtsevanos K, Banti A, Mihou D, et al. Incidence, risk factors and management of osteonecrosis of the jaw in patients with multiple myeloma: a single-centre experience in 303 patients. Br J Haematol (2006) 134:620–3.[CrossRef][ISI][Medline]
(18) Walter C, Groetz KA, Kunkel M, Al-Nawas B. Prevalence of bisphosphonate associated osteonecrosis of the jaw within the field of osteonecrosis. Support Care Cancer (2007) 15:197–202.[CrossRef][ISI][Medline]
(19) Durie BGM, Katz M, Crowley J. Osteonecrosis of the jaw and bisphosphonates. N Engl J Med (353) 1:99–100.
(20) Rubin R. Drug linked to death of jawbone. USA Today (2005 March 13) Available at: http://www.usatoday.com/news/health/2005-03-13-jawbone-deaths_x.htm. [Last accessed: June 3, 2007.].
(21) Kolata G. Drug for bones is newly linked to jaw disease (2006 June 2.) New York Times. Available at: http://www.nytimes.com/2006/06/02/health/02jaw.html. [Last accessed: June 3, 2007.].
(22) Hellstein JW, Marek CL. Bisphosphonate osteochemonecrosis (bis-phossy aw): is this the phossy jaw of the 21st century? J Oral Maxillofac Surg (2005) 63:682–9.[CrossRef][ISI][Medline]
(23) Cheng A, Mavrokokki A, Carter G, Stein B, Fazzalari NL, Wilson DF, et al. The dental implications of bisphosphonates and bone disease. Aust Dent J (2005) 50(Suppl 2):S4–13.[ISI][Medline]
(24) Du XL, Osborne C, Goodwin JS. Population-based assessment of hospitalizations for toxicity from chemotherapy in older women with breast cancer. J Clin Oncol (2002) 20:4636–42.
(25) Shahinian VB, Kuo YF, Freeman JL, Goodwin JS. Population-based analysis of fracture risk after androgen deprivation therapy for prostate cancer. N Engl J Med (2005) 352:154–64.
(26) Department of Health and Human ServicesNational Institutes for HealthNational Cancer Institute (NCI). SEER-Medicare linked database. Available at: http://healthservices.cancer.gov/seermedicare. [Last accessed: October 24, 2006.].
(27) Jones JP Jr. Osteonecrosis. In: Arthritis and allied conditions—Koopman WJ, ed. (1997) Baltimore (MD): Williams and Wilkins. 1923–42.
(28) Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. J Clin Epidemiol (2000) 53:1258–67.[CrossRef][ISI][Medline]
(29) Klein JP, Moeschberger ML. Survival analysis: techniques for censored and truncated data (1997) NewYork: Springer-Verlag.
(30) Allison JP. Survival analysis using the SAS system (2001) Cary (NC): SAS Publishing.
(31) Practice Management Information Corporation. International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) (2001) Los Angeles (CA): Practice Management Information Corporation.
(32) Zavras A, Zhu S. Bisphosphonates are associated with increased risk for jaw surgery in medical claims data: is it osteonecrosis? J Oral Maxillofac Surg (2006) 62:527–34.[CrossRef]
(33) Hitti M. FDA OKs first IV osteoporosis drug: intravenous Boniva is now alternative to pill form. WebMD Medical News 2006 Jan 13. Available at: http://www.webmd.com/content/article/117/112556. [Last accessed: June 19, 2006.].
(34) Guise TA. Bone loss and fracture risk associated with cancer therapy. Oncologist (2006) 11:1121–31.
(35) Novartis Pharmaceuticals Corporation. Aredia. In: 2006 physicians desk reference (2006) Montval (NJ): Thompson PDR. 2168–73.
Manuscript received December 14, 2006; revised April 26, 2007; accepted May 22, 2007.
Correspondence about this Article
Editorial about this Article
Related Articles in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 155.
J Natl Cancer Inst 2007 99: 986-987.
J Natl Cancer Inst 2007 99: 981.
J Natl Cancer Inst 2007 99: 981.
This article has been cited by other articles:
|
|
 |
|
  S. H. Giordano, S. Fang, Z. Duan, Y.-F. Kuo, G. N. Hortobagyi, and J. S. Goodwin Use of Intravenous Bisphosphonates in Older Women with Breast Cancer Oncologist, May 1, 2008; 13(5): 494 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Khamaisi and S. Elad Re: Intravenous Bisphosphonate Therapy and Inflammatory Conditions or Surgery of the Jaw: A Population-Based Analysis J Natl Cancer Inst, January 16, 2008; 100(2): 155 - 155. [Full Text] [PDF]
|
|
|
|
|
|
G. S. Wilkinson, Y.-F. Kuo, J. L. Freeman, and J. S. Goodwin Response: Re: Intravenous Bisphosphonate Therapy and Inflammatory Conditions or Surgery of the Jaw: A Population-Based Analysis J Natl Cancer Inst, January 16, 2008; 100(2): 155 - 156. [Full Text] [PDF]
|
|
|
|
 |
|
 V. M. Cartsos, S. Zhu, and A. I. Zavras Bisphosphonate Use and the Risk of Adverse Jaw Outcomes: A medical claims study of 714,217 people J Am Dent Assoc, January 1, 2008; 139(1): 23 - 30. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Bisphosphonates and Osteonecrosis of the Jaw Journal Watch Oncology and Hematology, August 14, 2007; 2007(814): 5 - 5. [Full Text]
|
|
|
|
 |
|
 Jaw Osteonecrosis in Cancer Patients Receiving IV Bisphosphonate Journal Watch (General), July 19, 2007; 2007(719): 2 - 2. [Full Text]
|
|
|
|
|
|
S.-B. Woo and D. H. Solomon Bisphosphonate Therapy for Cancer and Prevalence of Inflammatory Jaw Conditions J Natl Cancer Inst, July 4, 2007; 99(13): 986 - 987. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||