Bisphosphonate-induced osteonecrosis of the jaws: Prospective study of 80 patients with multiple myeloma and other malignancies

Summary

A prospective study was performed in 80 patients receiving bisphosphonates in order to determine frequency of occurrence, risk factors, clinical presentation, radiology, pathology and proper treatment of osteonecrosis of the jaw (ONJ). Of 80 patients, 22 (28%) developed ONJ. There were 11 male and 11 female patients. Median age was 65 years. Ten patients (46%) had multiple myeloma (MM), 5 (23%) had breast cancer and 7 (32%) had other malignancies. Of 22 patients with ONJ, 14 patients (64%) received zoledronate, 3 (14%) received pamidronate, 4 (18%) received pamidronate later followed by zoledronate and 1 patient received ibandronate later followed by zoledronate. The median time of exposure in ONJ group was 32 months compared with 27 months in patients without ONJ. The mean induction time until bone exposure was 26 months for patients who received zoledronate, 54 months for pamidronate and 48 months for pamidronate followed by zoledronate. Thirteen patients (59%) had ONJ with bone exposure of mandible, 6 (27%) of maxilla and 3 (14%) of both jaws. ONJ occurred spontaneously in 5 patients (23%) and in 17 patients (77%) occurred after tooth extractions and surgical tooth removals (P < 0.001). Nine patients (41%) had previous extractions of molars, 6 (27%) of premolars and 2 (9%) of front teeth. The cumulative hazard is significantly higher in zoledronate group (P = 0.015). It was 3.48 times higher than the other group (pamidronate alone; pamidronate followed by zoledronate; ibandronate alone; etidronate alone; ibandronate followed by pamidronate; ibandronate followed by zoledronate; ibandronate followed by pamidronate and zoledronate). There was no association of ONJ with age, sex, use of high-dose or conventional chemotherapy or the use of corticosteroids, thalidomide or bortezomib (P > 0.05). Patients diagnosed with multiple myeloma and breast cancer were found significantly associated with ONJ (P = 0.001 and P = 0.014, respectively). Long-term use of bisphosphonates (>2.5 years) increases the risk for development of ONJ. Intravenous application of zoledronate and previous dental extractions or surgical tooth removals are important risk factors of ONJ. Neither treatment with high-dose chemotherapy with autologous stem cell transplantation nor treatment with corticosteroids, thalidomide or bortezomib is a risk factor in this study.

Introduction

Bisphosphonates are used in the treatment of bone involvement by multiple myeloma (MM),¹ osteoporosis, Paget’s disease, primary and secondary hyperparathyroidism,² hypercalcemia related to malignancy and is useful in the management of bone metastases.³ The current benefits of bisphosphonates for metastatic bone disease include prevention of skeletal-related events, reduction in bone pain and improvement of quality of life.⁴ Bisphosphonate therapy is recommended to be used for all patients with multiple myeloma requiring chemotherapy, whether bone lesions are evident or not (grade A recommendation; level Ib evidence).⁵ Widely used bisphosphonates for multiple myeloma and metastatic bone disease are pamidronate or zoledronate.⁶

There is some evidence that bisphosphonates also might have anti-tumor activity,⁷ as indicated by reduced skeletal tumor burden and slower progression of bone lesions in animal models. Bisphosphonates may render the bone a less favorable microenvironment for tumor cell growth by reducing tumor-induced osteolysis and local release of growth factors. They also inhibit proliferation and induce apoptosis of a variety of human tumor cell lines in vitro.8, 9, 10

Bisphosphonates are compounds with a chemical structure based on a phosphorous–carbon–phosphorous spine, similar to endogenous pyrophosphate but with carbon replacing the central oxygen molecule, which allows the accommodation of two additional side chains. Carbon confers resistance to hydrolysis and degradation by osteoclasts and yet maintains the pyrophosphate-like inhibition of bone resorption. One of the side chains usually contains a hydroxyl group giving the compound a high affinity for calcium crystals and bone mineral, thus accounting for its avid binding to the bone matrix.¹¹ Different chemical groups attached to the second free carbon bond account for the variability in anti-resorptive potency of the bisphosphonates and increase the therapeutic index between the desired anti-resorptive effect² and the unwanted effect of demineralization of new osteoid.¹¹

Bisphosphonates are specific inhibitors of osteoclastic activity. They are non-hydrolysable analogues of inorganic pyrophosphate that bind avidly to hydroxyapatite crystals and are subsequently released during the process of bone resorption.¹² They have selective concentration at the interface of the active osteoclasts and the bone-resorption surface. The specific mechanism of this inhibition is not well understood, but there are several actions, including inhibition of osteoclast7, 13 development from precursor cells, increase in number of osteoclast,¹⁴ stimulation of osteoclast inhibitory factor, reduction of osteoclast activity, and down-regulation of matrix metalloproteinases. The net result in reduction of osteoclastic activity reduces bone resorption.¹⁵ However, oral bisphosphonates are poorly absorbed.²

Bisphosphonates are categorized into two main groups: The first-generation bisphosphonates, such as clodronate and etidronate have been in clinical use for more than 30 years. These groups have simple side chains attached to the central carbon atom and have comparatively weak actions on resorption. The second-generation bisphosphonates are nitrogen-containing aminobisphosphonates (N-BPs), such as ibandronate, pamidronate and alendronate. They have an additional aliphatic side chain containing a single nitrogen atom. Risedronate and zoledronate are the more potent second-generation bisphosphonates. Risedronate has a heterocyclic ring structure and zoledronate has a five-membered imidazole ring and two nitrogen atoms, which increases efficacy.¹¹

The newer nitrogen-containing bisphosphonates, such as zoledronate, pamidronate and ibandronate, have greater clinical activity than first-generation bisphosphonates that lack nitrogen such as etidronate and clodronate.¹² Zoledronate is 100 times more potent than pamidronate¹⁶ and has more potency in preclinical studies than any other current bisphosphonate.¹⁷ Pamidronate has significant effect of decreasing bone pain and is clinically safe.¹⁸ Monthly infusions of pamidronate provides significant protection against skeletal complications and so improve the quality of life of patients with stage III multiple myeloma.¹⁹

Although bisphosphonates have many clinical benefits, a number of cases of jaw osteonecrosis have recently been reported in patients receiving long-term intravenous bisphosphonate treatment, especially with zoledronate and pamidronate.20, 21, 22, 23, 24 The first clinical reports of this complication in the literature were published by Marx²⁰ in 36 cases at the same time with Migliorati²¹ in 5 cases in 2003, by Ruggiero et al.²³ in 63 cases in 2004, Bamias et al.²⁴ in 17 cases in 2005, by Schirmer et al.²⁵ in 6 cases in 2005, by Bagan et al.²⁶ in 20 cases in 2006 (Table 1). The average age of patients is more than 50 years or older.27, 28 Long duration of exposure to bisphosphonates seems to be the most important risk for complications. Bamias et al.²⁴ reported a median time of exposure of 39.3 months for patients with osteonecrosis, whereas it was 19 months for patients without osteonecrosis. In the study of Dimopoulos et al.,²⁹ a similar median time was found with 39 months for patients with osteonecrosis, but 28 months (P = .048) for patients without osteonecrosis. Migliorati et al.³⁰ presented that all patients had developed osteonecrosis after a mean exposure time of 25 months, in contrast Badros et al.³¹ presented that the mean exposure time was 8.37 years. Both jaws can be affected,20, 23, 30 but the most affected region is the mandible.31, 32 Oral surgical procedures or tooth extractions increase the incidence of osteonecrosis.23, 31, 33 However it can occur spontaneously without history of a recent dentoalveolar procedure.23, 33

Most patients required surgical procedures to remove the exposed jawbone, such as sequestrectomy, segmental resection of the jaw, decortication and antibiotic therapy.23, 34

The aim of this study was to describe the frequency of ONJ in patients receiving bisphosphonates, including clinical presentation, radiologic features, pathology and treatment. The risk factors for developing osteonecrosis of the jaw such as duration of administration, type of bisphosphonates, dental procedures during bisphosphonate therapy, were analysed. In contrast to previous retrospective studies the present study was designed as a prospective investigation.

We investigated 80 patients receiving bisphosphonates from the Department of Hematology and Oncology who were referred to the Oral Surgery Department for oral examination before and also after chemotherapy or high-dose chemotherapy with further stem cell transplantation including out-patients from July 2005 through December 2006. The patients were evaluated prospectively and were investigated at least twice. The two examinations were six months apart at least. All patients had previously been diagnosed with malignant diseases and received bisphosphonate due to osteolytic metastasis. They had received bisphosphonates up to September 2006 and received at least six infusions. The patients who had been treated with zoledronate received 4 mg intravenously over 15 min monthly, patients with pamidronate received 90 mg over 1 h intravenously at the same period and those with alendronate received 70 mg weekly (10 mg daily) orally. The time of exposure to bisphosphonates was defined in months from the initial infusion to the last recorded infusion in patients without osteonecrosis and to the first time of bone exposure in patients with bone involvement. The diagnosis of osteonecrosis of the jaw was carried out on the basis of digital panoramic radiological investigations and clinical diagnosis was based on the following criteria: exposed or necrotic bone of maxilla or mandible with or without pain, evidence of regional soft tissue inflammatory swelling or infection; exposed bone with pathologic fracture with pain, swelling or extraoral fistula. Biopsy of the bone was performed only when exclusion of metastatic disease was necessary. All medical records of patients were reviewed and patients with incomplete medical report were excluded. We included patients who received chemotherapy, but neither of the patients had been irradiated in the head and neck region nor had evidence of obvious osseous manifestations of the malignant disease in the jaws. The treatment of patients with ONJ included sequestrectomy, sequestrotomy, surgical debridement, wound debridement without surgical procedure, antibiotic administration (Clindamycin 600 mg, three times/day) and cessation of bisphosphonates in 20 patients.

SPSS statistical software was used for this analysis (SPSS for Windows, version 14.0). The χ² test was used for comparison of the diagnosis such as multiple myeloma, breast cancer, prostate cancer and other malignant diseases across ONJ. The Mann–Whitney U-test was used to compare sex, age, time of exposure, chemotherapy, high-dose chemotherapy followed by stem cell transplantation, use of corticosteroids, thalidomide or bortezomib, history of tooth extraction or surgical oral procedures, with ONJ. The t-test was used to compare the groups treated with zoledronate vs. pamidronate followed by zoledronate and pamidronate alone to compare the time for the onset of ONJ with each treatment. The hazard of developing ONJ according to type of bisphosphonate and the duration of exposure was estimated by survival analysis (cox regression method). In this study, P-value less than 0.05 was considered.