The Clinical Use of Bone Resorption Markers in Patients with Malignant Bone Disease

Transcription

1 2521 The Clinical Use of Bone Resorption Markers in Patients with Malignant Bone Disease Robert E. Coleman, M.D. Cancer Research Centre, YCR Academic Unit of Clinical Oncology, Weston Park Hospital, Sheffield, United Kingdom. BACKGROUND. Advanced tumors often metastasize to bone, resulting in a variety of skeletal complications. Bisphosphonates are potent inhibitors of osteoclast-mediated bone resorption that reduce the incidence and delay the onset of skeletal complications and reduce the need for radiation and surgery. Biochemical markers of bone resorption have been identified that can augment the imaging techniques used to diagnose bone metastases and assess response to bisphosphonate therapy. METHODS. In the current study, the available literature regarding bone resorption markers is reviewed and the clinical relevance of these data with respect to the treatment of bone metastases discussed. RESULTS. Urinary calcium and hydroxyproline have been widely used to assess bone metabolism, but do not appear to be well correlated with clinical outcome in patients with bone metastases. Several unique breakdown products of Type I collagen (including pyridinium crosslinks, pyridinoline, and deoxypyridinoline) and peptide-bound crosslinks (N-telopeptide and C-telopeptide) are more specific and sensitive markers of bone resorption. N-telopeptide and C-telopeptide have been identified as the most sensitive biochemical markers currently available for detecting bone metastases and for assessing response to therapy or disease progression. CONCLUSIONS. To the author s knowledge markers of bone resorption have not yet been recommended for routine clinical use. However, further research is needed to define their potential role in the diagnosis of bone metastases, the assessment of disease progression and response to bisphosphonate therapy, and predict the rate of bone loss and the potential for fracture. Suppression of bone resorption markers in response to bisphosphonate therapy appears to correlate with clinical outcome in patients with both osteolytic and blastic bone lesions; therefore, the goal of bisphosphonate therapy should be to suppress markers of bone resorption. Cancer 2002;94: American Cancer Society. DOI /cncr KEYWORDS: bisphosphonate, bone metabolism, bone metastases, bone resorption markers, pamidronate, zoledronic acid. Address for reprints: Robert E. Coleman, M.D., Cancer Research Centre, YCR Academic Unit of Clinical Oncology, Weston Park Hospital, Sheffield, S10 2SJ, UK; Fax: (011) 44 (0) , r.e.coleman@sheffield.ac.uk Received August 28, 2001; revision received November 26, 2001; accepted December 19, Advanced tumors frequently metastasize to the bone, and the resulting bone destruction is associated with a variety of skeletal complications, including pathologic fractures, bone pain, impaired mobility, spinal cord compression, and hypercalcemia. 1 It is estimated that 1.5 million cancer patients worldwide have bone metastases. Patients with multiple myeloma and patients with advanced carcinomas of the breast, prostate, thyroid, bladder, and lung are at the highest risk of developing skeletal lesions and associated complications (Table 1). 2 Patients may present with intractable bone pain and often suffer debilitating and painful fractures that can seriously erode their quality of life. Breast carcinoma patients are at particularly 2002 American Cancer Society

2 2522 CANCER May 15, 2002 / Volume 94 / Number 10 TABLE 1 Frequency of Skeletal Involvement among Patients with Advanced Cancer Cancer type Frequency of bone metastases (%) Myeloma Breast ca Prostate ca Thyroid ca 60 Bladder ca 40 Lung ca Renal ca Melanoma ca: carcinoma. Reprinted with permission from: Coleman RE. Skeletal complications of malignancy. Cancer 1997;80(Suppl): Copyright 1997 American Cancer Society. Reprinted by permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc. high risk, with pathologic fractures occurring in approximately 60% of patients with bone metastases with a median onset of 11 months from the initial diagnosis of bone involvement. 3,4 In a large clinical database, breast carcinoma patients were found to have a mean skeletal morbidity rate of four skeletalrelated events per year. 3 In patients with multiple myeloma, osteopenia of the lumbar spine is a common feature, resulting in a high incidence of vertebral fractures. 5,6 Current treatment options for patients with bone metastases include radiation therapy, surgery, bisphosphonates, and analgesics, in addition to standard anticancer therapy. The primary goal of therapy is to minimize bone pain and morbidity and improve mobility and quality of life. Radiotherapy stabilizes bone lesions and rapidly relieves bone pain from local tumor effects in up to 85% of patients, 7 and orthopedic surgery may be indicated to stabilize the bone, thus preventing fractures and relieving pain. Bisphosphonates, which are potent inhibitors of osteoclast-mediated bone resorption, have become an integral part of the current treatment of skeletal metastases because of their proven ability to reduce the incidence of skeletal complications and delay their onset Bisphosphonate therapy also can reduce the need for irradiation of and surgery to bone. In addition, bisphosphonates appear to reduce the skeletal tumor burden in animal models by making the bone microenvironment a less favorable site for tumor cell growth. 13 Consistent with this observation, the results of several clinical trials have suggested that bisphosphonate therapy may improve survival in certain subsets of patients, particularly when administered early (i.e., before there is clinical evidence of bone metastases). 3,9,14 In light of the significant morbidity associated with bone metastases and the well documented FIGURE 1. Fluoride-18 positron emission tomography scan (maximum pixelintensity projection) showing disseminated metastatic bone disease in a patient with breast carcinoma. Reprinted with permission from: Schirrmeister H, Guhlmann A, Kotzerke J, et al. Early detection and accurate description of extent of metastatic bone disease in breast cancer with fluoride ion and positron emission tomography. J Clin Oncol 1999;17: benefits of bisphosphonate therapy, the American Society of Clinical Oncology (ASCO) guidelines regarding the role of bisphosphonates in the treatment of breast carcinoma recommend the use of bisphosphonates in any patient with evidence of lytic bone destruction on plain-film radiographs. 15 Assessment of Bone Metastases Imaging techniques remain the standard method of assessing bone lesions. Plain-film radiography and technetium ( 99m Tc) bone scans are the standard methods for detecting bone lesions, and radiography is the standard method for assessing response to bisphosphonate therapy and progression of bone lesions. Osteolytic bone lesions typically appear as low-density areas on radiography (Fig. 1) and areas of enhanced activity on bone scans; however, the lesions associated with multiple myeloma often appear normal on a 99m Tc bone scan. Plain-film radiography and 99m Tc bone scans are limited by low sensitivity. Consequently, substantial damage to the bone must occur before a lesion can be detected by these techniques. Areas of healing within sites of previous lytic disease can be observed on 99m Tc bone scans 16 and may be confused with disease progression. Therefore, in patients with advanced disease, bone scans should be interpreted with caution within 6 months of a change in therapy. Bone scans most likely are most useful for restaging after disease recurrence to identify sites for radiologic assessment and those patients at risk of

3 Clinical Use of Bone Resorption Markers/Coleman 2523 FIGURE 2. N-telopeptide (NTX) values in patients with 1) newly diagnosed bone metastases, 2) progressive bone metastases, or 3) progressive bone metastases and hypercalcemia. Line indicates the geometric mean. pathologic fracture. Computed tomography (CT) is a sensitive imaging technique, offering three-dimensional information and high-quality images, although it is impractical to image more than a limited part of the skeleton. Therefore, CT is used primarily as a confirmatory technique and to assess healing of small lytic lesions. Both magnetic resonance imaging (MRI) and fluoride-18 positron emission tomography (18-F- PET) are reported to be more sensitive than CT or 99m Tc bone scans. MRI can be used to image the entire skeleton, may provide early diagnosis of bone metastasis, and can reliably detect changes in bone associated with response to treatment, disease progression, or fracture. 17 The newest imaging technique is 18-F- PET. A recent analysis of the sensitivity of 18-F-PET demonstrated that it could accurately detect small bone lesions in patients with breast carcinoma at an earlier stage than 99m Tc bone scans (Fig. 2). 18 Given the evidence demonstrating the sensitivity and accuracy of these alternative imaging techniques, the ASCO Bisphosphonates Expert Panel concluded that it is reasonable to administer bisphosphonates to any breast carcinoma patient who has bone pain and an abnormal 99m Tc bone scan, CT scan, or MRI scan. 15 However, these imaging modalities are considered expensive and time-consuming, and are confined largely to tertiary care centers. Biochemical markers Biochemical assessment of response is an attractive alternative approach. Tumor markers such as prostate specific antigen (PSA) are well established in the assessment of prostate carcinoma, and the tumor markers CA 15-3 and carcinoembryonic antigen (CEA) have been reported to be of some value in monitoring response in patients with advanced breast carcinoma. 19,20 However, biochemical assessment of bone metabolism is perhaps the most promising approach and is applicable to the range of tumors affecting bone. Markers of bone formation such as osteocalcin and alkaline phosphatase bone isoenzyme have also been evaluated. 16,19,21 However, the changes associated with bone healing, and the flare response are difficult, if not impossible, to differentiate from those caused by progressive disease. 21 Recently, highly specific biochemical markers of bone resorption have been identified that could augment these imaging techniques. These assays, based on the measurement of bone breakdown products, are straightforward and convenient, and emerging data suggest a correlation with clinical outcome. Because osteolytic lesions are associated with increased bone resorption, patients often have elevated levels of calcium and other components of the bone matrix, such

4 2524 CANCER May 15, 2002 / Volume 94 / Number 10 TABLE 2 Markers of Bone Metabolism Marker Urinary markers Calcium/creatinine ratio Hydroxyproline/creatinine ratio N-telopeptide/creatinine ratio C-telopeptide/creatinine ratio Pyridinoline/creatinine ratio Deoxypyridinoline/creatinine ratio Serum markers Bonealkalinephosphatase (bone formation) Osteocalcin (bone formation) N-telopeptide C-telopeptide Bone sialoprotein Normal range mm/mm creatinine mm/mm creatinine nm/mm creatinine a nm/mm creatinine a nm/mm creatinine a nm/mm creatinine a 4 20 g/l 3 13 ng/ml nm BCE nm/l g/l a BCE: bone collagen equivalents. a Range of the mean values for men (n 27), premenopausal women (n 30), and postmenopausal women (n 30). Data taken from: Woitge HW, Pecherstorfer M, Li Y, et al. Novel serum markers of bone resorption: Clinical assessment and comparison with established urinary indices. J Bone Miner Res 1999;14: as breakdown products of Type I collagen, in the serum and urine (Table 2). 22 Type I collagen accounts for approximately 85% of the total protein in bone tissue. 19 The byproducts of bone resorption provide surrogate markers that can be used to diagnose skeletal disease and to assess disease progression and response to therapy. Two traditional markers of bone resorption, urinary calcium and hydroxyproline, have been widely used for many years, particularly in clinical trials of bisphosphonate therapy, to assess response to treatment. Recently, several bone specific biochemical markers that constitute unique breakdown products of Type I collagen have been evaluated and have been found to demonstrate high levels of sensitivity and specificity. Biochemical markers of bone resorption could play a variety of roles in the management of metastatic bone disease. For example, they could be used to identify patients at risk of developing skeletal complications or patients who may or may not benefit from bisphosphonate therapy, and they may be useful for assessing response to treatment in clinical trials. 19 Recent studies suggest that bone resorption markers may predict the rate of bone loss, the risk of disease progression, and the potential for fracture in patients with lytic bone disease It also has been suggested that the newer bone specific markers may be sensitive enough to provide early diagnosis of bone metastases. 25,26 Nevertheless, to my knowledge, bone resorption markers have not yet been rigorously shown to be statistically significant predictors of prognosis or response to therapy. Therefore, the current ASCO guidelines for the use of bisphosphonates in patients with breast carcinoma recommend that these markers be used only within research protocols, and they do not yet play a role in routine practice. 15 The clinical utility of bone resorption markers in the diagnosis and treatment of patients with malignant bone disease is reviewed in detail herein. BONE RESORPTION MARKERS The hallmark of tumor-induced bone resorption is increased activity of osteoclasts, typically associated with a concomitant decrease in bone collagen synthesis by osteoblasts. 27 This uncoupling of the normal physiology of bone remodeling results in abnormal bone resorption. Osteoclasts produce a number of proteolytic enzymes capable of degrading the organic bone matrix, thus releasing calcium and a variety of collagen breakdown products into the serum. These byproducts of pathologic bone resorption are excreted primarily by the kidneys and therefore also can be measured in the urine. The molar ratio of their excretion relative to creatinine excretion provides a reproducible measurement of their rate of excretion that is independent of body size or urine dilution. Urinary calcium and hydroxyproline are traditional biochemical markers that have been widely used for many years to assess bone metabolism in patients with skeletal metastases and other metabolic disorders of the bone. However, neither is a highly specific marker for tumor-induced bone resorption. The calcium/creatinine ratio in an early morning urine sample after an overnight fast has been shown to be a reproducible method of quantifying calcium excretion, 31 and calcium excretion has been reported to be a useful marker of therapeutic response in patients with osteolytic bone lesions. 21,32 However, studies have demonstrated that in unselected groups of patients with bone metastases urinary calcium was not found to be increased significantly compared with controls or patients without bone metastases, 12,33 and there was no apparent correlation between urinary calcium and clinical findings or response to bisphosphonate treatment. Moreover, calcium excretion is affected by diet, renal function, uptake into bone, and circulating levels of parathyroid hormone and parathyroid hormone-related protein. 19 Hydroxyproline is a major amino acid constituent of collagen, and its excretion typically is elevated in the presence of abnormal bone resorption or formation. 20 Although much of the hydroxyproline released from the bone is oxidized in the liver, approximately 15% appears in the urine. 20 Measurements of hydroxyproline can be made on either a 24-hour urine

5 Clinical Use of Bone Resorption Markers/Coleman 2525 collection or on the second voided early morning urine sample after an overnight fast. Hydroxyproline is not strictly a bone specific marker; only approximately 50% of human collagen is localized to the bone. 19,34 It also is a major constituent of several other human proteins, including acetylcholinesterase, complement factor C1q, and elastin. Urinary excretion of hydroxyproline also is influenced strongly by diet, 35 age, and soft tissue destruction by tumor, 20 and has a circadian rhythm, with a peak between midnight and 8:00 a.m. 36 Although urinary hydroxyproline is a useful indicator of collagen breakdown in metastatic bone disease, 37 its utility for documenting disease progression and response to therapy has been questioned. 20 Several studies have failed to find a strong correlation between hydroxyproline levels and response to bisphosphonates or radiologic response to systemic therapy in patients with breast carcinoma In patients receiving oral or intravenous bisphosphonates, hydroxyproline levels demonstrated poor sensitivity. Indeed, urinary calcium was more sensitive in these studies. 39,41,42 In patients receiving only systemic anticancer therapy, hydroxyproline levels increased in 30 40% of patients despite evidence of stable disease or an objective radiologic response of bone lesions. 38,40 Therefore, the hydroxyproline/creatinine ratio is not a highly reliable or sensitive marker of bone resorption. The results from these studies suggest that these traditional markers of bone resorption are suboptimal in terms of selectivity and sensitivity. More recently, a number of biochemical markers have been developed that provide more specific and sensitive indications of bone resorption (Table 2). 22 These new markers include several unique breakdown products of Type I collagen, including the pyridinium crosslinks pyridinoline (PYD) and deoxypyridinoline (DPD), and the peptide-bound crosslinks N-telopeptide (NTX) and C-telopeptide (CTX). 22,43 46 In comparison with calcium and hydroxyproline, these collagen breakdown products are more specific to bone and do not appear to be influenced by diet or metabolism. 25 Pyridinium Crosslinks Pyridinoline and DPD crosslinks are both specific to bone, and can be quantitated in the urine using reverse-phase high-performance liquid chromatography and/or an enzyme-linked immunosorbent assay (ELISA), and their excretion relative to creatinine is affected only minimally by renal function. 47 However, their excretion varies substantially throughout the day with a circadian rhythm, and DPD also can vary from day to day. 19 Therefore, samples must be taken at the same time each day, and the collection of two samples on consecutive days is ideal to establish a reliable baseline DPD value. In patients with pathologic bone resorption, a good correlation has been observed between PYD and DPD excretion and radiologic or histomorphometric measurements of bone resorption. 48,49 Moreover, increased excretion of these crosslinks appears to be well correlated with bone resorption in a variety of pathologic conditions, including osteoporosis, Paget disease, and primary hyperparathyroidism, and as a result of fractures. 19 In the majority of patients with bone metastases, excretion of pyridinium crosslinks typically is increased by 2.5-fold compared with healthy controls and also is elevated significantly compared with cancer patients without bone metastases. Patients with breast carcinoma but without bone metastases also may exhibit slightly increased PYD and DPD excretion, attesting to the sensitivity of these markers. 50 The minor increases in crosslinks in these patients likely reflect systemic stimulation of bone resorption by circulating tumor-derived parathyroid hormone-related protein. In patients with bone metastases, the levels of PYD and DPD decrease rapidly in response to bisphosphonate therapy and have been shown to be correlated with pain scores. 12,51,62 DPD is the more sensitive indicator of response to bisphosphonate therapy. In a cohort of 51 patients, DPD was found to decrease significantly in patients treated with pamidronate compared with those treated with placebo, whereas PYD was not found to be different between treatment groups. 50 In a study of 36 breast carcinoma patients with bone metastases, both PYD and DPD excretion increased in those patients with progressive disease (P 0.03 at 8 weeks). 46 In contrast, levels of PYD and DPD did not appear to change significantly in responding patients. These bone resorption markers have demonstrated a good correlation with clinical outcome and appear to be much more sensitive to changes in bone metabolism compared with traditional bone markers. Peptide-Bound Crosslinks Immunoassays to measure the N-terminal and C-terminal peptide-bound crosslinks of Type I collagen in the urine and in serum have been developed. The CrossLaps (Osteometer Biotech A/S, Copenhagen, Denmark) and Osteomark (Ostex International, Inc., Seattle, WA) ELISA, which measure urinary CTX and NTX, respectively, were developed in the early 1990s, 43,44 and ELISA assays that can be used to measure serum levels of CTX and NTX recently have been developed. 22 A sensitive radioimmunoassay (RIA) for measuring CTX excretion (CrossLaps RIA) also is avail-

6 2526 CANCER May 15, 2002 / Volume 94 / Number 10 TABLE 3 Effects of Single Infusions of Pamidronate on Bone Resorption Markers (In All Cases, Results Are Expressed as a Percentage of the Baseline Values) a Bone marker Change after treatment with pamidronate (% of baseline) Body et al., Vinholes et al., Vinholes et al., C-telopeptide N-telopeptide ND Deoxypyridinoline 55 ND 36 Hydroxyproline Pyridinoline 77 ND 57 ND: not determined. a Values are approximated from graphic representations of the data. able. 53 These peptide-bound crosslinks constitute the major fraction of crosslinks from collagen degradation in both the serum and the urine. Therefore, changes in bone metabolism result in greater changes in serum and urine concentrations of NTX and CTX compared with PYD and DPD. 19 This is reflected in the greater decreases in NTX and CTX compared with PYD and DPD observed in response to bisphosphonate therapy. Both NTX and CTX typically are increased 2 7-fold in 70 80% of patients with bone metastases compared with healthy controls, and typically decrease by 60 25, % in response to bisphosphonate therapy. Several comparative trials have examined bone resorption markers in patients with bone metastases who were treated with bisphosphonates. 25,54,55 These studies have demonstrated significant correlations between baseline and posttreatment levels of PYD, DPD, NTX, and CTX. However, in every study, levels of these markers were found to correlate poorly with baseline urinary calcium. In a study of 19 breast carcinoma patients with extensive bone metastases, mean baseline levels of urinary calcium, hydroxyproline, CTX, and collagen crosslinks (PYD and DPD) were elevated in 47%, 74%, 83%, and 100% of patients, respectively. 54 All these markers decreased after pamidronate therapy, and the largest decrease was observed in the levels of CTX (reaching 13% of baseline levels 54 ) (Table 3). 25,54,55 In 29 breast carcinoma patients with progressing bone metastases, the mean baseline values of NTX, CTX, and DPD were elevated approximately 2-fold compared with age-matched controls, 55 and levels of NTX and CTX decreased significantly after pamidronate therapy (P 0.001) (Table 3). 25,54,55 In a double-blind study of 32 patients with hypercalcemia of malignancy, mean baseline levels of NTX were 7-fold above normal, and the mean DPD and CTX levels were each 5-fold higher than normal. 55 Again, NTX and CTX demonstrated the greatest decrease after pamidronate therapy, reaching 15% and 2% of the baseline values, respectively (P 0.01 vs. any other marker)(table 3). 25,54,55 N-telopeptide has been reported to be the most sensitive biochemical marker for the detection of bone metastases and for assessing response to therapy or disease progression, and NTX levels in patients with breast carcinoma are reported to correlate more closely with bone metastases than conventional tumor markers such as CA ,56 Increased NTX excretion also has been shown to correlate with the extent of bone metastases (Fig. 3). 57 Patients with advanced disease and hypercalcemia had dramatically higher mean NTX levels compared with patients with newly diagnosed bone metastases. Although collagen degradation products can be measured in the urine and therefore are convenient, noninvasive, and inexpensive, the clinical application of these assays is hampered by variability due to analytic and biologic factors ,58,59 Recently, immunoassays for serum NTX, CTX, and bone sialoprotein (BSP) have been developed that may be more sensitive and reproducible than measuring urinary excretion. 22 BSP is a bone matrix integrin-binding protein synthesized by osteoblasts and osteoclasts, and is expressed ectopically by malignant breast, prostate, and other tumor cells. 60 The observations that serum BSP levels were significantly higher in patients with bone metastases compared with those without (P 0.05) and that levels were 142% greater in postmenopausal women compared with premenopausal women 61 suggested that BSP measurement might be useful for monitoring bone resorption. The measurement of serum BSP levels currently is under evaluation as a prognostic factor for bone metastases, disease progression, and survival; however, results to date have been mixed. In a direct comparison, changes in the levels of urinary and serum NTX and CTX were comparable in patients with metabolic and malignant bone disease, including osteoporosis, primary hyperparathyroidism, Paget disease of bone, multiple myeloma, and breast carcinoma (Fig. 4) 22 In patients with breast carcinoma, levels of urinary DPD, urinary and serum NTX, and serum BSP were reported to be elevated dramatically in the presence of bone metastases compared with patients without bone metastases. Overall, urinary DPD and NTX were the most sensitive indicators of bone resorption in cancer patients, demonstrating significant mean increases compared with healthy controls.

7 Clinical Use of Bone Resorption Markers/Coleman 2527 FIGURE 3. Urinary (U) and serum (S) markers of bone resorption in patients with metabolic and malignant bone disease. Values are expressed as Z scores. The full lines represent the mean and the dotted lines represent 2 standard deviations around the mean of healthy controls. (*) indicates a P value 0.05; (**) indicates a P value 0.01; and (***) indicates a P value versus healthy controls. DPD: deoxypyridinoline; CTX: C-telopeptide; NTX: N-telopeptide; BSP: bone sialoprotein; OPO: primary vertebral osteoporosis; PHPT: primary hyperparathyroidism; PD: Paget disease; MM: multiple myeloma; BC : breast carcinoma without bone metastases; BC : breast carcinoma with bone metastases. Adapted from: Woitge HW, Pecherstorfer M, Li Y, et al. Novel serum markers of bone resorption: Clinical assessment and comparison with established urinary indices. J Bone Miner Res 1999; , with permission of the American Society for Bone and Mineral Research. 22 CLINICAL RELEVANCE OF BONE RESORPTION MARKERS Although biochemical markers of bone metabolism are widely used in clinical trials and have provided a wealth of important data, to my knowledge their clinical utility in the management of individual patients remains uncertain. New serum assays for NTX and CTX appear promising, but will require extensive validation before they will become used routinely in clinical practice. However, emerging data demonstrating a correlation between the levels of the new-generation, highly sensitive bone markers and clinical outcome may provide the rationale for their widespread acceptance. In clinical practice, markers of bone resorption play a potential role in the diagnosis of bone metastases, 25,26 the assessment of disease progression and response to treatment, 22 25,59 and predicting the rate of bone loss and the potential for fracture ,67 In addition, given that bone resorption markers are sensitive indicators of response to bisphosphonate

8 2528 CANCER May 15, 2002 / Volume 94 / Number 10 FIGURE 4. Mean maximum decrease from baseline in N-telopeptide (NTX) by dose of zoledronic acid. Adapted with permission from: Berenson JR, Vescio R, Rosen L. A phase I dose-ranging trial of monthly infusions of zoledronic acid for the treatment of osteolytic bone metastases. Clin Cancer Res 2001;7: therapy and appear to correlate with clinical outcome, it has been suggested that they potentially could be used to tailor the dose and schedule of bisphosphonates. Reliable definitions of response are required. A 50% reduction 12 or normalization of a previously elevated marker would appear to be a useful definition for response, 25 whereas a 50% increase would be an appropriate definition of disease progression. 56 Reporting changes in biochemical markers according to these criteria would facilitate the comparison of results from different studies. Diagnosis Bone resorption markers have been shown to be useful in the diagnosis of bone metastases in cancer patients. NTX, CTX, and DPD typically are increased in the majority of patients with bone metastases. 25,54 These data and similar reports from others 50,56 indicate that there are a few patients with indolent bone metastases, or an isolated one or two lesions, that do not break down sufficient bone matrix to influence serum or urinary concentrations of bone resorption markers significantly. In a study comparing breast carcinoma patients with bone metastases with those without bone metastases, skeletal involvement was associated with a significant increase (P 0.01) in both serum and urinary NTX. 22 In addition, PYD, NTX, and CTX appear to have a high degree of diagnostic validity for the differentiation of patients with or without bone metastases. 26 On the basis of radiographic and scintigraphic findings, 127 cancer patients were divided into 3 groups, including 83 patients with no bone metastases, 22 patients with 1 or 2 bone metastases, and 22 patients with 3 bone metastases. Pyridinoline, NTX, and CTX were found to be increased significantly in both groups of patients with bone involvement (1 to 2 and 3 lesions), indicating the specificity and sensitivity of these markers. Although bone metastases in prostate carcinoma patients primarily are osteosclerotic, increased bone resorption also is evident, and elevated CTX levels have been reported to correlate with skeletal involvement In a study of 39 prostate carcinoma patients with bone metastases, urinary serum CTX was found to be increased approximately 2-fold compared with 355 healthy, age-matched men. 68 Increases in bone resorption markers were not detected in prostate carcinoma patients without bone metastases (nine patients) or in patients with benign prostatic hyperplasia (nine patients). Levels of CTX have been reported to correlate with the extent of bone metastasis in patients with prostate carcinoma and may be important in predicting clinical outcome. 69,71,72 Biochemical markers of bone formation, including bone specific alkaline phosphatase and the amino-terminal and carboxy-terminal propeptides of Type I procollagen, also have been shown to be correlated with bone metastases in patients with prostate carcinoma.

9 Clinical Use of Bone Resorption Markers/Coleman 2529 Disease Progression and Response to Treatment In addition to diagnosing skeletal involvement in cancer patients, pretreatment levels of bone resorption markers also appear to be correlated with response to treatment. Pretreatment NTX values are useful in predicting clinical response to bisphosphonates. 25 Baseline NTX values of nonresponding patients were significantly higher (P 0.02) than those of the clinical responders. Indeed, patients with an initial NTX value 2 times the upper limit of normal rarely responded to therapy (13% response rate), whereas those patients with either normal NTX levels or a pretreatment level 2 times the upper limit of normal were much more likely to respond (70% response rate). 25 Although baseline values of CTX and DPD also were higher in nonresponding patients compared with responders, NTX provided the most significant correlation with clinical response (P 0.001). Bone resorption markers have been shown to decrease after bisphosphonate treatment and appear to be correlated with clinical outcome. In particular, NTX appears to be a highly sensitive marker for monitoring therapy. Among 25 cancer patients with lytic bone metastases who were treated with pamidronate monthly for 6 months, urinary NTX excretion was found to be reduced significantly (P 0.002) compared with 27 patients treated with placebo, 23 with maximum suppression achieved after 2 weeks of treatment. In contrast, there was no significant difference in PYD or DPD levels between patients treated with pamidronate and those treated with placebo. The most intriguing outcome of this trial was the observation that NTX levels were highly correlated with reduced risk of fractures and progression of bony disease. Among 21 pamidronate-treated patients with elevated baseline NTX levels, 12 patients completed the study with normal NTX levels and NTX levels remained elevated in 9 patients. Within the group of patients whose NTX levels normalized, there was a lower proportion of patients with fractures (42% vs. 89%; P 0.07) or bony disease progression (25% vs. 78%; P 0.03) compared with the group whose NTX levels remained elevated. Similarly, clinical benefit as indicated by improvement in a pain score was observed only in those patients whose bone resorption rate normalized after pamidronate (17 of 32 patients [53%]) with no responses observed in the 11 patients (34%) with persistently elevated levels (P 0.01). In a study of 37 breast carcinoma patients with newly diagnosed bone metastases who were receiving oral pamidronate, 56 NTX levels were significantly lower (P 0.05) at 1 month and 4 months compared with NTX levels in patients with progressive disease. Similarly, NTX levels at 4 months in patients with either partial or no response were significantly lower in those patients with a time to disease progression of 7 months compared with patients whose disease progressed in 7 months. Furthermore, an increase in NTX excretion of 50% was found to be predictive of disease progression in 78% of patients. These data suggest that the goal of bisphosphonate therapy should be the normalization of NTX excretion. 23 Suppression of CTX also has been shown to be correlated with response to bisphosphonate treatment in prostate carcinoma patients with blastic bone metastases. 68,73,74 In a study of 39 prostate carcinoma patients with bone metastases, a single injection of pamidronate (120 mg) significantly decreased (P ) urinary -CTX, urinary -CTX, and serum CTX. 68 Therefore, CTX may be an important biochemical marker for monitoring response to bisphosphonate treatment in patients with prostate carcinoma. Bone marker data are beginning to emerge from clinical studies investigating the latest generation of more potent heterocyclic nitrogen-containing bisphosphonates. Cancer patients with osteolytic metastases reportedly have been treated with doses of up to 16 mg of zoledronic acid in Phase I trials. 75,76 In the first study, patients received mg of zoledronic acid monthly for 3 months, 75 and sustained dose-dependent decreases in urinary calcium, hydroxyproline, PYD, and DPD were observed with doses of zoledronic acid 0.2 mg. Levels of NTX were suppressed even at the lowest dose, and decreased 70 80% below baseline at doses 0.8 mg (Fig. 5). 75 In the second study, 76 doses of zoledronic acid 2 mg suppressed all urinary bone resorption markers tested, including calcium, hydroxyproline, PYD, DPD, and NTX. It is important to note that, NTX levels were maintained 50% below baseline for 8 weeks after a single dose of zoledronic acid 2 mg (Fig. 6). 76 As discussed earlier, decreases in NTX levels after bisphosphonate therapy are associated with favorable clinical outcome. Similar clinical benefits would be expected after normalization of NTX levels with zoledronic acid. Zoledronic acid and pamidronate have been compared in a randomized, Phase II, multicenter trial for the prevention of skeletal complications (n 280). 77 Patients with osteolytic lesions due to metastatic breast carcinoma or multiple myeloma were randomized to double-blind treatment with either zoledronic acid (0.4 mg, 2 mg, or 4 mg) or pamidronate (90 mg) every 4 weeks for up to 10 months. Median NTX values decreased from baseline to the end of the study by 37%, 59%, and 61%, respectively, in the zoledronic acid dosing groups (0.4 mg, 2 mg, and 4 mg, respectively) and by 58% in the pamidr-

10 2530 CANCER May 15, 2002 / Volume 94 / Number 10 FIGURE 5. Kinetics of urinary N-telopeptide suppression after zoledronic acid treatment at doses of 1 16 mg. Levels of suppression are expressed as the median percent difference in the urinary N-telopeptide:creatinine ratio from baseline values at Week 0. Reprinted with permission from: Berenson JR, Vescio R, Henick K, et al. A Phase I, open label, dose ranging trial of intravenous bolus zoledronic acid, a novel bisphosphonate, in cancer patients with metastatic bone disease. Cancer 2001; 91: Copyright 2001 American Cancer Society. Reprinted by permission of Wiley-Liss, Inc., a subsidiary of John Wiley & Sons, Inc. onate group. Moreover, the 4-mg dose of zoledronic acid resulted in greater decreases in NTX at every time point throughout the duration of the study compared with 90 mg of pamidronate (P 0.05). Consistent with earlier observations, the greater reductions in NTX levels observed in the groups that received either 2 mg or 4 mg zoledronic acid or pamidronate were found to be correlated with a lower incidence of skeletal-related events, fractures, and the need for radiation to bone compared with the group of patients receiving 0.4 mg of zoledronic acid. The recently completed Phase III clinical trials of zoledronic acid used across the range of tumor types affecting bone included nearly 3000 patients and prospectively measured bone markers. Patients have been allocated randomly to receive anticancer treatment plus either zoledronic acid or pamidronate (breast carcinoma and myeloma patients), or anticancer treatment plus either zoledronic acid or placebo (patients with prostate carcinoma and those with other tumors affecting bone). The evaluation of these data should reliably determine whether changes in the bone resorption markers can be used in routine practice to assess response to systemic anticancer treatments and bisphosphonates. CONCLUSIONS The skeletal complications of cancer, including bone pain and fracture, are particularly debilitating to the patient. Imaging techniques, including plain-film radiography, 99m Tc bone scans, MRI, and 18-F-PET, are the current methods for identifying bone lesions and assessing responses to treatment. Unfortunately, these techniques are expensive and some are available only at major medical centers. Because osteolytic lesions are associated with increased bone resorption, patients often have calcium and other bone breakdown products in the serum and urine. Traditional bone resorption markers, including urinary calcium and hydroxyproline, have been used for many years primarily to assess responses to bisphosphonate therapy. However, these traditional markers have been shown to be suboptimal in terms of selectivity and sensitivity. A number of new biochemical markers involving unique breakdown products of Type I collagen have been developed that are more specific and sensitive indicators of bone resorption. These include PYD, DPD, NTX, and CTX. These markers are elevated in the majority of patients with bone metastases, suggesting a potential role in the diagnosis of skeletal involvement. In addition, the pretreatment and posttreatment levels of NTX appear to be useful in predicting clinical outcome. Bisphosphonate therapy results in substantial decreases in the levels of bone resorption markers, and NTX appears to be the most sensitive marker for the assessment of response to treatment. Indeed, the conclusion of several studies has been that the goal of bisphosphonate therapy should be the normalization of NTX excretion. Clearly, these new biochemical markers hold much promise for the management of osteolytic disease. In addition, the development of serum assays for NTX and CTX should provide the sensitivity, convenience, and reproducibility lacking in earlier assays. Ongoing clinical trials will provide the rigorous eval-

11 Clinical Use of Bone Resorption Markers/Coleman 2531 uation required for these newly developed biochemical assays to be accepted for routine clinical use. REFERENCES 1. Coleman RE, Rubens RD. The clinical course of bone metastases from breast cancer. Br J Cancer. 1987;55: Coleman RE. Skeletal complications of malignancy. Cancer. 1997;80(Suppl): Lipton A, Theriault RL, Hortobagyi GN, et al. Pamidronate prevents skeletal complications and is effective palliative treatment in women with breast carcinoma and osteolytic bone metastases: long term follow-up of two randomized, placebo-controlled trials. Cancer. 2000;88: Hortobagyi GN. Bone metastases in breast cancer patients. Semin Oncol. 1991;18(Suppl 5): Abildgaard N, Brixen K, Kristensen JE, Vejlgaard T, Charles P, Nielsen JL. Assessment of bone involvement in patients with multiple myeloma using bone densitometry. Eur J Haematol. 1996;57: Mariette X, Khalifa P, Ravaud P, et al. Bone densitometry in patients with multiple myeloma. Am J Med. 1992;93: Dodwell DJ, Howell A. The systemic treatment of bone metastases. In: Rubens RD, Fogelman I, eds. Bone metastases: diagnosis and treatment. London: Springer-Verlag, 1991: Berenson JR, Lichtenstein A, Porter L, et al. Efficacy of pamidronate in reducing skeletal events in patients with advanced multiple myeloma. N Engl J Med. 1996;334: Berenson JR, Lichtenstein A, Porter L, et al. Long-term pamidronate treatment of advanced multiple myeloma patients reduces skeletal events. J Clin Oncol. 1998;16: Hortobagyi GN, Theriault RL, Porter L, et al. Efficacy of pamidronate in reducing skeletal complications in patients with breast cancer and lytic bone metastases. Protocol 19 Aredia Breast Cancer Study Group. N Engl J Med. 1996;335: Hortobagyi GN, Theriault RL, Lipton A, et al. Long-term prevention of skeletal complications of metastatic breast cancer with pamidronate. Protocol 19 Aredia Breast Cancer Study Group. J Clin Oncol. 1998;16: Vinholes J, Guo CY, Purohit OP, Eastell R, Coleman RE. Metabolic effects of pamidronate in patients with metastatic bone disease. Br J Cancer. 1996;73: Mundy GR, Yoneda T. Bisphosphonates as anticancer drugs. N Engl J Med. 1998;339: Diel IJ, Solomayer EF, Costa SD, et al. Reduction in new metastases in breast cancer with adjuvant clodronate treatment. N Engl J Med. 1998;339: Hillner BE, Ingle JN, Berenson JR, et al. American Society of Clinical Oncology guideline on the role of bisphosphonates in breast cancer. J Clin Oncol. 2000;18: Coleman RE, Mashiter G, Whitaker KB, Moss DW, Rubens RD, Fogelman I. Bone scan flare predicts successful systemic therapy for bone metastases. J Nucl Med. 1988;29: Frank J, Ling A, Patronas N, Carrasquillo J, Horvath K, Dwyer A. Comparison of magnetic resonance imaging and radionuclide bone scan in the evaluation of primary and metastatic disease to the bone [abstract 25]. Proc Am Soc Clin Oncol. 1989;8: Schirrmeister H, Guhlmann A, Kotzerke J, et al. Early detection and accurate description of extent of metastatic bone disease in breast cancer with fluoride ion and positron emission tomography. J Clin Oncol. 1999;17: Vinholes J, Coleman R, Eastell R. Effects of bone metastases on bone metabolism: implications for diagnosis, imaging and assessment of response to cancer treatment. Cancer Treat Rev. 1996;22: Coleman RE. Assessment of response to treatment. In: Rubens RD, Fogelman I, eds. Bone metastases: diagnosis and treatment. London: Springer-Verlag, 1991: Coleman RE, Whitaker KB, Moss DW, Mashiter G, Fogelman I, Rubens RD. Biochemical prediction of response of bone metastases to treatment. Br J Cancer. 1988;58: Woitge HW, Pecherstorfer M, Li Y, et al. Novel serum markers of bone resorption: clinical assessment and comparison with established urinary indices. J Bone Miner Res. 1999;14: Lipton A, Demers L, Curley E, et al. Markers of bone resorption in patients treated with pamidronate. Eur J Cancer. 1998;34: Souberbielle JC, Cormier C, Kindermans C. Bone markers in clinical practice. Curr Opin Rheumatol. 1999;11: Vinholes JJF, Purohit OP, Abbey ME, Eastell R, Coleman RE. Relationships between biochemical and symptomatic response in a double-blind randomised trial of pamidronate for metastatic bone disease. Ann Oncol. 1997;8: Miura H, Yamamoto I, Takada M, et al. Diagnostic validity of bone metabolic markers for bone metastasis. Endocr J. 1997; 44: Bertolini DR, Nedwin GE, Bringman TS, Smith DD, Mundy GR. Stimulation of bone resorption and inhibition of bone formation in vitro by human tumour necrosis factors. Nature. 1986;319: Delaisse JM, Eeckhout Y, Neff L, et al. (Pro)collagenase (matrix metalloproteinase-1) is present in rodent osteoclasts and in the underlying bone-resorbing compartment. J Cell Sci. 1993;106: Drake FH, Dodds RA, James IE, et al. Cathepsin K, but not cathepsins B, L, or S, is abundantly expressed in human osteoclasts. J Biol Chem. 1996;271: Bossard MJ, Tomaszek TA, Thompson SK, et al. Proteolytic activity of human osteoclast cathepsin K. Expression, purification, activation, and substrate identification. J Biol Chem. 1996;271: Peacock M, Robertson WG, Nordin BE. Relation between serum and urinary calcium with particular reference to parathyroid activity. Lancet. 1969;1: Campbell FC, Blamey RW, Woolfson AM, Elston CW, Hosking DJ. Calcium excretion (CaE) in metastatic breast cancer. Br J Surg. 1983;70: 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. 1995;80: Deacon AC, Hulme P, Hesp R, Green JR, Tellez M, Reeve J. Estimation of whole body bone resorption rate: a comparison of urinary total hydroxyproline excretion with two radioisotopic tracer methods in osteoporosis. Clin Chim Acta. 1987;166: Gasser A, Celada A, Courvoisier B, et al. The clinical measurement of urinary total hydroxyproline excretion. Clin Chim Acta. 1979;95:

12 2532 CANCER May 15, 2002 / Volume 94 / Number Mautalen CA. Circadian rhythm of urinary total and free hydroxyproline excretion and its relation to creatinine excretion. J Lab Clin Med. 1970;75: Gasser AB, Depierre D, Courvoisier B. Total urinary and free serum hydroxyproline in metastatic bone disease. Br J Cancer. 1979;39: Blomqvist C, Elomaa I, Virkkunen P, et al. The response evaluation of bone metastases in mammary carcinoma. The value of radiology, scintigraphy, and biochemical markers of bone metabolism. Cancer. 1987;60: Body JJ, Dumon JC, Piccart M, Ford J. Intravenous pamidronate in patients with tumor-induced osteolysis: a biochemical dose-response study. J Bone Miner Res. 1995;10: Coombes RC, Dady P, Parsons C, et al. Assessment of response of bone metastases to systemic treatment in patients with breast cancer. Cancer. 1983;52: Glover D, Lipton A, Keller A, et al. Intravenous pamidronate disodium treatment of bone metastases in patients with breast cancer. A dose-seeking study. Cancer. 1994;74: O Rourke N, McCloskey E, Houghton F, Huss H, Kanis JA. Double-blind, placebo-controlled, dose-response trial of oral clodronate in patients with bone metastases. J Clin Oncol. 1995;13: Bonde M, Qvist P, Fledelius C, Riis BJ, Christiansen C. Immunoassay for quantifying type I collagen degradation products in urine evaluated. Clin Chem. 1994;40: Hanson DA, Weis MA, Bollen AM, Maslan SL, Singer FR, Eyre DR. A specific immunoassay for monitoring human bone resorption: quantitation of type I collagen cross-linked N-telopeptides in urine. J Bone Miner Res. 1992;7: Seibel MJ, Robins SP, Bilezikian JP. Urinary pyridinium crosslinks of collagen: specific markers of bone resorption in metabolic bone disease. Trends Endocrinol Metab. 1992;3: Walls J, Assiri A, Howell A, et al. Measurement of urinary collagen cross-links indicate response to therapy in patients with breast cancer and bone metastases. Br J Cancer. 1999; 80: Robins SP, Woitge H, Hesley R, Ju J, Seyedin S, Seibel MJ. Direct, enzyme-linked immunoassay for urinary deoxypyridinoline as a specific marker for measuring bone resorption. J Bone Miner Res. 1994;9: Eastell R, Hampton L, Colwell A, et al. Urinary collagen crosslinks are highly correlated with radioisotopic measurements of bone resorption [abstract]. Osteoporosis. 1990;2: Delmas PD, Schlemmer A, Gineyts E, Riis B, Christiansen C. Urinary excretion of pyridinoline crosslinks correlates with bone turnover measured on iliac crest biopsy in patients with vertebral osteoporosis. J Bone Miner Res. 1991;6: Demers LM, Costa L, Chinchilli VM, Gaydos L, Curley E, Lipton A. Biochemical markers of bone turnover in patients with metastatic bone disease. Clin Chem. 1995;41: Body JJ, Dumon JC, Delmas PD. Comparative evaluation of markers of bone resorption after a single dose of pamidronate in patients with breast cancer induced osteolysis. Bone Miner. 1994;25(Suppl 1):S Coleman RE, Houston S, James I, et al. Preliminary results of the use of urinary excretion of pyridinium crosslinks for monitoring metastatic bone disease. Br J Cancer. 1992;65: Bonde M, Qvist P, Fledelius P, Garnero P, Christiansen C. Effect of bisphosphonate therapies (pamidronate and ibandronate) on the excretion of degradation products of the C-telopeptides of type I collagen measured by a radioimmunoassay (Crosslaps RIA) [abstract 45]. Bone. 1995;17: Body JJ, Dumon JC, Gineyts E, Delmas PD. Comparative evaluation of markers of bone resorption in patients with breast cancer-induced osteolysis before and after bisphosphonate therapy. Br J Cancer. 1997;75: Vinholes J, Guo CY, Purohit OP, Eastell R, Coleman RE. Evaluation of new bone resorption markers in a randomized comparison of pamidronate or clodronate for hypercalcemia of malignancy. J Clin Oncol. 1997;15: Vinholes J, Coleman R, Lacombe D, et al. Assessment of bone response to systemic therapy in an EORTC trial: preliminary experience with the use of collagen cross-link excretion. European Organization for Research and Treatment of Cancer. Br J Cancer. 1999;80: Yamamoto I, Morita R, Fujimoto R, et al. Clinical studies using measurements of N-telopeptides of type I collagen (NTx) in patients with bone metastases. Comparison with bone scintigraphy and other metabolic markers. Kaku Igaku. 1995;32: Christenson RH. Biochemical markers of bone metabolism: an overview. Clin Biochem. 1997;30: Watts NB. Clinical utility of biochemical markers of bone remodeling. Clin Chem. 1999;45: Bellahcene A, Merville MP, Castronovo V. Expression of bone sialoprotein, a bone matrix protein, in human breast cancer. Cancer Res. 1994;54: Withold W, Armbruster FP, Karmatschek M, Reinauer H. Bone sialoprotein in serum of patients with malignant bone diseases. Clin Chem. 1997;43: Bellahcene A, Albert V, Pollina L, Basolo F, Fisher LW, Castronovo V. Ectopic expression of bone sialoprotein in human thyroid cancer. Thyroid. 1998;8: Bellahcene A, Kroll M, Liebens F, Castronovo V. Bone sialoprotein expression in primary human breast cancer is associated with bone metastases development. J Bone Miner Res. 1996;11: Bellahcene A, Menard S, Bufalino R, Moreau L, Castronovo V. Expression of bone sialoprotein in primary human breast cancer is associated with poor survival. Int J Cancer. 1996; 69: Karmatschek M, Maier I, Seibel MJ, Woitge HW, Ziegler R, Armbruster FP. Improved purification of human bone sialoprotein and development of a homologous radioimmunoassay. Clin Chem. 1997;43: Waltregny D, Bellahcene A, Van Riet I, et al. Prognostic value of bone sialoprotein expression in clinically localized human prostate cancer. J Natl Cancer Inst. 1998;90: Ross PD, Kress BC, Parson RE, Wasnich RD, Armour KA, Mizrahi IA. Serum bone alkaline phosphatase and calcaneus bone density predict fractures: a prospective study. Osteoporos Int. 2000;11: Garnero P, Buchs N, Zekri J, Rizzoli R, Coleman RE, Delmas PD. Markers of bone turnover for the management of patients with bone metastases from prostate cancer. Br J Cancer. 2000;82: Koizumi M, Yonese J, Fukui I, Ogata E. The serum level of the amino-terminal propeptide of type I procollagen is a sensitive marker for prostate cancer metastasis to bone. BJU Int. 2001;87: