Use of biochemical markers of bone turnover in osteoporosis

INTRODUCTION — In adults, bone is constantly being remodeled, first being broken down (bone resorption) and then being rebuilt (bone formation). The resorption and reformation of bone is important for repair of microfractures and to allow modification of structure in response to stress and other biomechanical forces. Bone formation is normally tightly coupled to bone resorption, so that bone mass does not change. Bone diseases occur when formation and resorption are uncoupled.

Several assays are available that measure bone turnover markers (BTMs). These assays measure collagen breakdown products and other molecules released from osteoclasts and osteoblasts during the process of bone resorption and formation. Markers that are specific to bone formation include bone-specific alkaline phosphatase (BSAP), osteocalcin, and serum procollagen type 1 N-terminal propeptide (P1NP); markers specific to bone resorption include N-telopeptide (NTX), C-telopeptide (CTX), and pyridinoline cross-links (table 1).

The use of BTMs in clinical trials has been helpful in understanding the mechanism of action of therapeutic agents. However, their role in the care of individual patients is not well established. Biologic and laboratory variability in BTM values has confounded their widespread use in clinical practice (table 2).

This topic will review issues surrounding the clinical use of biochemical BTMs. The physiology of BTMs, their relationship to the process of bone remodeling, and their use in other bone disorders are reviewed separately. (See "Bone physiology and biochemical markers of bone turnover" and "Osteoporosis in patients with chronic kidney disease: Diagnosis and evaluation", section on 'Bone-specific alkaline phosphatase' and "Investigational biologic markers in the diagnosis and assessment of rheumatoid arthritis", section on 'Bone-specific markers' and "Clinical manifestations and diagnosis of Paget disease of bone", section on 'Role of biochemical studies'.)

GENERAL PRINCIPLES — The measurement of bone turnover markers (BTMs) is complicated by large, random, within-patient variability; biologic variability (age, sex, body mass index [BMI], circadian, and menstrual variation); and poor standardization of most assays (table 2) [1,2]. These issues have confounded their widespread use in clinical practice. However, some clinical assays are now automated. In 2012, the National Bone Health Alliance (NBHA) initiated a project to standardize BTM sample collection procedures in the United States and to establish a reference range of serum procollagen type 1 N-terminal propeptide (P1NP, bone formation) and serum C-telopeptide (CTX, bone resorption), which are the markers that they, the International Osteoporosis Foundation (IOF), and the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) identified as most promising for clinical use [3,4]. This project is ongoing.

In the interim, if biochemical BTMs are used in clinical practice (eg, to monitor osteoporosis therapy), we suggest using BTMs that can be measured using automated technology and that have relatively small spontaneous variability, such as serum P1NP, CTX, or urinary N-telopeptide (NTX). Bone-specific alkaline phosphatase (BSAP) is also an option in patients without severe liver disease [5]. Baseline and post-treatment serum samples should be obtained under standardized conditions (fasting morning sample) and sent to the same laboratory [4,6,7]. Urinary collections should also be standardized (fasting second morning void is most common). (See 'Monitoring response' below.)

Each of the BTMs demonstrates different responses to treatment, such that the anticipated reduction varies for each marker. For an observed change in BTM to be clinically meaningful, it must exceed the least significant change, defined as a change that is 2.8 times the precision error for the assay. For urinary excretion of NTX, an approximately 50 percent decline is predictive of improvement in bone mineral density (BMD) and fracture risk [8-10]. For serum CTX, P1NP, and BSAP, a 30 percent decline is similarly predictive [11-13]. (See "Bone physiology and biochemical markers of bone turnover".)

BONE LOSS AND FRACTURE RISK — Biochemical markers of bone turnover (BTMs) are predictive of the rate of bone loss and, in some studies, risk of fracture [4]. However, there is no role for BTMs in selecting candidates for bone density testing or for osteoporosis therapy. The decision to measure bone density should be based upon age and the presence of clinical risk factors for fracture. Similarly, the decision to treat patients should be based upon fracture risk assessment using bone mineral density (BMD) and clinical risk factors. (See "Osteoporotic fracture risk assessment" and "Overview of the management of low bone mass and osteoporosis in postmenopausal women", section on 'Patient selection' and "Treatment of osteoporosis in men", section on 'Patient selection' and "Screening for osteoporosis in postmenopausal women and men", section on 'Candidates for BMD testing'.)

Bone density measurements at any skeletal site and with a variety of technologies can predict fracture risk [14]. However, a single measurement indicates only current density, not the anticipated rate of bone loss. Patients with a given bone density who are losing bone more rapidly will have a higher fracture risk.

Several studies have demonstrated that BTMs may be useful in populations in predicting rates of bone loss (figure 1) [15-21]. As examples:

●In a prospective cohort study of 1044 older women, subjects with the highest BTMs over five years suffered the greatest bone loss [21].

●In the control arm of a trial of 236 postmenopausal women randomly assigned to postmenopausal hormone therapy and calcium versus calcium alone (control), women with the highest quartile value of N-telopeptide (NTX) throughout the study had the greatest bone loss compared with women with the lowest quartile value [17].

●In a subset of 682 men participating in the Osteoporotic Fractures in Men (MrOS) study, higher baseline levels of bone turnover (serum procollagen type 1 N-terminal propeptide [P1NP], beta C-terminal telopeptide of type I collagen [betaCTX], and tartrate-resistant acid phosphatase 5b [TRACP5b]) were associated with greater hip bone loss over five years of follow-up [22].

In addition, several [23-30], but not all [22,31-33], studies have shown that elevated BTMs are associated with increased risk of vertebral and nonvertebral fracture in older individuals, independent of BMD. As examples:

●In the Epidemiology of Osteoporosis (EPIDOS) study, older women with urinary C-telopeptide (CTX) or free deoxypyridinoline (DPD) excretion above the normal limits for young women had twice the risk of hip fracture as compared with other women (figure 2) [24].

●Among osteopenic women followed in the Os des Femmes de Lyon (OFELY) prospective cohort study, low BMD, increased BTMs (bone-specific alkaline phosphatase [BSAP]), and prior fracture were independently associated with an increased fracture risk [28].

●In contrast, in a subset of placebo patients in the Multiple Outcomes of Raloxifene Evaluation (MORE) study, none of the BTMs (BSAP, osteocalcin, or urinary CTX) that were measured influenced fracture risk [31].

Although most epidemiologic studies show that BTMs are an independent risk factor for fracture, for a given BTM value, individual rates of bone loss and fracture are variable, limiting the usefulness of BTMs in predicting an individual's fracture risk.

OSTEOPOROSIS THERAPY — We do not routinely measure bone turnover markers (BTMs) in patients initiating osteoporosis therapy. However, for individual patients (eg, patients with conditions that might interfere with drug absorption or efficacy or patients who are reluctant to take anti-osteoporosis medications regularly), we sometimes measure fasting urinary N-telopeptide (NTX), serum C-telopeptide (CTX), or serum procollagen type 1 N-terminal propeptide (P1NP) before and three to six months after starting bisphosphonates or other antiresorptive therapy. A 50 or 30 percent reduction in urinary NTX excretion or serum CTX, respectively, provides evidence of compliance and drug efficacy. (See 'Monitoring response' below.)

Effect of osteoporosis therapy — BTMs typically show large and rapid responses to osteoporosis treatments. The effect of osteoporosis therapy on BTMs depends upon the mechanism of action of the specific therapy. Antiresorptive agents, such as bisphosphonates, cause a rapid decrease in markers of bone resorption, followed shortly thereafter by a decrease in bone formation markers [4]. In contrast, anabolic agents, such as recombinant human parathyroid hormone (rhPTH), cause a rapid increase in bone formation markers, followed by an increase in markers of bone resorption. The anabolic agent romosozumab uniquely increases bone formation markers and reduces bone resorption markers. 

Several trials have shown an association between the decrease in BTMs after initiation of antiresorptive therapy and long-term antifracture efficacy [8,11,34-38]. This association is better supported for vertebral fracture risk reduction, whereas an association with nonvertebral fracture risk reduction has not been clearly demonstrated [39]. As examples:

●In a post hoc analysis of the Fracture Intervention Trial (FIT), the greater the decline in bone-specific alkaline phosphatase (BSAP) and P1NP after initiation of alendronate, the greater the reduction in spine and hip fracture [11].

●In the Multiple Outcomes of Raloxifene Evaluation (MORE) trial, fracture risk reduction with raloxifene therapy correlated better with changes in BTMs than with improvements in bone mineral density (BMD) [34,35].

●In risedronate vertebral fracture trials, the greatest decrease in fracture risk was among subjects with a decrease in urine NTX of more than 40 percent and urine CTX of more than 60 percent [8]. In a subsequent analysis of the same data, measurement of a single urine CTX value while taking risedronate was predictive of fracture reduction [36]. CTX values less than or equal to the mean for premenopausal women were associated with the lowest fracture risk, and further suppression of turnover did not result in further reduction of fracture risk.

Thus, a reduction in BTMs after initiation of antiresorptive therapy is associated with a decrease in fracture. However, the optimal threshold for each marker is not well established. Based upon the above trials, a successful reduction in BTMs could be defined as an approximately 50 (urine markers) or 30 (serum markers) percent decline, or as reducing the BTM to within the lower one-half of the reference range for premenopausal women [8,36].

This approach (evaluating a reduction in BTMs) is only useful with antiresorptive therapy, not with rhPTH (markers would increase). An increase in markers of bone formation (C-terminal propeptide of type I procollagen [PICP] and BSAP) one month after initiation of PTH has been associated with improvement in bone structure [40].

There are no data on long-term outcomes when BTMs are decreased below the reference interval. Some investigators have hypothesized that such oversuppression of bone turnover may interfere with normal repair of microdamage and increased skeletal fragility. This topic is reviewed elsewhere. (See "Risks of bisphosphonate therapy in patients with osteoporosis", section on 'Atypical femur fracture'.)

Improving treatment adherence — Long-term adherence with anti-osteoporosis therapy is low. Some data suggest that sharing BTM results with patients as evidence of treatment effectiveness improves adherence to therapy [41-44]. However, data do not uniformly support this strategy [39]. In addition, occasional patients with osteoporosis have difficulty tolerating or are reluctant to take treatment. In such individuals, a BTM value above the upper limit of the reference range for premenopausal women indicates especially high risk of bone loss and fracture, which may further support the potential benefit of treatment.

Monitoring response

Early treatment response — For most patients initiating osteoporosis therapy, we do not measure BTMs. We perform a dual-energy x-ray absorptiometry (DXA) of the hip and spine after two years, and if BMD is stable or improved, monitor less frequently thereafter. For the subset of patients with conditions that might interfere with drug absorption or efficacy or for patients who are reluctant to take anti-osteoporosis medications regularly, we sometimes measure fasting urinary NTX, serum CTX, or serum P1NP before and three to six months after starting bisphosphonates or other antiresorptive therapy. A 50 or 30 percent reduction in urinary NTX excretion or serum CTX, respectively, provides evidence of treatment adherence and efficacy [9,10,12,13]; in such patients, therapy should be continued for two years, when bone density is typically remeasured. Monitoring osteoporosis therapy is reviewed separately. (See "Overview of the management of low bone mass and osteoporosis in postmenopausal women", section on 'Monitoring response to initial pharmacotherapy'.)

While there are a number of approaches to monitoring the response to antiresorptive therapy, no consensus exists on the optimal approach, and no prospective trials have defined how best to incorporate BTMs into monitoring strategies. Assessment of baseline and post-therapy BTMs may be useful for monitoring patients taking osteoporosis therapy for the following reasons:

●The early change in BTMs following therapy has been shown to be predictive of improvement in BMD and antifracture efficacy. (See 'Effect of osteoporosis therapy' above.)

●Demonstrating such changes reflects the degree of patient adherence and may improve patient persistence with therapy.

However, their role in monitoring osteoporosis therapy relies upon defining the threshold reduction in BTM to attain optimal treatment effects (ie, fracture reduction). Such thresholds are not universally accepted.

A less than 50 (NTX) or 30 (CTX, P1NP) percent reduction may not necessarily indicate treatment failure, since many patients with small changes in turnover have stable BMD on treatment [9]. However, in this setting, we evaluate for possible nonadherence or poor absorption (often related to an insufficient time interval between drug intake and food ingestion).

Suspected treatment failure — In individuals taking antiresorptive osteoporosis therapy, treatment failure due to nonadherence, gastrointestinal malabsorption, or other cause may be suspected based on an interim decrease in BMD or interim fragility fracture(s). In such individuals, measuring on-treatment BTMs may be a helpful component of the evaluation even in the absence of baseline measurement (algorithm 1). CTX, NTX, and/or P1NP values that are suppressed during antiresorptive therapy (eg, at or below the mean of the reference range for young adults) support treatment effectiveness, whereas values above the reference mean may suggest nonadherence or malabsorption. The approach to monitoring antiresorptive osteoporosis therapy in postmenopausal women is reviewed in detail separately. (See "Overview of the management of low bone mass and osteoporosis in postmenopausal women", section on 'Monitoring response to initial pharmacotherapy'.)

Duration of therapy

Initiating a drug holiday — No consensus exists on how long to continue bisphosphonate therapy. However, for some patients, stopping therapy after five years may be reasonable as there appears to be a residual benefit for BMD and fracture reduction for up to five years after treatment discontinuation. Although BTMs may help inform the decision to discontinue bisphosphonate therapy, limited data support this approach [38,45]. (See "Bisphosphonate therapy for the treatment of osteoporosis", section on 'Duration of therapy'.)

Resuming treatment after a drug holiday — Some clinicians, including the author of this topic, follow BTMs after discontinuing bisphosphonates and resume treatment when the BTM value reaches or exceeds the mid reference range for young adults [46,47]. While this approach makes physiological sense, supporting evidence is limited, and other UpToDate editors do not advocate using BTMs to determine whether to discontinue or resume bisphosphonate therapy.

In an observational study, 57 postmenopausal women with osteoporosis who had participated in oral bisphosphonate treatment trials were followed with serial measurement of BTMs and BMD after treatment discontinuation [48]. At 48 weeks after stopping bisphosphonate therapy, participants with CTX or P1NP values above the mean of the reference range exhibited greater decline in BMD at the total hip compared with those in whom BTMs remained below the reference mean (mean change in BMD -2.35 versus -0.26 percent, respectively, for CTX and -2.35 versus -1.09 percent, respectively, for P1NP).

Dental procedures — Until data support the ability of serum CTX to predict osteonecrosis of the jaw (ONJ), we do not recommend using serum CTX to determine whether it is safe to have an invasive dental procedure.

Some oral surgeons have proposed using a serum CTX level to assess risk and guide treatment of patients who are taking bisphosphonates and require invasive dental procedures [49]. They assign risk based upon CTX criteria and recommend withholding invasive dental procedures when the value is below a certain threshold. The CTX threshold was derived from 17 patients who developed ONJ while taking bisphosphonates. However, they did not measure CTX in a control group of unaffected bisphosphonate-treated individuals. Because bisphosphonates suppress bone resorption (which is why they are efficacious in reducing fracture), BTMs, including CTX, are reduced in patients taking bisphosphonates, and the vast majority of these patients do not get ONJ. It is impossible to identify a particular CTX level at which the risk of ONJ increases without also measuring CTX in a large cohort of bisphosphonate-treated individuals without ONJ [50,51].

GUIDELINES — The use of biochemical markers of bone turnover (BTMs) for managing osteoporosis is not a central component of most osteoporosis guidelines. When BTMs are addressed, guideline committees typically recommend against their routine use, due to the limitations of measuring and interpreting BTMs in individual patients [4,52,53] (see 'General principles' above). Most committees agree that a potential role of BTMs is monitoring osteoporosis therapy to identify nonresponders. However, prospective trials to define the most optimal approach for incorporating markers into management strategies are needed.

SUMMARY AND RECOMMENDATIONS

●Measurement of bone turnover markers – Several assays are available that measure bone turnover markers (BTMs). These assays measure collagen breakdown products and other molecules released from osteoclasts and osteoblasts during the process of bone resorption and formation. Markers that are specific to bone formation include bone-specific alkaline phosphatase (BSAP), osteocalcin, and serum procollagen type 1 N-terminal propeptide (P1NP), whereas markers specific to bone resorption include N-telopeptide (NTX), C-telopeptide (CTX), and pyridinoline cross-links (table 1). (See 'Introduction' above.)

●General principles – While the use of biochemical BTMs in clinical trials has been helpful in understanding the mechanism of action of therapeutic agents, their role in the care of individual patients is not well established. The measurement of BTMs is complicated by large, random, within-patient variability; biologic variability (age, sex, body mass index [BMI], circadian, and menstrual variation); and poor standardization of most assays (table 2). These issues have confounded their widespread use in clinical practice. (See 'General principles' above.)

●Bone loss and fracture risk – BTMs are predictive of the rate of bone loss and, in some studies, risk of fracture. However, there is no role for BTMs in selecting candidates for bone density testing or for osteoporosis therapy. (See 'Bone loss and fracture risk' above.)

●Monitoring response to antiresorptive therapy – We do not routinely measure BTMs in patients initiating osteoporosis therapy. If BTMs are used to monitor osteoporosis therapy, we suggest selecting BTMs that can be measured using automated technology and that have relatively small spontaneous variability, such as serum P1NP, CTX, or urinary NTX (table 2). (See 'Monitoring response' above and 'General principles' above.)

•Early treatment response – For individual patients (eg, patients with conditions that might interfere with drug absorption or efficacy or patients who are reluctant to take anti-osteoporosis medications regularly), we sometimes measure fasting urinary NTX, serum CTX, or serum P1NP before and three to six months after starting bisphosphonates or other antiresorptive therapy. A 50 or 30 percent reduction in urinary NTX excretion or serum CTX, respectively, provides evidence of treatment adherence and efficacy. This approach (with markers of bone resorption) is only useful with antiresorptive therapy. (See 'Monitoring response' above and "Overview of the management of low bone mass and osteoporosis in postmenopausal women", section on 'Monitoring response to initial pharmacotherapy'.)

•Suspected treatment failure – In individuals taking antiresorptive osteoporosis therapy, treatment failure due to nonadherence, gastrointestinal malabsorption, or other cause may be suspected based on an interim decrease in BMD or interim fragility fracture(s). In such individuals, measuring on-treatment BTMs may be a helpful component of the evaluation even in the absence of baseline measurement (algorithm 1).

●Duration of therapy – No consensus exists regarding the utility of BTM measurement for guiding the decision whether to discontinue antiresorptive therapy or to resume treatment after a drug holiday. (See 'Duration of therapy' above.)

●Assessing risk during dental procedures – In bisphosphonate-treated patients undergoing invasive dental procedures, data are insufficient to support any role for BTM measurement in assessing risk for osteonecrosis of the jaw (ONJ). (See 'Dental procedures' above.)