Differential Effects of Teriparatide and Alendronate on Bone Remodeling in Postmenopausal Women Assessed by Histomorphometric Parameters

Authors


  • Dr Eriksen was an employee and owns stock in Eli Lilly. Drs Donley and Dalsky are employees of Eli Lilly and Company. Dr San Martin was an employee at Eli Lilly and Company during the course of this study. Dr Meunier serves as a consultant for Aventis, Eli Lilly and Company, MSD, Nycomed, Procter & Gamble, and Servier. Dr Arlot received a grant from Eli Lilly and Company. Dr Boivin serves as a consultant for Eli Lilly and Company and Servier. All other authors have no conflict of interest.

Abstract

An 18-month randomized double-blind study was conducted in postmenopausal women with osteoporosis to compare the effects of once-daily teriparatide 20 μg with alendronate 10 mg on bone histomorphometry. Biopsies were obtained from 42 patients. Indices of bone formation were significantly higher after 6 or 18 months of teriparatide compared with alendronate treatment.

Introduction: Alendronate and teriparatide increased BMD, assessed by DXA, by different mechanisms of action, supported by changes in biochemical markers of bone turnover. The purpose of this cross-sectional study was to explore the differential effects of these two osteoporosis treatments at the bone tissue level by examining bone histomorphometric parameters of bone turnover after either 6 or 18 months of treatment.

Materials and Methods: Patients were a cohort from a randomized parallel double-blind study conducted to compare the effects of once-daily teriparatide 20 μg and alendronate 10 mg in postmenopausal women with osteoporosis. Transiliac crest bone biopsies were obtained after tetracycline double labeling from 42 patients treated for 6 months (n = 23) or 18 months (n = 14); 5 additional patients were biopsied from contralateral sides at 6 and 18 months. Biopsy specimens adequate for quantitative analysis were analyzed by 2D histomorphometry from 17 patients at 6 months (teriparatide, n = 8; alendronate, n = 9) and 15 patients at 18 months (teriparatide, n = 8; alendronate, n = 7). Data were analyzed by two-sample tests.

Results: Histomorphometric indices of bone formation were significantly and markedly greater in the teriparatide group than in the alendronate group at 6 and 18 months, whereas indices of bone resorption were only significantly greater in the teriparatide group than in the alendronate group at 6 months. Bone formation and activation frequency were significantly lower at 18 months compared with 6 months in the teriparatide group, returning to levels comparable with untreated postmenopausal women. In the teriparatide group, the peak in histomorphometric bone formation indices coincided with peak levels for N-terminal propeptide of type I collagen, a biochemical marker of bone formation. The degree of mineralization was lower at 18 months than at 6 months with treatment in both groups but was not different between groups.

Conclusions: These results confirm the opposite mechanisms of action of teriparatide and alendronate on bone remodeling and confirm the bone formation effect of teriparatide.

INTRODUCTION

HISTORICALLY, HIGH BONE turnover has been considered harmful to bone with adverse effects on the mechanical load-bearing competence of bone in untreated patients with osteoporosis.(1) High bone turnover has been associated with an increased loss of bone, deterioration of trabecular connectivity and subsequent increased risk for fracture.(1–6)

A negative bone remodeling balance in the presence of high bone turnover is found in untreated patients with osteoporosis, and the reduction of bone turnover and activation frequency has been the objective of osteoporosis treatments for many years.(7,8) Alendronate sodium, a second-generation aminobisphosphonate, achieves that objective by preferentially inhibiting bone resorption through its action on osteoclasts,(9) thereby decreasing bone turnover. Alendronate preserves existing architecture and reduces the incidence of osteoporotic fractures,(10–13) yet it neither improves nor restores architectural integrity associated with severe osteoporosis(14–17)

Teriparatide (rDNA origin) injection [recombinant human PTH(1-34)], a bone-forming agent that increases bone remodeling, represents a new therapeutic option for the treatment of osteoporosis.(18) Teriparatide, administered once daily through subcutaneous self-injection, is effective in reducing fracture incidence and increasing BMD.(19–23) However, in contrast to antiresorptive agents, teriparatide preferentially increases bone formation through direct early stimulation of osteoblasts.(24) This increase in new bone formation results in a positive bone balance at the level of individual bone multicellular units (BMUs) and improved bone microarchitecture and quality.(20–23,25–28)

The different effects of these two classes of drugs at the bone tissue level have been described in separate trials by analyzing iliac crest bone biopsies. However, no head to head histomorphometric trials have yet been conducted. Therefore, we conducted a cross-sectional study of a subset of patients enrolled in a multicenter, randomized, double-blind study in postmenopausal women with osteoporosis to compare and contrast the early (6 months) and later (18 months) effects of alendronate and teriparatide on bone histomorphometry parameters.

MATERIALS AND METHODS

Study participants

The effects of teriparatide and alendronate were measured cross-sectionally at 6 and 18 months in a subset of patients enrolled in a trial to compare teriparatide with alendronate. All patients were part of a randomized, parallel, double-blind, active-control 18-month clinical trial. Transiliac crest bone biopsies were obtained from 42 women at 6 clinical sites who were part of a larger study population of 203 postmenopausal women with osteoporosis, 45-85 years of age, who were randomized to treatment at 20 clinical sites globally.(29,30) Biopsies were obtained after tetracycline double labeling from patients treated for 6 (n = 23) or 18 months (n = 14). Five additional patients were biopsied from contralateral sides at 6 and 18 months; however, samples from only two patients were usable at both time-points, and both patients received teriparatide 20 μg. Biopsy specimens adequate for quantitative analysis were analyzed by 2D histomorphometry from 17 patients at 6 months (teriparatide, n = 8; alendronate, n = 9) and 15 patients at 18 months (teriparatide, n = 8; alendronate, n = 7). Briefly, all participants met the following criteria for randomization: ambulatory, 5 years or more past menopause, and a BMD T score between -2.5 and -4.0 at either the lumbar spine or femoral neck. Participants were free of disabling conditions other than postmenopausal osteoporosis and had normal or clinically insignificant abnormal laboratory values including serum calcium, PTH(1-84), 25-hydroxyvitamin D, and alkaline phosphatase.

The institutional review board at each clinical site approved the study protocol, which followed the principles of the Declaration of Helsinki. In addition to the written informed consent obtained from each participant before the start of screening tests, a separate consent was required of participants in this biopsy study.

Study treatments

Participants were enrolled in a randomized, parallel study and received either teriparatide 20 μg/day (Eli Lilly and Company, Indianapolis, IN, USA) or alendronate sodium 10 mg/day (Fosamax; Merck and Company, West Point, PA, USA). The patients self-injected teriparatide or placebo once a day using a disposable Gemini pen-injector with a prefilled, fixed-dose, cartridge delivery system. Each pen was used for up to 4 weeks; needles were changed daily. Alendronate or placebo was taken by mouth each morning after an overnight fast, following manufacturer's instructions. Each woman received daily supplementation of calcium (1000 mg) and vitamin D (400-800 IU) beginning at least 1 month before randomization and continued throughout the study.

The clinical trial endpoints included percent change from baseline in areal and volumetric BMD measured at the lumbar spine and proximal femur and percent change from baseline in biochemical markers of bone turnover, as described previously.(29,31) Biochemical markers of bone formation, serum procollagen type I N-terminal propeptide (S-PINP)], and bone resorption, urinary N-telopeptide corrected for creatinine (U-NTx), were assessed in patients at baseline and at 1, 3, 6, 12, and 18 months.

Histomorphometry

Transiliac crest bone biopsies were obtained from subsets of patients at 6 and 18 months of treatment using a Bordier needle trephine system after in vivo double labeling.(14) Approximately 20-22 days before biopsy, patients received 250 mg tetracycline hydrochloride four times daily for 2 days before and after an off period of 12 days (2:12:2-day sequence). All measured and derived parameters conform to the standard nomenclature and formulas recommended by the subcommittee on Bone Histomorphometry of the American Society of Bone and Mineral Research.(32) Biopsy specimens were transported in 70% ethanol to the central bone histomorphometry laboratory for assessment and analysis of biopsies and interpretation of results (PJ Meunier, Lyon, France), as previously described.(6,14,33) Histomorphometry analyses were conducted by laboratory personnel who were blinded to treatment assignments.

Biopsies were serially dehydrated in graded alcohols and embedded in methylmethacrylate. Serial sections were cut at three different levels sufficiently far apart (at least 250 μm) to avoid replicate sampling of a single surface event. Sections for Goldner's staining, Solochrom cyanin R, toluidine blue, May-Grunwald-Giemsa, and unstained sections were obtained from three separate planes from each biopsy specimen.

Qualitative study

All biopsy samples were assessed for the appearance of cellular components, presence or absence of woven bone, osteomalacia, marrow fibrosis, and any other notable features.

Quantitative study

The quantitative analyses were performed on complete, unbroken samples; however, a sample that was incomplete or broken or taken at the top of the crest was used when an adequate measurement was possible. These analyses were performed separately on the three bone envelopes (cortical, cancellous, and the endocortical bone). The limits of the cancellous bone were defined as twice the thickness of the largest trabecula coming from the cortices. A tissue volume of at least 20 mm2 (total of three planes) was needed for the quantitative analysis.

All static 2D parameters of remodeling activity, with the exception of wall thickness (W.Th, μm) were measured under light microscopy on 5-μm-thick biopsy sections stained with Goldner's trichrome. The following parameters were used to evaluate bone formation: osteoid volume/bone volume (OV/BV, %) measured in cancellous bone and expressed in percent of the cancellous bone volume; osteoid surface/bone surface (OS/BS, %) measured separately in cancellous and endocortical bone envelopes and expressed in percent of bone surface; osteoid thickness (O.Th, μm); mineral apposition rate (MAR, μm/day) measured on unstained sections under UV light; adjusted apposition rate (Aj.AR) (μm/day); bone formation rate (BFR/BS, μm/day); and formation period (days).(32)

The following parameters were measured to evaluate bone resorption: eroded surface (ES/BS, %); eroded volume (EV/BV, %); osteoclast surface (Oc.S/BS), the percentage of total bone surface that consists of active eroded surfaces; osteoclast number (N.Oc/BS), the number of osteoclasts per unit of bone surface; and erosion depth (E.De, μm), derived after rebuilding the resorption cavity on an image analyzer and analyzed with operator interaction.(34)

Bone turnover was assessed by the calculation of activation frequency (Ac.f, expressed as number of remodeling cycles per year). Changes in cancellous bone structure were assessed by measuring wall thickness (W.Th, μm.) of cancellous packets; cancellous bone volume (BV/TV, %); trabecular separation (Tb.Sp, μm); trabecular number (Tb.N, /mm), and trabecular thickness (Tb.Th, μm). Cortical bone structure was assessed by measuring cortical thickness (Ct.Th, μm) and cortical porosity (Ct.Po, %).

The technique of computerized quantitative contact microradiography(35) was used to measure the parameters reflecting the degree of mineralization of bone tissue (DMB). This method allows the analysis of each basic structure unit (BSU) of bone tissue, osteons in cortical bone, and cancellous packets (BSUs corresponding at the cancellous level to the osteons in cortical bone) in cancellous bone, within the limits that are imposed by the thickness of the biopsy sample. Using this computerized microdensitometric method, the quantity of mineral in a unit volume of bone tissue can be determined. The equipment used to perform the quantitative contact microradiography was a Philips X-ray diffraction unit (PW 1830/40; Limeil Brevannes) operated at 25 kV and 25 mA and equipped with a PW 2272/20 diffraction tube. A monochromatic X-ray (nickel-filtered copper K α radiation) was used, with a focus-to-film distance of 30 cm. An aluminum step wedge reference system was exposed on each micrograph with the biopsy section to allow for the quantitative evaluation of the absorption of X-rays. Using a combined contact microradiography-microdensitometry computerized method,(35,36) DMB parameters were independently quantified for both cortical and cancellous bone as well as for total bone (cortical + cancellous).

For the mean DMB (MDMB) section, the exact thickness of each biopsy section was measured with calipers, with a precision of 1 μm. For histomorphometric sections, an automatic microtome was used, with a precision of 1 μm. Using the calibration curve generated for each microradiograph from the aluminum reference step wedge, the grayscale values were converted to degree of mineralization and expressed as grams of mineral per cubic centimeter of bone (g mineral/cm3) adjusted for the exact thickness of each part of the biopsy section measured. For this final adjustment, the thickness of each section, generally 100 ± 1 μm, was verified at 10-15 sites, distributed both in cortical and cancellous bone tissues for each section. An advantage of this method is that all available bone tissue can be measured, thus reducing the error associated with selecting only a small sample area for analysis. The parameters measured in each group and reflecting the interindividual changes of DMB are MDMB, mean most frequent and highest DMB (DMB Freq. Max.), and mean interindividual heterogeneity index of the distribution of DMB (mean full width at half-maximum of the individual DMB curves). All the values are expressed in grams of mineral per cubic centimeter of bone.

Statistical analysis

Statistical analyses were based on a modified intent-to-treat (ITT) principle: the randomized, treated, and biopsied patients were analyzed according to the assigned treatments. For each quantitative parameter, separate two-sample tests were performed using the actual values to compare teriparatide and alendronate groups at each time-point and to compare the 6- and 18-month values within each treatment group. Monte Carlo methods were used to obtain estimates of the exact p values for each of the two-sample tests. For the qualitative study, the number of patients with woven bone, marrow fibrosis, or any other mineralization defect was tabulated. All analyses were performed using SAS software, version 8.0.

RESULTS

Baseline characteristics

Iliac crest bone biopsies were obtained from 42 patients (teriparatide, n = 21; alendronate, n = 21): 23 women at 6 months (teriparatide, n = 10; alendronate, n = 13); 14 women at 18 months (teriparatide, n = 6; alendronate, n = 8); 5 additional patients were measured at each time-point. Biopsy specimens adequate for quantitative analysis available from 17 patients at 6 months (teriparatide, n = 8; alendronate, n = 9) and 15 patients at 18 months (teriparatide, n = 8; alendronate, n = 7) were analyzed by 2D histomorphometry.

The biopsy samples collected at 6 months were obtained from sites in Mexico; ∼80% of the patients in the biopsy subset were Hispanic. Hispanic patients in the study were generally younger than the white patients. Thus, the subset of biopsy patients was significantly younger (63.2 ± 8.0 versus 66.6 ± 8.5 years) with fewer years past menopause (16.5 ± 8.8 versus 20.5 ± 10.1 years) than the subset of study patients who did not provide biopsies. The patients who were biopsied at 6 months were also younger and more predominantly Hispanic than the patients biopsied at 18 months. Within the biopsy subset, there were no significant differences in baseline characteristics between the teriparatide and alendronate treatment groups, including measures of BMD at the lumbar spine and femoral neck and bone remodeling activity, assessed by biochemical markers of bone turnover (Table 1).

Biochemical and densitometric changes after 18 months

The treatment effects on BMD (Table 1) in the biopsy subset were similar to those seen in the overall study.(30) BMD of the lumbar spine and proximal femur increased significantly from baseline with both teriparatide and alendronate treatment (Table 1). However, there was a significant difference between groups in the gain in BMD at the lumbar spine assessed by DXA and QCT (Table 1).

The pattern of response in markers of bone turnover in the biopsy subset was similar to the overall study population (Table 1; Fig. 1). There was a significant difference between groups in each of the markers. In the alendronate group, markers of bone formation (S-PINP) and resorption (U-NTx) decreased significantly by ∼70% after 18 months of treatment, comparable with the change in the alendronate group in the overall study. In patients treated with teriparatide, PINP was 135% above baseline after 18 months of treatment, whereas NTx increased by 32% from baseline.

Figure FIG. 1..

Percent change from baseline in markers of bone formation (PINP; ▪) and bone resorption (NTx; ○) in the alendronate (ALN) and teriparatide (TPTD) treatment groups of the biopsy subset. *p < 0.05; †p < 0.01; ‡p < 0.001 from baseline.

Qualitative assessment

There was no evidence of woven bone or primary mineralization defect in either treatment group at either 6 or 18 months. Single, small lymphoid nodes were observed in the marrow space of biopsies from two alendronate patients at 6 months and two teriparatide patients at 18 months. On gross examination, sections obtained from patients treated with teriparatide revealed extensive tetracycline labeling at 6 and 18 months (Figs. 2A and 2C), whereas labeling was sparse at both time-points in sections obtained from patients treated with alendronate (Figs. 2B and 2D).

Figure FIG. 2..

Tetracycline labels after 6 and 18 months in cancellous bone. Arrows indicate tetracycline labels and sites of new bone formation. (A) Six months, teriparatide. (B) Six months, alendronate. (C) Eighteen months, teriparatide. (D) Eighteen months, alendronate.

Quantitative assessment

There were a total of 32 biopsy specimens adequate for quantitative histomorphometric analyses: 17 specimens at 6 months (teriparatide, n = 8; alendronate, n = 9) and 15 specimens at 18 months (teriparatide, n = 8; alendronate, n = 7).

Cancellous bone formation

Most indices of cancellous bone formation (OS% BS, O.Th, OV/BV, MS/BS, Aj.AR, BFR/BS, Ac.f,) were significantly higher in the teriparatide group than in the alendronate group at both 6 and 18 months (Table 2; Fig. 3). MAR was significantly higher with teriparatide treatment compared with alendronate at 6 months but not at 18 months. There was no difference in W.Th between groups at either time period. In the teriparatide group, MAR, BFR/BS, and Ac.f were significantly higher at 6 months than at 18 months (Table 2; Fig. 3).

Table Table 2.. Quantitative Findings in Cancellous Bone Envelope From Biopsies Obtained After 6 or 18 Months of Teriparatide (TPTD) or Alendronate (ALN) Treatment
original image
Figure FIG. 3..

Activation frequency at 6 and 18 months in the alendronate (ALN) and teriparatide (TPTD) treatment groups of the biopsy subset. TPTD significantly greater than ALN at 6 and 18 months (p < 0.01). TPTD: 6 months significantly >18 months (p < 0.05). Reference values for untreated postmenopausal women with osteoporosis (mean ± SD).(40)

Cancellous bone resorption

Most indices of cancellous bone resorption were significantly higher in the teriparatide group compared with alendronate treatment at 6 months but not at 18 months (Table 2). Whereas eroded surfaces in the two treatment groups were similar at both time-points, volume-based resorption was significantly higher at 6 months in the teriparatide group than in the alendronate group (Table 2). However, maximum erosion depth did not exceed wall thickness in either treatment group. At 18 months, there were no differences in resorption parameters between treatment groups. Indices of cancellous bone resorption were relatively unchanged from 6 to 18 months in the teriparatide group, whereas most indices of cancellous bone resorption tended to be higher at 18 months compared with 6 months in the alendronate group, although only maximum erosion depth reached significance (Table 2).

Cancellous bone structure

There were no significant treatment effects for either treatment regimen on structural parameters in cancellous bone at either 6 or 18 months of treatment (Table 2).

Endocortical bone formation

After 6 months of treatment, mineralizing surface (MS/BS) and surface-based bone formation rate (BFR/BS) were significantly higher in the teriparatide group than in the alendronate group (Table 3). MS/BS and BFR/BS remained significantly greater in the teriparatide group compared with alendronate at 18 months.

Table Table 1.. Baseline Demographics and Treatment Effects on BMD and Markers of Bone Turnover for Patients in the Biopsy Subset
original image
Table Table 3.. Quantitative Findings in the Endosteal and Intracortical Bone Envelopes From Biopsies Obtained After 6 or 18 Months of Teriparatide (TPTD) or Alendronate (ALN) Treatment
original image

Endocortical bone resorption

Erosion surface (ES/BS) was significantly higher in the teriparatide group than in the alendronate group at 6 months but not at 18 months (Table 3).

Cortical bone structure

There were no significant differences between groups in cortical thickness or porosity at 6 months, although cortical porosity was significantly greater with teriparatide than with alendronate treatment at 18 months (Table 3), consistent with the increase in bone remodeling. Cortical porosity was reduced by one-third in the teriparatide group between 6 and 18 months, whereas it remained unchanged in the alendronate group (Table 3).

Mineralization parameters

There were no significant differences between groups in mineralization parameters in cortical, cancellous, or total bone at either 6 or 18 months (Table 4). The degree of mineralization was significantly lower at 18 months compared with 6 months for cancellous, cortical, and total bone without changes in the heterogeneity index in either the teriparatide and alendronate groups (Table 4).

Table Table 4.. Effects of Teriparatide (TPTD) and Alendronate (ALN) on the Parameters Reflecting the Degree of Mineralization of Bone (DMB) Measured by Quantitative Microradiography
original image

DISCUSSION

This cross-sectional histomorphometric study, the first to directly compare the effects of alendronate with teriparatide in a randomized design, confirmed that the increase in BMD is achieved by opposite effects on remodeling at the bone tissue level. In addition, this is the first report to examine, in a cross-sectional design, these agents at two different time periods during treatment. The stimulation of osteoblastic activity by teriparatide, suggested by the higher histomorphometric levels of cancellous and endocortical bone formation observed in bone biopsy samples after treatment for 6 and 18 months in these postmenopausal women with osteoporosis, is consistent with results from previous histomorphometric studies.(20,33,37,38) The inhibition of osteoclast activity by alendronate, suggested by the lower levels of bone remodeling and activation frequency found in this cross-sectional study, has been previously reported.(14,15,39)

The lower levels of histomorphometric indices of bone remodeling in alendronate-treated patients at 6 months were maintained through 18 months of treatment. These results confirm the distinct and opposing mechanisms of action at the BMU level for teriparatide and alendronate proposed by McClung et al.(30) suggested by changes in biochemical markers of bone turnover.

Histomorphometric values for untreated postmenopausal women with osteoporosis reported by the same laboratory(14,40) provide a reference point to interpret the findings in this cross-sectional study that did not include a placebo-treated group. Bone formation rate at the tissue level in the teriparatide group, as reflected in BFR/BS, was ∼150% of that compared with untreated postmenopausal women. In comparison, the alendronate-treated BFR/BS in this study was comparable with previously reported levels, which were 10-fold lower than the untreated women.(14) MAR and osteoid thickness, which reflect activity of single osteoblasts,(6) were also significantly higher in the teriparatide group compared with the alendronate group at 6 months, further highlighting the osteoblastic stimulation at the cellular level induced by teriparatide. The higher level of bone turnover with teriparatide treatment was reflected in the significantly higher mineralizing surface and activation frequency compared with placebo-treated patients.(14) Increased turnover paired with a positive BMU balance secures maximal accretion of new bone, because the number of remodeling units laying down new bone is increased. Our results found that biochemical markers of bone turnover reach maximal stimulation of bone formation after 6 months of teriparatide treatment, with a gradual decrease with continued treatment. The analyses of histomorphometric bone formation indices shown corroborate these findings.

Histomorphometric parameters from biopsies obtained at 6 and 18 months of teriparatide treatment, although from different patients, provided an opportunity to explore the responsiveness in dynamic histomorphometric parameters. Several indices of bone formation, including osteoid surface, mineralized surface, mineral apposition rate, adjusted apposition rate, bone formation rate, and activation frequency were ∼50% lower at 18 months compared with 6 months. Nevertheless lumbar spine and hip BMD continue to increase with teriparatide treatment beyond the point where dynamic histomorphometric indices begin to decrease.(22,29) Teriparatide leads to new bone tissue formation, which is not yet fully mineralized. Thus, BMD may not be the most appropriate measure to use in evaluating the response to therapy or in comparing bone formation agents with antiresorptive therapies.(41)

Previous histomorphometric studies have shown greater wall thickness in sections from patients treated with teriparatide.(37,38,42,43) In this smaller, active-control study, however, we were unable to show such differences. Using the classical method of measuring wall thickness used in this study, we were not able to detect the overfilling of resorption lacunae reported in other studies,(37,38,42) resulting in a probable underestimation of wall thickness and positive remodeling balance. In addition, wall thickness measurements are composed of existing thickness plus the addition of thickness developed with treatment. Our inability to detect a difference in cancellous bone volume, previously reported by Dempster et al.(37) and Jiang et al.,(33) show the limitations of assessing structural parameters by histomorphometry in small cohorts. The inability to detect improved trabecular structure as reflected in trabecular separation, number, and thickness is likely the combination of factors including a small sample size in an active-control cross-sectional study.

Bone resorption was greater in the teriparatide group, corroborating the increases in bone resorption markers reported previously.(21,37) The delay in increased bone resorption in the teriparatide group, evident in previous studies on bone marker changes after teriparatide, was also shown in this smaller cohort. As was the case for bone formation, there was a trend for bone resorption to decrease from 6 to 18 months, which parallels the changes seen for biochemical markers of bone resorption markers in this and other studies.(20–22)

The trend for cortical porosity to be greater in the teriparatide group relative to the alendronate group at 6 months reached significance at 18 months. Cortical porosity was approximately a third lower at 18 months compared with 6 months in the teriparatide group. This pattern was also seen in activation frequency, which was significantly lower (∼50%) at 18 months compared with 6 months. This dynamic effect on bone remodeling and the consequent effect on cortical porosity may explain the early decrease and subsequent increase in BMD at the more predominantly cortical regions in patients treated with teriparatide that was observed in several studies.(44–46)

The increases in MAR and the pronounced increase in turnover at the endosteal envelope as reflected in mineralizing surface, paired with the lack of observed labeling at the periosteum, suggest that the accretion of new cortical bone takes place primarily at the endosteum of the iliac crest.

Variability in periosteal surface may be related to skeletal site, sex, and age.(47,48) Previous studies(33,37) have shown that increased cortical thickness and cortical thickening are observed on most μCT images of bone biopsies obtained from patients treated with teriparatide. The absence of differences in cortical thickness with treatment in this study may be a result of the wide variability in this measurement. Because biopsy samples were not obtained at baseline for comparison purposes, it is possible that changes may have occurred within the first 6 months of treatment.

The concept of opposite mechanisms of action of antiresorptive and bone formation agents for the treatment of osteoporosis has been proposed previously.(29,31,49) Investigators have suggested that the gain in BMD with alendronate and other antiresorptive agents may be achieved by a reduction in remodeling spaces, that is, reducing bone turnover without a true stimulation of bone formation(39) or by an improvement in the mean degree of mineralization.(14,50) Teriparatide stimulates bone remodeling, with direct stimulation of bone formation. Because both agents lead to significant increases in BMD, the concept of the opposing mechanisms of action may be better understood by the response of biochemical markers of bone turnover and histomorphometric parameters of bone formation.

There are several limitations to consider in this study. Because of the cross-sectional study design, we were not able to make a comparison with a true change from baseline with treatment. The primary intent of this study was to profile the dynamic histomorphometric parameters after 6 or 18 months of treatment. Although the overall number of biopsies was comparable to previous studies, there were a comparatively small number of samples after allocation to two different treatment groups across two time periods. The inability to detect significant differences in static parameters of structure, mineralization, and wall thickness may be related to the rather small number of biopsy samples in each treatment group and time period and the lack of baseline values. The relatively short duration of the study, the lack of baseline samples and a placebo group, and the timing of the biopsies at 6 and 18 months may in part account for the unexpected results in mineralization in the alendronate group, contrasted with the expected decrease in the teriparatide group. We note that the DMB value measured at 6 months in the alendronate group is similar to the value obtained after 2 years of alendronate treatment,(15) but the decrease of this DMB between 6 and 18 months remains unexplained.

This study has shown the pronounced differences in mechanism of action between a potent antiresorptive agent like alendronate and a bone-forming agent like teriparatide. Both drugs increase bone mass, but do so by different and opposite mechanisms of action. The increase in BMD obtained with antiresorptive therapies is not the result of new bone being formed, but rather results from the preservation of existing bony structures, a closing down of the remodeling space, and increased mineralization of these structures over time. This explanation is corroborated by the pronounced reduction in bone turnover, equal to one remodeling cycle passing a point on the bone surface every 60 years. Because of the low bone formation rates in the alendronate-treated patients, a positive bone balance at each BMU would significantly increase the time to impact bone structure.(51) In contrast, teriparatide causes the opposite changes in bone remodeling with an increase in bone formation rates and increased turnover, securing a large number of active BMU laying down new bone. Teriparatide increases bone formation at both cancellous and endosteal envelopes. Stimulation of bone remodeling by teriparatide at both the cancellous and endosteal envelopes reaches a maximum after 6 months of treatment. Subsequently, bone turnover returned toward levels measured in untreated postmenopausal women, with formation still exceeding resorption.(40) The reduction in bone remodeling with alendronate was maintained at a constant level.

For many years, the traditional treatment of osteoporosis has been antiresorptive agents, which decrease remodeling spaces in size and number by a reduction in bone turnover.(39) Our results show that teriparatide, a bone-formation agent available for the management of osteoporosis, offers a mechanism for increasing bone mass by increasing bone turnover with preferential stimulation of bone formation over resorption.

Acknowledgements

The authors thank Jean-Paul Roux and Brigitte Burt-Pichat for technical assistance with the biopsy analyses and Catherine Simi and Delphine Farlay for the measurements of the parameters reflecting the degree of mineralization of bone. The authors thank Mary Ellen Perron for graphics. This study was funded by Eli Lilly and Company.

Ancillary