| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Studies |
Bone and Mineral Division, Henry Ford Hospital (H.G.B.), Detroit, Michigan 48202; the Endocrine Division, Medical College of Virginia (R.W.D.), Richmond, Virginia 23298; the Endocrine Division, Roger Williams Medical Center (J.R.T.), Providence, Rhode Island 02908; Osteoporosis Programs, University of California (S.T.H.), San Francisco, California 94117; Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Sciences (R.S.W.), Little Rock, Arkansas 72205, Osteoporosis/Metabolic Bone Disease Unit, Cleveland Clinic Foundation (A.A.L.), Cleveland, Ohio 44195; Osteoporosis Research Center (M.R.M.), Portland, Oregon 97213; Center for Hard Tissue Research, Creighton University (D.B.K.), Omaha, Nebraska 68131; and Merck Research Laboratories (B.J.G., E.H., W.J.P.), Rahway, New Jersey 07065
Address all correspondence and requests for reprints to: Henry G. Bone, M.D., Bone and Mineral Division K-15, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, Michigan 48202.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
This prospective, randomized, double blind, 2-yr multicenter study compared the effects of placebo with those of 1.0, 2.5, or 5.0 mg ALN daily. All subjects received supplemental calcium (500 mg daily) as the carbonate. We studied 359 women with lumbar spine BMD at least 2.0 SD below the peak young adult mean. Subjects were stratified by age, with 135 aged 60–69 yr and 224 aged 70–85 yr. Histomorphometry was performed on transiliac bone biopsies obtained from 104 subjects after 1 yr and from 83 subjects after 2 yr.
This study elucidated the previously uninvestigated lower region of the dose-response curve for ALN in osteoporosis. Over 2 yr, treatment with 1.0, 2.5, or 5.0 mg/day increased lumbar spine BMD, on the average, by 0.65%, 3.54%, and 5.67%, respectively, compared with that in the placebo group (P < 0.001 vs. placebo for the 2.5 and 5 mg groups). Significant dose-related increases were also seen in BMD at appendicular sites and in total body BMD. Dose-dependent reductions in bone turnover to new steady states were indicated by serum and urine biochemical markers as well as by histomorphometry. There was also a dose-related reduction in the proportion of subjects suffering nonvertebral fractures (P < 0.05). Safety profiles were similar for the ALN and placebo groups and for both age strata. Efficacy was similar for both age strata. There was no evidence of impaired mineralization or other histological abnormalities due to ALN treatment.
We conclude that treatment with ALN over a period of 2 yr was well tolerated and produced dose-dependent increases in BMD without evidence of a plateau over the dose range of 1.0–5.0 mg daily. One milligram daily did not result in a significant effect on BMD, and 5.0 mg daily produced favorable effects at all sites measured. Other studies have demonstrated somewhat greater effects on 10 mg daily. ALN, was equally effective and well tolerated in osteoporotic women over 70 yr old as in younger women with the same condition.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
Women between the ages of 60–85 yr were recruited at 15
clinical sites throughout the United States and were stratified so that
approximately two thirds would be in the 70–85 yr age group. Subjects
were required to be in generally good health apart from osteoporosis. A
screening evaluation of lumbar spine (L1–L4) bone mineral density
(BMD) was performed by dual energy x-ray absorptiometry (DXA). Patients
were accepted for entry if lumbar spine BMD was 0.824 g/cm2
or less by Hologic DXA or 0.944 g/cm2 or less by Lunar DXA.
These densities correspond to 2.0 SD below mean peak
levels. A physical examination and medical history, including dietary
calcium estimation based on a questionnaire, were obtained before
treatment of eligible subjects. Potential subjects were excluded if
they had more than 1 lumbar crush fracture or spinal anatomy was
otherwise unsuitable for DXA analysis. They were also excluded if they
had a history of recent major gastrointestinal disease, such as peptic
ulcer, esophageal disorder, or malabsorption, or had recently used a
drug to inhibit gastric acid secretion for more than 2 weeks. In
addition, patients receiving chronic nonsteroidal antiinflammatory
therapy or agents known to affect bone metabolism (such as etidronate,
estrogen, glucocorticoids, fluoride, or calcitonin) were excluded.
Subjects receiving thyroid hormone replacement were required to have
been on a stable dosage for at least 6 months before entry into the
study and euthyroid by ultrasensitive TSH assay. Clinically significant
vitamin D deficiency was similarly excluded or corrected. As
illustrated in Table 1
, the 4 treatment groups were
similar with respect to demographic, anthropometric and bone turnover
indexes. Baseline vertebral fracture prevalence was similar in the 4
groups. All centers carried out the study with appropriate
institutional review board approval, and all subjects gave informed
consent. Target enrollment was 90 women/treatment group to allow for
possible dropouts. A sample size of 60 evaluable subjects/treatment
group completing the full 2 yr was expected to provide 80% power to
detect a 1.8% between-group difference in the mean percent change from
baseline in lumbar BMD. Of the 359 subjects enrolled, 62.4% were in
the older stratum. Three subjects were later determined not to have
acceptable baseline density measurements. These subjects were followed
for the biochemical and safety measurements.
|
The subjects were randomized equally in a double blind fashion to four treatment groups, which received ALN (1.0, 2.5, or 5.0 mg/day) or placebo. Each patient was instructed to take a calcium supplement containing 500 mg elemental calcium (OsCal-500, Marion Merrell Dow, Kansas City, MO) each evening. Subjects were instructed to take their study medication in the morning, while fasting at least 30 min before breakfast or at least 2 h after breakfast and at least 30 min before the next meal. In practice, the dose was almost always taken before breakfast. Compliance was assessed at each clinic visit through patient reports and tablet counts.
DXA measurements
After the screening posterior-anterior (PA) lumbar spine BMD determination, subjects who met screening criteria and were enrolled underwent additional measurements of lumbar spine density (PA L1–L4 and lateral L3) as well as measurements of proximal femur (femoral neck, greater trochanter, and Ward’s triangle) BMD. Study sites using Hologic DXA equipment also performed determinations of distal forearm and intertrochanteric BMD. Sites with the capability for determining whole body BMD performed these measurements as well. BMD assessments were performed at baseline and after 3, 6, 9, 12, 18, and 24 months of treatment. All sites participated in a longitudinal DXA quality assurance program, which included periodic scanning of anthropomorphic phantoms. Adjustments were made for longitudinal drift in scanner performance by application of scanner-specific correction factors that were generated by the central quality assurance site (Department of Radiology, University of California-San Francisco; H. K. Genant, Director) which was blinded to treatment allocation. In addition, all DXA scan printouts were reviewed by the central site to ensure acceptability. The precision of measurement, expressed as the mean coefficient of variation for the two pretreatment PA lumbar spine densities, was 1.7% and did not differ significantly by densitometer manufacturer or center.
Biochemical measurements
Blood and urine samples were obtained from subjects in the morning, after an overnight fast, at clinic visits every 3–6 months for assay of biochemical markers of bone remodeling and calcium homeostasis. Except as noted, specimens were assayed as received throughout the study by Medical Research Laboratories (Highland Heights, KY). Serum for routine chemistries and 25-hydroxyvitamin D, osteocalcin, and bone-specific alkaline phosphatase determinations, plasma for PTH measurements, as well as aliquots of second voided morning urine specimens were frozen and shipped to the central laboratory. Serum creatinine, calcium, phosphate, and alkaline phosphatase and urinary calcium, creatinine, and phosphate were analyzed by standard methods on a Hitachi 747 analyzer (Hialeah, FL). Serum 25-hydroxyvitamin D was assayed using kits supplied by Nichols Institute (San Juan Capistrano, CA). Serum osteocalcin was measured by RIA using Incstar reagents (Stillwater, MN). Intact PTH was measured by immunoradiometric assay (Nichols Allegro kits). After completion of the study, urinary excretion of the type I collagen cross-links of N-telopeptide (NTx), a specific marker of bone resorption (6), was measured on specimens collected through each subject’s first 12 months of treatment. The Ostex Osteomark (Seattle, WA) enzyme-linked immunosorbent assay was employed.
The urinary deoxypyridinoline (DPyr) assays were performed by Nichols Institute, using high performance liquid chromatography with fluorescence detection (7). Although short term interassay coefficients of variation were less than 10%, there was progressive long term drift in the assay over the course of the study. Therefore, the percent changes for each dose and time point were adjusted based on the change in placebo at the corresponding time point, as described in Data analysis, and the adjusted values are reported.
Fracture ascertainment
Vertebral radiography. Lateral thoracic and lumbar spine radiographs were obtained at baseline and at annual clinic visits and sent to the central evaluation facility at the University of California-San Francisco. The films were reviewed and evaluated for prevalent (present on entry) and incident fractures according to a semiquantitative scoring methodology (8) by an expert radiologist (Harry K. Genant) whose analyses were considered definitive. Only those patients who had baseline and on-treatment x-rays were included in the analysis. For purposes of categorical analysis, vertebrae scored as unfractured or questionably fractured were considered intact, whereas those scored as mild (20–25% decrease in height), moderate (25–40% height loss), or severe (>40% height loss) were considered to represent fractures.
Nonvertebral fractures. Nonvertebral fractures were reported by each study center based on clinical presentation and confirmatory radiographs.
Bone biopsy/histology
Biopsy data were collected primarily to assure that no adverse histological effects were produced by treatment. Study sites were assigned to obtain transiliac bone biopsies at the end of either 1 or 2 yr of treatment. All biopsies were obtained, processed, and evaluated according to a prospectively defined plan that specified the criteria for specimen adequacy, end points for analysis, and hypotheses. It was hypothesized that in comparison with placebo, ALN would not affect mineralization, as indicated by osteoid thickness, or mineral apposition rate. It was further hypothesized that treatment would decrease turnover rate, as indicated by the mineralizing surface measurement, but would not produce marrow fibrosis or other pathological findings.
Subjects consenting to the biopsy procedure took demeclocycline orally as two separate labeling courses (2 days each) separated by a 10-day interval. Transiliac bone biopsy was obtained by 7.5 mm (inside diameter) trephine 4–7 days after the second label administration. The specimen was fixed in 70% ethanol overnight and then shipped to a central laboratory (Bone Histomorphometry Laboratory, Creighton University, Omaha, NE) for histomorphometric analysis (by D.B.K.).
Each specimen was dehydrated in graded ethanols and embedded in modified methyl methacrylate (9). Sections were prepared as previously described (10), and stained by the Goldner (11) or toluidine blue (12) method or left unstained. Sections from specimens were first assessed qualitatively. All specimens were surveyed for the presence of recent double fluorochrome label in both cortical and cancellous bone. Specimens were considered adequate if they had inner and outer cortexes, at least 20 mm2 of trabecular area (if two or more trabecular labeled surfaces were present), or 40 mm2 of trabecular area (if less than two trabecular labeled surfaces were present). When double label in cancellous bone could not be located, sections from additional levels in the specimen were prepared and examined either until trabecular labels were found or the specimen was exhausted.
A light/epifluorescent microscope with a camera lucida projecting onto a graphics pad interfaced to an IBM PCAT computer was used. Bioquant II Software (R&M Biometrics, Nashville, TN) was used to assist in the collection of the raw data used in the calculation of osteoid thickness, mineral apposition rate, and mineralizing surface, which was calculated as the sum of double labeled surface plus one half single labeled surface. The standard nomenclature for bone histomorphometry was applied (13), and customary equations were employed for the calculations (10).
Adverse experiences
Subjects were instructed to report any unwanted signs, symptoms, injuries, illnesses, or other medical events that occurred during the study and were queried about such experiences at each visit. These were classified as adverse experiences regardless of whether they appeared to be related to the investigational treatment. They were graded as to severity (symptomatic effects) and seriousness (potential or actual harm to the patient). Physical examinations (including weight, height, heart rate, and blood pressure) were performed at each clinic visit. For purposes of safety monitoring, routine serum chemistry, hematology, and urinalysis testing was performed at periodic intervals (Medical Research Laboratories), and abnormalities were assessed by the investigators. Electrocardiograms were performed at the beginning and end of the study.
Data analysis
Primary analyses of bone density data were performed on an intention to treat basis, including all subjects with a baseline and at least one follow-up BMD measurement. The actual sample sizes for lumbar spine BMD, using the intention to treat approach, were 90 (placebo), 81 (1.0 mg), 85 (2.5 mg), and 85 (5.0 mg). Per protocol analyses were performed to confirm the findings of the intention to treat analyses. These analyses were limited to those subjects who complied with all requirements and completed the study. There were 56, 50, 60, and 62 evaluable patients for the respective dosages. The per protocol analyses consistently agreed with the intention to treat analyses, and only the latter are reported here, with the single exception noted below.
Biochemical data were analyzed using a per protocol approach in view of the purpose of these measurements and their nature as secondary end points. The stepwise Tukey trend test, adjusted for center, age stratum effects, and multiplicity, was used to examine the dose-response relationships and make comparisons between active doses and placebo for efficacy and clinical safety parameters (14). A linear model (ANOVA) with factors for center, age stratum, and treatment was used to make comparisons among the active doses.
The natural log (fraction of baseline) was used to evaluate the biochemical data; however, the mean changes were back-transformed to represent the percent change from baseline in the tables.
To correct for the assay drift problem described above, changes from baseline in urinary dexoypyridinoline/creatinine ratios were adjusted by subtracting the mean natural log (fraction of baseline) observed in the placebo group from the mean natural log (fraction of baseline) in the ALN groups. The placebo-adjusted means were then back-transformed to represent the mean percent change from baseline. The SE of each placebo-adjusted mean was obtained using the pooled SE from a linear model that accounted for the terms treatment and study center.
Histomorphometric data were analyzed to determine the effect of ALN on bone mineralization, osteoid thickness, and mineral apposition rate. Because they were planned to evaluate safety, these data were tested for statistically significant changes only in the direction of toxicity. All adequate biopsies were included in the determination of osteoid surface and mineralizing surface. Due to the nonnormal distribution of the mineralizing surface data, normalized ranks were used to analyze those data. If a biopsy was adequate for analysis based on available area for inspection, but did not have sufficient double labeling in trabecular bone, the specimen was not included in the sample for calculation of the mineral apposition rate.
The Cochran-Armitage trend test (14) was performed to evaluate adverse experience data, including incident fractures.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
All measured changes in BMD are compiled in Table 2. Changes are expressed as a percentage of baseline due
to the differences between densitometers. Lumbar spine BMD (average of
L1–L4) increased in a dose-dependent fashion over the 2 yr of study
observation (Fig. 1 and Table 2). The increases of
1.21%, 4.10%, and 6.23% from baseline exceeded the change on placebo
by 0.65%, 3.54%, and 5.67% for the 1.0, 2.5, and 5.0 mg groups,
respectively. These changes correspond to 0.005, 0.025, and 0.041
g/cm2 vs. placebo (using the Hologic baseline
values). While these changes were all statistically significant
vs. baseline, only the changes on 2.5 and 5.0 mg were
significantly different from those observed on placebo treatment
(P < 0.001). The increases were not restricted to the
first year; between month 12 and month 24, lumbar BMD increased by
1.83% and 1.80% on 2.5 and 5.0 mg ALN, respectively
(P 
0.001). Lateral L3 density measurements increased
significantly for both the 2.5 mg group (+7.46%; P
0.01) and the 5.0 mg group (+8.96%; P 0.01), but
were not significantly changed for the 1.0 mg (+4.16%) or placebo
(-0.93%) groups.
|
|
. At the femoral
neck, the placebo group’s mean density declined significantly over the
24 months, whereas there was no significant change at that site on 1.0
or 2.5 mg ALN by intent to treat analysis. However, femoral neck
density on 2.5 mg ALN did increase significantly relative to placebo
when all protocol violators and incomplete studies were excluded (per
protocol analysis). The femoral neck density increased by 1.83% for
the 5.0 mg group, producing a difference from placebo of 3.35%;
(P < 0.001). It is noteworthy that for the 5.0 mg
group, the density increased by 1.21% (P = 0.003
vs. placebo) in the second year of the study. ALN (2.5 and
5.0 mg) produced significant, dose-dependent increases in total body
BMD. At the one third distal forearm site, only the 5.0 mg group
differed significantly from placebo at 24 months (P <
0.05), but there was a positive dose-related trend in responses at 24
months (P = 0.03). Thus, increases observed at the
spine and femoral neck did not occur at the expense of losses
elsewhere. There was no site assessed at which BMD decreased on 5.0 mg
ALN treatment. Treatment effects on BMD were consistently similar for
both age strata at the lumbar spine, femoral neck, trochanter, and
total body sites.
|
Biochemical markers were used to measure the effects of ALN on
bone resorption, bone formation, and mineral homeostasis, and the
results are summarized in Tables 3 and 4.
Reduced bone resorption was indicated by a decrease from baseline in
excretion rates of urinary deoxypyridinoline (adjusted as described) in
a dose-dependent manner as shown in Fig. 3. Similarly, a
dose-related reduction in NTx/creatinine ratios was noted over the
first year of treatment (Table 3). As expected, in concert with the
decrease in bone resorption markers, ALN also reduced markers of bone
formation (serum osteocalcin and serum alkaline phosphatase) in a
dose-related, but delayed, manner, reflecting decreased remodeling
activity. The changes in serum alkaline phosphatase over time are
displayed in Fig. 4. Serum alkaline phosphatase levels
declined from baseline for all groups over the 2 yr (P
0.01 for placebo; P 0.001 for all ALN treatment
groups), declining significantly more in the ALN groups than in placebo
recipients. There were no significant differences between age groups in
the effects noted on urinary DPyr/creatinine, serum osteocalcin, or
alkaline phosphatase.
|
|
|
|
As shown in Table 4
, ALN treatment produced a transient, dose-related
reduction in urinary calcium excretion consistent with a reduction in
bone resorption and the dose-related increases in serum PTH, which
peaked at month 1, but remained slightly increased up to month 24. In
contrast, the supplementary calcium probably accounts for the slight
decline in PTH in the placebo group.
Bone histomorphometry
Bone biopsies were obtained from 104 consenting patients at the
end of yr 1 and from 83 patients at the end of yr 2. Of these, 79 and
66, respectively, yielded adequate tissue for quantitative analyses, as
shown in Table 5. ALN (2.5 and 5.0 mg) treatment for 2
yr significantly decreased bone turnover vs. the effect of
placebo (P < 0.001), as indicated by the extent of
mineralizing surface, whereas 1.0 mg did not. Osteoid thickness and
mineral apposition rate were unaffected. There were one, two, and four
adequate specimens without double trabecular labels in the 1, 2.5, and
5 mg treatment groups, respectively. To determine whether this finding
was qualitative in nature, additional sections were reviewed from these
specimens. In every case but one (in the 5.0 mg group), double
tetracycline labeling was visualized within the specimen, indicating
that the bone was metabolically active, albeit at reduced levels, and
the finding of absent labeling in the original section was consistent
with the inherent limitations of sampling for rare labeled surfaces by
the methodologies employed. The proportion of unlabeled specimens was
consistent with previously reported findings for untreated osteoporotic
patients (15). In four additional cases, double labels were present but
were not sufficient for reliable measurement of MAR (Table 5).
Qualitative review showed normal lamellar bone in all specimens without
evidence of fibrosis or cytotoxicity.
|
Only 17 patients (6 in the placebo group; 4, 3, and 4 in the 1.0, 2.5, and 5.0 mg ALN groups, respectively) had new vertebral fractures; there were no statistically significant differences between groups. However, there was a significant dose-related reduction in the proportion of patients with nonvertebral fractures across treatment groups (P = 0.048). There were 16 patients in the placebo group (17.6%) who suffered a nonvertebral fracture, 15 in the 1.0 mg ALN group (17.4%), 9 in the 2.5 mg group (10.1%), and 9 in the 5.0 mg group (9.7%). The corresponding fracture rates were 10.4/100 and 11.3/100 patient yr for the placebo and 1.0 mg ALN groups and 6.1 and 5.9/100 patient yr for the 2.5 and 5.0 mg ALN groups, respectively. The fractures occurred at various sites, mainly in the upper and lower extremities, but also included clavicular, rib, and nasal fractures. The events were too few for analysis by anatomical site to be meaningful.
Safety profile and tolerability
Safety and tolerability were similar in the placebo and ALN treatment groups. Because of reports of upper gastrointestinal side-effects with pamidronate (16), special attention was paid to such reports. We observed no significant trend toward increasing frequency of upper gastrointestinal adverse experiences across treatment groups. The safety profile was also evaluated separately for the two thirds of patients 70 yr or older at entry into the study, and there was no evidence for a different pattern of adverse effects in this subpopulation. Even for those subjects whose adverse experiences were suspected by the on-site investigators of being drug related, there were no significant differences in incidence between treatment and placebo groups. For the placebo and 1, 2.5, and 5 mg ALN groups, the numbers of subjects so classified were 21 (23.1%), 17 (19.8%), 23 (25.8%), and 16 (17.2%), respectively. The numbers of serious adverse experiences suspected of being drug related were similar for all 4 groups. There were no significant differences between treatment groups in the number of patients withdrawn from treatment due to adverse events. (9, 8, 8, and 13 on placebo and 1.0, 2.5, and 5 mg ALN, respectively). Periodic laboratory safety assessments revealed no laboratory adverse experiences associated with ALN treatment.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
The geminal bisphosphonates are synthetic analogs of pyrophosphate, in which the linking oxygen is replaced by a carbon atom, rendering these compounds extremely resistant to hydrolysis in vivo. Bisphosphonates have been demonstrated to be potent inhibitors of bone resorption both in vitro and in vivo in animals and man (29) and have been used in the management of a broad range of skeletal disorders characterized by increased bone resorption. Etidronate, the first compound in this class to be used in clinical practice, was found to inhibit mineralization of new bone (30, 31), limiting the dose and duration of treatment. However, carefully limited cyclic administration of etidronate has been reported to increase BMD (32, 33, 34).
The more recently developed aminobisphosphonates have been found to be potent inhibitors of bone resorption, with no significant adverse effect on mineralization at therapeutic doses (35). ALN is an aminobisphosphonate that has been well characterized in extensive preclinical studies, demonstrating inhibition of bone resorption in a dose-dependent fashion without adverse effects on matrix mineralization (36). ALN treatment has been shown to maintain a normal relationship between the increases in bone strength and mass in rodent experiments (37, 38) and baboons (39). Histological studies have confirmed that normal microscopic morphology is preserved during ALN treatment (36, 39).
The present study sought specifically to evaluate the low end of the dose-response curve and to test the safety and efficacy of ALN in elderly women with osteoporosis. There were no demonstrable differences between patients 70 yr or older and those less than 70 yr old with regard to bone density, biochemical effects, or safety profile, indicating that ALN is beneficial in the general osteoporotic population without differential effects due to age. We found that ALN favorably affects bone mass in a dose-dependent manner, producing increases in density at both vertebral and appendicular sites. As expected, the effects on bone density were greater at sites with large proportions of cancellous bone, which typically have higher rates of turnover, than at predominately cortical sites, which have lower rates of turnover.
The increases in bone mass observed with increasing ALN doses demonstrate a well defined dose-response relationship, which did not reach a plateau over the range tested. Although the 1.0-mg dose did have some detectable biological effects, it did not produce significant differences from placebo in BMD at any site and thus may be considered a "no clinical effect" dose. The increase in lumbar spine BMD during 2 yr of treatment with 5 mg ALN compares favorably with changes reported after treatment with other available or experimental therapies for osteoporosis and is of sufficient magnitude to be clinically significant. Reduction of fracture risk and incidence is the ultimate goal of intervention in patients with osteoporosis, and clear relationships have been demonstrated between BMD and fracture risk in epidemiological studies in postmenopausal women (40, 41, 42). Much larger studies than ours would be required to have adequate power to reliably detect a reduction in vertebral fractures. The clinical significance of the effect on the appendicular skeleton is highlighted by the finding that 2 yr of treatment with ALN was associated with a significant dose-related reduction in the proportion of subjects suffering nonvertebral fractures. These findings are encouraging because they confirm that the gains in BMD produced by ALN treatment are associated with improved bone strength. The weakness of the relationship between biochemical changes and changes in BMD was not entirely expected and may reflect the variability in the measurements employed.
In earlier studies in which the drug was administered 2 h before breakfast, the effects of 5 and 10 mg were similar, and little additional effect was seen at higher doses (3, 4). Therefore, 5 mg was selected as the highest dose for the present study. However, in this study and other phase III studies (5, 43), ALN was administered 30 or 60 min before breakfast for practical reasons of patient convenience and compliance. Recently, it has been shown that when ALN is taken either 30 or 60 min before breakfast, its absorption is reduced by about 40% compared to doses taken 2 h before the meal (44). This information and the lack of a dose-response plateau in the present study suggest that the maximal effect of ALN would be achieved at a higher dosage when the drug is taken 30–60 min before breakfast. In fact, in concurrent studies in which 5 and 10 mg and higher daily dosages were evaluated under such circumstances, the effects of 5 mg were similar to those we observed, and a maximal effect was seen at the 10-mg dose level (5, 43). The 10-mg daily dose, to be taken at least 30 min before breakfast, has been approved by the US FDA for the treatment of postmenopausal osteoporosis.
The observed effects of ALN on BMD are consistent with its recognized antiresorptive mechanism of action. The rapid increase in BMD in the first 3–6 months is most likely due to a reduction of the remodeling space (26, 27, 28). Bone resorption decreases rapidly with the onset of therapy, but bone formation is not directly inhibited by ALN. Thus, for 3–6 months, formation greatly exceeds resorption, allowing previously initiated resorption cavities to be filled in. Eventually, formation decreases as a result of a decrease in the frequency of activation of new remodeling sites. The continued gains in BMD with ALN treatment over the second year may not be fully explained by the remodeling transient effect, leaving open the possibility of an additional favorable effect on focal remodeling balance.
In bone biopsies obtained after 1 or 2 yr of treatment, the data on osteoid thickness and mineral apposition rate indicate that continuous daily treatment with ALN impairs neither the rate of osteoid mineralization nor individual osteoblast function. Mineralizing surface measurements, which are indicative of the surface area undergoing remodeling, were decreased significantly by 2.5 and 5.0 mg ALN daily, demonstrating ALN’s ability to slow osteoclast recruitment and activity. Mineralizing surface did not differ between biopsies obtained from one group after 1 yr and those obtained from the other group after 2 yr of treatment, indicating that a new and persistent steady state of remodeling was established by the end of the first year. Thus, histological as well as biochemical data indicate that the decreases in turnover were maintained, but were not cumulative or progressive. This is consistent with the concept that once new bone formation occurs over an ALN-treated surface, the ALN deposited there is effectively sequestered. It may well be that the time to maximal effect is mainly related to remodeling kinetics, while the magnitude of the effect may also reflect the amount of drug deposited at each remodeling site.
The results from the biochemical analyses confirm that dose-related reductions in bone remodeling rates are observed with ALN therapy and that these altered rates appear to stabilize after about 1 yr of treatment. Biochemical indexes of mineral homeostasis (calcium, phosphate, and PTH) revealed the expected early transient rise in serum PTH, which was greatest with the 5.0-mg ALN dose. Presumably, the antiresorptive effect of the drug results in a tendency for serum ionized calcium to decline slightly, thereby stimulating PTH secretion. The effect would be expected to be maximal during the period of rapid net gain in total bone calcium and to be mitigated as a new equilibrium is approached. Eventually, PTH stabilized at a level slightly higher than baseline, which is not surprising, as ALN would be expected to render the skeleton slightly resistant to the stimulation of bone resorption by PTH. Potential benefits of mild increases in PTH on bone strength and connectivity have been described, and this mechanism may contribute to the beneficial effects of ALN (45, 46).
Bone loss may be retarded by calcium supplementation in late postmenopausal women (47, 48, 49). It is likely that the addition of a calcium supplement contributed to the slowing of the rate of bone loss in our study, as demonstrated by the relatively stable bone density in the placebo group over the study period as well as by evidence of some reduction of biochemical indexes of remodeling in the placebo group. In the setting of effective therapy such as we have described, the gain in bone mass leads to increased skeletal Ca requirements. Therefore, Ca is necessary to modulate the effects on mineral homeostasis and PTH described above. For all of these reasons, adequate Ca and vitamin D must be assured.
The tolerability of ALN in this elderly population was generally excellent. The rates of adverse event reporting and withdrawals from the study due to adverse events were similar across all treatment groups and were not different from those in the placebo group. Furthermore, the tolerability and safety profile did not differ by age stratum. In older women especially, ALN may offer significant tolerability advantages over alternative therapies.
Conclusion
ALN in daily oral doses of 1–5 mg was well tolerated in the population studied. Over the 2 yr of the study, ALN produced dose-dependent decreases in bone turnover and increases in mineral density in the lumbar spine, proximal femur, and skeleton as a whole. The dose-response relationship was similar for the two age strata. Although there were some indications of a biological effect at the lowest dose tested, from a clinical standpoint, 1.0 mg daily can be regarded as an effective dosage. The minimum dose demonstrating a significant increase in bone density at the femoral neck and forearm sites as well as the lumbar spine was 5.0 mg. The magnitude of effect we observed at 5.0 mg was similar to that seen at the same dosage in parallel studies (5, 43), in which the maximal increase in BMD was achieved at 10 mg daily. Women in the older stratum did not differ from those in the younger group with regard to efficacy, tolerability, or safety profile. The findings of increased bone density and normal bone quality together with a favorable effect on fracture incidence indicate that ALN is a useful new drug for the treatment of osteoporosis in older women.
|
Footnotes |
|---|
2 Additional investigators are listed in the appendix.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
Received April 23, 1996.
Revised August 9, 1996.
Accepted August 19, 1996.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionAppendix 1References |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. G. G. RUSSELL, Z. XIA, J. E. DUNFORD, U. OPPERMANN, A. KWAASI, P. A. HULLEY, K. L. KAVANAGH, J. T. TRIFFITT, M. W. LUNDY, R. J. PHIPPS, et al. Bisphosphonates: An Update on Mechanisms of Action and How These Relate to Clinical Efficacy Ann. N.Y. Acad. Sci., November 1, 2007; 1117(1): 209 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D.B. Kimmel Mechanism of Action, Pharmacokinetic and Pharmacodynamic Profile, and Clinical Applications of Nitrogen-containing Bisphosphonates J. Dent. Res., November 1, 2007; 86(11): 1022 - 1033. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. G. Bone, D. Hosking, J.-P. Devogelaer, J. R. Tucci, R. D. Emkey, R. P. Tonino, J. A. Rodriguez-Portales, R. W. Downs, J. Gupta, A. C. Santora, et al. Ten Years' Experience with Alendronate for Osteoporosis in Postmenopausal Women N. Engl. J. Med., March 18, 2004; 350(12): 1189 - 1199. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.M. Francis, S.P. Baillie, A.J. Chuck, P.R. Crook, N. Dunn, J.N. Fordham, C. Kelly, and A. Rodgers Acute and long-term management of patients with vertebral fractures QJM, February 1, 2004; 97(2): 63 - 74. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. M. Kiebzak, G. A. Beinart, K. Perser, C. G. Ambrose, S. J. Siff, and M. H. Heggeness Undertreatment of Osteoporosis in Men With Hip Fracture Arch Intern Med, October 28, 2002; 162(19): 2217 - 2222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. D. Nelson, M. Helfand, S. H. Woolf, and J. D. Allan Screening for Postmenopausal Osteoporosis: A Review of the Evidence for the U.S. Preventive Services Task Force Ann Intern Med, September 17, 2002; 137(6): 529 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cranney, G. Wells, A. Willan, L. Griffith, N. Zytaruk, V. Robinson, D. Black, J. Adachi, B. Shea, P. Tugwell, et al. II. Meta-Analysis of Alendronate for the Treatment of Postmenopausal Women Endocr. Rev., August 1, 2002; 23(4): 508 - 516. [Full Text] [PDF]
|
|
|
|
 |
|
 S. Palomba, F. Orio Jr., A. Colao, C. di Carlo, T. Sena, G. Lombardi, F. Zullo, and P. Mastrantonio Effect of Estrogen Replacement Plus Low-Dose Alendronate Treatment on Bone Density in Surgically Postmenopausal Women with Osteoporosis J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1502 - 1508. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Hochberg, S. Greenspan, R. D. Wasnich, P. Miller, D. E. Thompson, and P. D. Ross Changes in Bone Density and Turnover Explain the Reductions in Incidence of Nonvertebral Fractures that Occur during Treatment with Antiresorptive Agents J. Clin. Endocrinol. Metab., April 1, 2002; 87(4): 1586 - 1592. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Eastell, D.M. Reid, J. Compston, C. Cooper, I. Fogelman, R.M. Francis, S.M. Hay, D.J. Hosking, D.W. Purdie, S.H. Ralston, et al. Secondary prevention of osteoporosis: when should a non-vertebral fracture be a trigger for action? QJM, November 1, 2001; 94(11): 575 - 597. [Abstract] [Full Text] [PDF]
|
|
, P < 0.05
vs. baseline; 
, P 
, P 
