Romosozumab for the treatment of postmenopausal women at high risk of fracture

ABSTRACT

Introduction

Romosozumab is a monoclonal antibody that binds to sclerostin (an inhibitor of the Wingless-related integration site (Wnt) signaling pathway). It is a new osteoanabolic drug that simultaneously increases bone formation and decreases bone resorption. It has recently been approved by the US and EU authorities in postmenopausal women with at high risk of fractures.

Areas covered

The literature on romosozumab in preclinical and in phase II and III clinical studies has been reviewed about the effect on bone, bone markers, and fracture reduction and its safety.

Expert opinion

Compared to antiresorptive agents, its unique mechanism of action results in a quicker and greater increase in bone mineral density, it repairs and restores trabecular and cortical bone microarchitecture, and reduces fracture risk more rapidly and more effectively than alendronate, with persisting effects for at least two years after transition to antiresorptive agents. This finding has introduced the concept that, in patients at very high risk of fractures, the optimal sequence of treatment is to start with an osteoanabolic agent, followed by a potent AR drug. Recent national and international guidelines recommend the use of romosozumab as an initial treatment in patients at very high fracture risk without a history of stroke or myocardial infarction.

KEYWORDS:

• Romosozumab
• osteoporosis
• very high fracture risk
• osteoanabolic agents
• sequential treatment
• fracture prevention

1. Introduction

1.1. Epidemiology of fractures

In women older than 50 years, 50% sustain at least one clinical fracture [1,2] and 25% a vertebral fracture (VF) [3], of which most occur subclinical [4]. This lifetime fracture horizon varies over time. In the population of postmenopausal women, the incidence rates of any clinical first fracture increases from around 1/100 persons-years at the age of 50–60 years to >4/100 after the age of 80 years [1], and for morphometric VFs from 0.5% in 60–65 years old to 1.7% in 80+ old [2] and varies across countries [5]. The fracture incidence further increases with the presence of clinical risk factors (CRFs) [6], and in the presence of untreated low bone mineral density (BMD) and prevalent VFs, as shown in the placebo groups of randomized controlled trials (RCTs). After a recent fracture, the incidence of subsequent fractures is higher at short term than later, the so-called imminent fracture risk [7]. The main challenge for the clinician who takes care of fracture prevention in individuals 50 years and over at the Fracture Liaison Service (FLS) is to consider any clinical fracture as a signal for imminent subsequent fracture risk, except for pathologic, periprosthetic, head/neck, finger, and toe fractures. Classifying fractures according to the level of trauma (by rating or reporting ‘after high/low trauma’ or ‘associated with/without external injury’) can be misleading and is at best inaccurate as subsequent fracture risk is increased after a fracture, whatever level of rating of trauma [8–10]. Most non-VFs are easily clinically diagnosed, but the diagnosis of all VFs requires lateral imaging of the spine using dual-energy X-ray absorptiometry (DXA) or X-rays [11].

1.2. Fracture risk evaluation

Algorithms have been developed to calculate the 1-, 5- and 10-year risk of fractures, such as FRAX, Garvan, qFracture, and many others [12]. In a first step, fracture risk can be calculated in daily practice by evaluation of the presence of CRFs (including comorbidities and fall risks) [13]. In a second step, when the presence of CRFs indicates an increased risk of fractures, evaluation of BMD measured by DXA, vertebral fracture assessment (VFA) on lateral images of the spine, fall risk assessment, and laboratory investigation should be considered.

Of the many fracture risk algorithms, FRAX with and without BMD is the most calibrated and validated algorithm and is used in various ways for decisions about assessment and treatment in >70 guidelines worldwide [14,15].

In the absence of DXA, treatment thresholds have been recommended based on clinical risk evaluation alone [16]. When DXA is available, various assessment thresholds for measuring BMD with DXA have been recommended, to define treatment thresholds with inclusion of BMD [17]. Some guidelines also recommend performing imaging of the spine when available to evaluate the presence, number, and severity of VFs in the spine [18]. The additional advantage of such approach is that BMD and VF status are independent predictors of fracture risk and that their baseline data can then be used for follow-up [19,20].

1.3. Determining the fracture risk threshold for treatment decisions

None of the fracture prediction models is perfect [21]. As the risk of fracture is a continuous variable, there are no obvious natural thresholds for defining high risk [22]. Examples of propositions to define ‘very high’ fracture risk for treatment decisions include the presence of very low BMD (alone or combined with recent fractures or other CRFs), fracture history (their recency, and, for VFs, their presence, number, and severity), FRAX (above fixed, age-dependent or a mixture of both (‘hybrid’) thresholds), fall risk, and treatment failure during anti-resorptive (AR) treatment [19,23–28]. In subjects with a very high fracture risk, especially at short term, the goal is not to maintain but to improve bone structure and BMD [29], and to rapidly reduce the risk of VFs and non-VFs.

1.4. Overview of the market

In postmenopausal women with an increased risk of fractures, fracture prevention is recommended to prevent a first or subsequent fracture. This requires a lifelong treatment strategy that needs to be deliberated with the patient from the first treatment on [30,31], including lifestyle measures, adequate calcium and vitamin D intake, and fall prevention. In most subjects with an increased risk of fractures, AR agents are available as the first choice of treatment [32], but adherence often remains suboptimal [33].

In RCTs, it has been shown that the osteoanabolic agents (also referred as bone-forming agents) teriparatide (for two years) and romosozumab (for one year followed by ARs) are superior in fracture reduction of VFs, clinical, and non-VFs compared to ARs (both are approved by the European Medicine Agency (EMA) and the Food and Drug Administration (FDA)) [34,35], and that abaloparatide (for two years) is superior compared to placebo (approved by FDA) [36].

1.5. Unmet needs

A lifelong fracture prevention requires a strategy based on continuous, intermittent, and sequential drug treatment [30,31]. AR agents mainly preserve bone structure while osteoanabolic bone forming agents improve bone structure [37,38].

In view of the differences of action between osteoanabolic agents, their superiority over ARs, and differences in cost, the indication for osteoanabolic treatment will focus on subjects with very high fracture risk, and depend on the potency, speed, and superiority of these treatments on the effect on fractures compared to ARs.

With the introduction of romosozumab, a monoclonal antibody that binds to sclerostin, there is emerging evidence that osteoanabolic agents will change the approach of sequential treatment for fracture prevention, by using them as first treatment in patients at very high risk (Box 1).

2. Preclinical data

In the context of bone metabolism in humans, the canonical Wnt signaling pathway is a major regulator [39–41]. When Wnt ligands bind to LRP5 or LRP6 and Frizzled co-receptors at the cell surface, this stabilizes the accumulation of intracellular β-catenin. After translocation to the nucleus, β-catenin binds to transcription factors, so that the transcription of bone formation stimulating target genes is started. Bone remodeling is controlled by osteocytes, differentiated osteoblasts, which are embedded in the mineralized matrix and secrete sclerostin, a negative regulator of bone formation produced mainly in the skeleton, but also in chondrocytes, multiple myeloma cells and fibroblast-like synoviocytes [42–47]. When osteocytes produce sclerostin, the osteocyte dendritic network allows it to be transported to the bone surface. There it binds to LRPs, which antagonizes downstream signaling of β-catenin. As a result, the proliferation, differentiation, and survival of osteoblasts is inhibited. In addition, sclerostin also has an autocrine function by upregulating RANKL synthesis in osteocytes which then stimulates osteoclastogenesis. The regulation of sclerostin production is complex and regulated by several mechanisms, including hormones and mechanical load [48–53] (Figure 1).

Figure 1. Simplified schematic representation of sclerostin actions in bone cells. Sclerostin is secreted by osteocytes and binds to LRPs on osteoblasts, thereby inhibiting the Wnt signaling pathway and bone formation by degrading β-catenin. Sclerostin also upregulates RANKL synthesis in the osteocyte, which binds to the RANK receptor on osteoclasts, thereby stimulating osteoclastogenesis. Examples of up-regulators (a) and down-regulators (b) of sclerostin production are shown in the text boxes of the figure. Note the decrease of RANKL production by romosozumab in contrast to the increase by parathyroid hormone [48,52,53].

Sclerostin is encoded by the SOST gene. When SOST is knocked out in mice, they develop increased bone formation of trabecular bone, periosteal bone, as well as endosteal surface of cortical bone [54]. This leads to enhanced mineral apposition increasing bone strength. The lack of sclerostin increases osteoblast activity compared to osteoclast activity resulting in an uncoupling of bone formation and resorption [55]. Anti-sclerostin antibodies (Scl-Ab) demonstrated in small as well as larger animal models [56–60], an increase in BMD with increased serum levels of bone formation markers, but not resorption markers as suspected by the knockout models. After initiation of treatment with sclerostin antibodies modeling-based bone formation (on bone surfaces not undergoing bone remodeling) is observed. During long-term treatment, a positive remodeling balance was observed due to sustained suppressed bone resorption. No changes of mineral metabolism were observed after administration of the antibody except for a transient rise in plasma levels of PTH, which diminished with time and are more pronounced in animals then in humans.

In bone biopsies of the FRActure study in postMenopausal woMen with ostEoporosis (FRAME) study, histologic parameters showed increased bone formation and decreased bone resorption already within two months of romosozumab treatment [38].

In pharmacodynamic studies of romosozumab, the increase in bone formation markers declined over time whereas reduced bone resorption persisted. This might be due to the upregulation of antagonists of Wnt pathway and therefore, romosozumab will only be given for a limited period [54]. Pharmacokinetics were non-linear with dose, with peak serum concentrations occurring within the first week after subcutaneous administration. In addition, the magnitude of the increase in serum levels of P1NP decreased in time and reached the baseline after 12 months of treatment. Thus, after one year, the changes in markers of bone turnover were compatible with those of a mild antiresorptive agent.

In phase I and II clinical studies similar results were found, with early increases in biochemical markers of bone formation and reduction of biochemical markers of bone resorption, which was associated with early increases in BMD at the spine and the hip [61–65]. In a phase Ib study in women and men 40–80 years old with low BMD, romosozumab was administered for 3 months [66]. Using quantitative computer tomography (QCT) (at the lumbar vertebra L1 and L2, with voxel size of $\text{~}$ 600 $\mu$ m) and high-resolution QCT (at the thoracic vertebra T12, with voxel size of $\text{~}$ 187 $\mu$ m), already within 3 months large improvements were found in trabecular and cortical bone mass and structure, as well as whole-bone stiffness, which increased at T12 by a mean of 26.9%. These changes continued 3 months after the last romosozumab dose.

3. Clinical data

The phase III program of romosozumab consisted of four pivotal RCTs: FRAME, ARCH, BRIDGE and STRUCTURE [34,67–69].

The FRAME study was a RCT including 7180 postmenopausal women between 55 and 90 years, who had a BMD T-score of – 2.5 to – 3.5 at the total hip (TH) or femoral neck (FN) after well-defined washout periods of previous therapy with drugs affecting bone metabolism [67]. Patients were randomly assigned to placebo or romosozumab for 12 months followed by denosumab 60 mg every 6 months for 12 months. The primary endpoints of the study were the incidence of VFs at 12 and 24 months and the secondary endpoints the incidence of clinical and non-VFs. At 12 months, a significant relative risk reduction (RRR) was found in the romosozumab group for VFs (−73%) and clinical fractures (−36%), but not for non-VFs. At 24 months, a significant RRR was found for VFs (−75%) but not for clinical or non-VFs.

In a preplanned analysis, a significant treatment-by-geographic region interaction was observed for clinical and non-VFs [70]. At 12 months, the RRR for VFs was similar between Latin America and the rest-of-world (−70% and −74%, respectively). The RRR was significant for clinical and non-VFs in rest-of-world (−52% and −42%, respectively), but not in Latin America, where the background fracture risk was low. This is consistent with the finding that a greater effect on clinical and non-VFs may be observed in higher risk populations [71].

In a 36-month follow-up extension study with continuation of denosumab, significant RRRs were found for VFs (−66%), clinical (−27%) and non-VFs (−21%) [72].

In the Active-Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk (ARCH), the effectiveness of romosozumab for 12 months followed by alendronate for 24 months was compared with alendronate treatment alone after well-defined washout periods of previous therapy with drugs affecting bone metabolism [34].

Included were 4093 postmenopausal women 55–90 years with a BMD T score ≤ – 2.5 at the TH or FN and either one or more moderate or severe VFs or two or more mild VFs; or a BMD T score ≤ – 2.0 and either two or more moderate or severe VF or a fracture of the proximal femur sustained 3–24 months before randomization.

The primary endpoints of the study were the incidence of VFs at 24 months and clinical fractures at the time of primary analysis (after a mean of 2.7 years), and the secondary endpoints the incidence of non-VFs and hip fractures at that time (Table 1).

Table 1. Fracture risk reduction (RR) for VFs, non-VFs, clinical and Hip fractures in the ARCH study [34], which included postmenopausal women with a bone mineral density T score of – 2.5 or less at the total Hip or femoral neck and either one or more moderate or severe vertebral fractures or two or more mild vertebral fractures; or a bone mineral density T score of – 2.0 or less at the total Hip or femoral neck and either two or more moderate or severe vertebral fractures or a fracture of the proximal femur sustained 3–24 months before randomization.

	At 12 months			At 24 months (for VF) At 2.7 years (for other fractures in the primary analysis)
	ALN	ROMO	RR (95%CI) p-value	ALN-ALN	ROMO-ALN	RR (95%CI) p-value
Vertebral fracture	6.3%	4.0%	37% (15 to 53%) p = 0.003	11.9%	6.2%	48% (34 to 60%) p < 0.001
Non-vertebral fracture	4.6%	3.4%	26% (−1 to 46%) p = 0.06	10.6%	8.7%	19% (1 to 34%) p = 0.04
Clinical fracture	5.4%	3.9%	28% (4 to 46%) p < 0.05	13.0%	9.7%	27% (12 to 39%) p < 0.001
Hip fracture				3.2%	2.0%	38% (4 to 46%) p = 0.02

ALN: alendronate

ROMO: romosozumab

At 24 months, a significant RRR was found in the romosozumab group for VFs (−48%), clinical fractures (−27%), and at the time of primary analysis for non-VFs (−19%) and hip fractures (−38%).

In the BRIDGE study, 245 men between 55 and 90 years were included with a T-score ≤ −2.5 or ≤ −1.5 with a history of fragility fracture, to receive romosozumab (n = 163) or placebo (n = 82) [68]. The change in lumbar spine (LS) and TH BMD at 12 months was significantly higher for the romosozumab group than for the placebo group (LS: 12.1% vs 1.2%; TH: 2.5% vs −0.5%).

To further evaluate whether treatment sequence affects romosozumab response, the data from FRAME, ARCH and STRUCTURE were combined. Larger BMD increases and greater BMD responder rates were achieved when romosozumab was used before compared to after an AR agent [73].

Data regarding direct comparison between romosozumab and teriparatide are not available from RCTs with fractures as primary endpoint. In one-year studies, romosozumab was compared to teriparatide in a subset of participants of the phase II study (in postmenopausal women with low BMD) [65,74,75] and in the STRUCTURE study in postmenopausal women with prior AR treatment, still having a T-score $\le$-2.5 at the total hip, femoral neck, or lumbar spine and a history of non-vertebral fracture after age of 50 years or vertebral fracture [69], using DXA for evaluation of areal BMD and bone mineral content (BMC) and QCT for evaluation of integral, cortical and trabecular BMC, volumetric BMD (vBMD) and finite element analysis (FEA) to estimate bone strength. After one year, romosozumab was superior to teriparatide for most of the measured DXA and QCT parameters, including an increase in the estimated strength at the hip, compared to no changes by teriparatide [76]. No significant differences between romosozumab and teriparatide were found after one year at the spine for the increase in trabecular vBMD [74] and trabecular FEA [75], and at the hip, for the increase in trabecular vBMD [69].

4. Safety and tolerability

The overall incidence of adverse events and serious adverse events was similar between treatment groups in ARCH and FRAME, except for mild injection site reactions during the first 12 months.

4.1. Atypical femur fractures (AFF) and osteonecrosis of the jaw (ONJ)

In FRAME, one AFF (<0.1%) was reported after 3.5 months romosozumab, but with prodromes already before the study, and two ONJ cases (<0.1%) [67]. After 12 months of romosozumab treatment, one ONJ event occurred in the context of ill-fitting dentures. After 12 months of romosozumab treatment followed by one dose of denosumab, one ONJ event occurred and after a tooth extraction and was followed by osteomyelitis of the jaw. In ARCH, no AFF and ONJ events were reported in the 12-month double-blind period [34]. During the open-label period with alendronate, six AFF events occurred (four AFF events (0.2%) in the alendronate-to-alendronate group and two in the romosozumab-to-alendronate group (<0.1%)) and two events of ONJ (<0.1%) occurred (one after switch from romosozumab-to-alendronate and one in the alendronate-to-alendronate group).

4.2. Cardiovascular safety

During the overall study period, adjudicated serious cardiovascular events were balanced between the groups with and without romosozumab in FRAME (after two years: 2.3% versus 2.2%, after three years: 3.6% versus 3.5%, respectively) and in ARCH (after a mean of 2.7 years: 6.5% versus 6.1%, respectively) [34,67].

During year 1 in FRAME, adjudicated serious cardiovascular events were balanced between the groups with and without romosozumab (1.2% and 1.1%, respectively) [67].

During year 1 in ARCH, there was a numerical non-significant imbalance of all adjudicated serious cardiovascular adverse events between romosozumab and alendronate (2.5% vs 1.9%, Odds ratio (OR): 1.31 (95% CI 0.85–2.00) [34]. However, 0.8% in the romosozumab group and 0.3% in the alendronate group reported cardiac ischemic events (OR:2.65; 95% CI, 1.03-6.77), and 0.8% in the romosozumab group and 0.3% in the alendronate group reported cerebrovascular events (OR:2.27; 95% CI, 0.93-5.22), whereas heart failure, noncoronary revascularization, and peripheral vascular ischemic events not requiring revascularization were numerically lower in the romosozumab group.

These results were also analyzed posthoc as a traditional subgroup of events, using a composite 3-point major adverse cardiac events (MACE) endpoint, a combination of death, serious myocardial infarction, or serious stroke by removing non-coronary vascular events and heart failure [77,78]. In FRAME, the hazard ratio (HR) for MACE was close to 1.0 in the first year and in the overall study period of 36 months [77]. In ARCH during the first year, MACE occurred in 2.0% of patients treated with romosozumab compared to 1.1% of patients treated with alendronate, with a significant HR of 1.87 (95%CI 1.11–3.14) and of 3.21 (95% CI: 1.18–8.77) for myocardial infarction [77,78]. For the overall study period (mean duration of 36 months), all HRs were close to 1.0, except for stroke (HR: 1.75, 95%CI: 1.06–2.89) [77].

The reason for the difference in cardiovascular risk with romosozumab compared to alendronate in ARCH but not compared to placebo in FRAME is not fully elucidated.

In animals with and without atherosclerosis, pharmacodynamic effects of romosozumab were observed in bone, but no functional, morphological, or transcriptional effects on the cardiovascular system [79]. The cumulative incidence of MACE over the entire study period of the ARCH trial is not consistent with alendronate being protective or romosozumab being harmful and has been extensively reviewed [44,80–83].

The EMA concluded that romosozumab can address an unmet clinical need, that the MACE risk can be neither confirmed nor excluded, that the absolute risk difference is small and that detailed analysis suggest that the MACE risk is likely to be over-estimated in ARCH [78]. Therefore, in patients with any other cardiovascular risk, the likely benefits and potential risks should be openly discussed by the physician and the patient [78]. The FDA concluded that the benefit-risk of romosozumab in postmenopausal women at high risk of fracture is favorable, provided that the important potential risk of MI and stroke is effectively communicated [77]. The follow up of MACE during and after romosozumab treatment requires further high-quality, real-world evidence that accounts for sources of bias and confounding and has been proposed by the FDA and EMA and the scientific community [44,77,78,80–82]. Until such evidence is available, the restricted prescribing recommendations in the boxed warnings should remain, and patients with a history of stroke or MI (during last year according to FDA, and any history according to EMA) should not be considered for treatment. Further elucidation on the potential increased risk of MACE may open opportunities for further studies in men.

5. Conclusion

Romosozumab is new and unique osteoanabolic drug that simultaneously increases bone formation and decreases bone resorption. Compared to antiresorptive agents, this unique combination results in a more rapid and greater increase in BMD, and in repair and restoration of trabecular and cortical bone microarchitecture. Additionally, the reduction of fracture risk during treatment with romosozumab is more rapidly and more effectively than the AR alendronate, with persisting effects for at least two years after transition to AR agents. This finding has introduced the concept that in patients at very high risk of fractures may benefit from initial treatment with an osteoanabolic agent, followed by a potent AR agent, as recommended in recent national and international guidelines. It is therefore expected that romosozumab will be increasingly considered as first treatment when these guidelines are implemented, also in patients at the fracture liaison services with a recent fracture. Patients with a history of myocardial infarction or stroke should not be considered for treatment with romosozumab. Further data will be needed, such as long-term safety data, repeated use in patients with persisting or recurrent high fracture risk, and its effect on fractures in glucocorticoid users and in an ongoing study in pediatric patients with osteogenesis imperfecta [78].

6. Expert opinion

The introduction of romosozumab will have an important role in the treatment of patients with osteoporosis, especially in high-risk patients, for several reasons [22,29–31].

Romosozumab is a new osteoanabolic drug, with a unique mechanism of action by simultaneously increasing bone formation and decreasing bone resorption. Compared to AR treatment with alendronate, an active and reliable comparator, this results in a quicker and greater increase of BMD and of repair and restoration of trabecular and cortical microarchitecture, and a quicker and more substantial decrease of the risk of vertebral, clinical, and non-VFs, with persistent benefits after transition to ARs, including a decrease in the risk of hip fractures.

The studies with romosozumab have introduced the concept that in very high-risk patients the optimal sequence of treatment is starting with an osteoanabolic agent followed by a potent AR drug, because such sequence is building up new bone followed by preservation of the newly formed bone when followed by AR, instead of mainly preserving the existing disturbed microarchitecture, as in treatment with ARs alone.

The unraveling of a new pathophysiological Wnt-signaling pathway and the substantial change in the sequential treatment paradigm is increasingly supported by recent national and international guidelines [19,23–28]. There is a worldwide increase in educational sessions and teaching courses by the introduction of romosozumab, which is very likely to increase the awareness of osteoporosis and fracture prevention in high-risk patients [13,84,85].

6.1. What, if any, impact is this drug likely to have on current treatment strategies?

As mentioned above, romosozumab initiated a new and unique paradigm of sequential treatment by introducing the concept of using osteoanabolic drugs as the initial treatment in patients at very high risk of fractures.

Additionally, romosozumab can be prescribed in patients on treatment with AR, who are non-responders, such as after persisting BMD T score ≤ −2.5, persisting bone loss, or recurrent fractures during long-term AR treatment.

6.2. How likely are physicians to prescribe the drug?

Given the quick and favorable effects of romosozumab with a superiority versus alendronate, it can be expected that physicians will prescribe romosozumab in very high-risk patients. As the subsequent fracture risk is highest immediately following a fracture, a well-organized FLS, in which all patients older than 50 years with a recent fracture are evaluated with DXA and VFA, is a window of opportunity to identify very high-risk patients with a treatment indication for osteoanabolic agents, e.g. romosozumab. The same holds for patients with a recent fracture not attending the FLS but visiting their GP or specialist in the context of other diseases.

There is growing international consensus about identifying patients at very high risk for treatment with romosozumab, including those with very low BMD, multiple prior fractures (including VFs), very high FRAX, and treatment failure with AR.

The introduction of romosozumab also offers the opportunity to consider romosozumab in patients with side-effects on teriparatide, or in patients who prefer the injection regimen of 12 monthly double SC injections over one year above 730 daily SC injections with teriparatide over two years.

6.3. What data is still needed?

Since the use of romosozumab should be followed by ARs, we still need long-term data in extension studies (now up to 3 years) of the original RCTs with romosozumab, and studies comparing romosozumab to other ARs than denosumab and alendronate and evaluating the effect of other ARs after romosozumab.

In view of the effect of a second course of romosozumab on BMD and on bone turnover markers, the effect of repeated use of romosozumab on fractures and on safety will be needed [86].

Another potential indication is in patients treated on glucocorticoids (GC) with at high risk for fractures, but currently fracture data exist for teriparatide in GC-users [87], but not for romosozumab.

Since the FLS will play a dominant role in the starting with anabolic drugs in very high-risk patients, optimalization of existing FLSs and starting up an FLS will extend the identification of such patients.

The follow-up of MACE after romosozumab requires further high-quality, real-world evidence that accounts for sources of bias and confounding [88]. Until such evidence is available, the restricted prescribing recommendations in the boxed warnings should remain, in that patients with a history of stroke or myocardial infarction should not be considered for treatment. Further elucidation on the potential increased risk of MACE may open opportunities for further studies in men.

6.4. Where is the drug likely to be in five years’ time?

Starting with osteoanabolic agents has emerged as a first option in patients with very high risk for fractures. However, it can be expected that full implementation of romosozumab as first-line drug will take some time.

Sequential treatment and using osteoanabolics as first treatment in high-risk patients is starting to be recommended in new osteoporosis and fracture prevention guidelines. In view of this growing international consensus, it can be expected that implementation of these guidelines will result in better care of high-risk patients without a history of stroke or myocardial infarction.

Over the years, physicians and FLSs will therefore have increasing experience in identifying patients at very high risk, and to prescribe them romosozumab.

Finally, romosozumab can nowadays only be given for one year, but it is expected that in the coming years more data are becoming available on the effect of subsequent second or third courses of treatment to prevent subsequent fractures.

Box 1. Drug summary box

Drug name	Romosozumab
Phase	Launched
Indication	Osteoporosis in postmenopausal women at high risk of fracture
Mechanism of action	Sclerostin-inhibiting humanized monoclonal antibody
Route of administration	subcutaneously
Pivotal trials	FRAME: NCT01575834 ARCH: NCT01631214 BRIDGE: NCT02186171 STRUCTURE: NCT01796301

Article highlights

• Romosozumab, an inhibitor of sclerostin, is a new osteoanabolic drug that simultaneously increases bone formation and decreases bone resorption, in preclinical and clinical studies

• One year treatment with romosozumab reduces fracture risk more rapidly and more effectively than placebo or alendronate, with persisting effects for at least two years after transition to antiresorptive treatment.

• The overall incidence of adverse events and serious adverse events was similar compared to placebo or alendronate, except, during the first year, for mild injection site reaction. In comparison with alendronate, there was a numerical non-significant imbalance of all adjudicated serious cardiovascular adverse events. Patients with a history of myocardial infarction or stroke should not be considered for treatment with romosozumab.

• Recent national and international guidelines recommend the use of romosozumab as initial treatment in patients at very high fracture risk without a history of stroke or myocardial infarction.

Declaration of interest

P Geusens has worked on clinical studies/ advisory boards/ has received speaker fees from Abbvie, Amgen, Bristol Myers Squibb, Celgene, Janssen, Lilly, Merck Sharp & Dohme, Novartis, Pfizer, Roche, UCB, Fresenius, Mylan, Viatris, Sandoz, and Merck. NM Appelman-Dijkstra has worked on clinical studies/ received speaker fees from Amgen, UCB, Kyow a Kirin and Takeda as well as having worked on clinical studies/ advisory boards/ has received speaker fees from Amgen and UCB. W Lems declares advisory board and speaker fees from Amgen, UCB, Pfizer and Galapagos. J van den Bergh declares advisory board/ speaker fees from Amgen and UCB. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

A reviewer on this manuscript has received consulting fees and honorarium from Amgen and UCB, companies that market romosozumab. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Company review

UCB provided a scientific accuracy review of the sections of the paper relating to clinical data and safety and tolerability at the request of the journal editor.

Additional information

Funding

This paper was not funded.