Osteoanabolic and dual action drugs

Abstract

Teriparatide (TPTD) and abaloparatide (ABL) are the only osteoanabolic drugs available, at this time, for treatment of osteoporosis. TPTD is a 34‐amino acid fragment that is identical in its primary sequence to the 34 amino acids of full‐length human parathyroid hormone [hPTH(1‐84)]. ABL is identical to parathyroid hormone‐related peptide (PTHrP) through the first 22 residues with significantly different amino acids inserted thereafter, between residues 22 and 34. The osteoanabolic actions of PTH are due directly to its effects on cells of the osteoblast lineage and indirectly by stimulating IGF‐I synthesis and suppressing sclerostin and associated enhancement of Wnt signalling. Both TPTD and ABL are ligands that bind to and activate the PTH receptor type 1 (PTHR1) receptor but they appear to do so differently: ABL favours the transient, more anabolic configuration of the receptor. Both TPTD and ABL reduce the risk of vertebral fractures and non‐vertebral fractures. Both drugs are administered for a maximum of 24 months, and should be followed by an antiresorptive agent to maintain gains in bone mineral density (BMD). Romosozumab, a monoclonal antibody that binds to and inhibits sclerostin, appears to have dual actions by stimulating bone formation and reducing bone resorption. In the pivotal clinical trial, romosozumab, administered as a 210 mg monthly subcutaneous dose, significantly reduced new vertebral fractures and in a subsequent study reduced both vertebral and non‐vertebral fractures.

PTH and PTHrP analogues

In the 1990s Dobnig et al. highlighted the dual properties of PTH to harbour both anabolic and catabolic action on bone 1. Rats administered teriparatide [PTH (1‐34)] once daily (i.e. intermittent administration) enhanced osteoblastic activities while teriparatide administered by infusion pump over 24‐h periods (i.e., chronic administration) enhanced osteoclast activities. This classic demonstration delineated a key element of the potential of PTH analogues to be either anabolic or catabolic as dependent on the length of exposure of the peptides. Brief intermittent exposure would highlight anabolic properties of PTH, its peptides and analogues, while continuous, prolonged exposure would highlight catabolic properties 2, 3. PTH can be anabolic by virtue of stimulating bone remodelling (bone formation on active resorbing surfaces) or by bone modelling (bone formation on quiescent surfaces). Another proposed mechanism of PTH's anabolic effects has been called ‘overflow remodelling’ by which the resorptive pits, where remodelling takes place, are not only filled by the actions of the osteoblast but extend beyond the confines of the bone remodelling unit, spilling onto quiescent surfaces. The cellular mechanisms for these different descriptors of osteoanabolic actions are focused upon the osteoblast. This bone‐forming cell is stimulated to proliferate, to live longer (anti‐apoptotic) and to enhance itself by PTH's effect to convert quiescent lining cells into osteoblasts. These cellular actions include stimulation of osteoblast lineage cells directly and by indirect actions to induce IGF‐I synthesis and to suppress sclerostin, the latter effect of which enhances the canonic Wnt signalling pathway 4. An additional molecular switch is suppression of Notch signalling, an inhibitor of osteoblast differentiation 5. Through these various cellular and molecular mechanisms, PTH enhances osteoblastogenesis and increases the osteoblast cell pool, major elements of the osteoanabolic effect of PTH.

Knowledge of these differential properties led to the development of teriparatide for the treatment for osteoporosis. Teriparatide (TPDT) is the amino‐terminal, 34‐amino acid fragment that is identical in primary sequence to this region of the full‐length 84‐amino acid hormone. Parathyroid hormone‐related peptide (PTHrP) shares intense primary sequence homology to PTH over the first 15 amino terminus residues and maintains three‐dimensional function homology through residue 34. Both PTH and PTHrP bind to the same G protein‐coupled receptors. PTH receptor type 1 (PTHR1) is the receptor that mediates the skeletal effects of PTH, PTHrP and analogues of PTHrP. Abaloparatide (ABL), a new drug for osteoporosis, is identical to PTHrP through the first 22 residues but, thereafter, several substituent amino acids render this molecule very different in terms of its interactions at the receptor level 6. Both TPTD and ABL act on the same receptor but in different ways 7, 8. The difference in activity between the two molecules may be explained, at least in part, by the way in which these two peptides interact at the two conformational configurations of the PTHR1 receptor, namely R0 and RG 8. The affinity with which a ligand interacts at one or the other of these two receptor conformations may well help to describe the extent to which the ligand is anabolic or catabolic. Clearly, both teriparatide and abaloparatide are osteoanabolic but they appear to have different affinities for these two receptor conformations 9. Affinity for the R0 conformation is associated with longer lived effects (catabolic), while those that bind more efficiently to the RG conformation are associated with short‐duration (osteoanabolic) responses 10. Both teriparatide and abaloparatide have greater affinity for RG than for R0 but the differences in relative affinities are marked: 1600‐fold for abaloparatide and threefold for teraparatide 11, 12. A consequence of the more prolonged interaction with the R0 state of the PTHR1 receptor is internalization to intracellular compartments, where the bound ligand behaves potentially as a persistent and active ternary complex 13. Such a mechanism could explain the catabolic actions of PTH‐related ligands where internalization is associated with persistent cAMP generation and a greater catabolic response. In contrast, with only transient effects on cAMP generation, PTHrP and its associated ligands are less likely to be catabolic 8. While these different mechanisms are plausible and attractive, more studies are required to substantiate these hypothetical explanations.

Teriparatide

Teriparatide is the first approved anabolic, or bone building drug approved for the treatment of osteoporosis 14. The pivotal trial (Fracture Prevention Trial) 15 included 1637 postmenopausal women with prior vertebral fractures randomized to receive 20 or 40 mcg of PTH (1‐34) or placebo. The study showed that in postmenopausal osteoporosis, PTH (1‐34) decreases the risk of vertebral and non‐vertebral fractures. PTH (1‐34) at either dose reduced the risk of one or more new vertebral fractures by 65 and 69%, respectively; the risk of two or more fractures by 77 and 86%, respectively, and the risk of at least one moderate or severe vertebral fracture by 90 and 78%, respectively. Treatment with TPTD also reduced the total number of vertebral fractures. Increments in BMD were dose‐related at the spine and hip regions as well as in total‐body bone mineral. Modest declines in BMD at the forearm site were seen with the higher 40 μg dose. This peptide was associated with only minor adverse events (occasional nausea and headache). The lower TPTD dose of 20 mcg daily was approved because the 40 mcg dose was not more efficacious and was associated with more adverse events (e.g. hypercalcaemia).

The vertebral fractures of the pivotal trial were re‐analysed using a combination of quantitative morphometry (QM) and visual semiquantitative analysis (SQ) to define vertebral fractures. By these methodologies, vertebral fracture risk was reduced in the TPTD versus placebo group by 84%. The risk of two or more vertebral fractures was also significantly reduced by 94%. The absolute benefit of TPTD was greatest in those patients with the highest number and severity of prevalent vertebral fractures 16. After stopping study drug, 1262 patients were enrolled in a follow‐up study and lateral spine radiographs were repeated 18 months later 17. During the follow‐up period, other osteoporosis drugs were used by 47% of women, with greater use in the group that had been assigned to placebo. Fracture protection was more evident in those who had previously been treated with TPTD. The reduction in vertebral fracture risk associated with previous treatment with TPTD 20 μg day⁻¹ was 41% during the follow‐up study versus 13% in the group that had received placebo. Post hoc analysis also suggested that TPTD treatment substantially reduced the increased risk of subsequent fracture in women who had sustained a fracture during the study. The reduction of vertebral fracture by TPTD administration persists for at least 18 months after discontinuation of therapy. These results indicate that anabolic therapy with TPTD may alter the natural history of the progression of osteoporosis 18. The design of these analyses suggests the need for follow‐on therapy with an antiresorptive after teriparatide is stopped. Several large observational studies of osteoporotic patients treated with TPTD 20 μg day⁻¹ have confirmed its efficacy in the reduction of vertebral and non‐vertebral fractures 18, 19, 20, 21. A meta‐analysis 22 of eight randomized controlled trials (n = 2388) reported that teriparatide is associated with an increase in bone density of 8.14% in the spine and 2.48% at the hip. In trials that reported fracture as an outcome, treatment was associated with a 70% risk reduction in vertebral fractures and 38% risk reduction in non‐vertebral fractures.

Teriparatide vs. antiresorptives

The safety and efficacy of TPTD versus alendronate was analysed in a meta‐analysis of six trials involving 618 patients 23. The study demonstrated a significant increase in lumbar spine BMD, but not femoral neck BMD, in postmenopausal osteoporosis patients treated with TPTD compared with alendronate for 6–18 months. TPTD was not superior to alendronate in reducing fracture risk. Saag et al. compared TPTD with alendronate 24, 25 in glucocorticoid‐induced osteoporosis. The mean BMD increased to a greater extent in patients receiving TPTD than in those receiving alendronate. There was also a significant reduction in vertebral fractures when subjects who received TPTD were compared with those who were treated with alendronate.

The recently published VERO trial 26 is a double‐blind, double‐dummy trial, comparing TPTD 20 mcg daily vs. 35 mg of oral risedronate weekly. The study showed that the risk of new vertebral and clinical fractures is significantly lower in patients receiving TPTD than in those receiving risedronate. Previously, Hadji et al. 27 studied the effect of TPTD and risedronate on back pain and, although there were no differences in back pain‐related endpoints, patients receiving TPTD had greater densitometric benefits than those receiving risedronate. After 18 months, patients treated with TPTD showed a greater increase in lumbar spine and femoral neck BMD compared to risedronate. Meta‐analysis comparing TPTD and bisphosphonates has shown that TPTD increased the BMD of the lumbar spine, femoral neck and total hip to a greater extent than bisphosphonates 28, 29. However, no difference in risk of non‐vertebral fractures (and adverse events) was found 29.

The efficacy of TPTD vs. denosumab (Dmab) was evaluated in the DATA trial 30. In this randomized, controlled trial, postmenopausal women were assigned in a 1:1:1 ratio to receive 20 μg TPTD daily, 60 mg Dmab every 6 months, or both. After 12 months of treatment, mean spine BMD had increased significantly in all treatment groups, but increased significantly more in the combination‐therapy group than in the TPTD or Dmab groups (see below). Increase in BMD of the spine and the changes in femoral neck BMD did not differ significantly between the TPTD and Dmab groups. Femoral‐neck BMD increased more in women who received combination therapy than in those who received Dmab alone or TPTD alone. With regard to the monotherapeutic comparisons, TPTD vs. Dmab, there were no differences in BMD improvement between the two drugs.

Combination therapy

The availability of anabolic treatment led to hypotheses related to the possibility that combination therapy with anabolic and antiresorptive drugs would provide even greater benefits than either drug alone. The DATA trial, described above, is one of the latest attempts to study this issue.

Finkelstein et al. 31 randomized post‐menopausal women to receive alendronate alone (10 mg orally once daily, n = 31), TPTD alone (40 mcg subcutaneously once daily, n = 31), or both (group 3, n = 31). Spine, femoral neck and total hip BMD increased more in women treated with TPTD alone compared with combination therapy. In this study, alendronate reduced the ability of TPTD administration to increase BMD in postmenopausal women.

Cosman et al. 32 compared a single intravenous infusion of 5 mg zoledronic acid plus daily subcutaneous TPTD 20 mcg via monotherapy with zoledronic acid or TPTD. The study showed that at the 12‐month point, combination therapy was not more effective in increasing bone density at the spine or the hip regions than one of the monotherapeutic arms, but that with regard to increases in bone density at both lumbar spine and hip sites, combination therapy was advantageous.

In a double‐blind, oral placebo‐controlled, 6‐month trial, TPTD alone and combination TPTD and raloxifene were analysed 33. Lumbar spine, femoral neck and total hip BMD increased significantly in the combination group, and the increase in total hip BMD was significantly greater than in the TPTD group. Increases in lumbar spine and femoral neck BMD were not statistically different between treatment groups. The authors concluded that the combination treatment with raloxifene may enhance the bone‐forming effects of TPTD. This trial was small and, although promising, has not been followed up with a larger, rigorously designed trial.

A meta‐analysis of seven trials, which aimed to determine whether the concomitant combination therapy of anabolic agents and bisphosphonates produces greater effects on BMD than anabolic agents alone 34, showed that concomitant combination therapy significantly improved the BMD at the total hip and femoral neck with a shorter term (6–12 months) and produced similar benefits on BMD for the longer term (18–24 months). Another meta‐analysis confirmed that there was no evidence for the superiority of combination therapy, although significant change was found for hip BMD at 1 year in combination groups 35.

Other protocols tested the possibility that combination therapy could be enhanced if the two drugs were administered at different time points. Muschitz et al. 36 suggest that alendronate added to TPTD 9 months after initiation of TPTD results in a more robust increase in BMD (see below). Cosman et al. 37 suggested a regimen of three‐month cycles of parathyroid hormone alternating with three‐month cycles without parathyroid hormone in patients treated continuously with alendronate. Compared to alendronate plus PTH (1‐34) subcutaneously daily, the cyclic regimen stimulated the early phase of action of PTH (characterized by pure stimulation of bone formation) and not the later phase (activation of bone remodelling).

The DATA trial 30 and DATA extension trial 38 showed that combined TPTD and Dmab treatment increases BMD at the spine, hip and femoral neck in postmenopausal women with osteoporosis to a greater extent than either agent alone. Moreover, the BMD changes in the combined therapy group were greater than have been reported with any approved therapies.

The mechanisms that lead to the greater improvement of BMD are not clear. Dmab binds to and inactivates RANK‐L (receptor activator of NFkappaB ligand), a cytokine that stimulates osteoclast action and development. PTH and its analogues activate, as part of their mechanism of action, osteoclasts through RANK‐L 39. An attractive hypothesis is that by inhibiting RANK‐L, Dmab shifts endogenous PTH, which is increased by Dmab, towards its anabolic pathway. Thus, combination therapy with Dmab and TPTD would take advantage of Dmab's antiresorptive mechanism of action while amplifying teriparatide's anabolic pathway 40. There was also improvement in some microarchitectural indices with combination therapy as compared with monotherapy with TPTD or Dmab 41. For example, total volumetric bone mineral density at the radius and tibia significantly increased in the combination group compared to both TPTD and Dmab groups. Cortical porosity at both the radius and tibia was stable in the combination group, but increased progressively over the 24‐month treatment period in the teriparatide group. Bone strength estimated by finite element analysis improved in the combination group at both the radius and tibia. These observations show that combination treatment with Dmab and TPTD improves bone quality, in particular in the cortical bone, and may be especially attractive in patients who are at high fracture risk.

Teriparatide after antiresorptives

Many patients starting TPTD therapy have previously been treated with antiresorptives. In studies 42, 43 that have evaluated TPTD after alendronate or risedronate, differences in anabolic responsiveness to TPTD as a function of the specific kind of prior bisphosphonate exposure were noted. Prior treatment with alendronate appears to delay increases in BMD, particularly in the first 6 months. Cosman et al. 44 evaluated postmenopausal women with osteoporosis on alendronate or raloxifene therapy, added TPTD (Add groups) or switched to TPTD (Switch groups) for 18 months. In women with osteoporosis treated with antiresorptives, greater BMD increases were achieved by adding TPTD to the therapy.

Boonen et al. 45 evaluated 245 patients in the EUROFORS study who were treated with TPTD for between 12 months and 24 months after exposure to an antiresorptive drug for a minimum of 12 months. Lumbar spine BMD increased and hip BMD transiently declined in all groups. There were no differences in BMD responses among all previous antiresorptive groups. Duration of previous antiresorptive therapy and lag time between stopping previous therapy and starting TPTD did not affect the BMD response at any skeletal site 46.

Different results were found in patients previously treated with Dmab. In the DATA switch trial, postmenopausal osteoporotic women switching from TPTD to Dmab experienced impressive continued increases in BMD at both the lumbar spine and hip regions. Patients switching from Dmab to TPTD experienced initial declines in BMD the magnitude of which was dependent on the skeletal site. At the lumbar spine, the decline was minimal and transient. At the hip, the decline was more substantial but also readily reversed over time. Changes at the distal forearm site have led to some concerns about this sequence because BMD fell impressively when TPTD followed Dmab 47. The HRpQCT evaluation showed that switching from Dmab to TPTD is associated with a reduction in total and cortical volumetric BMD, thickness and estimated strength at tibia and radius. In contrast, switching from TPTD to Dmab is associated with an improvement in these parameters 48.

There are situations in which TPTD is an attractive approach to those who have previously been treated with an antiresorptive agent. These situations include the following: an inadequate response; fragility fracture on therapy; intolerance; or transition after long‐term antiresoprtive therapy. In all these situations, TPTD may be appropriately considered. The transient declines in BMD need to be recognized but do not contraindicate the use of TPTD in these settings. Adding TPTD to those on antiresorptive therapy may also be considered in particular clinical situations (e.g. patients with a recent hip fracture) 49.

Therapeutic decisions after 24 months of TPTD

When TPTD is stopped after 24 months, as generally required, measures are needed to maintain gains because BMD will fall rapidly if an antiresorptive does not follow. Bisphosphonates after TPTD withdrawal helps to maintain gains in BMD at the lumbar spine 50, 51, 52. The DATA‐switch trial showed further impressive increases in BMD after transitioning from TPTD to Dmab 47.

Anabolic window

TPTD increases bone density and bone turnover, improves microarchitecture, and changes bone size 53. Bone formation markers show a rapid increase, followed by increases in bone resorption markers. In particular, during the first month of treatment, there is an increase predominantly in the markers of bone formation, suggesting that bone formation exceeds bone resorption early during the treatment course 54. Thereafter, bone resorption markers rise, signalling an indication of bone remodelling. Serum levels of PINP and CTX peak after 6–12 months of TPTD therapy, followed by a gradual decrease in both markers 18. The sequence of changes in these bone turnover markers suggest an interesting therapeutic time course. PTH seems to initially stimulate processes associated with bone formation (bone modelling), followed several months later by a more prominent promotion of bone remodelling. This sequence of events has led to the concept of the anabolic window, the period of time when TPTD is maximally anabolic (Figure 1) 55, 56. There are different types of bone modelling, as shown by histomorphometric evaluation. If there are smooth cement lines underlying bone formation sites, modelling‐based formation is the predominant mechanism. If there are scalloped cement lines, activities at those sites are believed to be due to remodelling‐based formation. TPTD stimulates processes associated both with bone modelling and remodelling. There is even another mechanism that may help to explain the anabolic actions of PTH. After 3 months of treatment, evidence for bone formation can be seen directly from the resorption pits, now filled, onto adjacent quiescent bone surfaces 57. Microcomputed tomography (CT) of bone biopsy in patients treated with TPTD revealed that PTH treatment exerts anabolic action on cortical bone without increasing cortical porosity and can improve cancellous bone microarchitecture 58.

Figure 1.

Anabolic windows for teriparatide, abaloparatide and romosozumab. The windows represent the conceptual idea of the time period in which these drugs exert their maximum osteoanabolic effect

Safety

Teriparatide is well tolerated. Adverse events observed in patients include nausea, headache, dizziness and leg cramps 15. Both PTH and PTHrP analogues cause osteosarcoma in rats when administered in high doses for 18–24 months 59, 60, 61. For this reason, all PTH and PTHrP‐related drugs approved for human use [TPTD, ABL and rhPTH (1–84)] carried with them an Food and Drug Administration (FDA)‐mandated ‘black box’ warning. During the 16‐year experience with TPTD, no osteosarcoma toxicity signals have emerged and well over 2 million human subjects have been treated, to date. The post‐marketing study of safety has shown that the risk of osteosarcoma in humans did not differ from that in the general population 62, 63. The animal toxicity studies employing the non‐human primate cynomolgus monkey did not find any suggestion of osteosarcoma 62. Nevertheless, TPTD is not recommended in patients with any of the following conditions or history: Paget's disease of bone, unexplained alkaline phosphatase elevations, prior skeletal radiotherapy, primary or metastatic bone malignancy, a growing skeleton, or hypercalcaemic disorders, such as primary hyperparathyroidism 64. Limitations of TPTD include a requirement for refrigeration, daily self‐administered injections and cost.

Abaloparatide

Abaloparatide was approved in April 2017 by the FDA for therapy of postmenopausal women with osteoporosis at high risk for fracture. The FDA's approval of ABL was based on results at 18 months from the ACTIVE trial 65. ACTIVE was an international, randomized, placebo‐ and active‐controlled trial including postmenopausal women with osteoporosis. Women were randomized 1:1:1 to receive daily subcutaneous injections of ABL 80 μg, or matching placebo, or TPTD, 20 μg. The TPTD arm was open‐label but the ABL vs. placebo arms were double‐blinded. ABL significantly reduced the risk of new vertebral and non‐vertebral fractures as compared to placebo. Treatment with ABL, compared with both the placebo and TPTD groups, was also associated with higher gains in BMD, particularly at total hip and femoral neck. Reduction in the risk of major osteoporotic fracture (i.e. proximal humerus, wrist, hip or clinical vertebral fractures) or any clinical fracture in postmenopausal women at high risk of fracture, was significantly lower than placebo and was independent of baseline fracture probability assessed by FRAX 66. According to the original design of this study, after 18 months, and a 1‐month reconsent period, study subjects who were enrolled in either the ABL or placebo arms of the study were followed with 24 months of open label alendronate. The endpoint for the primary analysis of ACTIVExtend was 6 months with results after 24 months considered as exploratory endpoints. Following 18 months of ABL and 6 months of alendronate 67, fracture risk reduction was maintained: relative risk reduction of 87% for vertebral fracture and 52% for non‐vertebral fracture. Recently, the 24‐month experience with alendronate following 18 months of ABL was reported and continues to show risk reduction for vertebral, non‐vertebral, clinical and major osteoporotic fractures. There was an 84% reduction in relative risk for new vertebral fracture. BMD also increased further during the entire ACTIVExtend trial period 68.

Anabolic window

The concept of the anabolic window with osteoanabolic therapy is worth revisiting in the context of ABL because the patterns and kinetics of change in bone turnover markers differ from those of TPTD. The marked increase in bone formation markers occurs with the same rapid time course as with TPTD but the magnitude of the increase is somewhat less. After 1 month, the increase in P1NP was the same for both anabolic agents but, by 3 months, P1NP start to decrease with ABL treatment (P = 0.02), and remains lower during the following 18 months as compared to TPTD (P < 0.01). The major difference is seen in a much smaller and transient increase in the bone resorption marker CTX (P < 0.001 after 3 months compared to TPTD). These results suggest that there is a wider ‘anabolic’ window when ABL is compared to TPTD (Figure 1) 65. Both static and dynamic histomorphometric variables from bone biopsy performed in the ACTIVE trial show few differences among ABL, TPTD and placebo treatment groups 69. This bone biopsy study was conducted for safety and in this regard, there were no signals of concern. Unfortunately, the bone biopsy study was not designed to make comparisons among the three groups and, without baseline biopsy data, no conclusions with regard to potential differences among the groups can be reached. Histomorphometric data in monkeys showed that teriparatide increased cortical porosity without decreasing bone strength, while in the same animal model, abaloparatide did not increase cortical porosity 70, 71.

Safety

In the Phase 2 and 3 clinical trials, ABL showed a good safety profile and was well tolerated. In comparison with the TPTD arm, palpitations and nausea might have been somewhat greater with ABL 11. On the other hand, TPTD was associated with a statistically significantly higher incidence of hypercalcaemia than ABL: 6.4% vs. 3.4%. As noted earlier, the rat toxicity studies confirm that this analogue of PTHrP, like all forms of PTH and PTHrP, will cause osteosarcoma in rats when administered in high doses for prolonged periods of time. Also, like all forms of PTH and PTHrP that have been studied in the cynomolgus monkey, this toxicity is not seen.

The differences in the skeletal metabolism between rats, monkeys and humans may explain the manifestation of osteosarcoma only in rats. The adult human skeleton is characterized by remodelling and not bone growth. In contrast, skeletal growth occurs in rats for their entire lives. Furthermore, 2 years of treatment represents 80–90% of rats’ life span, compared to less than 2–3% of the life span for most postmenopausal women and men exposed to teriparatide for the same period of time. Finally, in rats the lack of Haversian systems determines a different modelling process compared to humans 72. Altogether, these differences help to explain the exaggerated anabolic response that occur in the cortical bone of rats and not in monkeys or in humans 73.

An advantage of ABL is that it does not require refrigeration after the first dose from the device. Refrigeration is needed prior to first use from the device.

Dual actions drug: romosozumab

The glycoprotein sclerostin, is a pivotal regulator of bone formation. It is a product of the osteocyte, responding to biomechanical stress. Sclerostin is an antagonist of the anabolic Wnt/β‐catenin signalling pathway in the osteoblast by interposing itself been Wnt ligands such as Wnt10 and its receptors such as LRP5/6 and the co‐receptor Frizzled. This inhibitory effect has cytoplasmic consequences that lead to the intracytoplasmic degradation of beta‐catenin, a key transcription factor. When sclerostin is inhibited or genetically absent (e.g., van Buchem disease, sclerosteosis), Wnt ligands engage more successfully at LRP5/6 receptor sites, which lead to the intracellular accumulation of beta‐catenin, the translocation into the nucleus and stimulation of bone formation (Figure 2) 74, 75, 76, 77. The therapeutic implications of sclerostin inhibition became attractive when it was observed that heterozygotic carriers of either the genes for van Buchem disease or sclerosteosis also experience high bone mass, as in the homozygous disease, but without the devastating neurological consequences that are seen in the full‐blown disorders.

Figure 2.

Sclerostin and Wnt/β‐catenin pathways (adapted from 74)

The sclerostin inhibitor being developed at this time is romosozumab, a human monoclonal antibody. With further experience from animal and early human studies, it became apparent that this antisclerostin agent not only stimulates bone formation, by the mechanism described above, but also leads to inhibition of bone resorption through OPG 78, 79. Indeed, romosozumab lead to an increase in OPG, thus reducing the RANKL‐OPG ratio and osteoclast activity 80. Thus, romosozumab can be best characterized as a dual agent, serving both osteoanabolic and antiresorptive functions.

The efficacy and safety of romosozumab was evaluated in a Phase 2, randomized, placebo‐controlled, study. A total of 419 postmenopausal women were randomized to receive subcutaneous romosozumab monthly (at a dose of 70 mg, 140 mg, or 210 mg) or every 3 months (140 mg or 210 mg), subcutaneous placebo, or an open‐label active comparator – oral alendronate (70 mg weekly) or subcutaneous TPTD (20 μg daily) for 12 months. The primary endpoint was the percentage change from baseline in BMD at the lumbar spine at 12 months. All dose levels of romosozumab were associated with significant increases in BMD at the lumbar spine, as compared with a decrease of 0.1% with placebo. The 210 mg monthly dose of romosozumab was associated with the greatest increase in BMD at the lumbar spine, significantly higher than that observed in the alendronate or TPTD groups 77. The increase seen at this dose after only 12 months was impressive, reaching an average of 11.3%. Not only was bone density impressively increased after only 12 months, but bone strength was also improved as showed by finite element analysis 81.

After the end of 12‐month treatment period, women in the romosozumab and placebo groups continued the same treatment for an additional 12 months. The alendronate group was transitioned to romosozumab 140 mg monthly for 12 months 82. After 24 months, study subjects were re‐randomized 1:1 within original treatment groups to placebo or Dmab 60 mg every 6 months, for an additional 12 months (total study period 36 months). Romosozumab significantly increased lumbar spine and total hip BMD through month 24, with largest gains observed with romosozumab 210 mg monthly (15.1%). Women receiving romosozumab who transitioned to Dmab continued to increase BMD, and similar benefits were observed in the alendronate/romosozumab group transitioned to Dmab; whereas BMD returned to pre‐treatment levels in the romosozumab/placebo group.

These early studies led to the design of the pivotal Phase 3 trial known as FRAME 83. The design of this study in which 7180 postmenopausal women were enrolled, limited exposure to romosozumab to 12 months vs. placebo, followed by Dmab 60 mg every 6 months for 1 year. The primary endpoint was vertebral fracture incidence at 12 months. Romosozumab was associated with a lower risk of vertebral fracture compared to placebo (73% lower risk at 12 months). After transition to Dmab, the risk of fracture was lower in the romosozumab/Dmab group than in the placebo/Dmab group (75% lower risk at 24 months). The BMD gains, at both lumbar spine and total hip, after 1 year of romosozumab followed by 1 year of Dmab were comparable to those obtained after 7 years of treatment with Dmab alone 84. The study also demonstrated the romosozumab reduced the incidence of clinical fractures, but non‐vertebral fracture reduction was not different from the control group. The study, which was pre‐planned and showed geographical heterogeneity, led to a post‐hoc subanalysis by regions. Despite the randomized design of the study, subjects recruited into the study from South America were at significantly lower risk of fracture by FRAX than the rest of the world. In retrospect, this heterogeneity could account for the lack of non‐vertebral fracture efficacy of FRAME because when the South American cohort was not included in the post‐hoc analysis, non‐vertebral efficacy was demonstrated 85.

Another key clinical trial is the head‐to‐head comparison of romosozumab with alendronate, known as ARCH 86. In this randomized, double‐blinded study, postmenopausal women with a fragility fracture were assigned to romosozumab 210 mg monthly or alendronate 70 mg weekly for 12 months, followed by open label alendronate in both groups. In these high‐risk postmenopausal women, romosozumab followed by alendronate resulted in a significantly lower risk of fracture, by 48% after 24 months, than alendronate alone. In ARCH, the group that received romosozumab showed a significant reduction in non‐vertebral fracture (6.2% vs. 11.9%).

Another noteworthy clinical trial with romosozumab was based upon the reality that most patients who are considered candidates for romosozumab therapy will have previously received alendronate or other bisphosphonate. The STRUCTURE study 87 enrolled postmenopausal women with osteoporosis who had taken an oral bisphosphonate for at least 3 years before screening and alendronate the year before screening. Patients were randomized to receive romosozumab 210 mg once monthly or TPTD 20 μg once daily. The primary endpoint was the change in the total hip BMD after 12 months. Romosozumab led to a significantly greater gain in total hip BMD than the TPTD treatment group (2.6% vs. −0.6%, 95% CI 2.7–3.8, P < 0.0001).

A meta‐analysis 88 of six trials showed that romosozumab significantly reduces the risk of new vertebral fracture, non‐vertebral fracture and hip fracture compared with other treatments, and improves BMD. The dose of 210 mg monthly showed the largest gains in BMD.

Recently, the efficacy of romosozumab was also demonstrated in men in a double‐blind placebo trial (BRIDGE trial) 89. Men treated with romosozumab 210 mg monthly for 12 months experienced a significant improvement of lumbar spine and total hip BMD compared to placebo.

Anabolic window

It is instructive to consider the anabolic window concept in connection with romosozumab, recognizing its dual effect to increase bone formation and also to reduce bone resorption. Romosozumab is associated with a rapid but very transient increase in bone formation markers (P1NP, osteocalcin, and bone‐specific alkaline phosphatase). After only 1 month when these markers peak, there is a rapid decline that ultimately leads to levels slightly below baseline at months 2–9 depending on romosozumab dose. Equally impressive is an immediate decline in serum CTX during the first week to about 50% of baseline followed by a slight recovery towards baseline and then a return to levels substantially below baseline at 1 year. During most of this first year, therefore, the anabolic window would appear to be wider than monotherapy with TPTD or ABL. After 1 year, the markers suggest that the drug's actions have been converted essentially to the behaviour of a pure antiresorptive. With this is mind, it was prescient to design FRAME and ARCH to be a 1‐year‐only exposure to romosozumab, which corresponds to the period when it is acting as an osteoanabolic and antiresorptive agent together (Figure 1) 90.

Safety

SOST gene expression, and therefore sclerostin production, is generally limited to skeletal tissue 75. Therefore, the effects of romosozumab would be expected to be limited to the skeletal system. Nevertheless, during the ARCH trial, serious cardiovascular adverse events were observed more often with romosozumab than with alendronate (50 of 2040 patients [2.5%] vs. 38 of 2014 patients [1.9%]) 86. This observation has given cause for concern and has led to a complete independent adjudication of the entire study, currently in progress. There are several interpretations of the data. Since there was no such increase in cardiovascular risk during FRAME when there was a placebo control and there was an apparent imbalance in ARCH in which the comparator was alendronate, not placebo, the possibility has been raised that the observation can be explained by a beneficial effect of alendronate to reduce cardiovascular risk, not an increase in risk due to romosozumab 83. However, also during the BRIDGE trial (romosozumab vs. placebo), there was a numerical difference in the cardiovascular serious adverse events (romosozumab 4.9% vs. placebo 2.5%) 89. The results of the review are awaited with interest.

As a bone‐forming agent and activator of the canonical osteoanabolic Wnt‐signalling pathway, a potential for carcinogenesis has been raised with romosozumab. The WNT‐signalling pathway has been related to different human cancers 91. In human osteosarcoma tissue, the presence of Wnt receptor LRP5 correlates with a worse prognosis 92. Blocking Wnt/LRP5 signalling can reduce tumour invasiveness in vitro 93 and in an animal model 94. However, these observations are not consistently seen in high‐grade human osteosarcoma 95, 96. Nevertheless, no romosozumab‐related effects on tumour incidence were found in rats 97.

Summary

Teriparatide and abaloparatide are osteoanabolic therapies for osteoporosis. In postmenopausal women, the risk of vertebral fractures and non‐vertebral fractures is significantly reduced. The drugs are well tolerated. They are approved for a maximum 24 months, and should be followed with an antiresorptive agent to maintain the BMD gains. Romosozumab is a monoclonal antibody that binds to and inhibits sclerostin. It is a dual action drug because bone formation is stimulated while bone resorption is inhibited. Ultimately, the effects of romosozumab over 12 months of exposure become primarily those of an antiresorptive agent. Romosozumab reduces the risk of new vertebral fracture and non‐vertebral fractures. At this time, it is not clear whether the drug is associated with an increase in serious cardiovascular adverse events.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY 98, and are permanently archived in the Concise Guide to PHARMACOLOGY 2017/18 99.

Competing Interests

There are no competing interests to declare.

Tabacco G., and Bilezikian J. P. (2019) Osteoanabolic and dual action drugs, Br J Clin Pharmacol, 85, 1084–1094. doi: 10.1111/bcp.13766.