Renal toxicity and osteonecrosis of the jaw in cancer patients treated with bisphosphonates: a long-term retrospective analysis
- First Online:
- Cite this article as:
- Bonomi, M., Nortilli, R., Molino, A. et al. Med Oncol (2010) 27: 224. doi:10.1007/s12032-009-9195-y
- 132 Views
Background: Bisphosphonates (BPs) are the mainstay of bone-directed therapy for bone metastases from multiple myelomas and a wide range of solid tumours, but some patients experience renal toxicity or osteonecrosis of the jaw (ONJ). Patients and methods: We reviewed data relating to 398 patients treated with intravenous BP for bone metastases, checking their serum creatinine levels throughout the treatment period in order to assess renal function, and seeking any signs and symptoms of ONJ recorded in their medical records. We also analysed other risk factors for renal toxicity and ONJ in patients who developed them. Results: The median treatment period was 14 months (range 1–119); 108 patients received BP for more than 1 year, and 112 for more than 2 years. Sixteen patients (4%) developed renal toxicity after a median of 24 months of BP treatment, eight of them had been treated for more than 2 years. Ten patients (2.5%) were diagnosed as having ONJ after a median of 39 months on BP, only three of them had been treated for less than 2 years. Two patients experienced both ONJ and renal toxicity. Conclusions: The low incidence of ONJ and renal toxicity indicates the safety of BP. However, prevention and early detection are still the “first-line therapy” for decreasing their occurrence further.
KeywordsBisphosphonateNephrotoxicityOsteonecrosis of the jawPamidronateZoledronic acid
Bisphosphonates (BP) are widely used to treat hypercalcemia due to malignancy, and are part of the standard of care in patients with bone metastases in order to reduce the skeleton-related events (SREs) that increase morbidity and greatly affect their quality of life .
The efficacy of pamidronate in reducing the incidence of SRE was first demonstrated in the 1990s in patients with advanced multiple myeloma (MM) or breast cancer (BC) with bone metastasis [2–6]. A few years later, a comparative trial involving MM and BC patients showed that zoledronic acid (zoledronate) was equivalent to, if not more effective than, pamidronate [7, 8]; subsequently, zoledronic acid was found to be clinically effective in prostate cancer  and other solid tumours .
The intravenous administration of BP may be associated with side effects such as inflammatory reactions at the infusion site, nausea and vomiting, flu-like symptoms and electrolyte imbalances (hypocalcaemia and hypophosphatemia) ; till 2002, nephrotoxicity was considered the principal long-term side effect [7, 8, 11]. Cancer patients, especially advanced or metastatic, have an increase baseline risk of renal failure. Possible causes are: MM or a cancer primarily affecting the urogenital tract (prostate, kidney), previous or concurrent use of non-steroidal anti-inflammatory drugs (NSAIDs), concurrent nephrotoxic chemotherapy (e.g. cisplatin), elderly patients with an advanced stage cancer and other comorbidities, such as diabetes, hypertension or dehydration. Prevention plays a key role in controlling the incidence of nephrotoxicity and includes checking serum creatinine (SCr) levels before each administration, avoiding infusion times of less than 15 min for zoledronic acid or 2 h for pamidronate, reducing doses in patients with impaired renal function, ensuring adequate hydration, and taking care over the concurrent administration of other nephrotoxic drugs. In addition to discontinuing BP, the treatment of nephrotoxicity is mainly supportive, but full recovery (i.e. the normalisation of SCr levels) is possible, as is the resumption of BP therapy.
Osteonecrosis of the jaw (ONJ) was first described by Marx in 2003  and, although it is most frequently reported in patients with MM, it has also been observed in BC and other solid tumours . Chemotherapy, antiangiogenetic therapy, corticosteroids, head and neck radiotherapy, previous dental procedures, poor oral hygiene, systemic or local infections, immunodepression, due to specific therapy or cancer itself, can increase the risk of development of ONJ and of the sovrainfection of the necrotic bone. Although it is usually asymptomatic (exposed bone in the oral cavity is the only clinical presentation), it can cause serious pain due to inflammation of the involved sites or oral soft tissue trauma . Its pathogenesis and the interactions of risk factors are not yet clear [12–16], and so there is a lack of complete guidelines based on clinical trials and various treatment protocols have been proposed [16–20]. The aim of our retrospective analysis is to assess the incidence of nephrotoxicity and ONJ in all our patients treated with intravenous BPs for bone metastases, with particular regards to the long-term treated group (>24 months) and to highlight the presence of other risk factors for both ONJ and nephrotoxicity in the patients who developed them.
Patients and methods
Patients baseline characteristics
Number of patients (%)
Median age at diagnosis of tumour
Median age at diagnosis of bone M+
Median age at the beginning of BP
Number of patients (%)
Patients with bone metastases at diagnosis
Extra-osseous M + at diagnosis (%)
No extra-osseus M + at diagnosis (%)
Patients without bone metastases at diagnosis
Extra-osseous M + at diagnosis (%)
No extra-osseus M + at diagnosis (%)
Median age at the start of BP therapy was 63 years (range 22–84). The median overall duration of BP treatment was 14 months (range 1–119); 112 patients (28%) received BP for more than 2 years and 108 (27%) for more than 12 months. There was no treatment discontinuation amongst patients who treated for up to 12 months, but 17 temporary discontinuations among those treated for longer. The main causes of discontinuation were: side effects (nausea, vomiting, myalgia, flu-like syndrome), renal toxicity and ONJ. The duration of BP therapy was calculated as the number of months of actual treatment, but the patients were also monitored during the temporary discontinuation.
Regarding the BP treatment, 247 patients received zoledronic acid (Z) alone, 37 pamidronate (P) alone, and 100 both. Other patients received clodronate, either alone (three patients) or followed by pamidronate (four patients) or by pamidronate and zoledronic acid (five patients); two patients received also ibandronate (preceded by pamidronate and zoledronic acid). Most of the patients treated with both pamidronate and zoledronic acid received pamidronate before being switched to zoledronic acid (P → Z group) at the end of 2002, when the newer nitrogen-containing BP was found to be equivalent if not superior to the older one (some patients did an “inverse switch”, or a “double switch”, i.e. from zoledronic acid to pamidronate, or from pamidronate to zoledronic acid and back to pamidronate again because of supposed nephrotoxicity caused by zoledronic acid or intolerance to the newer BP). The median duration of treatment was 10 months (range 1–48) in the zoledronic group, 13 months (range 2–33) in the pamidronate group and 35 months (range 2–97) in the P → Z group (with a median time of 22 months of pamidronate and 16 months of zoledronic acid for the 51 patients affected by breast cancer who did the “simple” switch, i.e. from pamidronate to zoledronic acid, without discontinuations, nor referred toxicity and treated at least 3 months with each of the 2 BPs). Most of the patients received zoledronic acid (4 mg over 15 minutes) or pamidronate (90 mg over 120 min) each month. Therefore, the median cumulative dose for each patient was 1170 mg for pamidronate group, 40 mg for the zoledronic one and, for the 51 patients who did the “simple switch” (see above) 1980 mg and 64 mg of pamidronate and zoledronic acid, respectively.
Almost a half (47%) of the zoledronic acid group patients were still on treatment at the time of analysis (which means that the 10-month mean duration was not a definitive result), against 0% of the pamidronate group and 19% of the P → Z one.
During BP treatment, 240 patients underwent chemotherapy for a median of six months (range 1–33), 267 received hormone therapy for a median of 14 months (range 1–97), 147 received both, and 31 did not receive either. One hundred and thirty-four patients underwent radiotherapy to bone: 105 received 8 Gy flash a median of 7 months (range 0–88) after starting BP and 30 received fractioned treatment a median of 3 months (range 0–33) after starting BP (one received both), none of the patients received radiotherapy on the mandible. Considering other SREs, 19 (4.8%) episodes of hypercalcemia of malignancy have been signalled after a median period of 11 months (range 0–88) of treatment, 29 (7.3%) pathologic fractures after 17 months (range 1–77), 5 (1.2%) spinal cord compressions after 22 months (range 18–37). Surgery was used in seven (1.7%) cases all following pathologic fractures or spinal cord compressions.
We checked SCr levels throughout the treatment period, and defined as nephrotoxic event an increase of 0.5 mg/dl if the baseline level recorded at the start of BP therapy was normal, or an increase of 1.0 mg/dl if the baseline level was >1.4 mg/dl . SCr levels were checked at least once a month for patients who were receiving also chemotherapy or for patients who experienced an increase in SCr levels during treatment (until they returned to baseline levels); for patients who were receiving only BP and best supportive care (or ormonotherapy), SCr were checked less frequently (every 1–3 ms).
Diagnosis of ONJ was mainly clinical (i.e. made by the oncologist or by the dentist).
During treatment with BP, 37% of the patients achieved osseous and extra-osseous disease stabilisation, 43% experienced bone progression (31% osseous alone; 12% osseous + extra-osseous), after a median of 11 months (range 1–77) from BP treatment onset, 20% had only extra osseous progression. The median time of PD to bone was of 11 months (range 1–77), while the extra-osseous PD time was 10 months (range 1–90).
One hundred and fifty-one patients were still in follow-up at the end of August 2006; 93 had died and 154 were considered lost to follow-up because the time from their last visit was 6 months or longer (these patients had a poor performance status or were only receiving supportive care).
Three hundred and fifty-seven of the 398 patients (90%) retained a normal renal function; 25 (6%) had altered baseline levels that did not significantly increase during treatment; and 16 (4%) developed a nephrotoxic event in which BP therapy may have played a role (although it was the only nephrotoxicity risk factor in only one case). Eight of these 16 patients were receiving NSAIDs or other nephrotoxic drugs (such as cisplatinum), and eight developed nephrotoxicity shortly before death or their last follow-up visit (when their performance status was rapidly worsening due to progression of their disease). Underlying MM (one patient) or a cancer primarily affecting the urogenital tract (prostate in three patients, kidney in one patient, bladder in one patient) may have affected the renal function of some patients. Six patients switched from zoledronic acid to pamidronate because of the presumed lesser nephrotoxic effect of pamidronate: any significant renal function improvement was detected. The median time from the start of BP treatment and the development of the nephrotoxicity was 24 months (range 2–87); eight patients developed a nephrotoxic event after more than 2 years of BP. Considering the main tumour types, it occurred in 7 of the 238 BC patients (2.9%) after 30 months (range 2–41), in 3 of the 46 patients with prostate cancer (6.5%) after 15 months (range 14–23), and in 1 of the 15 patients with kidney cancer (6.7%) after 25 months.
Regarding the type of BP administered, the incidence of nephrotoxicity was 2.7% in the pamidronate group (1/37, after 2 months of therapy), 2.4% in the zoledronic group (6/247, after 19 months of therapy, range 9–33 months), and 8% in the P → Z group (8/100, after a median treatment period of pamidronate of 19.5 ms—range 2–39 months—followed by 9.5 months of zoledronic acid—range 1–39 ms); in the latter population the toxic event was diagnosed during the zoledronic acid treatment. In one patient, SCr level increased either on zoledronic or on pamidronate therapy. She was first treated with pamidronate for 10 months before developing a mild nephrotoxicity (SCr started to increase shortly after the beginning of therapy); the renal function improved, without anyway returning to basal values, after the placement of a urethral stent; subsequently she went on with pamidronate for other 11 months, while the SCr assessed around 1.5 mg/dl; then the treatment was stopped for 6–7 months and the renal function did not improve; because of the worsening of metastatic bone disease zoledronic acid was started; after another 8 months the renal function got worst (1.7 mg/dl); the substitution of urethral stent and of the zoledronic acid with pamidronate did not obtain any result.
Osteonecrosis of the jaw
Ten cases of ONJ were reported between February 2002 and April 2006: seven females with BC, two males with MM, and one male with prostate cancer. Five of the seven patients for whom information concerning risk factors was entirely available had received chemotherapy, two corticosteroids and one both; three patients had undergone dental extractions before developing ONJ. All five of the patients for whom complete information regarding ONJ treatment was available received antibiotics, three underwent surgery (curettage, hemimandibulectomy), and two hyperbaric oxygen therapy with poor results. BP were stopped in only three out of seven patients.
The other three patients (one with prostate cancer and two with BC) had an incomplete follow-up. One patient with prostate cancer, who was initially treated in our Department and then moved elsewhere, received a diagnosis of ONJ, but no other information is available. The second case involved a BC patient who was treated in our Department until April 2004, and returned in February 2006 with a diagnosis of ONJ mainly due to a dental extraction, after 53 months of BP therapy. The third patient experienced ONJ after a dental extraction and 51 months of BP treatment. Both of the BC patients were surgically treated, but BP therapy was not stopped until 2006, when they returned to our Department.
The mean time to the induction of ONJ in the seven patients with a complete follow-up was 36 months (range 11–51). The first case was reported in February 2002 in a patient with MM.
The two patients with MM had only received pamidronate before the diagnosis of ONJ, and developed it after 39 and 51 months from the beginning of the BP therapy. Three of the seven BC patients had only received zoledronic acid before developing ONJ after, respectively, 11, 15 and 16 months. The other four (including the two patients with an incomplete follow-up) had received both pamidronate and zoledronic (one received also clodronate): three of them developed ONJ after a median of 36 months of zoledronic acid treatment (range 20–39) preceded by 5 months of pamidronate (range 3–33); the fourth developed ONJ after being treated for 30 months with pamidronate, 11 months with zoledronic acid and a further 10 months with pamidronate (due to mild nephrotoxicity probably caused by zoledronic acid).
In the last few years, life of people with metastatic bone disease has changed. Chemotherapy and other aetiological therapies have improved, leading sometimes to a better prognosis for this group of patients. Although not effective in the overall survival, BP have decreased SREs that often compromise the quality of life of these patients.
The American Society of Clinical Oncology guidelines for BP in BC suggests treating patients for as long their performance status does not deteriorate , but there are still few data concerning their safety and efficacy in patients treated for more than 2 years [22–24]. Ali et al.  found BP safe and well tolerated in 22 patients treated for a median of 3.6 years. In an other study, 57 patients were treated for a median of 34 months: grade 1 nephrotoxicity was reported in 12.2% of them  Bujanda et al.  found an overall 9% incidence of nephrotoxicity in 67 patients analysed (18% in the group treated for >2 years).
In our population, the overall incidence of renal toxicity was 4% as a whole, and 7% (8/112) in the patients treated for more than 2 years. These data are similar to those reported from the randomised studies comparing zoledronic acid with pamidronate or placebo in the treatment of bone metastases [7–10, 25]. The incidence of renal toxicity regarding the type of BP used was 2.7% in pamidronate group, 2.4% in zoledronic group, 4% (14/347) in the combination group. Patients experiencing renal dysfunction during treatment with zoledronic acid did not show any improvement when they were switched to pamidronate.
The second main concern raised by BP therapy is ONJ whose incidence among our patients was 2.5%. It was first diagnosed in February 2002, when attention was initially drawn to the problem; before this time it is possible that some asymptomatic cases may have been misdiagnosed. The real incidence of ONJ remains unknown yet [12, 24, 26–31]. One of the largest study reported in the literature described 119 cases of ONJ: 26% occurred after pamidronate, 40.3% after zoledronic acid, 30.2% in patients treated with both drugs, and 2.5% in patients who had received alendronate due to osteoporosis . The mean induction times from the onset of BP were, respectively, 14.3, 9.4, 12.1 and 36 months.
Long-term BP treatment increases the risk of developing ONJ, which may explain why this side effect was never reported in the randomised studies of BP, usually limited to 24 months of treatment; furthermore, most of them were conducted, and closed, before 2002–2003 when ONJ was first described. Sixty-six per cent of our patients affected by ONJ were treated for more than 36 months.
Although it is difficult to compare pamidronate with zoledronic acid, which has been available for a shorter period, the latter does seem to cause ONJ more often and after a shorter period of treatment. In our population, the incidence of ONJ was 5.4% in the pamidronate group, 1.2% for patients only treated with zoledronic acid, and 4% in the P → Z group; the median time on treatment was 15 months for zoledronic acid alone, 45 months for pamidronate and 5 months of pamidronate plus 36 of zoledronic acid for patients who received both.
Our retrospective analysis of patients treated with BP found a low incidence of both renal toxicity and ONJ, although it was higher for patients receiving long-term treatment. Clinicians should pay attention to possible additional risk factors and early symptoms during follow-up in an attempt to avoid and prevent the occurrence of these events, which can compromise greatly patients’ quality of life.