Consensus on the utility of bone markers in the malignant bone disease setting

Abstract

Biochemical markers of bone turnover provide insight into ongoing rates of skeletal metabolism and tumor–bone interactions in patients with malignant bone disease. This article reviews the available recent evidence assessing the potential of bone markers for detecting and monitoring malignant bone lesions in patients with advanced cancers, and for assessing overall skeletal health and response to antiresorptive therapies in patients at all stages of cancer progression. Most data thus far are for urinary N-terminal cross-linked telopeptide of type I collagen (NTX) in predicting risks of skeletal morbidity and death and monitoring response to zoledronic acid in patients with bone metastases. Ongoing studies are evaluating such correlations for other markers and therapies. Emerging evidence suggests that bone markers may help identify patients at high risk for bone metastasis or bone lesion progression, thereby allowing improved follow-up. Results from ongoing clinical trials evaluating such potential applications of bone markers are awaited.

Introduction

Bone is a common site for distant metastasis from solid tumors (especially breast cancer [BC] and prostate cancer [PC], wherein approximately 70% of patients with advanced disease develop bone metastases [1]), and bone lesions can also develop in patients with hematologic malignancies, especially multiple myeloma (MM)—which colonizes the bone marrow [2], [3], [4]. In addition to its vast surface area, bone has many features that make it a favored site for cancer cell growth and render it susceptible to potentially devastating ramifications. Maintenance of the skeleton involves osteoclast-mediated osteolysis of old or damaged bone tissue, resulting in a resorption cavity on the bone surface, and osteoblast-mediated osteogenesis, which fills in the resorption cavity with nascent bone matrix. In patients with malignant bone lesions, the interplay between tumor and bone dysregulates these otherwise balanced and spatially coupled activities, resulting in increased rates of osteolysis and osteogenesis and release of high levels of bone matrix-derived factors that can stimulate tumor growth [5], [6], [7], [8], [9]. The processes of osteolysis and osteogenesis are associated with the release of distinct biochemical markers [10] that are amenable to noninvasive measurement in blood or urine (Fig. 1). Therefore, biochemical markers of bone metabolism, such as the cross-linked collagen peptides that are breakdown products from osteolysis (e.g., the amino [N]- and carboxy [C]-terminal cross-linked telopeptides of type I collagen, or NTX and CTX) and the terminal peptides that are cleaved from procollagen before its integration into new bone matrix (e.g., procollagen type I N-terminal and C-terminal peptides, or PINP and PICP), can provide meaningful insight into the ongoing effects of tumor growth on bone turnover [11]. This review investigates the theoretical basis for the clinical utility of these markers and evaluates the current evidence for their emerging roles in evaluating and monitoring patients in the oncology setting.

Serum levels and urinary concentration of NTX and CTX reflect ongoing rates of osteolysis, whereas bone-specific alkaline phosphatase (bone ALP) levels in serum reflect ongoing rates of osteogenesis. In addition, some markers of bone metabolism may be associated with both osteolysis and osteogenesis (e.g., osteocalcin) [10]. Biochemical markers of bone metabolism reflect ongoing rates of bone resorption and formation in the body as a whole [11]. Therefore, bone marker assessments do not provide information specific to individual lesion sites. Moreover, changes in bone marker levels are not disease specific, but are associated with alterations in skeletal metabolism independent of the underlying cause [10]. For example, hormonal therapies for BC and PC, as well as bone metastases, can all increase bone resorption marker levels [12], [13]. However, some of the more recently evaluated markers, including the non-isomerized form of CTX and the C-terminal cross-linked telopeptide of type I collagen generated by matrix metalloproteinases (CTX-MMP or ICTP), may preferentially reflect different biologic processes that may be specific to certain disease states [14]. Standardized assays are now becoming available for many bone turnover markers, and “normal” or reference ranges for several markers have been established (Table 1) [15]. Selection of the appropriate reference value is critical for data interpretation—for example, normal serum CTX levels are similar in men and premenopausal women, but markedly higher in postmenopausal women (Table 1) [15].

Several studies have explored associations between bone markers and underlying skeletal pathology in patients with bone metastases. Factors released during osteolysis (e.g., breakdown products of type I collagen, enzymes such as tartrate-resistant acid phosphatase 5b [TRAcP-5b], pyridinium crosslinks) have been associated with the development and progression of bone lesions from MM and bone metastases from solid tumors [5], [7], [8], [9], [11], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54]. Similarly, elevations of osteogenesis markers such as bone ALP have been associated with malignant bone lesions from BC, PC, lung cancer, and MM [5], [16], [22], [27], [31], [32], [34], [43], [50], [55], [56], [57], [58], [59], [60], [61]. Although there is a continuum of radiologic appearances for bone metastases, ranging from the predominantly osteolytic lesions of renal cell carcinoma (RCC) to the predominantly osteoblastic lesions of PC, all are associated with elevations in both osteolytic and osteoblastic marker levels [62]. Moreover, regulatory and signaling factors involved in bone homeostasis (e.g., parathyroid hormone-related hormone, osteoprotegerin [OPG], and osteopontin) may also correlate with clinical parameters, including the extent of bone disease, bone pain, palliative response to bone-directed therapy, and with established prognostic factors in patients with bone metastases [7], [8], [22], [33], [34], [41], [48], [50], [53], [55], [61], [63], [64], [65], [66], [67].

Data from patients with osteoporosis provide additional support for the utility of bone markers to assess ongoing rates of bone turnover outside the oncology setting. For example, in a retrospective analysis of 1593 women with postmenopausal osteoporosis, baseline urinary deoxypyridinoline (DPD) levels above the median value for normal premenopausal women were associated with an increased incidence of new vertebral fractures [68]. In a smaller correlative study, urinary NTX levels were significantly higher in postmenopausal women with hip fractures (n = 21) compared with healthy postmenopausal women (n = 20; P < .05) [69]. Moreover, various studies have reported correlations between early changes in bone turnover markers during therapy and long-term bone mineral density (BMD) increases in women with osteopenia and osteoporosis [70], [71], [72], [73], and monitoring bone turnover levels might provide additional tools to assist clinicians in making decisions regarding antiresorptive treatment in this setting [74].

Tumor and metabolic markers have been shown to have prognostic significance and predictive value for determining response to targeted therapy in patients with cancer [75], [76], [77], [78], [79], [80], [81], [82]. For example, prostate-specific antigen (PSA) doubling time during biochemical relapse is associated with risks of local and systemic disease recurrence in patients with PC [79], [80]. However, expression of tumor markers may be altered by distant, local, and soft-tissue metastases or by anticancer therapies and has shown no correlations with bone metabolism [79], [83]. Therefore, tumor-specific markers have limited utility in monitoring bone disease progression and response to therapy such as bisphosphonates.

Markers of bone metabolism may reflect the extent and progression of bone lesions, thereby providing prognostic insight for patients with bone metastases. This is particularly relevant when the objective assessment of bone metastases is complicated by relatively slow and sometimes equivocal radiologic changes during disease progression. Some studies in patients with bone metastases from solid tumors suggest that alterations in the levels of bone resorption markers such as NTX and ICTP can predict progression of bone metastases with greater specificity than tumor markers [41], [84]. Moreover, such bone markers may identify patients who are most likely to benefit from antiresorptive therapies such as bisphosphonates.

Observational studies and retrospective analyses have revealed correlations between bone markers and clinical parameters such as the presence, extent, and progression of bone lesions (Table 2) [18], [20], [32], [37], [65], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], suggesting that these factors may provide predictive and prognostic insights in patients with malignant bone disease. The largest body of evidence to date, much of which is based on the prospective trial database for the phase III controlled trials of zoledronic acid (ZOL) [85], [86], [87], [88], [89], [99], [100], [104], [105], [106], [107], supports a role for NTX in patients with bone metastases from solid tumors (Table 2) [18], [20], [32], [37], [65], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103]. These ZOL trial databases have provided the means to retrospectively evaluate correlations between bone marker levels, disease outcomes, and risks of skeletal morbidity. Based on the current evidence, NTX has the best-established correlations with clinical outcomes and response to bone-directed therapies (Table 3). The correlations between on-treatment NTX levels and risks of skeletal-related events (SREs) or disease progression are especially striking. However, there is little evidence supporting a role for NTX in predicting or detecting bone metastasis, and the theoretical concept of bone marker level-directed bisphosphonate therapy awaits validation from ongoing prospective trials.

Although bone marker assessments recently were included in clinical trials of bisphosphonates other than ZOL, current evidence for the utility of bone markers with these agents is limited and requires further analysis. However, recent trials of denosumab, an investigational monoclonal antibody targeted against the receptor-activator of nuclear factor kappa-B ligand (RANKL), include bone marker assessments [108], [109], and bone marker studies continue to be components of ongoing ZOL and denosumab trials. In addition, bone marker assessments have an established role in elucidating the pharmacodynamics of antiresorptive agents, and have been extensively used in phase I/II dose-finding trials of new antiresorptives in the oncology setting (including denosumab and saracatinib, an inhibitor of Src kinase) [109], [110]. Therefore, the role of bone markers in oncology is expected to evolve rapidly over the next few years.