The Oncologist, Vol. 13, No. 3, 330-336, March 2008; doi:10.1634/theoncologist.2007-0159 © 2008 AlphaMed Press
Osteonecrosis of the Jaw in Patients with Bone Metastases Treated with Bisphosphonates: A Retrospective StudyaOsteo-oncology Center and Department of Medical Oncology, Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori, Meldola, Italy; b Oral Surgery and Dentistry Department, Otorhinolaryngology Unit, Morgagni Pierantoni Hospital, Forlì, Italy Key Words. Bisphosphonates • Bone metastasis • Jaw • Osteonecrosis Correspondence: Dino Amadori, M.D., Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori, Via Maroncelli 40, 47014 Meldola, Italy. Telephone: 39-0543-739252; Fax: 39-0543-739249; e-mail: segronco{at}ausl.fo.it Received September 4, 2007; accepted for publication January 14, 2008. Disclosure: No potential conflicts of interest were reported by the authors, planners, reviewers, or staff managers of this article.
Access and take the CME test online and receive 1 AMA PRA Category 1 CreditTM at CME.TheOncologist.com
Objective. Bone metastases are a major cause of morbidity in cancer patients. Treatment includes bisphosphonates, which are also associated with avascular osteonecrosis of the jaw (ONJ). Our aim was to evaluate the correlation between bisphosphonates and ONJ. Patients and Methods. Of the 539 patients with bone metastases treated in our department from June 2002 to December 2006 with i.v. bisphosphonates, eight (1.5%) developed ONJ. Results. The eight patients with ONJ had all been given zoledronic acid, and two had also been treated with pamidronic acid. In four of the patients, ONJ was diagnosed during treatment, while in the remaining four it was diagnosed several months after therapy with bisphosphonates had ended. Six of these patients received local noninvasive treatment, of whom five progressed. Two showed apparent autolimitation of the disease. The remaining two patients underwent surgical resection and currently show no signs of relapse. All eight ONJ patients presented with various concomitant risk factors such as paradontopathy, dental extraction, or spontaneous avulsion. Conclusions. Our results show that ONJ can appear months after interruption of treatment and that a surgical approach can be used in suitable cases. Closer cooperation is needed among specialists to define guidelines for the prevention of ONJ in patients with bone metastases.
Bone metastases are a major cause of morbidity in cancer patients for epidemiological and clinical reasons [1]. Indeed, about 80% of metastases arise from cancers with a high incidence and mortality, such as breast, prostate, and lung cancer [2]. The frequency of complications in these patients depends mainly on the site of metastasis, type of lesion (lytic or blastic), and treatment [3]. Treatment of bone metastases should be multidisciplinary, and bisphosphonates play a key role [1]. These drugs have a favorable impact on the natural history of the disease, improving quality of life and decreasing pain and skeletal-related events [4–8]. However, in addition to the well-known side effects, such as flu-like syndrome, creatinine alteration, and hypocalcemia, the use of bisphosphonates has also recently been linked to the development of avascular osteonecrosis of the jaw (ONJ) [9]. The first report of bisphosphonate-related ONJ was published in 2003 [10]. In the present study, we aimed to further our knowledge of this problem by carrying out a retrospective study of patients with bone metastases who were treated with i.v. bisphosphonates, and focusing on those who developed ONJ.
We searched medical records from the Department of Medical Oncology from June 2002 to December 2006 for patients with all the following characteristics: confirmed cytological or histological diagnosis of a malignancy, radiological or cytological/histological diagnosis of bone metastases, treatment with bisphosphonates after diagnosis of bone metastases, and ability to access the full medical history. The bisphosphonates used were zoledronic acid (ZA) (Zometa®; Novartis Pharmaceutical Corporation, East Hanover, NJ), administered at a dosage of 4 mg via a 15-minute infusion, and pamidronic acid (PA) (Aredia®; Novartis Pharmaceutical Corporation), given at a dosage of 90 mg via a 2- to 3-hour infusion. Patients with jaw metastases or those treated with radiotherapy were excluded from the study. Patients with a diagnosis of ONJ made by a dentist and/or a maxillofacial surgeon were reviewed. In some cases it was necessary to confirm the diagnosis with radiological or histological exams. When the phenomenon was first documented, ONJ was defined as a nonspecific necrotic lesion of bone tissue characterized by slow progression and by failure to heal spontaneously. The diagnosis of ONJ was also based on the presence of refractory osteomyelitis, a nonhealing extraction socket, exposed bone, or chronic osteomyelitis with sequesters or purulent material. The definition has obviously, over time, become more accurate for these patients, who are subject to repeated bacterial infections associated with varied symptomatology. We used the above widely accepted definition in our study [11].
From June 2002 to December 2006, 539 patients with bone metastases were treated with bisphosphonates in our department. Patient characteristics and the type of bisphosphonate administered are reported in Tables 1 and 2, respectively. Five hundred fifteen patients were treated with ZA, receiving a mean of 6.7 cycles (range, 1–65). PA was administered to eight patients, who received a mean of 2.8 cycles (range, 1–8). Sixteen patients were given both PA and ZA, with a mean of two cycles (range, 1–7) and 11 cycles (range, 2–34) administered, respectively (Table 2). The median follow-up duration was 10.8 months (range, 1–65 months).
Eight (1.5%, 95% confidence interval [CI], 0.7%–2.8%) cases of ONJ were observed in the entire sample, representing 2.3% (95% CI, 0.8%–4.8%) of all the breast cancers, 12.5% (95% CI, 0.9%–43.3%) of the hepatocarcinomas, and 3.4% (95% CI, 0.5%–10.3%) of the multiple myelomas treated. The incidence rate per year was 0.01 (95% CI, 0.005–0.02). The full patient characteristics are described in Table 3
Bisphosphonate treatments are shown in Table 3
The approaches used for ONJ diagnosis and predisposing factors to the disease other than the use of bisphosphonates are reported in Table 3 Upon oral examination, all these patients had areas of bone necrosis (Fig. 1) at the jaw level and two also showed necrosis in the jowl area, with formation of an oroantral fistula (Fig. 2). One patient presented multiple sites of ONJ with bilateral necrosis of the left jaw and at the jowl level. The mandible was more commonly affected than the maxilla (2:1 ratio).
Clinically, ONJ manifested itself as an area of bone tissue with varying degrees of necrosis, often associated with events of bacterial overinfection. These episodes often resulted in pain, swelling, and pus discharge. Holes were often present between exposed bone and gingival tissue, where food frequently became lodged. This problem was further aggravated by the spongy morphology of exposed bone and by incorrect oral hygiene. Computed tomography scans and orthopantomography revealed images of a rearranged bone structure [9].
The treatment administered in the eight patients is shown in Table 3 Bone spicules were removed and examined, and oral swabs of lesions were taken and cultured in two patients, revealing a bacterial infection associated with a mycotic infection. Over a follow-up period of 1 year, we observed an extension of exposed bone areas in five patients. However, two cases of spontaneous detachment (patient 1 at the jaw and patient 5 at jowl level) of necrotic bone were observed, indicating a sort of autolimitation of the event, with formation of granulation tissue to cover the area (Fig. 3). These patients are being closely monitored. The two patients treated with surgical resection at another institution (patients 4 and 7) are currently disease free.
ONJ is a nonspecific necrotic lesion of bone tissue characterized by a slow progression and by nonspontaneous recovery [11]. The main cause of the development of bone necrosis is generally considered to be radiotherapy in the head and neck area (radio-osteonecrosis) [11, 13, 14], but other factors, such as mycotic or herpetic infections or necrotizing sialolytic metaplasias, may also be involved [15]. Generally speaking, however, any pathologic or traumatic event capable of determining tissue hypoxia, hypocellularity, or hypovascularization is considered to be a potential etiological factor for bone necrosis. Although there have been several reports in the literature suggesting that ZA has antiangiogenic activity [16], this, in itself, cannot be held accountable for the increased incidence of ONJ. Indeed, studies carried out on patients with multiple myeloma and treated with antiangiogenic drugs, such as thalidomide, did not highlight a greater risk for ONJ [17]. The detection of ONJ in cancer patients during treatment with ZA recently induced Novartis to issue a warning letter to physicians concerning ONJ [18, 19]. The introduction of new drugs such as bortezomib has prolonged survival in patients with multiple myeloma, thus increasing the exposure to bisphosphonates [20, 21]. The same thing has happened for several other tumors, including breast and prostate cancer. The length of time patients are exposed to the drug and the number of infusions administered seem to be the most significant risk factors for the development of osteonecrosis [22]. In our experience, ONJ was observed after a median exposure time of 12 months (range, 9–24) and a median of 11 (range, 8–25) treatment cycles. This differs from data reported in other studies in which none of the patients who received fewer than 12 bisphosphonate infusions developed ONJ [22, 23]. A prospective study published in 2005 by Bamias and coworkers reported an ONJ incidence of 6.8% for 252 patients who were treated with bisphosphonates [22]. In their study, length of exposure was associated with the development of the disease, with ONJ patients receiving a median of 35 infusions (range, 13–68). A small prospective study recently published by Bujanda and coworkers reported a 9% incidence rate of ONJ in 67 patients treated with bisphosphonates [24]. Four of the six ONJ patients received prolonged treatment (median exposure time of case series, 22 months). A number of other works report results from retrospective analyses, with an estimated incidence varying in the range of 0.8%–5.2% [25–28]. These widely differing incidence rates could be a result of the different number of patients involved in each study, the length of exposure, a variety of other possible risk factors, or the different follow-up times observed. Our case series is much larger than most of the above works. The retrospective review by the MD Anderson Cancer Center encompassed all patients who received i.v. bisphosphonates in 1996–2004 [26]. An ONJ incidence rate of 0.8% was observed, which is lower than ours (1.5%, 95% CI, 0.7%–2.8%). Furthermore, of the eight cases of ONJ we observed, four were diagnosed several months after interruption of bisphosphonate treatment (median, 8.5 months; range, 5–13). This would seem to indicate that the drug is still present in bone tissue months after its suspension. It is known that the maximum plasma concentration of ZA is about 1 µM and that approximately 55% of the initially administered dose is retained in the skeleton and slowly released back into the circulation, with a terminal half-life of about 200 days. This demonstrates that long-lasting accumulation in bone is a common feature of nitrogen-containing bisphosphonates [29–31], and would seem to support the observation of ONJ diagnosis several months after treatment suspension.
In addition to bisphosphonate administration, literature data show that other factors may also be involved in the development of ONJ, such as tooth extraction, poor oral hygiene, the absence of regular dental checkups, periodontopathies, inadequate partial dentures, loose teeth, chemotherapy, oral infections, tobacco, diabetes, and anemia [20, 26, 32]. The retrospective nature of our study precludes the possibility of verifying that these conditions were indeed risk factors, because of the missing values for patients who did not develop ONJ. However, we know that all our ONJ patients received prior chemotherapy, compared with 69% of patients who did not develop the disease. Furthermore, paradontopathy (six of eight patients) and dental extraction or spontaneous avulsion (six of eight patients) were frequently observed in our ONJ patients (Table 3 The increase in the number of patients with bone metastases and the need for a multidisciplinary approach to this disease encouraged us to set up an osteo-oncology center in our hospital involving a team of 17 specialists. The aims of the center are multidisciplinary care, research, training, and above all, improvement in the management of this disease and the quality of life of patients [33]. A multidisciplinary protocol was designed by our center in collaboration with the Oral Surgery and Dentistry Department of our hospital. It recommends a dental checkup, including orthopantomography, for all patients with suspected ONJ who are about to start treatment with bisphosphonates. Dental checkups are repeated every 4–6 months, avoiding all invasive maneuvers, and treatment for patients who have undergone tooth extraction can begin only when mucous membranes have fully recovered. We have not observed any cases of ONJ since this preventive oral protocol was activated in June 2006. Sixty-three patients have been recruited to date but it is still too early to understand whether activation of the oral protocol will lead to a decrease in ONJ. Monitoring is ongoing. On the basis of our experience, and in view of the important role of bisphosphonates in metastatic bone disease [4–8], we believe that this treatment should be administered together with anticancer therapy and that the development of ONJ can be regarded as a possible complication of metastatic bone disease. The limitations of our retrospective study are the potentially missed cases of ONJ resulting from the fact that, in the first year of accrual, ONJ had not yet been associated with the use of bisphosphonates, and the short follow-up. However, we observed only eight cases of ONJ in 4.5 years, involving patients with a high number of local risk factors. Nevertheless, patients must be informed of the need for regular dental checkups and of the importance of reporting oral cavity problems immediately to their oncologist. Many issues remain to be clarified about ONJ—its real incidence, its etiopathogenesis, the most effective treatment for the condition, and above all, whether it is correct to stop bisphosphonate administration when ONJ is diagnosed. In our experience, this complication also appeared after months of nonexposure to the drug. In our opinion, it is important to evaluate cases individually, taking into account general health, prognosis, and in particular, the condition of the oral cavity. Furthermore, response to anticancer therapy and to treatments administered for ONJ should also be taken into consideration. This can be achieved only through close cooperation among the specialists who treat patients with bone metastases and through the application of well-defined protocols for the prevention and the biological/clinical monitoring of the disease.
The authors wish to thank Gráinne Tierney for editing the manuscript. This work was supported by Istituto Scientifico Romagnolo per lo Studio e la Cura dei Tumori (I.R.S.T.), Meldola, Italy.
Conception/design: Toni Ibrahim, Francesca Barbanti, Dino Amadori Provision of study materials or patients: Toni Ibrahim, Francesca Barbanti, Gianluca Giorgio-Marrano, Laura Mercatali, Sonia Ronconi, Claudio Vicini, Dino Amadori Collection/assembly of data: Toni Ibrahim, Laura Mercatali Data analysis and interpretation: Toni Ibrahim, Francesca Barbanti Manuscript writing: Toni Ibrahim, Francesca Barbanti, Gianluca Giorgio-Marrano Final approval of manuscript: Toni Ibrahim, Francesca Barbanti, Gianluca Giorgio-Marrano, Laura Mercatali, Sonia Ronconi, Claudio Vicini, Dino Amadori
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||