Bisphosphonate Treatment does not Affect Serum Levels of Osteoprotegerin and RANKL in Hypercalcemic Cancer Patients

Transcription

1 Bisphosphonate Treatment does not Affect Serum Levels of Osteoprotegerin and RANKL in Hypercalcemic Cancer Patients NIKLAS ZOJER 1, KARIN BRENNER 1, DORA BEKE 1, STEFAN KUDLACEK 2,4, GERHARD HAWA 3, WOLFGANG WOLOSZCZUK 3,4, LORENZ C. HOFBAUER 5 and MARTIN PECHERSTORFER 1 1 First Department of Medicine and Medical Oncology, Wilhelminenspital der Stadt Wien, Vienna; 2 Department of Medicine, Krankenhaus der Barmherzigen Brüder, Vienna; 3 Biomedica Group, Vienna; 4 Boltzmann Institut für Altersforschung, Vienna, Austria; 5 Department of Gastroenterology, University of Marburg, Germany Abstract. Bisphosphonates are the standard treatment for hypercalcemia of malignancy. We hypothesized that bisphosphonate treatment and the subsequent fall in serum calcium might induce changes in the RANK/RANKL/OPG system, which plays a pivotal role in the regulation of bone resorption. Soluble RANKL and OPG levels were measured in the serum of 15 hypercalcemic patients at baseline and on 5 consecutive days following treatment with the aminobisphosphonate ibandronate. At day 0, the median soluble OPG level was elevated (p=0.0021) in the hypercalcemic group as compared to normal controls, while the median serum RANKL level was not significantly different. Ibandronate treatment and the resulting decrease (p<0.0001) in serum calcium levels did not affect the serum concentrations of OPG, serum RANKL, or the serum RANKL/OPG ratio. In comparison with day 0, these factors did not change significantly at any time-point analyzed. In healthy individuals, the total serum calcium concentration is maintained within the narrow range of mmol/l. Parathyroid hormone (PTH) is the crucial regulating hormone in this process, with the PTH serum concentration varying inversely with the serum calcium level (1). PTH does not stimulate osteoclasts directly, but interacts with PTH receptors on osteoblasts, thereby increasing the expression of nuclear factor-î B (RANK) ligand (RANKL) in these cells. RANKL then binds to its receptor RANK on Correspondence to: Martin Pecherstorfer, MD, Professor of Medicine, 1st Department of Medicine and Medical Oncology, Wilhelminenspital der Stadt Wien, Montleartstrasse 37, A-1171 Vienna, Austria. Tel: , Fax: , martin.pecherstorfer@wienkav.at Key Words: Hypercalcemia, osteoprotegerin, RANKL, bisphosphonates. osteoclast precursors, inducing the formation of osteoclasts and subsequent bone resorption (2). There is a soluble form of RANKL (srankl), which can be released from the osteoblast precursor cell surface by the metalloproteasedisintegrin-tnf- -converting-enzyme. Osteoprotegerin (OPG) is the soluble decoy receptor of RANKL (3). When it binds to OPG, RANKL and srankl can no longer interact with RANK. Thus, OPG blocks the activation of osteoclasts by RANKL and srankl. In neoplastic bone disease, osseous destruction is mediated by the uncoupling of bone resorption and bone formation. In this uncoupling process, the RANK/RANKL/OPG system plays a pivotal role: a decrease in the ratio of the inhibiting OPG molecule to the stimulating RANKL molecule leads to an excess of bone resorption and underlies lytic bone lesions in both hematological and solid tumors (4). In hematological neoplasms, calcitriol and cytokines, such as interleukin-6 and TNF-, lead to the increased expression of RANKL on cells of the bone marrow microenvironment (5). A further possibility is that the tumor cells (e.g., myeloma cells) directly produce RANKL (6). In solid tumors, parathyroid hormonerelated peptide (PTHrP) is the critical mediator of metastatic bone destruction (7). This proteohormone closely resembles parathyroid hormone (PTH) and interacts with the same receptor on osteoblasts, as described above. Regardless of the type of tumor, hypercalcemia of malignancy is characterized by abnormally increased bone resorption, which releases a large amount of calcium into the circulation in an uncontrolled fashion. In tumors producing PTHrP, increased renal calcium reabsortion further contributes to the development of hypercalcemia (8). Hypercalcemia of malignancy is diagnosed in 20-40% of cancer patients at some time during the course of their disease. The condition is life-threatening if left untreated. However, normal serum calcium levels can be restored rapidly in most patients by re-hydration and application of potent bisphosphonates. Bisphosphonates have structural /2005 $

2 similarity to inorganic pyrophosphate and a high affinity for bone mineral. After i.v. application, bisphosphonates are rapidly cleared from the circulation and localize to hydroxylapatite bone mineral surfaces, especially at sites of increased osteoclast activation (9). The molecular mechanisms subsequent to localization of bisphosphonates to the bone which ultimately results in restoration of normal serum calcium levels are as yet not fully understood. However, it is clear that the inhibition of osteoclast activity and the resulting reduction of bone resorption are the major events leading to normocalcemia in initially hypercalcemic patients treated with bisphosphonates. In the present investigation, we tested the hypothesis that the restoration of normocalcemia is associated with changes in serum concentrations of osteoprotegerin and soluble RANKL following treatment of hypercalcemic cancer patients with the bisphosphonate ibandronate. Materials and Methods Patients and treatment. Fifteen hypercalcemic (serum calcium >2.7 mmol/l) cancer patients were treated with 6 mg ibandronate on day 0 of days 0-5. The duration of the study was determined by the fact that the majority of hypercalcemic patients are normocalcemic by the 5th day after treatment with bisphosphonates. The median patient age was 68 years (range years); 7 patients were male, 8 female. Five patients had breast cancer and 5 lung cancer. The other 5 patients suffered from esophageal cancer, prostate cancer, renal cancer, endometrial cancer and B-cell-non-Hodgkin s lymphoma, respectively. Bone metastases were present in 9 patients and absent in 4; bone involvement had not been determined in the remaining 2 patients. Nine patients had been previously treated with bisphosphonates, the last bisphosphonate treatment having been administered at least one month before day 0 of the study period. Six mg of ibandronate (1-hydroxy-3-[1-methylpentylamino]- propylidene bisphosphonic acid; Roche, Basel, Switzerland) was dissolved in 500 ml of normotonic saline solution and infused over the course of 1 hour. Blood samples were drawn at baseline (day 0) and on each of the next 5 days. Serum creatinine and serum calcium were determined by routine automatic methods. To measure OPG, srankl and PTH, aliquots were centrifuged and frozen immediately after the blood samples were collected. The study was performed in accordance with the Declaration of Helsinki, amended by the 29th (Tokyo) and the 35th (Venice) World Medical Assembly, and in accordance with pertinent Austrian laws. The protocol was approved by the local ethics committee. Patients and healthy controls gave their informed consent prior to enrolment in the study. Healthy controls. Sex- and age-matched healthy adults, acting as controls in the present investigation, participated in a large investigation of normative values of bone metabolism in the Austrian population. The criteria for recruitment of individuals for this investigation have been published elsewhere (10-11). Normative values stated below in the sections on OPG and srankl assays refer to these publications. The 15 control persons in our study were selected from the entire cohort based on agreement in sex and age with the cancer patients. OPG assay. The osteoprotegerin test kit (Biomedica Group, Vienna, Austria) is an enzyme immunoassay designed to determine the total amount (bound and unbound) of OPG in biological fluids (serum, plasma, cell culture supernatants). In the first step, the assay buffer, the sample and the biotinylated OPG detection antibody were added simultaneously to the wells. Osteoprotegerin, if present in the sample, binds to the precoated capture antibody and is sandwiched by the detection antibody. After a washing step, which removes all non-specific bound material, streptavidin-hrp conjugate was added to the wells. After removal of unbound conjugate by washing, tetramethylbenzidine (TMB) was added to the wells as substrate. Osteoprotegerin was quantitated by an enzyme-catalyzed color change detectable by a standard ELISA reader. The intensity of the color developed is directly proportional to the amount of osteoprotegerin present in the sample. In 447 healthy males, the median serum OPG concentration was 1.74 pmol/l. In 687 healthy females, the median serum OPG was 1.95 pmol/l. Based on the entire collective of healthy adults, the upper limit of normal range (mean + 2SD) of serum OPG was 7.67 pmol/l for males and 7.73 pmol/l for females (10). srankl assay. The srankl test kit (Biomedica Group) is an enzyme immunoassay designed to determine soluble, uncomplexed human RANKL in biological fluids (serum and cell culture supernatants). In the first step, samples and the biotinylated anti-srankl detection antibody were pipetted into the wells. Human srankl, if present in the sample, binds to the precoated recombinant OPG and forms a sandwich with the detection antibody. Following a wash to remove all non-specific bound material, streptavidin-hrp conjugate was added to the wells. After removal of unbound conjugate by washing, TMB was added to the wells as substrate. srankl was quantitated by an enzyme-catalyzed color change. The intensity of the color developed is directly proportional to the amount of srankl present in the sample. The color development was stopped after 30 minutes and the intensity of the color was measured with a standard ELISA reader. In 394 healthy males, the median srankl concentration was 0.94 pmol/l. In 635 healthy females, the median srankl was 0.37 pmol/l. Based on the entire collective of healthy adults, the upper limit of normal range (mean + 2SD) of srankl was 1.38 pmol/l for males and 1.09 pmol/l for females (11 and unpublished results). Statistical analysis. At baseline, the comparison of laboratory values in cancer patients and in healthy adults was performed using the nonparametric Mann-Whitney rank sum test. The intraindividual changes in laboratory values after ibandronate treatment were analyzed with the nonparametric Friedman test followed by Dunn s test. All tests were two-sided. A p-value <0.05 was considered significant. Results Comparison of baseline laboratory values in hypercalcemic cancer patients and healthy controls. As shown in Table I, hypercalcemic cancer patients had a higher median serum OPG concentration (p=0.0021) and a lower serum PTH level (p<0.0001) than did sex- and age-matched healthy adults. The serum srankl concentration did not differ 3608

3 Zojer et al: Bisphosphonates do not Affect OPG and RANKL in Hypercalcemic Patients Table I. Laboratory values (median; minimum maximum) in healthy adults and hypercalcemic cancer patients at day 0. Controls Hypercalcemic Controls versus hypercalcemic patients at day 0 patients at day 0 Serum srankl (pmol/l) 0.58 ( ) 0.15 ( ) n.s.* Serum OPG (pmol/l) 3.18 ( ) 7.93 ( ) p= srankl/opg 0.06 ( ) 0.01 ( ) p= Serum calcium (mmol/l) 2.46 ( ) ( ) p< Serum PTH (pg/ml) ( ) 3.10 (1.0 21) p< Serum creatinine (mg/dl) n.d.* 1.10 ( ) n.d. n.s.: not significant n.d.: not done None of the cancer patients had a serum srankl level above the sex-specific upper limit of normal range, whereas 8/15 (53%) had an increased serum OPG concentration. Course of laboratory values following treatment with ibandronate. As shown in Figure 1 and Table II, the infusion of ibandronate on day 0 led to a decrease in the median serum calcium concentration (p<0.0001) and an increase in the median serum PTH level (p<0.0001). At day 5, 12/15 patients (80%) were normocalcemic (serum calcium level < 2.6 mmol/l). In 3 patients (suffering from carcinomas of the lung, esophagus and breast, respectively), the serum calcium levels decreased but did not normalize within the observation period. The initial serum calcium levels in these patients were 4.53 mmol/l, 2.90 mmol/l and 3.36 mmol/l. The corresponding values at day 5 were 2.78 mmol/l, 2.65 mmol/l and 2.75 mmol/l. Ibandronate treatment and the resulting normalization, or at least decrease in serum calcium levels, did not affect the serum concentrations of OPG, srankl (see Table II, Figure 1) or the srankl/opg ratio (see Table II). In comparison with day 0, these values did not change significantly at any time-point analyzed. Discussion Figure 1. Course of the serum concentration of calcium (A), OPG (B) and srankl (C) following ibandronate treatment on day 0. While the bisphosphonate therapy significantly (p<0.0001) lowered the serum calcium concentration, OPG and srankl levels did not change. between cancer patients and controls. The median srankl/opg ratio was lower in tumor patients than in the control group (p<0.01). In vitro studies on the effects of bisphosphonates on the RANK/RANKL/OPG system have been controversial. Viereck et al. (12) demonstrated an up-regulation of OPG mrna in primary trabecular human osteoblasts during exposure to pamidronate and zoledronate. Pan et al. hypothesized that the effect of zoledronate on the recruitment and differentiation of osteoclast precursor cells is mediated indirectly by increasing OPG protein secretion and decreasing transmembrane RANKL expression in osteoblasts (13). In contrast, Kim et al. (14) found that the bisphosphonates pamidronate and alendronate left OPG and RANKL expression unchanged in osteoblasts, while they decreased osteoclast formation in co-culture experiments. There have 3609

4 Table II. Course of laboratory values (median; minimum-maximum) in initially hypercalcemic cancer patients. Day 0 Day 1 Day 2 Day 3 Day 4 Day 5 RM- ANOVA* Serum 0.15 ( ) 0.12 ( ) 0.14 ( ) 0.14 ( ) 0.14 ( ) 0.15 ( ) n.s. srankl (pmol/l) Serum 7.93 ( ) 8.13 ( ) 7.66 ( ) 7.64 ( ) 8.92 ( ) 8.59 ( ) n.s. OPG (pmol/l) srankl/ 0.01 ( ) 0.01 ( ) 0.01 ( ) 0.01 ( ) 0.01 ( ) 0.01 ( ) n.s. OPG Serum 3.10 ( ) 3.06 ( ) 2.81 ( ) 2.60 ( ) 2.53 ( ) 2.43 ( ) p< calcium (mmol/l) Serum 3.10 (1.0 21) n.d. n.d ( ) n.d ( ) p< PTH (pg/ml) Serum 1.10 ( ) 1.20 ( ) 1.10 ( ) 1.00 ( ) 1.00 ( ) 1.15 ( ) p=0.022 creatinine (mg/dl) *A repeated measurement analysis of variance on ranks (RM-ANOVA) was used to evaluate the intraindividual changes in laboratory values following ibandronate treatment on day 0. p<0.05 when compared to values on day 0 (Dunn s test) been only a few studies on the effects of bisphosphonates on the RANK/RANKL/OPG system in vivo so far. In patients with osteoporosis associated with beta-thalassaemia, repeated treatment with pamidronate was shown to decrease OPG serum levels over a period of 12 months (15). Treatment was associated with an increase in bone mineral density and a decrease in markers of bone turnover. RANKL levels remained unchanged during the study period. Similar data were obtained for patients with Paget s disease of the bone: daily oral treatment with tiludronate for 3 months led to a decrease in serum OPG concentrations, while srankl levels remained unchanged (16). Our study is the first to investigate the effects of bisphosphonates on the RANK/RANKL/OPG system in patients with hypercalcemia of malignancy. As evaluated by serial serum measurements, it was shown that srankl and OPG levels as well as the srankl/opg ratio remained unchanged during the first 5 days after treatment of hypercalcemia of malignancy with ibandronate, while the serum calcium level normalized in 80% of patients and decreased in all patients. Our findings thus support the in vitro findings by Kim et al. (14). It seems that an alternative mechanism independent of the serum srankl/opg ratio mediates the bisphosphonate action on osteoclasts. Nitrogen-containing bisphosphonates inhibit the mevalonate pathway, thus reducing protein prenylation in osteoclasts. This affects small guanosin-tri-phosphatases and signal transduction and, ultimately, may lead to osteoclast apoptosis (17). However, we cannot entirely rule out changes of RANKL/OPG in the bone microenvironment, which may not be reflected fully in changes in serum levels. Interestingly, serum OPG levels were elevated at baseline in our patient cohort versus age- and sex-matched normal controls. The elevation of serum OPG levels in hypercalcemic cancer patients probably reflects a mechanism counterregulating excessive bone resorption, although increased OPG production ultimately fails to prevent hypercalcemia. Indeed, as revealed by immunohistochemistry, RANKL and OPG co-localize in osteolytic lesions (18). The situation in myeloma is different, because OPG binds to syndecan-1 on malignant plasma cells and is rapidly internalized and degraded (19). The low level of OPG then contributes to the increased srankl/opg ratio in myeloma, which correlates very positively with the extent of bone disease and inversely with prognosis (20). 3610

5 Zojer et al: Bisphosphonates do not Affect OPG and RANKL in Hypercalcemic Patients In a murine model of hypercalcemia, OPG infusions led to a decrease in serum calcium level and osteoclast activity (21). However, whether pharmacological doses of OPG can exert a protective effect on bone when baseline OPG levels are high is unclear. A pronounced effect might be expected in myeloma, where bone destruction is associated with a decreased OPG level. In a phase I study, infusion of AMGN-0007, a recombinant OPG construct, led to a rapid, dose-dependent decrease of serum markers of bone resorption in patients with myeloma and breast cancerrelated bone metastases (22). The effect was comparable to that exerted by a single infusion of 90 mg pamidronate. It should be noted that OPG levels were not measured at baseline in these patients, and it remains unknown whether the efficacy of this therapy is compromised in patients with high baseline OPG levels. In contrast to the serum OPG levels, srankl was within normal range in our patients at baseline. This finding is somewhat surprising given the importance of RANKL in mediating the effects of PTHrP. In multiple myeloma, malignant plasma cells and tumor-infiltrating lymphocytes were shown to produce RANKL (6, 23), and the ratio of srankl to OPG was demonstrated to be of prognostic significance (20). Increased RANKL expression was shown to be present in bone marrow cells from postmenopausal women (24) and in osteolytic bone metastases (25) by flow cytometry and immunohistochemistry, respectively. This, at first, does not seem to translate into a higher serum level of RANKL in these disorders (26). However, it should be noted that only unbound srankl is measured by the test used in this and other studies. The higher OPG level in our patients might mask an increase in total srankl due to binding of srankl to OPG. This notion seems to be supported by the finding that, although srankl levels were in the normal range, srankl levels correlated with the severity of hypercalcemia at baseline in our patients (data not shown). In conclusion, we showed that serum srankl and OPG remain unchanged during successful treatment of hypercalcemia with ibandronate. Alternative mechanisms, independent of the RANK/RANKL/OPG system, seem to be responsible for the rapid calcium-lowering action of bisphosphonates. Acknowledgements We thank Jane Neuda for her editorial review of the manuscript. References 1 Parfitt AM, Mundy GR, Roodman GD, Hughes DE and Boyce BF: A new model for the regulation of bone resorption, with particular reference to the effects of bisphosphonates. J Bone Miner Res 11: , Lacey DL, Timms E, Tan HL et al: Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93: , Simonet WS, Lacey DL, Dunstan CR et al: Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 89: , Goltzman D: Osteolysis and cancer. J Clin Invest 107: , Mundy GR and Guise TA: Hypercalcemia of malignancy. Am J Med 103: , Sezer O, Heider U, Zavrski I, Kühne CA and Hofbauer LC: RANK ligand and osteoprotegerin in myeloma bone disease. Blood 101: , Roodman GD: Mechanisms of bone metastasis. N Engl J Med 350: , Pecherstorfer M, Zimmer-Roth I, Schilling T et al: The diagnostic value of urinary pyridinium cross-links of collagen, serum total alkaline phosphatase, and urinary calcium excretion in neoplastic bone disease. J Clin Endocrinol Metab 80: , Theriault RL and Hortobagyi GN: The evolving role of bisphosphonates. Semin Oncol 28: , Kudlacek S, Schneider B, Woloszczuk W, Pietschmann P and Willvonseder R: Austrian Study Group on Normative Values of Bone Metabolism. Serum levels of osteoprotegerin increase with age in a healthy adult population. Bone 32: , Hawa G, Brinskelle-Schmal N, Glatz K, Maitzen S and Woloszczuk W: Immunoassay for soluble RANKL (receptor activator of NF-Î B ligand) in serum. Clin Lab 49: , Viereck V, Emons G, Lauck V et al: Bisphosphonates pamidronate and zoledronic acid stimulate osteoprotegerin production by primary human osteoblasts. Biochem Biophys Res Commun 291: , Pan B, Farrugia AN, To LB et al: The nitrogen-containing bisphosphonate, zoledronic acid, influences RANKL expression in human osteoblast-like cells by activating TNFalpha converting enzyme (TACE). J Bone Mineral Res 19: , Kim YH, Kim GS and Jeong-Hwa B: Inhibitory action of bisphosphonates on bone resorption does not involve the regulation of RANKL and OPG expression. Exp Mol Med 34: , Voskaridou E, Terpos E, Spina G et al: Pamidronate is an effective treatment for osteoporosis in patients with betathalassaemia. Br J Haematol 123: , Alvarez L, Peris P, Guanabens N et al: 2003 Serum osteoprotegerin and its ligand in Paget's disease of bone: relationship to disease activity and effect of treatment with bisphosphonates. Arthritis Rheum 48: , Rogers MJ, Gordon S, Benford HL et al: Cellular and molecular mechanisms of action of bisphosphonates. Cancer 88(12 Suppl): , Grimaud E, Soubigou L, Couillaud S et al: Receptor activator of nuclear factor kb ligand (RANKL)/osteoprotegerin (OPG) ratio is increased in severe osteolysis. Am J Pathol 163: , Standal T, Seidel C, Hjertner O et al: Osteoprotegerin is bound, internalized, and degraded by multiple myeloma cells. Blood 100: ,

6 20 Terpos E, Szydlo R and Apperley JF: Soluble receptor activator of nuclear factor kappab ligand-osteoprotegerin ratio predicts survival in multiple myeloma: proposal for a novel prognostic index. Blood 102: , Capparelli C, Kostenuik PJ, Morony S et al: Osteoprotegerin prevents and reverses hypercalcemia in a murine model of humoral hypercalcemia of malignancy. Cancer Res 60: , Body JJ, Greipp P, Coleman RE et al: A phase I study of AMGN-0007, a recombinant osteoprotegerin construct, in patients with multiple myeloma or breast carcinoma related bone metastases. Cancer 97(3 Suppl): , Giuliani N, Colla S, Sala R et al: Human myeloma cells stimulate the receptor activator of nuclear factor-kappa B ligand (RANKL) in T lymphocytes: a potential role in multiple myeloma bone disease. Blood 100: , Eghbali-Fatourechi G, Khosla S, Sanyal A, Boyle WJ, Lacey DL and Riggs BL: Role of RANK ligand in mediating increased bone resorption in early postmenopausal women. J Clin Invest 111: , Jung K, Lein M, Stephan C et al: Comparison of 10 serum bone turnover markers in prostate carcinoma patients with bone metastatic spread: diagnostic and prognostic implications. Int J Cancer 111: , Schett G, Kiechl S, Redlich K et al: Soluble RANKL and risk of nontraumatic fracture. JAMA 291: , Received March 8, 2005 Accepted June 6,