Osteoporosis

Osteoporosis is a progressive systemic skeletal disease characterised by low bone mass and micro-architectural deterioration of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture.^1,2

Bone turnover (remodelling)

Bone turnover is the process of resorption followed by replacement by new bone with little change in shape, and it occurs throughout a person’s life. Osteoclasts break down bone (bone resorption), releasing the minerals, resulting in a transfer of calcium from bone fluid to the blood. The osteoclast attaches to the osteon (layers of compact bone tissue surrounding a central canal), and secretes collagenase and other enzymes. Calcium, magnesium, phosphate and products of collagen are released into the extracellular fluid as the osteoclasts tunnel into the mineralised bone. Osteoblasts are mature bone cells responsible for bone formation and ossification. They produce the organic portion of the matrix of bone tissue, osteoid, which is composed mainly of type I collagen, and are responsible for mineralisation of the osteoid matrix. Ossification fixes circulating calcium in its mineral form, removing it from the bloodstream. Repeated stress, such as weight-bearing exercise or bone healing, results in the bone thickening at the points of high stress.

Remodelling in adults repairs micro-damage to bone and plays a role in the regulation of calcium homeostasis. An imbalance in the bone remodelling processes in adults is thought to impact on bone strength as a result of reductions in bone volume and mineralisation, loss of trabeculae, deterioration of trabecular connectivity, and the formation of resorption cavities and trabecular perforations.^3,4 Therefore, an increase in bone turnover where resorption exceeds formation is not only inversely correlated with bone mineral density (BMD), but may also alter bone architecture and porosity, increasing the risk of fracture beyond that due to reduced BMD, and can therefore be an independent predictor of fracture risk.^3–6

Diagnosis

Osteoporosis causes no symptoms until a bone is broken. As osteoporosis is associated with low bone density, bone density scanning [using dual-energy X-ray absorptiometry (DXA)] has become the most commonly used diagnostic technique.² There are accepted diagnostic criteria based on DXA: osteopenia (low bone mass) is present when the BMD is between 1 and 2.5 standard deviations below the mean value for young adults (BMD T-score of –1 to –2.5); osteoporosis is diagnosed when BMD is < 2.5 standard deviations below young adults’ (BMD T-score of < –2.5).⁷

Risk of fracture

A reduction in BMD results in the thinning of the trabeculae and an increase in the fragility of the bones.⁸ Therefore, people diagnosed with osteoporosis have an increased risk of suffering low trauma (fragility) fractures. When BMD is measured by DXA, a reduction of 1 standard deviation in BMD is reportedly associated with a 50–150% increase in the risk of osteoporotic fracture.⁹ Increasing age is one of the major risk factors for osteoporosis; after 35 years of age bone loss increases gradually as part of the natural ageing process.² By 75 years of age, approximately half of the population will have osteoporosis. In addition, there is an increased risk of falling which increases the risk of fracture; one in two women and one in five men over the age of 50 in the UK will fracture a bone, mainly as a result of skeletal fragility.^2,10 The most common fractures in people with osteoporosis are of the wrists, hips and spinal bones; these are most common in older people, but younger people can sometimes be affected.^8,11

According to recent National Institute for Health and Care Excellence (NICE) guidance [clinical guideline (CG) 146], assessment of the risk of fragility fractures should be considered in:¹²

• all women aged 65 years and over and in all men aged 75 years and over
• in women aged under 65 years and in men aged under 75 years in the presence of risk factors, for example:
  • previous fragility fracture
  • current use or frequent recent use of oral or systemic glucocorticoids
  • history of falls
  • family history of hip fracture
  • other causes of secondary osteoporosis
  • low body mass index (BMI) (< 18.5 kg/m²)
  • smoking
  • alcohol intake of more than 14 units per week for women and more than 21 units per week for men.

An assessment tool for assessing fracture risk, FRAX^®, has been developed by the World Health Organization (WHO).¹³ The factors taken into account are age, gender, weight, height, previous fracture, parental history of hip fracture, smoking status, the use of oral glucocorticoid steroids, a diagnosis of rheumatoid arthritis, the presence of a disorder strongly associated with osteoporosis and alcohol consumption, with or without BMD as determined using DXA.^12,14

Treatments for osteoporosis

Diet and exercise can be modified to improve a person’s fracture risk. Exercises considered best for people with osteoporosis are those that (1) are thought to have an effect on density and strength, such as weight-bearing exercises that cause force on the bones like jogging, stair climbing, walking briskly and resistance exercises, and (2) can reduce the risk of falling, such as balance training (e.g. tai chi), leg strengthening and flexibility training (e.g. yoga). Exercises that people with osteoporosis are advised to avoid are those that might increase the risk of falling, those that involve twisting the spine or bending from the waist, high-impact activities such as high-intensity aerobics or jumping and the use of excessive weight during resistance exercise. A diet containing foods rich in calcium and vitamin D (vitamin D is required for the absorption of calcium) or the use of calcium and vitamin D supplements can also improve bone strength.

The most common medical therapies for osteoporosis are bisphosphonate drugs. Bisphosphonates inhibit the activity of mature osteoclasts and reduce the rate of resorption.⁴ The most commonly prescribed bisphosphonate is generic alendronate; other bisphosphonates include etidronate, risedronate (now available in generic form), ibandronate, and zoledronate. The recommended dose of alendronate is one 70-mg tablet per week, rather than 10 mg daily as originally prescribed, to reduce the incidence of gastrointestinal adverse effects and increase adherence. A strict technique must be adhered to when taking oral bisphosphonates to ensure satisfactory absorption. They must be taken on an empty stomach first thing in the morning, while remaining upright to prevent reflux, at least 30 minutes before the first food, drink or other medication of the day. The tablet should be taken with plain water only; other drinks (including mineral water), food and some medicines are likely to reduce the absorption of bisphosphonates.¹⁵ Intravenously administered bisphosphonates are available; the recommended doses are 3 mg 3-monthly of ibandronate, or 5 mg annually of zoledronate. Pamidronate is not licensed for the treatment of osteoporosis but has been widely used off-licence at a dose of 30 mg quarterly.

Other medical therapies available include:

• raloxifene (Evista^®, Eli Lilly and Company Ltd): a selective oestrogen receptor modulator (SERM), which is a synthetic hormone that copies the effects of oestrogen on the bones
• strontium ranelate (Protelos^®, Servier Laboratories Ltd): a strontium(II) salt of ranelic acid, which is a dual-action bone agent that stimulates new bone growth and reduces bone loss
• teriparatide (Forsteo^®, Eli Lilly and Company Ltd): a recombinant form of parathyroid hormone (PTH 1–34) that helps regulate calcium levels and the activity of cells involved in bone formation
• denosumab (Prolia^®, Amgen Ltd): a monoclonal antibody that targets the RANK ligand
• hormone replacement therapy (HRT): a mix of hormones (oestrogens, progesterone or progestins, and sometimes testosterone) prescribed to post-menopausal women (natural or surgically induced) to reduce the symptoms caused by reduced circulating oestrogen and progesterone. The risk of development and progression of osteoporosis can therefore be reduced by the maintenance of oestrogen levels.

Burden of the disease on the NHS

Approximately 3 million people in the UK have osteoporosis, with about 20% of women aged 60–69 affected. There are thought to be about 230,000 osteoporotic fractures every year, with broken wrists, hips and spinal bones being the most common. Of the 60,000 people who suffer osteoporotic hip fractures each year, 15–20% are likely to die within a year from causes related to the fracture.²

As stated in Diagnosis, above, there are a range of treatments available for osteoporosis, and the costs of these vary (pamidronate has not been costed as it is not licensed for use in osteoporosis):¹⁶

• Generic sodium alendronate: a 28-tablet pack of 10-mg tablets is £1.44 (approximately £19 annually); a four-tablet pack of 70 mg for once-weekly administration is £1.10 (approximately £14 annually). Fosamax^® (MSD) costs £23.12 for 28 10-mg tablets and £22.80 for four 70-mg once-weekly tablets.
• Generic sodium risedronate: a 28-tablet pack of 5-mg tablets is £17.99 (approximately £220 annually); a four-tablet pack of 35 mg for once-weekly administration is £19.12 (approximately £230 annually).
• Zoledronate: Zometa^® (Novartis) costs £174.17 for 4 mg in 5 ml, and Aclasta^® (Novartis) costs £253.38 for 5 mg in 100 ml – 5 mg administered once annually.
• Strontium ranelate (Protelos^®, Servier) costs £25.60 for 28 sachets each containing 2 g of granules daily (approximately £330 annually).
• Denosumab (Prolia^®, Amgen Ltd) costs £183.00 for 60 mg/ml in a 1-ml prefilled syringe – 60 mg administered 6-monthly (£366 annually).
• Raloxifene (Evista^®, Daiichi Sankyo) costs £17.06 for 28, and £59.59 for 84, 60-mg tablets – 60 mg daily (approximately £220 annually).
• Teriparatide (Forteo^®, Eli Lilly and Company Ltd) costs £271.88 for 250 µg/ml in a 3-ml pre-filled pen – 20 µg self-administered daily (approximately £3540 annually).

According to Hospital Episode Statistics (HES), in 2005–6 in England there were 5759 consultations and 4034 admissions (2368 emergency) for osteoporosis with a pathological fracture, and a further 8725 consultations and 8313 admissions (716 emergency) without pathological fracture.¹⁷ For surgical interventions for fractures of the spine and hip (not only those associated with osteoporosis), there were 809 consultations and 667 admissions (353 emergency) for fixations of spinal fractures (approximately 26% in patients 60 years and older), and 46,812 consultations and 46,191 admissions (1611 emergency) for primary total prosthetic replacement of hip joint [depending on method used, approximately 50% (not using cement) to 85% (using cement) 60 years and older].¹⁷ Given the discrepancies in the numbers of hip replacements in the elderly and consultations of osteoporotic fractures, the incidence/consultation rate for osteoporosis may be underestimated. A recent report published by the Royal College of Physicians stated that only 32% (1933 out of 6083) of non-hip fracture and 67% (2324 out of 3484) of hip fracture patients had a clinical assessment for osteoporosis/fracture risk.¹⁸ Osteoporosis reportedly costs the NHS and government £2.3B per year (£6M per day).²

National Institute for Health and Care Excellence guidance

NICE has produced a number of technology appraisals (TAs) and CGs that have some relevance to this area. Three relevant TAs have been published: TA160 (Osteoporosis – primary prevention; postmenopausal women),¹⁹ TA161 (Osteoporosis – secondary prevention including strontium ranelate; postmenopausal women)²⁰ and TA204 (Osteoporotic fractures – denosumab).²¹

For the primary prevention of fractures, alendronate is recommended as the first-line treatment for most women at risk of fractures. Risedronate, etidronate and strontium ranelate are alternative treatments for post-menopausal women who cannot adhere to the required alendronate regimen, or those women with pre-specified combinations of T-score, age and number of independent clinical risk factors; strontium ranelate is not recommended as a first-line treatment for osteoporosis. Raloxifene is not a recommended treatment for the primary prevention of osteoporotic fragility fractures.¹⁹ The recommendations for the secondary prevention of fractures are similar to those for primary prevention. The two differences are that (1) strontium ranelate can be used as a first-line treatment and (2) raloxifene is recommended as an alternative treatment for post-menopausal women who cannot adhere to alendronate, or in women with pre-specified combinations of T-score, age and number of independent clinical risk factors.²⁰ Denosumab has now also been added to the list of alternative second-line treatments for the primary or secondary prevention of fractures.²¹

There are also four potentially relevant CGs available that deal with the management of independent risk factors for fracture: CG146 (Osteoporosis fragility fracture),¹² CG21 (Falls: the assessment and prevention of falls in older people),²² CG59 (Osteoarthritis: the care and management of osteoarthritis in adults)²³ and CG79 (Rheumatoid arthritis: the management of rheumatoid arthritis in adults).²⁴

This review will focus on patients being treated for osteoporosis with any of bisphosphonate, raloxifene, strontium ranelate, teriparatide or denosumab.

Bone turnover markers

Biochemical markers of bone turnover are used to monitor treatment response and may prove to be more useful than serial BMD measurements as they are non-invasive, relatively cheap compared with DXA, and there is an increased availability of auto-analysers in clinical chemistry laboratories.

Variability in bone turnover markers

Several factors can impact on the bone turnover marker levels, causing variability across samples, which can reduce repeatability and comparability, both within patients and between patients. These include specimen collection and storage;^25,31–35 differences between analytical methods used;^32,34 temporal variations (diurnal, menstrual, seasonal);^25,31–35 diet and fasting;³⁶ patient characteristics (age, gender or ethnicity);^25,31,33,35 concomitant medication other than osteoporosis medications [HRT, anabolic agents, glucocorticoids, anticonvulsants, gonadotropin-releasing hormone (GnRH) antagonists or oral contraception];^25,31 and comorbid conditions (renal impairment, liver disease, diabetes, thyroid disease, osteomalacia, systematic inflammatory diseases, degenerative joint disease, conditions causing immobility, or eating disorders).^25,31,33,35

Intrapatient variability for serum markers is lower than for urinary markers.³⁴ Some tests are more accurate when monitoring the response to specific treatments (e.g. CTX with bisphosphonates). Some tests have the advantage of not requiring the patient to fast prior to sampling (e.g. P1NP), or are less affected by diurnal variations (P1NP and BALP), and/or have lower overall intraindividual variability (BALP) than other bone turnover markers.³⁷ Each of these tests also has disadvantages: CTX has a large circadian rhythm, and therefore repeat sampling must be done at the same time of day, fasting is required prior to sampling, and the marker requires freezing soon after sampling as it can be unstable; BALP is affected by cross-reactivity with the liver form of alkaline phosphatase, limiting its use in patients with liver disease; and P1NP has a higher cost compared with other bone turnover markers.³⁷ Given the advantages that CTX, P1NP and BALP offer, and the availability of NTX, these are the bone turnover markers that will be investigated in the current review.

Use of bone turnover markers

The use of bone turnover markers varies greatly across the UK, in terms of both the test used and the frequency of its measurement. Several factors will need to be considered when choosing the bone turnover marker to be used, not least the availability of the assay methods. Bone turnover markers have a number of potential uses, including:^6,37

1. predicting bone loss
2. identifying people at risk of primary or secondary osteoporosis and fracture
3. predicting treatment response prior to commencement
4. monitoring the response to osteoporosis treatment; identifying non-responders, which will include those not adhering with osteoporosis treatment (including patients not taking the medication or not following the instructions for administration)
5. identifying oversuppression of bone turnover in patient on long-term osteoporosis therapy
6. monitoring of people who have been on long-term treatment, or shown signs of oversuppression, and are taking a ‘treatment holiday’.

The main focus of this systematic review will be role 4: monitoring the response and non-response to osteoporosis therapy (and change in fracture risk).

Monitoring response to treatments for osteoporosis

There is currently no standard practice for the monitoring of patients receiving treatment for osteoporosis. The options include the use of repeated DXA, repeated measures of bone turnover markers, clinical review, or a combination of these. The use of DXA to monitor the response to osteoporosis treatment has limitations. Firstly, detectable changes in bone density due to treatment can take up to 2 years to become apparent;³⁸ therefore, the identification of non-responders to treatment is delayed. Secondly, there is limited access to the technology and the test is relatively expensive (average £72 per scan). Thirdly, there is evidence that there is limited value in the regular monitoring of BMD in patients on bisphosphonate therapy.^39,40

As stated earlier, the relationship between bone turnover and bone density and architecture means that the rate of bone turnover may be an independent predictor of fracture risk;^3–6 this can be measured using one or more of the bone turnover markers listed above. However, it is still unclear whether or not changes in bone turnover detected by bone turnover markers are reliable surrogate measures for improved bone density and architecture, and consequently accurate predictors of future fracture risk. Two studies have suggested that bone turnover markers can have independent predictive value in assessment of fracture risk.⁴¹ If biochemical markers of bone turnover are reliable indicators of future fracture risk, their use may prove advantageous compared with serial BMD measurements, as not only are they non-invasive, relatively cheap compared with DXA, and the availability of auto-analysers in clinical chemistry laboratories is increasing, but a response to treatment can be detected much earlier than with DXA.

Changes in bone turnover rates have been detected in post-menopausal women within as early as 2 weeks after starting HRT,⁴² although the peak accuracy of changes in bone turnover markers to predict fracture risk in response to osteoporosis treatment may be later than this, between 3 and 12 months after initiating treatment, depending on the treatment and bone turnover marker used.^43–46 The ability to identify non-responders early within the treatment can be beneficial for patients by allowing early changes in management strategy if deemed necessary. The definition of treatment success varies depending upon the baseline risk of the patient being treated; in some patients a reduction in bone turnover would be considered a treatment success, but in others success may be a stabilisation of bone turnover. For all patients a continued increase in bone turnover rates would be considered a treatment failure. The definitions used throughout this project will reflect clinical practice and be based upon evidence for least clinical significant change.

There is a complex association between changes in bone turnover and fracture risk that is influenced by the treatment–bone turnover marker combination; the observed change in bone turnover markers will depend upon the treatment being administered. In studies of raloxifene, risedronate, alendronate and zoledronic acid, bone turnover markers have been reported as explaining between 28% and 77% of fracture risk reduction.⁴⁷

Bisphosphonates are antiresorptive therapies, and therefore they reduce the rate of bone resorption. Bone resorption is closely coupled to bone formation; consequently, there is usually a subsequent reduction in the rate of bone formation. This results in a transient uncoupling of bone turnover, which leads to a small increase in BMD. This increase in BMD may account in part for the decrease in fracture risk, but the reduction in bone turnover may independently improve bone strength by improving bone architecture and porosity.⁴⁸ Both raloxifene and denosumab reduce bone resorption, and therefore act as antiresorptive therapies; decreases in both bone resorption markers and subsequently bone formation markers should be observed in treatment responders as with bisphosphonates.

Teriparatide causes a small, transient increase in serum calcium, mainly due to the stimulation of tubular reabsorption of calcium from the proximal kidney tubules and increased calcium absorption from the bowel, but in a small part by increasing bone resorption (hence chronically elevated PTH can deplete bone). However, intermittent administration of PTH (i.e. daily injections of teriparatide) activates osteoblasts more than osteoclasts, stimulating new bone formation and increasing BMD. Therefore, a positive response in bone formation markers would be expected in treatment responders, with a subsequent increase in bone resorption markers due to the coupling of the processes, the opposite response to that seen with antiresorptive therapies.

Strontium ranelate increases new bone formation as well as reducing bone resorption and is classed as a dual-action bone agent. These effects are more modest than those seen with anabolic and antiresorptive treatments, with smaller positive changes in bone formation and negative changes in bone resorption markers, respectively. However, strontium ranelate appears to lead to persistent uncoupling of bone turnover.

The interpretation of changes in bone turnover markers is also influenced by the type of sample used: serum or urine. The intraindividual variability is greater for urine markers, giving serum markers a better signal to noise (S/N) ratio;³⁴ the percentage change in a urinary biomarker needed to indicate a treatment response (least significant change) is greater than that required for a serum biomarker.