Clodronate Reduces the Incidence of Fractures in Community-Dwelling Elderly Women Unselected for Osteoporosis: Results of a Double-Blind, Placebo-Controlled Randomized Study
Abstract
A 3-year prospective, randomized, placebo-controlled trial of oral clodronate 800 mg showed that the incidence of clinical fractures was decreased by 20% in 5596 elderly women unselected for osteoporosis. The effect occurred in the absence of systematic calcium and vitamin D supplementation and was observed across a wide range of BMDs.
Introduction: To date, most studies with bisphosphonates have reported on their use in individuals selected to be at high risk for fracture usually by the presence of low BMD or a prior fragility fracture, usually of the spine. We wished to determine the effect of the bisphosphonate, clodronate, on the rate of fractures in women ⩾75 years of age living in the community.
Materials and Methods: Women ⩾75 years of age living in the general community in South Yorkshire and North Derbyshire, identified from general practice registers, were recruited by letter of invitation to a randomized, double-blind, controlled trial of 800 mg oral clodronate (Bonefos) or matching placebo daily over 3 years. The main outcomes were the incidences of hip and any clinical fracture.
Results: Of the 5579 elderly women included in the intention-to-treat analysis of efficacy, 114 had a new hip fracture during the 3-year treatment phase: 56 (2.0%) women in the clodronate group and 58 (2.1%) women in the placebo group (hazard ration [HR], 1.02; 95% CI, 0.71–1.47). Clodronate did, however, decrease the incidence of any clinical fracture by 20% (264 women [9.5%] versus 337 [12.1%] in the placebo group; HR, 0.80; 95% CI, 0.68–0.94). The incidence of osteoporosis-associated nonhip fractures was also significantly decreased by 29% (5.2% versus 7.4%; HR, 0.71; 95% CI, 0.57–0.87). The ability of clodronate to reduce the risk of osteoporotic fracture was independent of baseline BMD, but the number needed-to-treat was lower in the presence of osteoporosis.
Conclusions: Oral daily clodronate can prevent fractures without significant adverse effects in elderly women living in the general community. The effect on hip fracture risk is not significant, but an effect similar to that at other nonvertebral sites cannot be excluded. This study suggests that antiresorptive therapies can reduce fracture incidence in high-risk individuals even in the presence of a normal or osteopenic BMD.
INTRODUCTION
Fractures among elderly women are a major cause of morbidity and mortality. For example, hip fractures account for ˜20% of orthopedic bed occupancies in the United Kingdom, and based on current population trends, the number of hip fractures may rise to 120,000/year by 2015.1 Bisphosphonates have been shown to significantly reduce fracture risk by decreasing bone turnover and maintaining or increasing BMD.2-11 To date, most studies with bisphosphonates have reported on their use in individuals selected to be at high risk for fracture usually by the presence of low BMD or a prior fragility fracture, usually of the spine. This study design is unique because it examined the efficacy of a bisphosphonate, clodronate, in a population cohort of elderly women, randomly selected from general practice lists, regardless of underlying osteoporosis or fracture risk.
MATERIALS AND METHODS
The study was a double-blind, prospective, randomized, placebo-controlled, single center study in elderly community-dwelling women ⩾75 years of age. In addition to examining the efficacy of clodronate, the study was also designed to determine risk factors for fracture in elderly women in the United Kingdom. Participants were therefore recruited randomly from general practice lists and did not need to have proven osteoporosis or any other known risk factors for fracture. Exclusion criteria was comprised of concurrent treatment for a malignancy, concurrent medication likely to influence skeletal metabolism (other than calcium supplements ⩽ 500 mg daily), bilateral hip arthroplasties, known malabsorptive states, impaired mental state or concurrent illness that would impede informed consent or compliance with the study, significant impairment of renal or hepatic function (serum creatinine > 300 μM or alanine aminotransaminase [ALT] more than twice the upper limit of the reference range, respectively), and serum biochemistry consistent with underlying metabolic bone disease (e.g., osteomalacia) or calcium disorders other than primary hyperparathyroidism.
Baseline assessments
All baseline assessments were carried out during a single visit to the WHO Collaborating Center for Metabolic Bone Diseases in Sheffield. All visits thereafter were conducted in the community by a team of study nurses at 6-month intervals to undertake collection of fracture data, adverse events, and hospitalizations, as well as to collect and dispense study medication.
Each participant underwent a detailed and comprehensive assessment of their general health, fracture history, and a number of measurements of BMD, muscle strength, and postural stability. BMD was measured by DXA at the hip using a Hologic QDR4500 Acclaim densitometer in the majority of women (main study), with a Hologic QDR2000plus densitometer used for a small number recruited to an earlier study (feasibility study). Morphometric X-ray absorptiometry (MXA) was undertaken at the thoracic and lumbar spine for vertebral morphometry at entry in all main study participants using a Hologic QDR 4500 Acclaim densitometer. Prevalent vertebral fractures were defined according to vertebral morphometry method based on the McCloskey algorithm and using reference data from lateral spine radiographs.12, 13 MXA scans were planned to be repeated in a subset of women with baseline vertebral fractures at the end of the 3 years of treatment.
None of the results of the baseline assessments of fracture risk, including BMD values, were communicated to the participants.
Incident fracture assessments
Incident fractures and hospital admissions were inquired about at each visit. All reported fractures were confirmed by hospital notes, discharge/general practitioner (GP) letters, copies of radiographic reports, or review of radiographs if necessary. Deaths were checked against death registry reports from the NHS Central Registry. Only verified fractures were included in statistical analyses and were defined as “clinical fractures” because they had presented symptomatically and triggered radiological study. Within these clinical fractures, further categories were derived including hip fractures, osteoporotic fractures (excluding fractures of the skull, nose, face, hand, finger, feet, toe, ankle, and patella fractures)14 and nonhip osteoporotic fractures. The accuracy of fracture reporting was validated by comparing the study data to fractures recorded on a database at the Northern General Hospital (NGH) Accident and Emergency department. Of the 172 fractures on the NGH database, the study site had missed 3 fractures (two finger fractures and one wrist fracture). None of the incident hip fractures were missed.
Study treatment
After randomization, the women received either clodronate 800 mg daily (two BONEFOS 400-mg tablets once daily or one tablet twice daily) or an identical placebo. Women were randomized 1:1 in each group using the SAS/PLAN procedure for one site, two treatments, and a block size of 10. Study medication was taken on an empty stomach with a drink of water at least 1 h before breakfast. It could also be taken in the middle of the night if desired by the women after fasting for ˜5–6 h. The intervention was continued for 3 years. Concomitant calcium and vitamin D supplementation was not given. No direct measures of calcium intake were collected at entry to the study apart from a simple question about current milk intake (graded as never, occasional, 1–2 glasses/day, 3–4 glasses/day, and >5 glasses/day). Subjects with biochemical evidence of calcium and/or vitamin D deficiency (serum calcium adjusted for albumin < 2.12 mM and hypophosphatemia) were excluded from the trial.
Statistical analysis
The primary statistical analysis was conducted by Schering, but the entire dataset of individualized data was provided for independent statistical evaluation and confirmation to Dr McCloskey, Prof Nicholl (for analysis on behalf of the MRC), and Prof Anders Oden (Consulting Statistician, Gothenburg, Sweden). All authors had full unrestricted access to the data.
The primary endpoint was the incidence of hip fractures. All clinical fractures during the 3 years of the study were also analyzed. The secondary efficacy endpoints included the incidence of all osteoporosis associated nonhip fractures.
The sample size calculations were based on the assumptions that the rate of hip fracture is 140 per 10,000 person-years in this age group of women and that clodronate would cause a 40% decrease in fracture rate. The sample size calculations allowed for a 13.7% mortality rate (estimated from the OPCS life table mortality data for a cohort of women ⩾75 years of age) and 20% loss-to-follow-up for other reasons over a total treatment period of 3 years. With statistical power of 80% and a significance level of 0.05, the needed number of subjects was estimated to be 2849 subjects in each treatment group (total, 5698 subjects).
The analyses of fracture incidence were undertaken by Cox regression modeling that included treatment as a factor and baseline age and weight as covariates. Poisson regression was used to examine the relationship between baseline hip BMD and the reduction in fractures during clodronate treatment. Changes in femoral neck BMD were compared by ANOVA for repeated measurements.
All analyses were undertaken on an intent-to-treat basis so that all fractures were included in the analyses regardless of whether the women were taking study medication or not.
RESULTS
Approximately 35,000 women ⩾75 years of age identified through local GP listings were approached by letter of invitation (Fig. 1). Altogether, 5873 elderly women attended the screening visit, of which 281 did not fulfill the inclusion/exclusion criteria; 5592 women were finally randomized at the single study site.
Study overview.
The two study groups were well matched at entry (Table 1). The only difference noted was a slightly but significantly higher prevalence of self-reported osteoporotic nonvertebral fractures in the clodronate group (24.4% versus 21.1%, p = 0.004). The prevalence of radiological vertebral fractures, detected on MXA, was similar in both groups.
|
Follow-up data were available in 5579 women (99.8% of those randomized) and comprised the intent-to-treat population. The total follow-up during the 3-year study was similar in both study groups (7761 and 7628 person-years for placebo and clodronate, respectively), with a median treatment exposure of 2.9 years for both treatment groups. Women completing the 3-year study were followed under blinded conditions until the last patient had completed the original study; for the safety evaluation, events were collected over this entire study period (median duration, 4.0 years). There were no major differences between women assigned to clodronate or and those assigned to placebo with respect to the reasons for discontinuation of the study (Fig. 1).
Efficacy of clodronate to prevent hip bone loss and reduce the risk of hip fracture
Of the 5579 elderly women included in the intention-to-treat analysis of efficacy, 114 had a new hip fracture during the 3-year treatment phase, 56 (2.0%) women in the clodronate group and 58 (2.1%) women in the placebo group, giving an overall hazard ratio (HR) of 1.02 (95% CI: 0.71, 1.47; p = 0.918). There was a significant interaction between treatment and the study duration so that the treatment effect changed significantly during the 3-year study (p = 0.032). During the first study year, new hip fractures occurred in 24 women (0.86%) in the clodronate group and 17 women (0.61%) in the placebo group (HR, 1.31; 95% CI, 0.84, 2.03) compared with 14 women (0.60%) in the clodronate group and 23 women (0.95%) in the placebo group during the third study year (HR, 0.49; 95% CI, 0.23, 1.06).
Repeat hip BMD measurements obtained at 36 months in a subset of 692 women showed that, whereas total hip BMD decreased significantly in both study groups, the mean decrease was 61% lower in the 336 women assigned to clodronate (−0.012 g/cm2, −1.6%) than in the 356 women in the placebo group (−0.029 g/cm2, −4.1%; p < 0.0001 between groups; Fig. 2).Figure 3, Figure 4
Mean ± SE changes in total hip BMD (g/cm2) observed over 3 years in women assigned to placebo or clodronate. Compliance was determined by tablet counts (compliant ⩾ 60% of tablets taken). ITT, intention-to treat analysis. The numbers above each bar represent the numbers of subjects with paired BMD measurements at baseline and 3 years in each group.
Kaplan-Meier plot of incident clinical fractures over entire study duration.
Relationship between reduction in risk of fracture caused by clodronate therapy and total hip BMD at entry to the study. No significant relationship was observed.
Using tablet counts in both groups, women were categorized as fully compliant (defined as taking at least 60% of study medication), as noncompliant, or as having stopped therapy completely during the 3 years of medication (Fig. 2). Differences in compliance in the placebo group showed little impact on the rates of bone loss, with only a slightly lower rate of bone loss in patients deemed to be compliant. Clodronate treatment in the compliant patients was associated with prevention of bone loss at the hip, whereas noncompliance was associated with rates of loss similar to that observed in the placebo group (Fig. 2). The majority of women who discontinued therapy did so within the first year; the change in BMD over 3 years was almost identical to that observed in similar women in the placebo group (Fig. 2).
Efficacy of clodronate to reduce the risk of any clinical fracture and nonhip osteoporotic fractures
During the 3 years of treatment, 264 (9.5%) women in the clodronate group and 337 (12.1%) women in the placebo group sustained an incident clinical fracture (HR, 0.80; 95% CI, 0.68, 0.94), indicating that clodronate treatment reduced the risk of any clinical fracture by 20% (Fig. 3). There was no significant interaction between treatment and study duration. In contrast to hip fractures, the incidence of osteoporosis-associated nonhip fractures was significantly lower in the clodronate group (p = 0.001) during the 3-year treatment period (Table 2); there were 144 (5.2%) women with such fractures in the clodronate group compared with 207 (7.4%) women in the placebo group (HR, 0.71; 95% CI, 0.57, 0.87), indicating a 29% decrease in the risk of sustaining a new osteoporotic nonhip fracture in the clodronate group (Table 2). The effect was observed at the spine (22 versus 31 women with incident clinical vertebral fractures), other axial sites (31 versus 42 women), and the appendicular skeleton (102 versus 143 women). A similar effect to reduce vertebral fracture risk (8.7% versus 12.2%; HR, 0.73; 95% CI, 0.36, 1.49) was observed in the subgroup of 297 women with repeat MXA scans at 3 years.
|
Interestingly, the ability of clodronate to reduce relative fracture risk was not significantly influenced by baseline hip BMD (Fig. 3). The reduction in fracture risk was similar in women with normal or osteopenic BMD at baseline to that observed in women with osteoporosis (defined as a total hip BMD T score < −2.5, equivalent to a value of 0.64g/cm2; Table 3). The higher event rate resulted in lower numbers needed-to-treat in the presence of osteoporosis.
|
Safety and tolerability
Adverse events (AEs) were reported for 97% of the women in both treatment groups. Premature termination of the study drug caused by an AE was reported for 593 women in the clodronate group and for 471 women in the placebo group, largely because of more common occurrences of diarrhea, abdominal pain, dyspepsia, nausea, and vomiting in the clodronate group. However, the incidences of esophagitis confirmed by gastroscopy (clodronate: 22 women versus placebo: 36 women) and gastritis (clodronate: 48 women versus placebo: 54 women) were higher in the placebo group. Serious AEs (SAEs) were reported by 1386 (49.6%) women in the clodronate group and 1397 (50.0%) in the placebo group. During the 3-year treatment period, a similar number of women died in both study groups: 264 women in the clodronate group and 249 in the placebo group.
DISCUSSION
Our study shows that clodronate was able to significantly reduce the incidence of all clinical fractures by 20% during 3 years of treatment in elderly, community-dwelling women unselected for osteoporosis. Indeed, the efficacy of clodronate on clinical fractures in this study seems comparable with that of other bisphosphonates10 and hormone replacement therapy (HRT)15 and is somewhat more effective than raloxifene.16 The current result is unusual in that, despite a highly significant reduction in clinical fractures, clodronate treatment showed no apparent efficacy on the incidence of hip fracture. The apparent lack of efficacy of clodronate is a consequence of a small excess (n = 7) of incident hip fractures in the first year with a progressive reduction in the incidence of hip fracture over the second and third years. The incidence of hip fractures observed in the study was approximately one half that expected at the study outset (76/10,000 person-years in the placebo group versus 140/10,000 person-years expected in the general population) and indicates that the study participants were at lower risk of fracture. The observation of a healthy participant bias is supported by a lower prevalence of hip osteoporosis at entry (19.6% versus an expected prevalence of ˜30%) and a mortality rate that was approximately one half that of the age-matched population. Bearing in mind the effect on all clinical fractures, it is important to note that a 20% reduction in the incidence of hip fracture during clodronate therapy cannot be excluded, because the lower confidence limit of the estimate is 0.7. Given the lower incidence of fractures in the placebo group, a total in excess of 11,500 or 33,000 women would have been required to detect a 33% or 20% decrease in fracture risk, respectively.
This study design differs substantially from that used in most other studies of fracture prevention in the elderly with bisphosphonates. In previous studies, participants were usually selected on the basis of increased risk (previous fragility fracture and/or underlying low BMD and/or increased fall risk), but the women taking part in our study were unselected for risk. There was therefore no opportunity for the knowledge of risk to positively influence compliance, an effect that has been shown previously in osteoporosis prevention.17 There is evidence within this study of significant interactions between compliance, assessed by tablet counting, and the prevention of bone loss. However, despite these observations, no direct interaction could be shown between compliance and the ability of clodronate treatment to reduce the incidence of hip fractures.
A small number of other population-based placebo-controlled studies have been conducted to determine the effects of non-bisphosphonate therapies on fracture risk, with most examining the effects of calcium and/or vitamin D supplementation.18-20 In a relatively small study of 199 men and 246 women ⩾65 years of age, 3 years of supplementation with calcium 500 mg and vitamin D 700 IU seemed to reduce the incidence of nonvertebral fracture by 50% (5.9% versus 12.9%; relative risk [RR], 0.5; 95% CI, 0.2–0.9),18 although the number of observed events was relatively small. A smaller but statistically significant reduction was observed in a 5-year double-blind placebo-controlled study of the effect of 4-month vitamin D supplementation on the rate of fractures in men and women ⩾65 years of age living in the community in the United Kingdom.20 The observed reduction in the risk of any fracture of 22% is very similar to that obtained in this study, in which concomitant calcium and vitamin D supplements were not a standard part of the protocol. No effect of vitamin D supplementation at a dose of 400 IU of vitamin D daily (without calcium) was observed in the study by Lips et al.19 More recently, two studies of calcium and/or vitamin D have failed to show any significant reduction in fracture risk in elderly persons with recent osteoporotic fractures21 or risk factors for hip fracture.22
Finally, some previous studies have suggested that clinical or hip fracture risk reduction is largely confined to those women whose BMD values met the criteria for osteoporosis.8, 10 This study shows for the first time that the interaction between bisphosphonate efficacy and BMD may not be as marked as first suggested by analyses of studies in which only relatively small numbers of events occurred in patients with BMD lying above the osteoporotic threshold or who were selected on the basis of risk factors largely related to falls risk.8, 10 This suggests that strategies that do not require BMD measurements in all subjects to identify those at highest risk of fracture may be used to target bisphosphonate therapy.23
In conclusion, oral clodronate 800 mg daily can significantly reduce by 20% the incidence of clinical fractures in elderly women unselected for the risk of osteoporosis. A similar magnitude of effect on hip fracture cannot be excluded because of the lower than expected event rate and reduced power of the study. Whether these results would pertain to younger women than those included in the study is not known. This study has shown, however, that the reduction in fracture incidence is largely independent of, and occurs across a wide range of, BMD. However, the use of risk factors, including low BMD, to target therapy to those at highest risk and decrease the number needed to treat seems to be the most appropriate use of clodronate and similar therapies.
Acknowledgements
This study was funded by Schering Oy and the Medical Research Council.
