SEARCH

SEARCH BY CITATION

Keywords:

  • BONE HISTOMORPHOMETRY;
  • OSTEOPOROSIS;
  • CLINICAL TRIALS

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Denosumab reduced bone resorption, increased bone mineral density (BMD), and decreased new vertebral, hip, and nonvertebral fracture risk in postmenopausal women with osteoporosis in the FREEDOM trial. Consistent with its mechanism of action, transiliac crest bone biopsies from subjects treated with denosumab for 1 to 3 years demonstrated reduced bone turnover that was reversible upon treatment cessation. Long-term denosumab treatment for up to 6 years in the FREEDOM extension provides sustained bone turnover reduction and continued low fracture incidence. Here, we evaluate 5 years of denosumab treatment on bone remodeling at the tissue level. Transiliac crest bone biopsies were obtained from 41 subjects (13 cross-over and 28 long-term from the FREEDOM placebo and denosumab groups, respectively) at year 2 of the FREEDOM extension, representing up to 5 years of denosumab treatment. Demographics for this subset were comparable to the overall extension cohort. The mean (SD) duration from the last denosumab dose to the first dose of tetracycline was 5.7 (0.5) months. Qualitative bone histology assessed in all biopsy samples was unremarkable, showing normally mineralized lamellar bone. Structural indices, including trabecular bone volume, number, and surface, were similar between cross-over and long-term groups. Bone resorption was decreased as reflected by eroded surface in cross-over and long-term subjects. A total of 11 of 13 (85%) cross-over subjects and 20 of 28 (71%) long-term subjects had specimens with double or single tetracycline label in trabecular and/or cortical compartments; specimens from 5 cross-over subjects and 10 long-term subjects were evaluable for dynamic trabecular bone parameters. Dynamic remodeling indices were low for both groups and consistent with reduced bone turnover with denosumab. In conclusion, denosumab treatment through 5 years resulted in normal bone quality with reduced bone turnover. These observations are consistent with its mechanism of action and associated with continued BMD increases and low fracture incidence. © 2014 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Denosumab is a potent antiresorptive agent that inhibits receptor activator of NF-κB ligand (RANKL), a cytokine essential for osteoclast formation, activity, and survival.[1-4] In clinical trials, denosumab treatment rapidly and significantly decreased bone turnover, increased bone mineral density (BMD), and reduced the risk of new vertebral, hip, and nonvertebral fractures compared with placebo in postmenopausal women with osteoporosis.[5-8] Bone biopsies obtained after 1 to 3 years of denosumab treatment have demonstrated the potent and sustained inhibition of bone turnover by denosumab at the tissue level.[9] Analysis of bone biopsies obtained approximately 2 years after discontinuation of denosumab treatment confirmed the effects of denosumab are fully reversible at the tissue level,[10] consistent with observations of bone turnover markers and BMD.

Available data indicate that long-term treatment with denosumab produced continued BMD gains for up to 8 years in a phase 2 dose-ranging study and a low incidence of new vertebral and nonvertebral fractures for up to 7 years in the FREEDOM extension study.[11, 12] Bone biopsies were planned to be obtained at years 2 and 7 of the FREEDOM extension study (representing up to 5 and 10 years of denosumab treatment, respectively), and data from year 2 are now available. The present analysis evaluated bone biopsies obtained from women who received denosumab for up to 5 years in the FREEDOM extension to determine tissue-level effects of long-term denosumab treatment. In addition, we compared findings with histology and histomorphometry from subjects treated with denosumab or placebo in the previously published FREEDOM bone biopsy substudy.

Materials and Methods

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Study participants

Subjects included in this study were enrolled in the FREEDOM trial and its extension, which have been previously described in detail.[5, 8] Briefly, FREEDOM was a 3-year international, randomized, double-blind, placebo-controlled trial in postmenopausal women with osteoporosis.[5] Subjects received subcutaneous denosumab 60 mg every 6 months or placebo for 3 years. All subjects received daily calcium (≥1000 mg) and vitamin D (≥400 IU) supplementation. Subjects who completed the FREEDOM trial and did not miss more than 1 dose of their assigned treatment were eligible to enroll in the extension study that will continue to evaluate the safety and efficacy of denosumab for an additional 7 years. All subjects enrolled in the extension study receive open-label denosumab 60 mg every 6 months. In the extension study, subjects who had received placebo during FREEDOM are referred to as the cross-over group, and subjects who had received denosumab during FREEDOM are referred to as the long-term group. Subjects were eligible to enroll in the bone biopsy substudy if they were enrolled at a clinical trial center that was participating in the bone biopsy substudy and had no sensitivity to tetracycline or its derivatives.

The study protocols for the FREEDOM trial and its extension were approved by the institutional review board or ethics committee at each study center. All subjects provided written informed consent.

Bone biopsy procedures

The methodology for the bone biopsy procedure and analyses is consistent with the approach from prior denosumab clinical trials and has been previously described in detail.[9] Briefly, bone biopsies of the iliac crest were obtained within 56 days of the year-5 visit (year 2 of the extension study) from subjects in the cross-over and long-term groups. All subjects scheduled for biopsy followed a standard double tetracycline derivative labeling procedure, as previously described.[9] Urine samples were collected within 24 hours of the last dose of the first tetracycline labeling period to confirm compliance.

Bone biopsies were obtained from the anterior iliac crest using a Bordier/Meunier or Rochester-type trephine (internal diameter 7 to 8 mm). Specimens were fixed and shipped in 70% ethanol and then dehydrated, embedded in glycol methylmethacrylate, and sectioned at a thickness of 5 µm at a central histomorphometric facility (Mayo Clinic, Rochester, MN, USA). Sections were mounted unstained for analysis of the tetracycline labels. If tetracycline labels were present, the adjacent sections were stained with Goldner trichrome or toluidine blue for qualitative and histomorphometry analyses by a hematopathologist. Histomorphometry variables were analyzed using a semi-automated OsteoMeasure system (OsteoMetrics, Decatur, GA, USA) and calculated and expressed according to the ASBMR recommendations.[13] Assessment of presence or absence of tetracycline label in trabecular and cortical bone compartments was performed as previously described.[9]

Statistical analysis

Consistent with the ASBMR committee recommendations for evaluating biopsies with only single label in trabecular bone within the measureable region, a value of 0.3 µm/d was imputed for mineral apposition rate (MAR).[13, 14] Because no control group was available from this cohort, a descriptive comparison with data from biopsies obtained from placebo- and denosumab-treated subjects at years 2 and/or 3 in FREEDOM[9] is presented.

Results

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Subjects

Forty-one subjects participated in the bone biopsy substudy (13 cross-over and 28 long-term subjects). The baseline demographics of the bone biopsy cohort were comparable with the overall population of the FREEDOM extension[8] (Table 1). Nearly all subjects underwent transiliac crest bone biopsy within 30 days before their next dose of denosumab; 2 subjects received the dose of denosumab 7 and 21 days before the biopsy procedure. Across groups, the mean (SD) duration from the last dose of denosumab to the first dose of tetracycline was 5.7 (0.5) months. Tetracycline compliance was evaluated through presence/absence of urinary tetracycline derivatives. Positive urine samples were collected in all cross-over (13 [100%]) and nearly all long-term (24 [86%]) subjects; urinary specimen status was unknown in the remaining 4 long-term subjects.

Table 1. Baseline Demographics and Clinical Characteristics for Subjects Enrolled in the Extension Study and Subjects Who Underwent Biopsy
 Cross-over denosumab extension subjectsLong-term denosumab extension subjects
Overall (n = 2207)Bone biopsy (n = 13)Overall (n = 2343)Bone biopsy (n = 28)
  1. Data are mean (SD) unless otherwise specified.

  2. IQR = interquartile range.

  3. a

    Bone turnover markers were evaluated in a subset of subjects in the extension study (36 cross-over and 76 long-term; the number of subjects who participated in the bone biopsy study and had bone turnover marker evaluations were 0 cross-over and 11 long-term subjects).

Age (years)74.8 (5.1)76.7 (4.2)74.9 (5.0)75.5 (3.9)
Years since menopause26.7 (7.4)30.4 (5.2)26.7 (7.3)26.1 (5.3)
sCTx-1a (ng/mL), median (IQR)0.57 (0.43, 0.73)0.18 (0.09, 0.56)0.22 (0.07, 0.64)
Lumbar spine BMD T-score–2.8 (0.8)–2.9 (0.7)–2.1 (0.8)–2.3 (0.9)
Total hip BMD T-score–1.9 (0.8)–1.6 (1.0)–1.5 (0.8)–1.7 (0.8)
Femoral neck BMD T-score–2.2 (0.7)–1.9 (0.8)–1.8 (0.8)–1.9 (0.8)

Biopsies

All biopsies (41) were evaluable for histology, and 38 were evaluable for structural and static histomorphometric indices (13 of 13 and 25 of 28 samples from cross-over and long-term groups, respectively). The search for tetracycline labeling revealed evidence of double or single labels in trabecular and/or cortical bone in 85% and 71% of subjects in the cross-over and long-term groups, respectively (Table 2). Of these, 6 of 13 and 12 of 28 samples from cross-over and long-term subjects, respectively, had double or single labels in trabecular bone within the measurement area and were evaluable for full histomorphometric analyses, including dynamic formation indices. Seven subjects (1 cross-over subject and 6 long-term subjects) had undergone bone biopsy at year 2 and/or year 3 in the FREEDOM trial and had a sequential biopsy in year 5. Although nearly all of these subjects had presence of tetracycline labeling in trabecular or cortical compartments (Table 3), quantitative comparisons between paired biopsies were limited because of labels outside of the prespecified measurement field (total assessed tissue area of 20 mm2 from 3 sets of unstained sections, 250 µm apart).[9]

Table 2. Tetracycline Labeling Status
FREEDOM (year 2 and/or 3)[9]TrabecularCorticalTrabecular or cortical
PlaceboDenosumabPlaceboDenosumabPlaceboDenosumab
Evaluable biopsies625362536253
Any label, n (%)61 (98)18 (34)62 (100)30 (57)62 (100)34 (64)
Double label, n (%)58 (94)10 (19)61 (98)16 (30)62 (100)21 (40)
Single label only, n (%)3 (5)8 (15)1 (2)14 (26)0 (0)13 (25)
No label, n (%)1 (2)35 (66)0 (0)23 (43)0 (0)19 (36)
Extension (year 5)Cross-over denosumabLong-term denosumabCross-over denosumabLong-term denosumabCross-over denosumabLong-term denosumab
  1. n = number of biopsies. Percentages are calculated based on number of evaluable biopsies from year 2 and/or 3 in FREEDOM trial and year 5 in the extension.

Evaluable biopsies132813281328
Any label, n (%)6 (46)12 (43)10 (77)18 (64)11 (85)20 (71)
Double label, n (%)6 (46)8 (29)8 (62)11 (39)10 (77)14 (50)
Single label only, n (%)0 (0)4 (14)2 (15)7 (25)1 (8)6 (21)
No label, n (%)7 (54)16 (57)3 (23)10 (36)2 (15)8 (29)
Table 3. Tetracycline Labeling Status in Subjects Who Underwent Bone Biopsy in the FREEDOM Trial and Extension Study
Subject biopsy yearTrabecularCorticalTrabecular or cortical
  1. Biopsies in year 2 and 3 of the FREEDOM trial; biopsy in year 5 of the extension study. Subject 1 was in the cross-over group; all other subjects were in the long-term group.

Cross-over subject
Subject 1
Year 2Single and double labelsSingle and double labelsSingle and double labels
Year 5No labelSingle and double labelsSingle and double labels
Long-term subjects
Subject 2
Year 3No labelNo labelNo label
Year 5Single labelNo labelSingle label
Subject 3
Year 2Double labelSingle labelDouble label
Year 3Double labelSingle and double labelsSingle and double labels
Year 5Single labelSingle and double labelsSingle and double labels
Subject 4
Year 3No labelNo labelNo label
Year 5Single and double labelsSingle labelSingle and double labels
Subject 5
Year 3Single labelNo labelSingle label
Year 5No labelNo labelNo label
Subject 6
Year 3Single labelSingle labelSingle label
Year 5Single and double labelsSingle labelSingle and double labels
Subject 7
Year 2No labelNo labelNo label
Year 3No labelNo labelNo label
Year 5No labelSingle and double labelsSingle and double labels

Qualitative bone histology as assessed in all samples was unremarkable, showing normally mineralized lamellar bone and no evidence of pathologic findings, including osteomalacia, woven bone, or marrow fibrosis (Table 4). Biopsies from 5 subjects in the long-term group did not have visualizable osteoid; one of these specimens had a crush artifact in the trabecular compartment, and 2 of these specimens had evidence of tetracycline labeling.

Table 4. Bone Histology
 FREEDOM (year 2 and/or 3)[9]Extension (year 5)
Placebo (n = 45)Denosumab (n = 47)Cross-over denosumab (n = 13)Long-term denosumab (n = 28)
  1. a

    Evaluable biopsies from year 2 and/or 3 in FREEDOM trial and year 5 in the extension.

  2. b

    No osteoid in 5 subjects at year 2.

  3. c

    No osteoid in 5 subjects at year 5.

Evaluable biopsiesa62531328
Normal lamellar bone, n (%)62 (100)53 (100)13 (100)28 (100)
Normal mineralization, n (%)62 (100)53 (100)13 (100)28 (100)
Present osteoid, n (%)62 (100)48 (91)b13 (100)23 (82)c
Osteomalacia, n0000
Marrow fibrosis, n0000
Woven bone, n0000

Structural histomorphometric indices for the cross-over and long-term groups are shown in Table 5. Structural indices that assess the amount of bone, including trabecular bone volume, thickness, separation, and number, and cortical width were similar between the cross-over and long-term groups. Eroded surface was low, consistent with the mechanism of action of denosumab. Because bone resorption and formation are coupled, indicators of bone formation, including osteoid surface and osteoid width, were also low in both the cross-over and long-term groups. When MAR was assigned a value of 0.3 µm/d for samples with only single labels, dynamic remodeling indices including MAR, bone formation rate (MAR* [MS/BS]), and activation frequency were low for the cross-over and long-term groups (Table 5). When MAR and mineralizing surface expressed in relation to bone surface (MS/BS) were assigned zero values in samples with no labels in the measurement area, the medians and interquartile ranges for the cross-over and long-term groups, respectively, were as follows: MAR 0.0 (0.0, 0.5) and 0.0 (0.0, 0.3) µm/d; bone formation rate 0.0 (0.0, 0.7) and 0.0 (0.0, 1.0) %/yr; and activation frequency 0.00 (0.00, 0.01) and 0.00 (0.00, 0.01) per year.

Table 5. Histomorphometric Indices
 FREEDOM (year 2 and/or 3)[9]Extension (year 5)
Placebo (n = 45)Denosumab (n = 47)Cross-over denosumab (n = 13)Long-term denosumab (n = 28)
  1. Data are median (interquartile range).

  2. a

    Evaluable biopsies from year 2 and/or 3 in FREEDOM trial and year 5 in the extension.

  3. b

    In biopsies with single labels, values of 0.3 µm/d were imputed.

Evaluable biopsiesa62531325
Structural indices
Trabecular bone volume (BV/TV %)12.5 (9.4, 17.4)13.5 (10.3, 15.9)13.9 (10.6, 18.2)14.2 (8.4, 20.2)
Trabecular thickness (µm)145.8 (118.8, 165.8)138.3 (121.4, 160.1)139.9 (128.4, 152.4)148.3 (119.7, 185.8)
Trabecular separation (µm)1001 (749, 1291)896 (738, 1033)878 (662, 972)1000 (718, 1357)
Trabecular number (mm−1)0.88 (0.70, 1.08)0.96 (0.85, 1.17)0.98 (0.93, 1.23)0.89 (0.69, 1.06)
Cortical width (mm)0.77 (0.54, 1.1)0.66 (0.44, 1.07)0.53 (0.48, 0.74)0.69 (0.55, 0.81)
Static formation indices
Osteoid surface (%)6.8 (3.6, 10.1)0.4 (0.2, 1.2)0.5 (0.2, 0.7)0.1 (0.0, 0.8)
Osteoid width (µm)8.7 (6.4, 11.0)5.4 (4.4, 7.4)5.6 (3.3, 6.6)3.3 (0.0, 7.4)
Wall thickness (µm)47.5 (39.6, 56.1)43.6 (39.4, 53.5)40.6 (38.2, 48.1)51.9 (42.4, 56.3)
Dynamic formation indices
Mineral apposition rateb (µm/d)0.8 (0.7, 0.8)0.3 (0.3, 0.5)0.6 (0.5, 0.7)0.4 (0.3, 1.1)
Bone formation rate, volume based (%/yr)14.6 (8.6, 21.8)0.4 (0.2, 0.8)1.2 (0.7, 1.3)2.2 (0.2, 4.7)
Activation frequency (year−1)0.200 (0.120, 0.330)0.002 (0.001, 0.004)0.017 (0.011, 0.020)0.031 (0.001, 0.071)
Static resorption indices
Eroded surface/bone surface (%)1.0 (0.6, 1.9)0.2 (0.0, 0.7)0.2 (0.0, 0.4)0.1 (0.0, 0.3)
Osteoclast number, length based (mm−1)0.08 (0.04, 0.12)0.00 (0.00, 0.04)0.03 (0.00, 0.13)0.03 (0.00, 0.07)

Structural, static, and dynamic variables at year 5 of the study extension were consistent with the histologic and histomorphometric data in biopsy samples obtained after 2 and/or 3 years of treatment with denosumab in the FREEDOM trial (Tables 4 and 5).[9]

Discussion

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

This substudy documents maintenance of qualitatively normal bone histology after up to 5 years of denosumab treatment in postmenopausal women with osteoporosis. Normal trabecular and cortical microarchitecture was maintained, mineralization was normal, and bone marrow unremarkable. The bone tissue was lamellar without any woven bone. The structural histomorphometric indices showed preservation of structural integrity compared with subjects treated with denosumab for 2 and/or 3 years.[9] The static resorption indices showed low eroded surface, reflecting a reduction in bone turnover, which may represent an important contributing mechanism for the maintenance of the observed low fracture incidence in the extension study through 6 years.[15] The dynamic indices of bone formation at the tissue level were low, regardless of how MS/BS, MAR, and bone formation rate were calculated and were coupled with the reduction of bone resorption, which is consistent with the mechanism of action of denosumab.

There has been speculation that oversuppression of bone remodeling, often acknowledged by the absence of double tetracycline labeling, could result in increased bone fragility and fractures.[16] However, no definitive link has been established between low remodeling rates and adverse skeletal consequences in postmenopausal women with osteoporosis. Thus, neither assigning a specific rate of bone remodeling considered to be “oversuppression” nor defining a low remodeling state linked to impairment in bone strength has a scientific basis within this population. Recker and colleagues noted that in the absence of double tetracycline labeling, low remodeling may be assumed when bone formation rate/bone surface and activation frequency are reported as zero, but this may not actually be zero in either the specimen or the skeleton.[17] Previous histomorphometric studies have revealed that up to 30% of untreated postmenopausal women with osteoporosis had no double tetracycline labels.[18, 19] Furthermore, in a previous report of subjects undergoing bone biopsies after denosumab treatment for up to 3 years, levels of bone turnover markers were similar in subjects with no label, single label, or double label.[9] Assessment of label status in 7 subjects who had undergone biopsy during FREEDOM and again in the study extension showed that tetracycline label was present in trabecular or cortical compartments despite continued low remodeling as reflected by serum CTx-1 (sCTx-1). Furthermore, it is important to point out that these results derive from evaluation of the iliac crest, a site that may not reflect the tissue level response in other relevant skeletal regions.

Animal and human studies suggest that reduction in remodeling is associated with improvement in bone strength and fracture risk reduction.[20-23] Denosumab-treated cynomolgus monkeys with the lowest levels of fluorescent labeling exhibited the greatest structural bone strength values at weight-bearing (proximal femur, vertebrae, and tibial diaphysis) and non-weight bearing (iliac crest) sites.[24] Similarly, the consistent reduction in bone turnover in the FREEDOM extension study through 6 years has been associated with low incidence of vertebral, hip, and nonvertebral fracture.[8, 15]

This study has several limitations. Not all subjects who underwent biopsy had concurrent bone turnover marker assessment, which limited our ability to correlate findings globally with the tissue level. Because there was not a defined control group in this substudy, data were compared with that obtained from biopsies from placebo- and denosumab-treated subjects at years 2 and/or 3 in FREEDOM. Our evaluation focused on the trabecular compartment, whereas the endocortical envelope has been shown to be highly remodeled,[25] leading to the expansion of the medullary cavity at the expense of cortical thinning.[26] Given the decreased cortical porosity that has been observed with denosumab,[27] it will be important to understand its effects on periosteal, subcortical, and endocortical compartments, and this work is in progress. Finally, bone biopsies of the iliac crest, while standard, may not be fully representative of bone histomorphometry at other important skeletal sites.

In conclusion, the histology and histomorphometry findings were similar in subjects treated for 2, 3, or 5 years with denosumab. Bone histomorphometry confirmed sustained reduction of bone remodeling at the tissue level and maintenance of normal bone microarchitecture in postmenopausal women with osteoporosis treated with denosumab for up to 5 years. Taken in the context of its demonstrated low fracture incidence through 6 years, these data provide further evidence for the safety of denosumab through this time period and advance a new paradigm regarding the relevance of remodeling to bone strength in the postmenopausal and aging skeleton.

Disclosures

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

JPB has received research grants from Amgen Inc., Eli Lilly, Merck, Novartis, and Warner Chilcott, board membership for Amgen Inc., Eli Lilly, and Merck, and consulting and speaking fees from Amgen Inc. and Eli Lilly. IRR has received research grants and speaking fees from Amgen Inc. RBW and ND are employees of Amgen Inc. and may own stock and/or stock options in Amgen Inc. DK has received research grants, consulting, and speaking fees from Amgen Inc., Eli Lilly, GSK, J&J, Merck, Novartis, Pfizer, and Roche. PDM has received research grants Amgen Inc., Boehringer-Mannheim, Lilly, Merck, Radius, and Takeda, and consultancy fees from Amgen Inc., Lilly, Merck, and Radius. JEBJ has board membership for and has received speaking fees from Amgen Inc., Eli Lilly, MSD, Novartis, and Nycomed. MAB has received speaking fees from Amgen Inc. IV has no conflicts of interest. CAFZ has received research grants from Amgen Inc., board membership for Merck and Lilly, consulting and/or speaking fees from Lilly, Pfizer, and Sanofi, and development of educational presentations for Pfizer. DWD has received research grants from Eli Lilly, board membership, development of educational presentations for, consulting and speaking fees from Amgen Inc., Eli Lilly, and Merck, and provided expert testimony for Merck.

Acknowledgments

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Amgen Inc. sponsored this study. Medical writing support was provided by Erica Rockabrand and Mandy Suggitt, who are employees of Amgen Inc. The authors thank the trial investigators and participants.

Authors' roles: JPB, RBW, IRR, DK, PDM, MAB, IV, JEBJ, CAFZ, ND, and DWD contributed to the conception and design of the study, the acquisition of data, and the analysis and interpretation of the data. JPB and DWD participated in drafting the manuscript. All authors contributed to revising the manuscript, and all authors approved the final version of the manuscript for submission.

References

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References