Ian H. Thorneycroft, MD, PhD

steoporosis is recognized as a significant contributor to morbidity and mortality in postmenopausal women. Several effective strategies, including calcium supplementation, weight-bearing exercise, and, most importantly, hormone replacement

, have been developed to prevent or at least delay clinically significant bone loss.

Densitometry has been utilized increasingly for osteoporosis screening and can be quite useful for absolute measurements of bone density at fracture risk sites, such as the hip and lumbar spine. The complementary use of urinary bone markers of bone collagen

 to measure bone resorption and assess bone loss rate provide a dynamic indicator of response to prescribed prophylactic regimens. These techniques have the potential to predict response to therapy earlier than bone mineral density measurements, which generally should be performed no less than 12 months apart.

Once the clinician learns how to apply these techniques, appropriate bone density measurement, as well as quantitation of bone resorption with urinary bone markers, can help direct therapeutic decision making and reinforce compliance with prescribed regimens.

A woman's relative risk of hip fracture is equal to her combined risk for developing breast, uterine, and ovarian cancers.

 Measurement of bone density is an excellent way to predict an individual's probability of fracture.

 In fact, the relationship of serum cholesterol levels to myocardial infarction, for example, is much weaker than the association between low bone density and high fracture probability. Moreover, the overall risk of osteoporotic fracture in postmenopausal women is extremely high-one in two women will sustain at least one fracture by the age of 70.

Bone density measurement, or densitometry, is used in postmenopausal women to screen for osteoporosis

, defined by WHO criteria as more than 2.5 standard deviations below the mean for young adult women.

 It is also used sequentially to assess the efficacy of antiresorptive therapy over the long-term. Such assessments have been, and continue to be, useful in establishing a baseline in individual patients and measuring progress over time. Caution should be exercised, however, in the selection of sites for measurement. Peripheral sites do not necessarily predict hip density or response to therapy. The achilles or calcaneous appears to be best suited to peripheral measurement.

Urinary bone marker measurements should be viewed as complementary to bone density measurement. They do not provide, however, an absolute measure of bone mineral density (BMD). Markers of bone formation are not clinically useful at the present time. Biochemical markers of resorption, in contrast, are relatively new tests that detect highly specific products of bone breakdown in urine. When the level of bone marker is high, it indicates a rapid rate of bone loss. If antiresorptive therapy is effective, the rate of bone loss should be slower and the level of bone marker should be decreased from baseline.

By learning how to use urinary bone marker testing appropriately in postmenopausal women, clinicians will be better able to predict who is at greatest risk of rapid bone loss, and whether patients are responding to therapy within a few months. Repeat bone density measurement can only be performed at 1- to 2-year intervals to reliably detect changes. These tests may also be considered tools for monitoring compliance (Is the patient taking her therapy properly?), as well as encouraging compliance.


 When collagen breaks down, as part of the ongoing physiologic process of bone remodeling, it releases specific products into the urine, such as hydroxyproline, galactosyl hydroxylysine, and pyridinoline cross-links. The products that are specific to type I collagen-which accounts for about 90% of the organic matrix of bone-are proving to be the best markers of this process.

 Since many connective tissues that contain collagen are subject to turnover, the ideal marker for monitoring the effects of antiresorptive agents, such as exogenous estrogen and bisphosphonates, would be one that is specific to bone and/or the osteoclastic resorption process.

The collagen molecule is cross-linked by two amino acids, pyridinoline and deoxypryrindinoline, with pyridinoline predominating. The telopeptides (short sequences of amino acids) at each end of the collagen molecule, which are critical to cross-linking, are called N-telopeptides and C-telopeptides. Since these short peptide fragments have been isolated from urine, it is presumed that the cross-link has protected the peptide sequence from degradation during osteoclastic resorption and passage through the body. One example of urinary bone markers that has been developed based on these findings is NTx (Osteomark®, Ostex International, Seattle), which measures the 8-amino acid sequence of the N-telopeptide. This sequence, which originates in the B2(1) collagen chain and has to be cross-linked (ie, it originates from a degraded fibril rather than a newly made collagen molecule), is generated by cleavage at a particular peptide linkage in the parent collagen molecule.

 NTx measurements in urine, therefore, provide a selective index of osteoclastic activity on bone collagen.

CLINICAL ROLE OF URINARY BONE MARKERS IN MENOPAUSAL WOMEN


 The primary clinical role of urinary bone markers is to identify patients who are likely to experience rapid bone loss. Even when a woman has a normal result on baseline bone densitometry (eg, dual-energy x-ray absorptiometry or DEXA), a high level of urinary breakdown bone marker is predictive of substantial and/or rapid bone loss within 1 year. A practical clinical study published recently by Chesnut et al documented the ability of a urinary bone marker test such as NTx to monitor and predict the therapeutic effects of hormone replacement therapy (HRT) in recently postmenopausal women.

 227 healthy postmenopausal women were randomized to receive either HRT (conjugated estrogens plus medroxyprogesterone acetate cyclically or continuously) plus 500 mg supplemental calcium, or supplemental calcium carbonate alone. In the HRT group, NTx values decreased significantly: overall mean decrease of 52% in NTx and significant gain in BMD of 2.5% (mean) at the lumbar spine and 1% at the hip. In contrast, NTx values in the calcium-only group remained relatively constant over the 12-month study period (mean, 7% decrease), with a 1.1% BMD loss (mean) at both the spine and hip.

When baseline NTx measurements were stratified into quartiles, women with the highest values (68 to 188 units) or the greatest decrease in NTx from baseline (-66% to -87%) at 6 months derived the greatest benefit from HRT in terms of BMD gain. For every increase of 30 units in baseline NTx, the odds of BMD gain in response to HRT increased by a factor of 5.0. For every 30% decrease in NTx values through 6 months, the odds of gaining BMD in response to HRT increased by a factor of 2.6. Likewise, in the control group, an increase of 30 units in mean NTx values indicated a higher risk of losing bone by a factor of 3.2.

We can conclude on the basis of these findings that determining NTx levels prior to initiating antiresorptive therapy with agents such as exogenous estrogens can help identify patients who would benefit most from this therapy. Moreover, by using a urinary bone marker test, the clinician can predict the type of BMD response likely to result from therapy.

In our own experience, we have noted that while standard laboratories provide normal ranges to help in interpretation of urinary bone marker tests results, these ranges are generally not useful in managing patients. The Chesnut study demonstrated that NTx values greater than 50 units should be considered indicative of bone loss.

Since all current preventive strategies-estrogens, bisphosphonates, calcitonin-inhibit bone breakdown, urinary bone marker tests can be used universally to monitor therapy better and earlier than bone density testing, which requires a long interval (12 to 18 months) to observe any change. In addition, since the natural variation in bone density is about 1% to 3%, the difference between a baseline densitometry test and a subsequent one must be greater than 3%.


 NTx testing has been utilized recently in a number of studies designed to evaluate preventive therapies for osteoporosis. In a trial comparing an estrogen-only patch with a new combination transdermal system containing both estradiol (E2) and norethindrone acetate (NETA) at three dose levels, NTx levels fell rapidly in all treatment groups. When densitometry was performed after 24 of months of treatment, it was clear that both types of patches were effective in maintaining bone. The fact that patients were responding to treatment, however, did not require 24 months to determine. Periodic NTx measurement showed significant reductions from baseline in all treated groups within the first months of therapy.

In another study that evaluated the effects of several dose levels of alendronate compared with placebo, NTx levels began to drop within 2 to 3 months of beginning treatment. Differences between all alendronate doses and placebo were significant.

Finally, in the Early Postmenopausal Interventional Cohort (EPIC) study, which compared placebo vs conjugated estrogens vs aldenronate, monitoring with urinary bone markers yielded much more information about patients' responses to therapy more rapidly than bone densitometry.

Based on these and other studies, we believe that practitioners should obtain a baseline urinary bone marker level just prior to prescribing an antiresorptive agent. Patients may be reassessed as early as 2 to 3 months after starting therapy. Response to treatment in this setting has been defined as a decrease of 30% or more in the NTx urinary bone marker level from baseline (Case Studies 

An additional role for urinary bone markers is for monitoring bone loss with GnRH agonist therapy. Prior to the availability of bone markers, agonist therapy was monitored by measuring estradiol levels. Unfortunately, different levels of estradiol were variably associated with bone breakdown. Urinary bone markers may be used to determine individual thresholds and may be useful outside of the menopause setting when a patient requires long-term GnRH agonist therapy.


 Bone resorption markers can serve as important aids to patient counseling and compliance. Therapies such as alendronate and, to a lesser extent, some hormone replacement regimens, are fairly complicated. By measuring a bone marker, the clinician should be able to determine whether the patient is taking her medication correctly, and, of course, whether bone loss is slowing (see 


 Since the use of urinary bone markers adds costs to the delivery of health care, it becomes incumbent on the medical profession to demonstrate how such agents can prevent long-term disability in a relatively cost-effective manner. Defining bone density by machine measurement provides documentation for insurers that deny reimbursement for antiresorptive therapy. Moreover, demonstrating a concrete risk, such as a low bone density, increases a patient's likelihood of initiating estrogen replacment therapy.

Since it is known that bone mass, or the lack of it, predicts fracture risk, and urinary bone markers can identify patients with rapid bone loss, it seems reasonable to assume that such bone markers can be used in association with densitometry in efforts to prevent osteoporosis.

One of the problems with bone density measurement has been the variability of results and interpretation. To address concerns about standardization of bone density assessments, the International Society for Bone Densitometry is performing site and physician certification similar to that performed for mammography. Until recently, for example, there were differences in the normal ranges for BMD such that a patient could be osteoporotic on one machine but not another. This problem has been resolved by all manufacturers using the same normative data.

As the population ages, the number of women at risk for osteoporosis will continue to increase. The use of urinary markers of bone collagen breakdown can help identify those patients at highest risk of bone loss. Once appropriate therapy is initiated, bone markers can rapidly demonstrate and characterize response to the treatment. Those rapid results, available within 4 to 12 weeks of beginning treatment, can be used as a powerful reinforcing tool to persuade patients to continue preventive regimens. They may also have a role in providing documentation of the value of antiresorptive therapy in the managed care setting.

 A 48-year-old premenopausal woman presented with osteoporosis (significant T-scores). Since she was already on oral contraceptives and was premenopausal, estrogen replacement was not an option.

Prior to beginning therapy with alendronate, the NTx test was used to establish a baseline. Interestingly, the baseline level, which was 30 units, was not remarkably high. Nevertheless, when this woman was reassessed a few months after starting alendronate, the NTx level dropped rapidly to 12.

In this case, it is presumed that frank bone loss may have been halted and that some gain in bone may have begun. These results were available in a few months and helped both the clinician and the patient decide that they were pursuing the correct approach. A second bone density test, in contrast, cannot be performed for almost a year.

 A 50-year-old, premenstrual woman with a fibroid uterus presented in 1995, requesting bone density measurement. Initially, the clinician resisted ordering the test since there weren't many therapeutic options for a premenopausal woman with low bone density. Nevertheless, the patient insisted and indeed, although measurement at the spine was normal, hip measurement was signficantly below the mean. The following year, this woman's bone density measurement revealed additional bone loss.

When alendronate became available, two urinary bone marker tests were performed to establish baseline measurements. Within 2 months, the NTx level had decreased considerably. Since that time, densitometry has confirmed a gain in bone.

It is always useful to know as rapidly as possible whether a selected therapy is having the desired effect. Moreover, since alendronate was used outside of labeling in this setting, and therefore without data to substantiate a predicted response, the decrease in NTx level represented extremely valuable clinical information.

 A patient for whom alendronate had been prescribed insisted that she would absolutely not be able to wait 30 minutes for her morning orange juice and coffee. Due to poor absorption, patients are generally instructed to take alendronate in the morning and to wait 30 minutes before consuming food or beverages. It was decided to let her take her medication and then go ahead with her morning beverages. Urinary bone markers were used to establish a baseline and to monitor change. Even though she admitted to taking her medication "incorrectly," bone markers in this patient decreased. The practitioner was thus reassured that therapy was likely to be effective, despite this patient's unorthodox method of administration.

    1. National Osteoporosis Foundation. Boning Up on Osteoporosis: A Guide to Prevention and Treatment. Washington D.C., National Osteoporosis Foundation, 1991.

    2. Riggs B, Melton LJ III. Involutional osteoporosis. N Engl J Med 1986;314(26):1676-1686.

    3. Jensen GF, Christiansen C, Boesen J, Hegedus V, Transbol I. Epidemiology of post-menopausal spinal and long bone fractures: a unifying approach to post-menopausal osteoporosis. J Orthopaedic Research Register 1982;16675-81.

    4. Report of a WHO Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. WHO Technical Report Series 1994;843:1-129.

    5. Eyre DR. How biochemical markers measure bone remodeling activity, in: Bone Resorption Markers in Menopause. OBG Management Oct 1997; pp 3-5.

    6. Chesnut CH III, Bell NH, Clark GS, et al. Hormone replacement therapy in postmenopausl women: Urinary N-telopeptide of type I collagen monitors therapeutic effect and predicts response of bone mineral density. Am J Med 1997;102:29-37.

    7. Hosking D, et al. Prevention of bone loss with alendronate in postmenopausal women under 60 years of age. Early Postmenopausal Intervention Cohort Study Group 1998;338(8):485-492.

    8. Dmowski WP, Rana N, Pepping P, et al . Excretion of urinary N-telopeptides reflect changes in bone turnover during ovarian suppression and indicates individually variable estradiol threshold for bone loss. Fertil Steril 1996;66(6):929-936.

    9. Marshall LA, Cain DF, Dmowksi WP, Chesnut CH III. Urinary N-telopeptides to monitor bone resorption while on GnRH agonist therapy. Obstet Gynecol 1996;87(3):350-354. 

Prepared from a symposium held May 1998 in New Orleans, Louisiana, and was made possible by an unrestricted educational grant from RhÃ´ne-Poulenc Rorer.

Originally published in International Proceedings Journal Volume 2 Number 1, September 1998 (Obstetrics and Gynecology), and reprinted with kind permission of the publisher, Worldwide Medical Communications, Inc.

Articles

Osteoporosis is recognized as a significant contributor to morbidity and mortality in postmenopausal women. Several effective strategies, including calcium supplementation, weight-bearing exercise, and, most importantly, hormone replacement, have been developed to prevent or at least delay clinically significant bone loss.

Ian H. Thorneycroft, MD, PhD

Articles

Assessing Bone Resorption Levels to Predict Skeletal Responses to HRT