The purpose of the study was to evaluate the efficacy and safety of a bone-targeted regimen consisting of zoledronate, thalidomide, and interferon-γ in patients with renal cell carcinoma and bone metastases.
The purpose of the study was to evaluate the efficacy and safety of a bone-targeted regimen consisting of zoledronate, thalidomide, and interferon-γ in patients with renal cell carcinoma and bone metastases.
Eligible patients had radiographic evidence of bone metastasis. Impending pathologic fractures or spinal cord compressions must have been controlled by surgery or radiation therapy before enrollment. Zoledronate (4 mg) was given intravenously every 4 weeks, thalidomide (300 mg) was given orally once a day, and interferon-γ (100 μg) was given subcutaneously once a week. Patients were evaluated for time to skeletal-related events, the appearance of calcification in osteolytic metastases, and levels of the bone formation/resorption markers.
Fifteen patients were treated between November 2002 and November 2003; 12 had previously undergone surgery, radiation, or embolization for their bone metastases; 11 had more than 3 sites of bone involvement; and 9 also had nonosseous metastases in the lung, liver, lymph node, pancreas, or adrenal gland. The median time to progression was 8.3 weeks (range, 2.1–48 weeks). The median time to a skeletal-related event was 12.0 weeks (range, 3.9–46.4 weeks). Two patients discontinued treatment because of adverse drug reactions (1 deep venous thrombosis and 1 myocardial infarction). Two patients experienced pain improvement and developed calcification in osseous metastases; these patients also showed favorable changes in bone marker levels.
In this pilot study a bone-targeted regimen combining zoledronate, thalidomide, and interferon-γ was well tolerated and might provide clinical benefit for a small subset of patients with renal cell carcinoma and bone metastases. Cancer 2006. © 2006 American Cancer Society.
Renal cell carcinoma (RCC) displays a striking propensity to form osteolytic metastases. About half of patients with RCC develop bone metastases, and 80% of those metastases become evident within 3 years of diagnosis. The median interval time from diagnosis of RCC to the appearance of first skeletal metastasis is about 8.5 months.1 Complications of osseous metastases include pathologic fracture and spinal cord compression. RCC accounts for 10% of all pathologic fractures and 5% of all spinal cord compressions.2 The median survival time of patients with multiple skeletal metastases from RCC is only about 1 year, but patients with a solitary bone lesion can survive for longer periods, with a 5-year survival rate of 30%.1
Attempts to treat metastatic RCC have generally produced disappointing results. RCC tends to be resistant to standard cytotoxic agents and radiation therapy. Some modest success has been achieved with immunotherapy, with interleukin (IL)-2 or interferon (IFN)-α producing response rates of approximately 5% to 20%.3 Recently, novel targeted agents (e.g., bevacizumab, sorafenib, sunitinib, temsirolimus, AG-013736) have shown promising results.4–8 Treatment of bone metastases poses special challenges. Because of difficulty in quantifying the response of a tumor within the bone, patients with metastases confined to the bone are usually excluded from clinical trials. Consequently, little information is available on the potential benefit of many treatments for bone metastases from RCC despite the clear need for such information. To address this gap, we investigated the safety and efficacy of a unique combination of agents—the bisphosphonate zoledronate, the antiangiogenic compound thalidomide, and the cytokine IFN-γ—for the treatment of bone metastasis in RCC.
Bisphosphonates are analogs of endogenous pyrophosphate that preferentially bind to mineralized bone matrix, particularly in areas of high bone turnover (i.e., increased resorption or formation). Its mechanisms of action include inhibition of tumor adherence to the bone matrix, reduction of osteoclast development from precursor cells, disruption of bone resorption after internalization by osteoclasts, induction of apoptosis of both osteoclasts and tumor cells (by preventing prenylation of ras and rho), inhibition of angiogenesis, and reduction of IL-6 production from bone stromal cells.9–11 Zoledronate recently was demonstrated to reduce the incidence of skeletal-related events in patients with metastatic RCC.12
Thalidomide's antitumor effects include inhibition of angiogenesis by reducing levels of vascular endothelial growth factor and basic fibroblast growth factor. Thalidomide increases the number of CD8+ T-cells and induces the secretion of IFN-γ and IL-2 by CD8+ T-cells.13 It modulates adhesion molecules and influences interactions between tumor cells and bone stromal cells so as to decrease the growth and survival of tumor cells.14 It also modulates various cytokines (such as IL-6, IL-1, IL-10, and tumor necrosis factor-α) and inhibits the growth and survival of tumor cells, bone stromal cells, or both.15 Interestingly, thalidomide (100–1200 mg/day) was shown to elicit a response in 5 of 37 (13%) patients with metastatic RCC.16, 17
IFN-γ is a pleotropic cytokine produced by activated lymphocytes. It is a potent inducer of MHC Class I and II antigens, can enhance the activity of antigen-presenting cells, and is required for the development of cytotoxic T-cells and the activation of macrophages. IFN-γ inhibited osteoclast activity by promoting the destruction of TRAF6, and thus may oppose the effect of RANK/RANKL signaling on NF-κB18 and overcome increased osteoclastic activity in osteolytic bone metastases. Although IFN-γ does not generally benefit patients with metastatic RCC,19 it may exert a favorable effect on selected patients with predominantly osseous metastases. In 2 studies that specifically reported patients with bone metastases (in addition to measurable lesions at other sites), 2 of 5 patients responded to IFN-γ treatment.20, 21
Our goal in this pilot study was to enhance the bone-targeting effects of zoledronate with potential antitumor or bone-modulating activities of thalidomide and IFN-γ for the treatment of bone metastasis in RCC patients. The endpoints of the study were 1) to delay skeletal-related events; 2) find evidence of calcification of osteolytic lesions; and 3) note changes in the levels of 3 bone markers (urinary concentrations of N-telopeptide [NTX] and deoxypyridinoline [DPD] and serum concentrations of bone-specific alkaline phosphatase [BSAP]). We report here results from the first 15 patients enrolled on a planned Phase II study of this regimen.
Patients in this study had histologically confirmed RCC. They had evidence of bone metastases on X-rays, bone scans, magnetic resonance images (MRI), or computed tomography (CT) scans. Bidimensionally measurable lesions were not required. Patients who had impending complications (such as pathologic fracture or spinal cord compression) from skeletal metastases were included only if those complications were controlled by surgery or radiation therapy with or without pain medications. Patients with brain metastases that were controlled by surgery or radiosurgery for at least 6 weeks and requiring no further therapy were eligible for the study. Prior chemotherapy or immunotherapy, including treatment with thalidomide, bisphosphonates, or IFN-γ, was allowed, but patients must not have received bisphosphonates during the 6 months before enrollment and thalidomide or IFN-γ during the 12 months before enrollment. Concurrent chemotherapy, surgery, or radiotherapy was not permitted. Patients with life-threatening illness unrelated to the tumor (e.g., active congestive heart failure, uncontrolled angina, or myocardial infarction) within 6 months before entry, and those with concurrent serious infection, were excluded. Concomitant use of steroids was not allowed, except for topical formulations. Patients of childbearing potential were required to practice adequate contraception. Life expectancy had to be at least 12 weeks and Zubrod performance score had to be 2 or less. Other eligibility criteria were having a platelet count greater than 70,000/μL, a neutrophil count of at least 1000/μL, transaminase and conjugated bilirubin levels less than twice the upper limit of normal, and a serum creatinine level of less than 2.0 mg/dL (or, if creatinine level was 2 mg/dL or higher, then creatinine clearance [estimated by Cockcroft formula] had to be at least 50 mL/min). Patients with hypercalcemia were eligible for the study. All patients gave written informed consent to participate before enrollment. The study had been approved by the institutional review board of the University of Texas M. D. Anderson Cancer Center.
Treatment consisted of zoledronate, 4 mg, administered intravenously once every 4 weeks; IFN-γ 100 μg given subcutaneously once a week; and thalidomide, 300 mg taken orally once a day. Each course of treatment lasted for 4 weeks. Bone marker levels were checked every 2 months (NTX and DPD in the urine, BSAP in serum), whereas imaging studies were obtained every 3 months. The patients' clinical histories, laboratory results, and treatment responses were collected from their medical records and from M. D. Anderson Cancer Center's computer data management system (NETPASS or ClinicStation). Survival data were obtained from the clinical records or from the Social Security Death Index Interactive Search Web Site (http://ssdi.genealogy.rootsweb.com/). Time to a skeletal-related event was defined as the time from registration to radiation therapy or a surgical procedure for a bone metastasis or to the development of a pathologic fracture or spinal cord compression. Time-to-event analysis was performed and survival curves generated by using the methods of Kaplan and Meier.22 Survival was calculated from the time of registration until death from any cause or last follow-up visit.
Fifteen patients were treated between November 14, 2002, and November 7, 2003. Baseline characteristics of these patients are summarized in Table 1. The median age of the patients was 59 years (range, 43–72 years). The male-to-female ratio was 6.5:1. Thirteen patients (87%) had previously undergone a radical nephrectomy. The pathologic diagnosis made from the nephrectomy specimens was conventional type in 12 cases and papillary type in 1 case. Two of the conventional-type tumors had sarcomatoid or rhabdoid features. Four patients (27%) had 3 or fewer focal sites of bone metastasis. Metastases were confined to the bone in 6 cases (40%). Nine patients (60%) had been diagnosed with RCC within the 12 months preceding enrollment (median, 5 months; range, 1–78 months). All patients had hemoglobin levels less than the lower limit of normal (15 g/dL for men and 12 g/dL for women; median, 12.2 g/dL; range, 8.7–13.8 g/dL). One (7%) patient had a corrected serum calcium level of more than 10 mg/dL (median, 8.8 mg/dL; range, 8.3–10.5 mg/dL). One patient (7%) had a serum lactate dehydrogenase level more than 1.5 times the upper normal limit of 927 IU/L (median, 441 IU/L; range, 288–1,419 IU/L). Ten (67%) patients had metastases in more than 1 organ site (median, 2 sites; range, 1–6 sites). Six patients (40%) had at least 3 of these clinical features, indicating a poor prognosis. Six (40%) patients had undergone surgery to repair pathologic fractures before registration; 8 (53%) had received external beam radiation therapy; and 1 (7%) had undergone embolization to control the bone metastases or associated symptoms before enrollment. Three patients (20%) had had prior systemic treatments, 1 with fluorouracil/leucovorin and high-dose IL-2, 1 with IFN-α and capecitabine/gemcitabine, and 1 with CP675,206, a monoclonal anti-CTLA4 antibody.
|Characteristic||No. of Patients (n = 15)|
|Age, y (median, range)||59 (43–72)|
|Conventional (clear cell)||12|
|Only site of metastasis||6|
|Surgery on bone metastasis||6|
|XRT to bone metastasis||8|
|<1 year from initial diagnosis||9|
|Zubrod performance status ≥2||1|
|Hemoglobin below LLN (male 15, female 12 g/dL)||15|
|Corrected serum calcium >10 mg/dL||1|
|LDH >1.5 time ULN (927 IU/L)||1|
|More than one site of metastasis||9|
The 15 patients received a total of 41 courses of treatments (median, 2 courses; range, 1–11 courses) (Table 2). The median time to disease progression was 8.3 weeks (range, 2.1–48 weeks). Three patients (20%) showed disease progression in the bone, 4 (27%) in nonosseous sites, and 6 (40%) in both bone and nonosseous sites. The median time to a skeletal event was 12.0 weeks (range, 3.9–46.4 weeks). Six patients received radiation to a symptomatic osseous metastatic site. One patient developed a pathologic fracture in the rib. One patient received an intrathecal pump and another underwent embolization/radiofrequency ablation to control bone pain during the study. Baseline urinary NTX or DPD levels were elevated in 2 of 13 (15%) patients and 8 of 10 (80%) patients, respectively; whereas baseline serum BSAP level was low in 2 of 12 (17%) patients. In terms of change in bone marker levels, urinary NTX or DOD concentrations decreased by at least 50% from baseline values in 4 of 11 patients (36%) and in 1 of 7 patients (14%), respectively; whereas BSAP increased by at least 33% in 2 of 8 (25%) patients (Fig. 1, Table 2). Two (13%) patients had evidence of sclerosis formation within osteolytic lesions, suggestive of a clinical response (Fig. 2). Those 2 patients also experienced a decrease in pain and required less pain medication during treatment. In 1 of those patients the response was difficult to interpret because of previous irradiation to the bone lesion in question after surgery. In general, patients who derived the most benefit from treatment had metastases confined to the bone and 3 or fewer lesions in the bone (Table 3). Those patients also showed a decrease in urinary NTX level (1 by 52% and 1 by 77% from baseline) or an increase in serum BSAP level (by 33% from baseline) during treatment. The median overall survival time for all 15 patients was 58 weeks (range, 29–89 weeks) (Fig. 3).
|Characteristic||No. of Patients|
|No. treatment courses received (median, range)||2 (1–11)|
|Time to progression, wk (median, range)||8.3 (2.1–48)|
|Site of progression|
|Both bone and nonosseous||6|
|Time to skeletal-related event, wk (median, range)||12.0 (3.9–46.4)|
|Type of skeletal event (n = 9 patients)|
|Intrathecal pump placement||1|
|Spinal cord compression||0|
|Urinary NTX (n-telopeptide)|
|Baseline above normal (n = 13)||2|
|Decreased from baseline by >50% (n = 11)||4|
|Urinary DPD (deoxypyridinoline)|
|Baseline above normal (n = 10)||8|
|Decreased from baseline by >50% (n = 7)||1|
|Serum BSAP (bone-specific alkaline phosphatase)|
|Baseline below normal (n = 12)||2|
|Increased from baseline by >33% (n = 8)||2|
|Overall survival, wk (median, range)||58 (29–89)|
|Patient ID||Site of Bone Metastases||Prior Treatments||Other Metastases||TTP, Weeks and Site||SRE, Weeks and Site||Survival, Weeks|
|1||Multiple||XRT T2-6, T12-L2||—||4.4 brain||10.6 rib fracture||24|
|4||Multiple||Surgery Calvarium||Lung||9.0 Lung, bone||—||88|
|5||Multiple||XRT/surgery femur, XRT L3-SIJ||Lung||3.9 bone||3.9 XRT T9-12||10|
|6||Ileum, sacrum||—||—||8.3 Lung, bone||44.9 EMB/RFA ileum||98|
|7||Multiple||XRT T12-L2, humerus||Mediastinum LN, lung||2.1 bone||5.0 IT cath||9|
|8||Multiple||—||Liver, lung||10.3 liver||15.4 XRT BOS||38|
|9||Humerus, sacrum,* rib||Surgery humerus||—||15.0 brain||—||89|
|10||Multiple||XRT humerus, T12-L4||Hilar LN, liver, lung||8.4 lung, liver, LN, bone||—||31|
|11||Pubis||—||Lung, pancreas||12.0 chest wall, bone||12.0 XRT pelvis||72|
|12||Multiple||Surgery/XRT humerus XRT knee||Adrenal, local||17.0 adrenal, bone||42.0 XRT T7||57|
|13||Multiple||XRT shoulder||Brain, Lung, skin, LN||8.0 brain, bone||7.3 XRT T12||21|
|14||Multiple||EMB sacrum||Adrenal, buttock, pleura, liver, LN||7.0 adrenal, buttock, bone||—||9|
|15||Multiple||XRT T7-11, L5, sacrum, femur||—||4.0 bone||46.4 XRT T9||71|
Seven (47%) patients developed NCI Common Toxicity Criteria Grade 3 or 4 toxic effects during treatment (Table 4), and 2 (13%) discontinued treatment as a result of worsening adverse reactions. Overall, 3 patients (20%) developed thrombotic complications from the treatment (2 lower-extremity deep venous thrombosis and 1 myocardial infarction) despite prophylactic anticoagulation with aspirin and warfarin. Four patients experienced Grade 3 or 4 dyspnea, although none showed any evidence of pulmonary embolism. Of these 4 patients, 1 was hospitalized for pneumonia, another experienced relief from dyspnea after a bowel movement, and the other 2 cases could be attributed to treatment. The 1 case of Grade 4 anemia and 2 cases of Grade 3 fatigue could have resulted from the disease and may have been exacerbated by the treatment. The 1 case of Grade 3 nausea could have been caused by narcotics. The skin rashes (1 Grade 1 and 1 Grade 3) and sensory neuropathy (1 Grade 1) could be attributed to thalidomide; the renal insufficiency (Grade 2) and hypocalcemia/hypophosphatemia (Grade 2) could be attributed to zoledronate. One patient experienced fever (Grade 2) and another dyspnea (Grade 3) with arthralgia (Grade 2) that could have resulted from either zoledronate or IFN-γ. Only 1 patient reported constipation (Grade 1), which could have been caused by thalidomide or narcotics; the low incidence of this symptom, however, might have been due to most patients having been given prophylactic laxatives.
|Grade 4||Grade 3||Grades 1-2|
|Cardiac ischemia/infarction||(1)||Thrombosis/ embolism||(2)||Fatigue||(4)|
The principal objective of this exploratory Phase II trial was to acquire data on the effects of a unique bone-targeted regimen on defined parameters of bone response in RCC. Specifically, the bone-response endpoints were: 1) time to development of skeletal-related events; 2) radiographic evidence of calcification in osteolytic lesions; and 3) baseline levels and patterns of change in bone marker levels during treatment. Although no formal guidelines exist for assessing response of bone lesions, evidence of calcification associated with clinical improvement suggests bone response. Results from the present study suggest that a bone-targeted regimen consisting of zoledronate, thalidomide, and IFN-γ benefits only a small fraction of patients with RCC and bone metastasis.
Our experience indicates that the proportion of patients with RCC whose osteolytic lesions show evidence of calcification during treatment is generally low (i.e., <10%). A calcification rate of 20% or higher, accompanied by evidence of therapeutic activity or clinical benefit (e.g., normalization of bone markers or freedom from skeletal events for 12 months) would suggest that the treatment warrants further study. Accrual to the planned trial was halted after the first 15 patients were enrolled, because fewer than 2 patients had evidence of bone response and it was unlikely that an acceptable rate of bone response (20%) could be achieved.
Finding an effective bone-targeted therapy could potentially improve the quality of life for patients with osseous metastases from various malignancies. At least two-thirds of the 570,000 patients with cancer who die in the US annually have bone metastases.23 Bone metastases can cause substantial morbidity in the form of severe pain, pathologic fracture, spinal cord compression, nerve impingement, or bone marrow failure. Optimizing treatment of bone metastasis requires the enlistment of expertise from a variety of specialties, including orthopedic surgery, neurosurgery, radiation oncology, medical oncology, interventional radiology, and pain service. An effective bone-targeted therapy is a critical component of this multidisciplinary therapeutic approach.
One way of assessing the success of bone-targeted therapy is based on a delay in the occurrence of skeletal-related events (e.g., the need for radiation therapy or surgical intervention or the development of complications such as pathologic fracture or spinal cord compression). In this study, 9 patients developed symptomatic progression of bone disease that required further treatment for palliation. Six of these patients required radiation therapy, 1 required placement of an intrathecal pump, and 1 underwent embolization/radiofrequency ablation. One patient developed a pathologic fracture in the rib. Two patients who were free of any skeletal events or complications at 89 and 112 weeks after enrollment had also achieved evidence of clinical response in the bone metastases, as described below.
Improvement of bone pain accompanied by the presence or an increase in the extent of sclerosis in osteolytic lesions also suggests a treatment response. Normally, sclerosis in an osteolytic lesion takes 3–6 months to become apparent on X-ray.24 In this study, a response based on this criterion became apparent on a CT scan within 2 months of treatment in 1 patient. In another patient, this criterion for response also applied, except that the sclerosis overlapped with a previously irradiated site and thus might not have reflected a treatment response solely to the current treatment. Our results suggest that the patients who appeared to benefit from this regimen had had metastases confined to the bone and had a relatively low metastatic disease burden (i.e., no more than 3 bone metastases). The value of giving zoledronate combined with other “bone-modulating” agents in the context of bone-targeted therapy for RCC or other malignancies needs to be further validated in clinical trials.
The advent of improved bone markers represents another means of assessing the efficacy or monitoring the clinical effects of bone-targeted therapy.25 Osteolytic metastases typically show increased osteoclastic activity. The rate at which osteoclasts erode the bone matrix can be determined by measuring bone matrix components released into the circulation during degradation. Several studies have reported that increased urinary levels of DPD, a collagen pyridinium cross-link, and NTX, a type I collagen telopeptide, correlate with the presence or extent of bone metastases.26, 27 Changes in these resorptive bone markers also seem to signal response to therapy or progression of bone metastases.25, 28, 29 Conversely, the rate of bone-matrix formation that occurs with healing or repairing of osteolytic metastases can be determined by measuring the enzymatic activity of the bone-forming cells (osteoblasts). BSAP, an osteoblast differentiation marker, is currently the most specific and sensitive marker of bone formation available.
In our study, baseline urinary DPD levels were elevated in 80% of patients, and baseline urinary NTX levels were elevated in 15% of patients; 14% of patients had an improvement (i.e., a drop by 50% or greater) in the urinary DPD levels, whereas 36% of patients had improvement in the urinary NTX levels during treatment. Although our study included far too few patients to allow us to draw firm conclusions, these results in the context of therapeutic effects suggest that urinary DPD level was more sensitive and reliable than NTX for both detecting osteolytic metastases and monitoring their response to therapy in RCC. Interestingly, urinary levels of DPD and NTX continued to increase in a patient who showed disease progression in spite of treatment (Patient 5, Table 3). Conversely, Patient 3, who had responded to treatment with improved pain control and development of calcification on radiographic studies, had decreased urinary NTX and increased serum BSAP levels. Clearly, more cases need to be studied and improved bone markers need to be evaluated to confirm the value of bone markers for de tecting bone metastases and monitoring the response of osseous metastases to therapy.
In terms of toxicity, the treatment regimen used in this study was tolerable overall. Some of the adverse effects (e.g., anemia, bone pain, fatigue) were likely to have been caused by the disease itself, and others (e.g., nausea, vomiting, constipation) could have been aggravated by the use of other medications, such as narcotics. The most serious adverse effect observed was thrombosis/embolism, with 1 patient developing myocardial infarction and 2 others developing deep venous thrombosis; moreover, it is possible that an occult pulmonary embolism could have caused the Grade 4 dyspnea in another patient. The thalidomide may have been responsible for these complications, despite prophylaxis with low-dose warfarin (2 mg orally daily) and aspirin. The risk of thrombotic or embolic complications from thalidomide could be minimized by reducing the dose to 50 to 200 mg per day and by selecting patients who may be less disposed to such complications, i.e., those without a history of cardiovascular disease, active tobacco use, or sedentary habits.
In conclusion, the results from the present study suggest that the therapeutic benefit of zoledronate was minimally enhanced when combined with thalidomide and IFN-γ. The clinical efficacy of bone-targeted therapy might be improved by selecting appropriate patients (e.g., those with a small burden of disease confined to the bone) for such treatment. Although the results of this exploratory study did not provide evidence to support the routine use of this regimen for bone-targeted therapy, they did help to establish a platform for the development of future studies that might lead to improved bone-targeted regimens for RCC and other malignancies.
We thank Christine Wogan for editorial assistance.