3meun.oa

The new england journal of medicine Pierre J. Meunier, M.D., Christian Roux, M.D., Ph.D., Ego Seeman, M.D., Sergio Ortolani, M.D., Janusz E. Badurski, M.D., Tim D. Spector, M.D., Jorge Cannata, M.D., Adam Balogh, M.D., Ernst-Martin Lemmel, M.D., Stig Pors-Nielsen, M.D., René Rizzoli, M.D., Harry K. Genant, M.D., b a c k g r o u n d
Osteoporotic structural damage and bone fragility result from reduced bone formation From the Department of Rheumatology and and increased bone resorption. In a phase 2 clinical trial, strontium ranelate, an orally Lyons, France (P.J.M.); the Department of active drug that dissociates bone remodeling by increasing bone formation and decreas- Rheumatology, Cochin Hospital, René Des-ing bone resorption, has been shown to reduce the risk of vertebral fractures and to in- cartes University, Paris (C.R.); the Endocrine Unit, Austin Hospital, University of Mel- bourne, Melbourne, Australia (E.S.); theCenter for Metabolic Bone Disease, Istituto Auxologico Italiano, Milan, Italy (S.O.); theCenter for Osteoporosis and Osteo-Artic- To evaluate the efficacy of strontium ranelate in preventing vertebral fractures in a phase ular Diseases, Bialystok, Poland (J.E.B.); the 3 trial, we randomly assigned 1649 postmenopausal women with osteoporosis (low Department of Rheumatology, St. Thomas’bone mineral density) and at least one vertebral fracture to receive 2 g of oral strontium Hospital, London (T.D.S.); the Mineral and ranelate per day or placebo for three years. We gave calcium and vitamin D supplements turias, Oviedo, Spain (J.C.); the Department to both groups before and during the study. Vertebral radiographs were obtained annu- of Obstetrics and Gynecology, Universityally, and measurements of bone mineral density were performed every six months.
of Debrecen, Medical and Health SciencesCenter, Debrecen, Hungary (A.B.); MaxGrundig Klinik, Innere Medizin–Rheuma- tology, Bühl, Germany (E.-M.L.); the De- New vertebral fractures occurred in fewer patients in the strontium ranelate group than partment of Clinical Physiology, Hillerød Hospital, Hillerød, Denmark (S.P.-N.); the in the placebo group, with a risk reduction of 49 percent in the first year of treatment Division of Bone Diseases, Department and 41 percent during the three-year study period (relative risk, 0.59; 95 percent confi- of Internal Medicine, University Hospital,dence interval, 0.48 to 0.73). Strontium ranelate increased bone mineral density at month Geneva (R.R.); the Osteoporosis and Arthri- tis Group, University of California, San Fran- 36 by 14.4 percent at the lumbar spine and 8.3 percent at the femoral neck (P<0.001 for cisco (H.K.G.); and the World Health Or- both comparisons). There were no significant differences between the groups in the ganization Collaborating Center for Publicincidence of serious adverse events.
Health Aspects of Osteoarticular Disorders,Department of Epidemiology and PublicHealth, University of Liège, Liège, Belgium c o n c l u s i o n s
(J.Y.R.). Address reprint requests to Dr. Meu- Treatment of postmenopausal osteoporosis with strontium ranelate leads to early and nier at the Department of Rheumatology and Bone Disease, Pavillon F, Edouard Her- sustained reductions in the risk of vertebral fractures.
riot Hospital, 69437 Lyons CEDEX 03, France,or at [email protected].
*Investigators participating in the study N Engl J Med 2004;350:459-68.
Copyright 2004 Massachusetts Medical Society. Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine sequence of osteoporosis, lead to acute and chronic back pain, spinal deformity, and study subjects
a reduction in survival equivalent to that caused by We recruited postmenopausal women from Novem-hip fractures.1 The health care burden in financial ber 1996 through July 1998 at 72 centers in 11 Euro-terms is substantial.2 Vertebral deformities predict pean countries and Australia for this prospective,further vertebral fracture, even within one year, and randomized, double-blind, placebo-controlled trial.
also predict nonvertebral fractures.1,3,4 Women were eligible for the study if they were at The bone fragility that characterizes osteoporo- least 50 years old, had been postmenopausal for at sis after menopause results both from an imbalance least five years, had had at least one fracture con-in bone remodeling at the cellular level, which caus- firmed by spinal radiography (after minimal trau-es bone resorption to exceed bone formation, and ma), and had a lumbar-spine bone mineral densityfrom an increase in the rate of remodeling at the of 0.840 g per square centimeter or less (measuredtissue level.5 Antiresorptive therapies reduce the rate with Hologic instruments). Women were ineligibleof bone remodeling and lower the fracture rate by if they had severe diseases or conditions that could30 to 50 percent. Antiresorptive agents, however, do interfere with bone metabolism or if they used anti-not increase bone tissue mass. Instead, the increase osteoporotic treatments (fluoride salts and bis-in bone mineral density observed in clinical trials phosphonates taken for more than 14 days withinof antiresorptive drugs is the result of a more com- the previous 12 months, or estrogen, calcitonin, orplete secondary mineralization of the existing (but calcitriol taken for more than 1 month in the previ-reduced) bone tissue mass.6-10 Restoration of bone ous 6 months). All participants provided written in-tissue mass and bone structure is not achieved with formed consent before enrollment in the study; theantiresorptive drugs and requires the use of anabol- study was approved by the institutional review boardic agents.11,12 Strontium ranelate is a new orally active agent consisting of two atoms of stable strontium and an treatment protocol and follow-up
organic moiety (ranelic acid). It stimulates the for- Throughout the study, subjects received daily cal-
mation of new bone tissue and decreases bone re- cium supplements at lunchtime (up to 1000 mg of
sorption, as has been shown in vitro and in experi- elemental calcium, depending on their dietary cal-
ments in animals.13-16 Strontium ranelate prevents cium intake), to maintain a daily calcium intake
bone loss in ovariectomized rats, increases bone above 1500 mg, and vitamin D (400 to 800 IU, de-
mass in osteopenic animals, and increases bone pending on the base-line serum concentration of
strength in normal animals.13,17,18 To date, no del- 25-hydroxyvitamin D). After a run-in period of 2 to
eterious effects on the primary or secondary miner- 24 weeks, depending on the severity of the defi-
alization of bone in laboratory animals18,19 or ciency of calcium and vitamin D, the subjects were
humans20 have been reported. Results from a two- randomly assigned to receive 2 g a day of strontium
year placebo-controlled, phase 2, dose–response ranelate (two packets a day of a powder that they
study involving 353 postmenopausal women with mixed with water) or placebo powder for 3 years.
osteoporosis suggested that ingestion of 2 g a day Subjects were instructed to take the study drug once
of oral strontium ranelate reduced the incidence of daily, at bedtime, or twice daily (one packet 30 min-
vertebral fractures during the second year of treat- utes before breakfast, and one at bedtime). Most
ment and simultaneously increased bone mineral subjects (87 percent) chose the once-daily regimen.
density.20 In order to confirm these results, we de-
Three lateral radiographs of the spine (thoracic signed the Spinal Osteoporosis Therapeutic Inter- and lumbar radiographs and an image of the thora-vention study to test the safety of strontium rane- columbar junction) were obtained at base line andlate and its efficacy against vertebral fracture when annually, according to standardized procedures, andgiven orally at a dose of 2 g per day for three years in if there were indications of symptomatic vertebralpostmenopausal women with osteoporosis and a fracture (acute back pain, a decrease in body heighthistory of vertebral fracture.
of at least 1 cm, or both). At base line, anteroposte- Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. s t r o n t i u m r a n e l a t e a n d v e r t e b r a l f r a c t u r e i n p o s t m e n o p a u s a l o s t e o p o r o s i s rior radiographs of the spine were also obtained. centrations of bone-specific alkaline phosphataseAll radiographs were assessed at a central facility; (a marker of bone formation) were measured withradiologists were told the time sequence of each ra- an immunoradiometric assay (Tandem-R Ostase,diograph but were unaware of the treatment as- Hybritech), serum concentrations of C-telopep-signment.
tide cross-links (a marker of bone resorption) with Two methods of assessing vertebral fracture an enzyme-linked immunosorbent assay (Serum were used. First, a semiquantitative visual assess- CrossLaps, Osteometer Biotech), parathyroid hor-ment of each vertebra, from T4 to L4, was performed mone with an immunoradiometric assay (N-tact,by the same reader throughout the study. The semi- DiaSorin), 25-hydroxyvitamin D with a radioimmu-quantitative grading scale was as follows: grade 0, noassay (DiaSorin), and calcitonin with an immu-normal; grade 1, a decrease in the height of any ver- noradiometric assay (Biosource). Also measuredtebra of 20 to 25 percent; grade 2, a decrease of 25 was 1,25-dihydroxyvitamin D, by means of a radi-to 40 percent; and grade 3, a decrease of 40 percentor more.21,22 For primary analysis, a new fracturewas defined by a change in the score of a vertebrafrom grade 0 at base line to a subsequent grade of 1 or more. Second, quantitative assessment was also performed: anterior, middle, and posterior verte-bral heights were measured for each vertebra, fromT4 to L4. A new fracture was defined by a decrease in height of at least 15 percent (or 3 mm) on a ver-tebra graded 0 at base line and with a grade on thesemiquantitative scale of 1, 2, or 3.23 Nonvertebral fractures were confirmed by a ra- diologic evaluation or by a report from a hospital-ization. (The study did not have sufficient power foradequate statistical comparison of the two groups.) Skull, face, finger, toe, and coccygeal fractures were not considered to be osteoporotic fractures. Stand-ing height was measured at base line and every six months with a Harpenden stadiometer, according Bone mineral density at the lumbar spine and proximal femur was measured by dual-energy x-rayabsorptiometry at base line and at six-month inter-vals (Hologic). All the scans were analyzed centrally.
A quality-control program was conducted through-out the study.24 Lumbar-spine bone mineral densi-ty, adjusted for the strontium content of the bone, was calculated at month 36 according to the fol- lowing formula20: adjustment factor = 1÷[1+(esti-mated bone strontium content ¬ 0.61)], with bonestrontium content defined as the value measured inbone-biopsy samples obtained in some subjects atmonth 36.
Blood and urine samples were collected at base Completed follow-up at 3 yr, 1260 patients line, three months, and six months, and then ev- ery six months; the samples were stored at ¡80°Cand analyzed centrally. Biochemical tests were per-formed by Bio Analytical Research Corporation Figure 1. Study Populations.
(BARC) according to standard methods. Serum con- Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine oreceptor assay (DiaSorin). The strontium con- pare percent changes from base line in independenttent of serum and bone was measured with induc- samples, the Pearson chi-square test was used totively coupled plasma-emission spectrophotometry compare the numbers of patients with at least two(BARC), and the results were released only after the new vertebral fractures and the numbers of patientsrandomization code had been broken.
with a specific adverse event, and 95 percent confi- Biopsies of transiliac bone were carried out after dence intervals were determined for the differences tetracycline double labeling in 20 consenting pa- between the groups with respect to mean changestients to assess bone strontium content and safety- in serum calcium, phosphorus, and parathyroid hor-related histomorphometric variables.
mone levels. For percent changes from the base-linevalue in bone mineral density at each subsequent s t a t i s t i c a l a n a l y s i s
visit, a step-down hierarchical procedure was per- The main efficacy analysis was performed on an in- formed on the basis of the increasing treatment ef-tention-to-treat basis and included patients who fect over time. The mean values in the two groupsunderwent randomization, who had taken at least at each visit were compared with the use of one-sid-one packet of treatment, and for whom at least one ed Student’s t-tests with a type I error rate of 2.5 per-spinal radiograph was obtained after base line. Kap- cent. The P values presented correspond to a two-lan–Meier product-limit estimates of the incidence sided test at the 5 percent threshold. (One-sidedof new vertebral fractures were calculated at the time P values were doubled.)each year when radiography was performed. An un- The data and assessments collected in this study adjusted Cox model served as the main statistical were held by Servier, and statistical analyses wereanalysis for the comparison of groups and the esti- performed by Servier. Data concerning bone min-mation of relative risks and 95 percent confidence eral density, biochemical markers and other bio-intervals.
chemical variables, and the evaluation of spinal ra- Two-sided Student’s t-tests were used to com- diographs were collected centrally by independent investigators and then transferred to Servier for sta-tistical analysis. The authors had access to all the Table 1. Base-Line Characteristics of the 1442 Patients in the Intention-to-Treat
data and take responsibility for the veracity of the Population.*
Strontium
This study was coordinated and organized un- Ranelate
der the control of an independent advisory com- Characteristic
mittee, whose members were not directly involved in the study, and the international coordinator (Dr.
Meunier), who monitored the scientific quality of the studies, patient compliance and adherence to the protocol, results, and conclusions. A steering committee planned the study, its conduct, and the statistical analysis, and it oversaw any scientific or technical issues. The members of an adverse-events committee were independent of the sponsor and of s t u d y s u b j e c t s
Of 1649 women who underwent randomization, Serum bone-specific alkaline phosphatase 87.4 percent (1442 women) made up the popula- Serum C-telopeptide cross-links (pmol/liter) tion for the intention-to-treat analysis (Fig. 1). Thebase-line characteristics of the two groups were * Plus–minus values are means ±SD. There were no significant differences be- similar both in the intention-to-treat population (Table 1) and among all patients randomly assigned † The body-mass index is the weight in kilograms divided by the square of the to treatment groups (data not shown). In the inten- tion-to-treat population, 87.4 percent of the place- Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. s t r o n t i u m r a n e l a t e a n d v e r t e b r a l f r a c t u r e i n p o s t m e n o p a u s a l o s t e o p o r o s i s bo group and 87.3 percent of the strontium ranelate period (relative risk, 0.90; 95 percent confidencegroup completed three years of follow-up.
v e r t e b r a l f r a c t u r e s , b o d y h e i g h t ,
b o n e m i n e r a l d e n s i t y , s e r u m s t r o n t i u m
a n d n o n v e r t e b r a l f r a c t u r e s
l e v e l s , a n d m a r k e r s o f b o n e t u r n o v e r
At the end of the first year of treatment, there was a Bone mineral density was similar at base line in the49 percent lower risk of a new vertebral fracture in two groups and increased continuously at the spine,the strontium ranelate group than in the placebo femoral neck, and total hip in the strontium rane-group (incidence, 6.4 percent vs. 12.2 percent; rela- late group over the three-year period, with no trendtive risk, 0.51; 95 percent confidence interval, 0.36 toward a plateau (Fig. 3). Over a period of threeto 0.74; P<0.001), and a 52 percent lower risk of years, bone mineral density in the strontium rane-symptomatic fracture (3.1 percent vs. 6.4 percent; late group had increased from base line by 12.7 per-relative risk, 0.48; 95 percent confidence interval, cent at the lumbar spine, 7.2 percent at the femoral0.29 to 0.80; P=0.003). Over the entire three-year neck, and 8.6 percent at the total hip (P<0.001 forstudy period, the strontium ranelate group had a 41 all three comparisons with base-line values), corre-percent lower risk of a new vertebral fracture than sponding to differences between the placebo andthe placebo group (20.9 percent vs. 32.8 percent; the treatment groups at three years of 14.4 percent,relative risk, 0.59; 95 percent confidence interval, 8.3 percent, and 9.8 percent, respectively. At three0.48 to 0.73; P<0.001) (Fig. 2). On the basis of these years, the bone mineral density at the lumbar spine,data, 9 patients would need to be treated for three adjusted for the strontium content, showed an in-years with strontium ranelate in order to prevent crease over the base-line value of 6.8 percent in the1 patient from having a vertebral fracture (95 per-cent confidence interval, 6 to 14).
Quantitative assessment confirmed by the semi- quantitative evaluation of vertebral fractures showed that 17.7 percent of patients receiving strontium ranelate for three years and 28.4 percent of patientsreceiving placebo had one new vertebral fracture that met the study criteria (relative risk in the stron- tium ranelate group, 0.58; 95 percent confidenceinterval, 0.46 to 0.73; P<0.001). The proportion of patients with more than one new vertebral fracture over the three-year period was 6.4 percent in the strontium ranelate group and 9.8 percent in the pla- Probability of Vertebral Fracture (%)
cebo group (relative risk, 0.64; 95 percent confi- dence interval, 0.44 to 0.93; P=0.02). Symptomatic vertebral fractures were detected in 192 patients(11.3 percent of the strontium ranelate group and Figure 2. Proportion of Patients in the Intention-to-Treat
17.4 percent of the placebo group), corresponding Population Who Had One or More New Vertebral Frac-
to a 38 percent lower risk over a period of three years tures, Assessed According to the Semiquantitative
in the strontium ranelate group than in the control group (relative risk, 0.62; 95 percent confidence in- Analysis at month 12 was restricted to patients with as-sessable radiographs at base line and at month 12 (686 patients in the strontium ranelate group and 699 in the Over the three-year treatment period, fewer pa- placebo group). The relative risk of fracture in the treat- tients had height loss of at least 1 cm in the stron- ment group at 12 months was 0.51 (95 percent confidence tium ranelate group (30.1 percent) than in the pla- interval, 0.36 to 0.75; P<0.001). Analysis for the three-year cebo group (37.5 percent, P=0.003). Back pain was period was restricted to patients with assessable radio-graphs at base line and after base line (719 patients in reported by 17.7 percent of the women in the stron- the strontium ranelate group and 723 in the placebo tium ranelate group and by 21.3 percent in the pla- group). The relative risk of fracture over 36 months cebo group (P=0.07). Nonvertebral fractures oc- was 0.59 (95 percent confidence interval, 0.48 to 0.73; curred in 234 women (112 in the strontium ranelate group and 122 in the placebo group) over the study Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine µmol per liter and then reached a plateau. At the Lumbar Spine
third month of therapy, the serum concentration of bone-specific alkaline phosphatase was higher inthe strontium ranelate group than in the placebo group (a treatment-related increase of 8.1 percent, P<0.001), and this difference persisted at each eval- uation during the three years (Fig. 4A and 4C). The concentration of serum C-telopeptide cross-links was lower in the strontium ranelate group than in the placebo group at month 3 (a treatment-related difference of 12.2 percent, P<0.001) and at each sub-sequent evaluation during the three years (P<0.001)(Fig. 4B and 4C).
Femoral Neck
h i s t o m o r p h o m e t r i c f e a t u r e s o f b o n e
Bone biopsies were performed at 24, 36, or 48 months in 20 consenting patients, resulting in 14 samples that could be assessed. All biopsy speci-mens consisted of lamellar bone. None of the biop- sies showed evidence of osteomalacia or any sign of a primary mineralization defect. As compared withthe placebo group, the strontium ranelate group had no increase in osteoid thickness (P=0.64) or in the mineralization lag time (P=0.76) and no decreasein the mineral apposition rate (P=0.93).
Mean (±SD) Percent Change in Bone Mineral Density
Total Hip
a d v e r s e e v e n t s a n d m e t a b o l i c r e s u l t s
During the three years of the study, the rate of com- pliance with therapy was 83 percent in the stron- tium ranelate group and 85 percent in the placebo group. The rates of adverse events, serious adverse events, and withdrawal due to adverse events were similar in the two groups. Diarrhea was the most frequent gastrointestinal adverse event (occurring in 6.1 percent of the strontium ranelate group and 3.6 percent of the placebo group, P=0.02). How- ever, this effect disappeared after the first threemonths. There was a lower incidence of gastritis, Figure 3. Effects of Strontium Ranelate on Bone Mineral Density in All Pa-
as diagnosed clinically by the investigators, in the tients Receiving 2 g a Day of Oral Strontium Ranelate.
strontium ranelate group than in the placebo group P<0.001 for all comparisons, with the use of a step-down hierarchical procedure.
(3.6 percent vs. 5.5 percent, P=0.07). Serum calci-um concentrations were lower and serum phos-phate concentrations higher in the strontium rane- strontium ranelate group and a decrease of 1.3 per- late group than in the control group at month 3, withcent in the placebo group (P<0.001); these changes a plateau thereafter (95 percent confidence intervalcorrespond to a treatment-related increase of 8.1 for the difference between the groups, ¡0.32 to ¡0.20percent.
mg per deciliter [¡0.08 to ¡0.05 mmol per liter] for At base line, the median serum strontium con- calcium and 0.25 to 0.37 mg per deciliter [0.08 to centration was 0.3 µmol per liter in both groups. At 0.12 mmol per liter] for phosphorus). There was athree months, the median serum strontium con- slight reduction in serum parathyroid hormonecentration in the treated group had risen to 117.9 from the level at month 6 in both groups (decrease Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. s t r o n t i u m r a n e l a t e a n d v e r t e b r a l f r a c t u r e i n p o s t m e n o p a u s a l o s t e o p o r o s i s A Bone-Specific Alkaline Phosphatase
B C-Telopeptide Cross-Links
Change from Base Line (ng/ml)
Change from Base Line (pmol/liter)
C Difference over Time between Groups
Bone-Specific Alkaline Phosphatase
(pmol/liter)
C-Telopeptide Cross-Links
Figure 4. Strontium Ranelate–Induced Changes in Serum Biochemical Markers of Bone Metabolism.
Panel A shows absolute changes from base-line values in bone-specific alkaline phosphatase, Panel B shows absolute changes from base-line values in C-telopeptide cross-links, and Panel C shows differences over time in biochemical markers between the two groups. Data shown are mean (±SE) values in the strontium ranelate group minus mean val-ues in the placebo group. Comparisons were performed with analyses of covariance in which base-line values were used as covariates.
Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. The new england journal of medicine from base-line to six-month values, ¡3.4±10.7 pg bone formation and bone resorption.26 When para-per milliliter in the strontium ranelate group and thyroid hormone and alendronate are combined,¡1.3±15.8 pg per milliliter in the placebo group). there is, unexpectedly, no potentiation of their ef-No changes were observed for serum 25-hydroxyvi- fects on biochemical bone markers.27 The mecha-tamin D, 1,25-dihydroxyvitamin D, or calcitonin.
nism of action of strontium ranelate is probably dif- Serum creatine kinase concentrations in the ferent from those of these drugs. Each time the skeletal muscle increased in some patients; with a patients were evaluated during our study, bone for-high concentration defined as a value that was at mation had increased in the group assigned to stron-least twice the upper limit of the normal range (145 tium ranelate, on the basis of serum concentrationsIU per liter), high levels were detected in 3.4 per- of bone-specific alkaline phosphatase, and bone re-cent of patients in the strontium ranelate group at sorption had decreased, on the basis of serum con-least once during the study and in 1.8 percent of centrations of C-telopeptide cross-links, as com-patients in the placebo group. No increase in mus- pared with the values in the placebo group. Thecular symptoms or other biologic abnormalities changes in biochemical markers of bone resorptionwere observed. Most of the increases in creatine and formation were most pronounced during thephosphokinase were transient, and in more than first six months; the dissociation between the bone88 percent of the patients who had high concentra- markers was evident throughout the study. Thetions, the values returned to the normal range dur- mechanisms for the apparent dissociation betweening treatment.
reduced bone resorption and increased bone forma-tion are not yet understood, but they probably dif-fer from the mechanisms of current treatments.6-9 Any metal with an atomic number higher than Prevention of fractures is the primary aim of anti- that of calcium can be expected to influence boneosteoporotic treatment. In the present study, stron- mineral density.28 In experiments, strontium intium ranelate ingested daily reduced the risk of new bone correlated with strontium in plasma.29 In thisvertebral fractures by 49 percent at one year and by study, strontium ranelate increased bone mineral41 percent over a three-year period among post- density at three years by 14.4 percent at the lumbarmenopausal women with osteoporosis. Although spine, by 8.3 percent at the femoral neck, and bydata from direct comparisons with other antios- 9.8 percent at the total hip, and bone strength isteoporotic treatments are lacking, the reduction in directly proportional to bone mineral density. Al-the risk of vertebral fracture seems similar to the though data from directly comparable trials are notreduction reported with alendronate (47 percent),6 available, the treatment effect after adjustment for5 mg of risedronate (49 percent),7 60 mg of ralox- the strontium content of bone was an 8.1 percentifene (30 percent),8 and parathyroid hormone (65 increase in bone mineral density at the lumbar spinepercent after 21 months of treatment).12 The meth- at 3 years, which compares favorably with a 9 per-ods of assessing vertebral fractures, involving both cent increase with parathyroid hormone (20 µg) atsemiquantitative and morphometric evaluations, 21 months and with the increases with other drugswere similar to the methods used in these other re- (5.9 percent with risedronate, 6.2 percent with alen-ports.6-8,12 Strontium ranelate also reduced the risk dronate, and 2.6 percent with raloxifene). Moreover,of multiple vertebral fractures and the risk of symp- this increase is consistent with the results of pre-tomatic fractures. In this group of women, who had vious phase 2 studies involving strontium rane-established osteoporosis and a history of fractures, late.20,30 In summary, strontium ranelate given oral-the number needed to treat to prevent one fracture ly at a dose of 2 g daily appears to reduce the riskwas nine.
of vertebral fractures rapidly, effectively, and safely Most antiresorptive agents prevent bone destruc- among postmenopausal women with osteoporosis.
tion by reducing the rate of bone remodeling, as reflected by a decrease in both markers of bone re- Drs. Balogh, Meunier, Nielsen, Ortolani, Reginster, Roux, See- man, and Spector report having received consulting fees from Servi- sorption (more than 50 percent with bisphospho- er; Drs. Meunier and Roux consulting fees from Eli Lilly; Dr. Baloghnates and about 30 percent with raloxifene) and lecture fees from Lilly Hungaria; Dr. Genant consulting and lecturemarkers of bone formation (about 50 percent with fees from Servier, Wyeth Laboratories, Eli Lilly, GlaxoSmithKline, Novartis, Pfizer, and Procter & Gamble and grant support from Pfi- bisphosphonates and 20 percent with raloxifene).25 zer, Procter & Gamble, and Eli Lilly; Dr. Ortolani consulting feesTreatment with parathyroid hormone increases both from Roche and Janssen, lecture fees from Italfarmaco, and grant Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. s t r o n t i u m r a n e l a t e a n d v e r t e b r a l f r a c t u r e i n p o s t m e n o p a u s a l o s t e o p o r o s i s support from Servier, Novartis, and TEVA Pharmaceutical Indus- Merck, Teijin, Anelis, Theramex, Nycomed, and Novo Nordisk, tries; Dr. Meunier lecture fees from Servier, Eli Lilly, Merck, Procter and grant support from Bristol-Myers Squibb; Dr. Rizzoli consult- & Gamble, and Aventis; Dr. Reginster consulting fees from Novar- ing fees from Novartis and Wyeth Laboratories, lecture fees from tis, Negma Laboratories, Eli Lilly, Wyeth Laboratories, Amgen, Merck, Servier, and Aventis, and grant support from Proskalia and GlaxoSmithKline, Roche, Merck Pharmaceuticals, Nycomed, NPS Servier; Dr. Roux consulting and lecture fees from Procter & Gam- Pharmaceuticals, and Theramax, lecture fees from Merck, Eli Lilly, ble and lecture fees from Novartis and Eli Lilly; and Dr. Spector lec- Rotta Pharmaceuticals, Institut Biochimique, Novartis, Servier, ture fees from Procter & Gamble.
a p p e n d i x
The Spinal Osteoporosis Therapeutic Intervention study investigators were as follows: Australia: J. Graham, Ashford Specialist Centre, Ash-ford; K.W. Ng, St. Vincent’s Hospital, Fitzroy; R. Prince, Sir Charles Gairdner Hospital, Nedlands; J. Prins, Princess Alexandra Hospital,Wooloongabba; E. Seeman, Austin and Repatriation Medical Centre, Heidelberg; J. Wark, Royal Melbourne Hospital, Parkville; Belgium: J.Y.
Reginster, Brull Hospital, Liege; J.P. Devogelaer, Catholic University of Louvain, Brussels; J.M. Kaufman, University of Gent, Gent; F. Rae-man, Jan Palfijin Ziekenhuis, Merksem; M. Walravens, Medical Centre, Tessenderloo; Denmark: S. Pors-Nielsen, Hillerød Hospital, Hill-erød; H. Beck-Nielsen, University Hospital Osteoporosis Centre, Odense; P. Charles, Aarhus Amtssygehus, Aarhus; O.H. Sorensen,Osteoporosis Research Center, Hvidovre; France: P.J. Meunier, Edouard Herriot Hospital, Lyons; J.P. Aquino, Medical Clinic La Porte Verte,Versailles; C. Benhamou, La Madeleine Hospital, Orleans; F. Blotman, Lapeyronie Hospital, Montpellier; O. Bonidan, E. Muller Hospital,Mulhouse; P. Bourgeois, Pitié-Salpétrière Hospital, Paris; M.C. De Vernejoul Lariboisière Hospital, Paris; J. Dehais, Pellegrin-Tondu Hospi-tal, Bordeaux; P. Fardellone, Nord Hospital, Amiens; A. Kahan, Cochin Hospital, Paris; J.L. Kuntz, Hautepierre Hospital, Strasbourg; C.
Marcelli, Regional Hospital and University Centre, Caen; A. Prost, Hôtel Dieu Hospital, Nantes; B. Vellas, La Grave-Casselardit Hospital,Toulouse; G. Weryha, Brabois Hospital, Vandoeuvre; Germany: E.M. Lemmel, Max Grundig Clinic, Buhl; D. Felsenberg, Benjamin FranklinClinical University, Berlin; J. Hensen, Medical Clinic 1, Erlangen; H.P. Kruse, Medical Clinical University, Hamburg; W. Schmidt, St. JosefHospital, Bochum; J. Semler, Immanuel Hospital, Berlin; G. Stucki, Clinic for Physical Medicine and Rehabilitation, Munich; Greece: C.
Phenekos, Red Cross Hospital, Athens; Hungary: A. Balogh, University of Debrecen, Medical and Health Sciences Centre, Debrecen; R. DeChatel, Semmelweis Orvostudomanyi Egyetem, Budapest; Italy: S. Ortolani, Centre Auxologico Italiano, Milan; S. Adami, Hospital andClinical Centre of Medecine, Valeggio sul Mincio; G. Bianchi, Hospital La Colletta, Arenzano; M.L. Brandi, General Hospital Careggi, Flo-rence; D. Cucinotta, S. Orsola Malpighi Hospital, Bologna; C. Fiore, Medical Clinic, Vittorio Emanuel Hospital, Catania; C. Gennari, Insti-tute of Medicine Clinic, New General Hospital Le Scotte, Siena; G. Isaia, University of General Medicine, Turin; G. Luisetto, University In-stitute of Semeiotica Medica, Padua; R. Passariello, Institute of Radiology, La Sapienza University, Rome; M. Passeri, Parma University,Parma; G. Rovetta, E. Bruzzone Institute, Genoa; L. Tessari, San Raffaele Hospital, Milan; Poland: J.E. Badurski, Center for Osteoporosisand Osteo-Articular Diseases, Bialystok; K. Hoszowski, Medical Centre, Warsaw; R.S. Lorenc, Osteoporosis Centre of Krajowe, Warsaw; A.
Sawicki, Warsawian Center of Osteoporosis and Calcium Metabolism Osteomed, Warsaw; Spain: A. Díez, El Mar Hospital, Barcelona; J.B.
Cannata, General Hospital of Asturias, Oviedo; M. Díaz Curiel and A. Rapado, Jiménez Díaz Foundation, Madrid; J. Gijon and A. Torrijos,La Paz Hospital/Hospital de Rehabilitacion y Traumatologia, Madrid; J.M. Padrino, Hospital 12 de Octubre, Madrid; A. Roces Varela, Hos-pital Nuestra Sra de la Candelaria, Santa Cruz de Tenerife, Canary Islands; Switzerland: J.P. Bonjour and R. Rizzoli, Geneva University Hospi-tal, Geneva; United Kingdom: T. D. Spector, Department of Rheumatology, St. Thomas’ Hospital, London; M. Clements, North London Clin-ical Studies Centre, Mount Vernon Hospital, Northwood; D.V. Doyle, Silverthorn Centre, Chingford; P. Ryan, Medway Hospital Trust,Gillingham; I.G. Smith, Synexus Ltd. Clinical Research Centre, Manchester; R. Smith, Banbury Clinical Studies Centre, Banbury.
r e f e r e n c e s
Nevitt MC, Ettinger B, Black DM, et al.
12. Neer RM, Arnaud CD, Zanchetta JR, et
The association of radiographically detected et al. Randomised trial of the effects of ris- al. Effect of parathyroid hormone (1-34) on vertebral fracture with back pain and func- fractures and bone mineral density in post- tion: a prospective study. Ann Intern Med with established postmenopausal osteoporo- Ettinger B, Black DM, Mitlak BH, et al.
13. Marie PJ, Amman P, Boivin G, Rey C.
III. Medical expenditures for the treatment Reduction of vertebral fracture risk in post- of osteoporotic fractures in the United States menopausal women with osteoporosis treat- tential of strontium in bone. Calcif Tissue in 1995: report from the National Osteoporo- ed with raloxifene: results from a 3-year ran- sis Foundation. J Bone Miner Res 1997;12: domized clinical trial. JAMA 1999;282:637- 14. Marie PJ, Garba MT, Hott M, Miravet L.
Effect of low doses of stable strontium on Lindsay R, Silverman SL, Cooper C, et al.
bone metabolism in rats. Miner Electrolyte Risk of new vertebral fracture in the year fol- et al. Treatment of postmenopausal osteopo- lowing a fracture. JAMA 2001;285:320-3.
sosis with transdermal estrogen. Ann Intern 15. Canalis E, Hott M, Deloffre P, Tsou-
deros Y, Marie PJ. The divalent strontium Atkinson EJ, O’ Fallon WM, Riggs BL. Risk 10. Boivin GY, Chavassieux PM, Santora AC,
salt S12911 enhances bone cell replication of hip fracture in women with vertebral frac- Yates J, Meunier PJ. Alendronate increases and bone formation in vitro. Bone 1996;18: ture. J Bone Miner Res 1994;9:599-605.
gree of mineralization of bone tissue in os- 16. Buehler J, Chappuis P, Saffar JL, Tsou-
teoporotic women. Bone 2000;27:687-94.
deros Y, Vignery A. Strontium ranelate in- 11. Chavassieux PM, Arlot ME, Reda C, Wei
L, Yates AJ, Meunier PJ. Histomorphometric bone formation in alveolar bone in monkeys al. Randomised trial of effect of alendro- assessment of the long-term effects of alen- (Macaca fascicularis). Bone 2001;29:176-9.
nate on risk of fracture in women with ex- dronate on bone quality and remodeling in 17. Marie PJ, Hott M, Modrowski D, et al.
isting vertebral fractures. Lancet 1996;348: patients with osteoporosis. J Clin Invest 1997; prevents bone loss by depressing bone re- Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved. s t r o n t i u m r a n e l a t e a n d v e r t e b r a l f r a c t u r e i n p o s t m e n o p a u s a l o s t e o p o r o s i s Comparison of semiquantitative visual and women on oestrogen with osteoporosis. Lan- in estrogen-deficient rats. J Bone Miner Res prevalent and incident vertebral fractures 27. Black DM, Greenspan SL, Ensrud KE,
18. Grynpas MD, Marie PJ. Effects of low
in osteoporosis. J Bone Miner Res 1996;11: et al. The effects of parathyroid hormone doses of strontium on bone quality and quan- and alendronate alone or in combination in 23. Wu CY, Li J, Jergas M, Genant HK. Com-
postmenopausal osteoporosis. N Engl J Med 19. Boivin G, Deloffre P, Perrat B, et al. Stron-
parison of semiquantitative and quantitative tium distribution and interactions with bone techniques for the assessment of prevalent 28. Nielsen SP, Slosman D, Sorensen OH,
mineral in monkey iliac bone after strontium and incident vertebral fractures. Osteoporos et al. Influence of strontium on bone miner- salt (S 12911) administration. J Bone Miner al density and bone mineral content measure- 24. Slosman DO, Provvedini DM, Meunier
ments by dual X-ray absorptiormetry. J Clin 20. Meunier PJ, Slosman DO, Delmas PD, et
PJ, et al. The use of different dual X-ray ab- al. Strontium ranelate: dose-dependent ef- sorptiometry brands in a multicenter clini- 29. Dahl SG, Allain P, Marie PJ, et al. Incor-
fects in established postmenopausal verte- cal trial. J Clin Densitom 1999;2:37-44.
poration and distribution of strontium in bral osteoporosis — a 2-year randomized 25. Delmas PD. Markers of bone turnover for
placebo controlled trial. J Clin Endocrinol monitoring treatment of osteoporosis with 30. Reginster JY, Deroisy R, Dougados M,
antiresorptive drugs. Osteoporos Int 2000; Jupsin I, Colette J, Roux C. Prevention of ear- 21. Genant HK, Wu CY, van Kuijk C, Nevitt
MC. Vertebral fracture assessment using a 26. Lindsay R, Nieves J, Formica C, et al. Ran-
ranelate: the randomized, two-year, double- semiquantitative technique. J Bone Miner Res domised controlled study of effect of para- thyroid hormone on vertebral-bone mass and PREVOS trial. Osteoporos Int 2002;13:925-31.
22. Genant HK, Jergas M, Palermo L, et al.
Copyright 2004 Massachusetts Medical Society. Downloaded from www.nejm.org at AMERICAN UNIVERSITY OF BEIRUT on April 21, 2008 . Copyright 2004 Massachusetts Medical Society. All rights reserved.

Source: http://www.osteos.org.lb/admin/uploads/strontium1-2004.pdf