|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Clinical Endocrinology & Metabolism , doi:10.1210/jc.2008-2324 Copyright © 2009 by The Endocrine Society Decreased Bone Turnover Despite Persistent Secondary Hyperparathyroidism during Prolonged Treatment with ImatinibSusannah O'Sullivan, Anne Horne, Diana Wattie, Fran Porteous, Karen Callon, Greg Gamble, Peter Ebeling, Peter Browett and Andrew GreyDepartments of Medicine (S.O., A.H., D.W., K.C., G.G., A.G.) and Molecular Medicine and Pathology (F.P., P.B.), The University of Auckland, Auckland 1142, New Zealand; and Department of Medicine (P.E.), Western Hospital, The University of Melbourne, Parkville, Victoria 3050, Australia Address all correspondence and requests for reprints to: Dr. O'Sullivan, Department of Medicine, The University of Auckland, Private Bag 92019, Auckland 1142, New Zealand. E-mail: s.osullivan{at}auckland.ac.nz.
Context: The tyrosine kinase inhibitor imatinib mesylate has an established role in the management of a number of malignant and proliferative conditions. Cross-sectional and short-term prospective studies have demonstrated secondary hyperparathyroidism during imatinib therapy, and variable changes in markers of bone turnover. Objective: Our objective was to determine the biochemical and skeletal effects of imatinib during long-term therapy. Design: This was a 2-yr prospective study. Setting: The study was performed at an academic clinical research center. Patients or Other Participants: Nine patients with bcr-abl positive chronic myeloid leukemia were included in the study. Interventions: Patients received Imatinib mesylate 400 mg/d. Main Outcome Measures: Serum and urine biochemistry, markers of bone turnover, and bone mineral density were measured. Results: Participants developed mild secondary hyperparathyroidism, with significant decreases in serum calcium and phosphate (P < 0.05 and P < 0.0001 vs. baseline, respectively) and an increase in PTH (P < 0.0001 vs. baseline). Biochemical markers of bone turnover demonstrated a biphasic response, with an initial increase in markers of bone formation being followed by a decrease in markers of both formation and resorption. Bone density at the lumbar spine increased [mean (95% confidence interval) change from baseline 3.6% (1.6, 5.5); P = 0.003] as did that at the total body [1.4% (0.2, 2.5); P = 0.065], whereas that at the proximal femur did not change [–0.12% (–3.0, 2.7); P = 0.93]. Body weight and fat mass increased significantly (P < 0.0001 vs. baseline). Conclusions: Long-term treatment with imatinib leads to persistent mild secondary hyperparathyroidism. Despite this, bone turnover is decreased, and bone density is stable or increased. Evaluation of the skeletal actions and safety of imatinib during longer-term therapy is warranted.
Imatinib mesylate (Glivec; Novartis International AG, Basel, Switzerland) is an orally active tyrosine kinase inhibitor with activity against c-abl (including bcr-abl), c-kit, and the platelet-derived growth factor receptor (1). It has an established role in the management of patients with bcr-abl positive chronic myeloid leukemia (CML) (2), gastrointestinal stromal cell tumor (3), and several disorders that are characterized by activated platelet-derived growth factor receptor signaling (4), and is being investigated as a treatment for other metabolic, inflammatory, and proliferative conditions (5, 6, 7, 8, 9, 10). "Bystander" effects of imatinib, attributable to inhibition of its molecular targets in normal tissues, include hypopigmentation, diarrhea, hypogonadism in men (11, 12), and cardiotoxicity (13). Recently, several studies have demonstrated altered bone and mineral metabolism in patients taking imatinib (14, 15, 16, 17). In a short-term (6 months) prospective study, we found that patients with CML treated with imatinib developed secondary hyperparathyroidism, with uncoupling of markers of bone turnover, such that bone formation was increased, without change in bone resorption (14). Cross-sectional studies have found similar biochemical changes, but markers of bone turnover (both formation and resorption) were low normal (15, 17). In vitro, imatinib has complex effects on skeletal tissue: it stimulates differentiation, and inhibits proliferation and survival, of osteoblasts, and inhibits osteoclastogenesis (18). Thus, whereas initiation of imatinib therapy may lead to an increase in bone formation in the short term, there is potential for long-term therapy to cause decreased bone turnover, as a result of inhibition of osteoclastogenesis and osteoblast proliferation and survival. Such complex skeletal actions make it difficult to predict the effect of imatinib on the skeleton in vivo. To date, there are no prospectively collected biochemical data beyond 6 month treatment, and no bone mineral density (BMD) data collected during imatinib therapy. A recent retrospective analysis of iliac crest trephine biopsies suggested increased trabecular bone volume in CML patients taking imatinib (16). Because patients who respond to imatinib continue this treatment indefinitely, there is a need for evaluation of the biochemical and skeletal effects of the drug during long-term therapy. Here, we report the results of the first long-term prospective assessment of bone and mineral metabolism in patients treated with imatinib.
Subjects The study population has previously been described (14). Nine subjects (six male, three female, mean age 46 yr, range 32–60) with newly diagnosed bcr-abl positive CML were studied. None had metabolic bone disease, or was receiving medications known to influence bone or calcium metabolism. The median daily dose of imatinib over 24 months was 400 mg daily (range 375–488). Five subjects received 400 mg imatinib daily throughout the study. The imatinib dose in three subjects was temporarily lowered (to 100–375 mg daily) because of drug-related side effects (pancytopenia, liver dysfunction, gastrointestinal symptoms, and rash), but in each case the dose was increased back to 400 mg daily within 10 wk. Two patients ceased imatinib therapy before completing 24 month treatment: one after 18 months due to the development of imatinib-resistant disease, and one after 23 months due to a lack of cytogenetic response. The latter patient received 600 mg daily for 9 months before discontinuation of therapy. Biochemistry
All subjects provided serum, plasma, and urine samples, after overnight fast, before initiation of imatinib therapy, and at 3, 6, and 18 months after commencing imatinib treatment. Samples were stored at –70 C until the end of the study, and then batch analyzed. Serum calcium, albumin, phosphate, and creatinine, and urine phosphate and creatinine were measured on a Roche Modular autoanalyzer (Roche Diagnostics, Mannheim, Germany). Glucose and amino acid levels were measured in urine samples collected at baseline and 3 months. Urinary amino acids were measured by thin-layer chromatography (19). Albumin-adjusted serum calcium was calculated using the formula: adjusted total Calcium = measured total Calcium + [0.8 x (4.0-measured serum Albumin)]. Tubular maximum for phosphate reabsorption (TmP)/glomerular filtration rate (GFR) was calculated as previously described (20). Serum 25 hydroxy vitamin D (25 OHD) was measured by RIA (DiaSorin Inc., Stillwater, MN). 1,25 dihydroxyvitamin D [1,25(OH)2 D] levels were measured by a Bone turnover Serum levels of β-C-terminal telopeptide of type I collagen (βCTX), osteocalcin, and procollagen type-I N-terminal propeptide (P1NP) were measured as previously described (21). Coefficients of variation of these markers are as follows: osteocalcin, 5.5%; βCTX, 5.1%; and P1NP, 1.9%. BMD BMD was measured at baseline, and 6, 12, 18, and 24 months using a Lunar Prodigy densitometer (General Electric Lunar, Madison, WI). Separate scans of the lumbar spine, proximal femur, and total body were obtained at each time point. Data obtained using this densitometer in a placebo-controlled clinical trial conducted over the same time period as the current study demonstrated stable measurements in the placebo group (21, 22). Fat mass was measured using the total body scans. Body weight was measured using electronic scales Statistical analyses Statistical analyses were performed using a mixed models approach to repeated measures (PROC Mixed, SAS version 9.1; SAS Institute Inc., Cary, NC). Tests for linear and quadratic trend were performed on the time course data. Significant time effects were investigated post hoc using the method of Dunnett to preserve an overall 5% significance level. All tests were two tailed. The study was approved by the Auckland Ethics Committee, and all participants provided written, informed consent.
Baseline demographical, biochemical, and bone density data are shown in Table 1  .
Biochemistry
Mild secondary hyperparathyroidism was present throughout the study, although mean levels of serum calcium, phosphate, and PTH remained within the respective normal ranges (Fig. 1
Bone turnover
Levels of the bone formation markers P1NP and osteocalcin were higher than baseline values during the first 6 months but declined subsequently (Fig. 2
BMD
At 24 months, BMD at the spine was significantly higher than at baseline: mean (95% CI) change 3.6% (1.6, 5.5); P = 0.003; Fig. 3A
This study is the first to assess prospectively the biochemical and skeletal effects of long-term imatinib therapy. Previous cross-sectional studies and a short-term prospective study demonstrated that imatinib therapy is associated with the development of secondary hyperparathyroidism (14, 15, 17). The current data demonstrate that these biochemical changes persist, but do not progress, during long-term therapy with imatinib. There is a biphasic effect on bone turnover, such that in the early phase of treatment, bone formation is increased, but with continuation of therapy, both components of bone remodeling are decreased. The latter finding is consistent with data from cross-sectional studies, in which markers of bone turnover were low normal (15, 17). We found that BMD increases at the spine and total body, and remains stable at the proximal femur, during 2 yr imatinib therapy. The biphasic response of bone formation to treatment with imatinib is consistent with in vitro data, which demonstrate that the drug both stimulates osteoblast differentiation and inhibits osteoblast proliferation and survival (16, 18). The initial increase in bone formation in vivo likely reflects stimulation of differentiation of a preexisting pool of osteoblast precursors, although we cannot exclude the possibility that recovery from illness contributed to this effect. With ongoing therapy, levels of bone formation decrease, as the pool of osteoblast precursors is depleted by the antiproliferative and pro-apoptotic effects of the drug. Imatinib also decreases osteoclast development and survival in vitro (18, 23, 24, 25), and the long-term effect of imatinib to decrease bone resorption in vivo is in keeping with these observations. Although bone resorption is decreased after long-term imatinib therapy, the pattern of change in markers of bone turnover is not typical of treatment with antiresorptive drugs, such as bisphosphonates (26). It is noteworthy that the coupled decrease in bone turnover markers occurs despite the presence of secondary hyperparathyroidism. In the setting of biphasic changes in bone remodeling, it is difficult to predict the in vivo effects on bone mass. The current study reports the first prospectively collected BMD data in imatinib-treated subjects. The results of these analyses demonstrate stable or increased BMD at all sites. It is likely that direct skeletal effects of imatinib, namely the initial increase in bone formation, and later inhibition of bone resorption, are the principal contributors to the positive effects on BMD. However, our data do not exclude the possibility of indirect actions of imatinib on BMD. Body weight and fat mass are positive regulators of BMD (27), so the significant increases we observed in each of these variables during imatinib therapy might contribute to the positive effects on BMD. We could not find published data regarding the effect of imatinib on weight and/or fat mass, and it is possible that these increases are attributable to treatment with imatinib rather than reflecting recovery from illness. This notion is supported by the finding in our study that body weight and fat mass continued to increase beyond 12 months, well after the symptomatic improvement that occurs in response to initiation of imatinib therapy. Our BMD data are consistent with the report of increased trabecular bone volume in a retrospective analysis of trephine biopsy samples in a group of imatinib-treated patients (16). The current data also provide some reassurance in regard to the biochemical effects of long- term therapy with imatinib. Although serum calcium and phosphate remain lower at 18 months than at baseline, mean levels do not decrease below the normal range, and values of the latter do not approach those that are associated with impaired mineralization (28). We also investigated other potential contributing factors to the phosphaturia induced by imatinib. There has been one case report of partial Fanconi syndrome associated with imatinib therapy (29), but we did not find evidence of generalized proximal renal tubule dysfunction in our patients. Neither did we detect changes in circulating levels of the phosphatonin FGF-23 (30). However, we cannot exclude the possibility that imatinib might directly affect renal tubular phosphate reabsorption, independent of changes in PTH. It is likely that the skeletal actions of imatinib contribute substantially to the persistent mild secondary hyperparathyroidism. Thus, the increase in bone formation relative to resorption in the first 6 months, and the subsequent decrease in bone resorption relative to formation are probably pivotal to the ongoing secondary hyperparathyroidism. A possibility that we are not able to address in this study is that imatinib also decreases intestinal calcium absorption, thereby contributing to the development of secondary hyperparathyroidism. To our knowledge there is no evidence from preclinical studies that imatinib alters intestinal calcium absorption. Our study has some limitations. The small sample size is a reflection of the low incidence of CML. However, the effects we observed were both internally consistent, and consistent with findings from existing cross-sectional studies. Second, there was no control group because it would have been unethical to leave patients with CML untreated. In summary, the current data demonstrate that therapy with imatinib causes a biphasic change in bone turnover, with an initial stimulation of bone formation being followed by coupled suppression of bone resorption and formation. Stable mild secondary hyperparathyroidism persists for at least 18 months. BMD is stable or increases during 2 yr treatment. There is no evidence of either proximal renal tubular dysfunction or altered production of FGF-23. These data suggest that routine monitoring of serum biochemistry and BMD in imatinib-treated patients may not be necessary during the first 2 yr of therapy. Evaluation of the skeletal actions and safety of imatinib during longer-term therapy is warranted.
This work was funded by the Health Research Council of New Zealand. S.O. is the recipient of postgraduate scholarships from the Australia and New Zealand Bone and Mineral Society, and the University of Auckland. A.G. is the recipient of a University of Auckland Early Career Excellence Award. Disclosure Summary: S.O., A.H., D.W., F.P., K.C., G.G., P.E., and A.G. have nothing to declare. P.B. has received research funding and consulting fees from Novartis. First Published Online January 27, 2009 Abbreviations: BMD, Bone mineral density; CI, confidence interval; CML, chronic myeloid leukemia; βCTX, β-C-terminal telopeptide of type I collagen; 1,25(OH)2 D, 1,25 dihydroxyvitamin D; FGF-23, fibroblast growth factor 23; GFR, glomerular filtration rate; 25 OHD, 25 hydroxy vitamin D; P1NP, procollagen type-I N-terminal propeptide; TmP, tubular maximum for phosphate reabsorption. Received October 24, 2008. Accepted January 21, 2009.
This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
-B 1,25(OH)2 D RIA after immunoextraction from serum (Immunodiagnostic Systems Ltd., Bolden, UK). Intact PTH and total testosterone were measured using electrochemiluminescence immunoassays (E170; Roche Diagnostics). Fibroblast growth factor 23 (FGF-23) levels were measured by ELISA (Kainos Laboratories, Tokyo, Japan). 
