Sclerostin Inhibition for Osteoporosis [Romosozumab]-
A New Approach - Editorial
Download the PDF
Download the PDF
see original study below
Carolyn B. Becker, M.D.
N Engl J Med 2014
Effective new therapies are still needed for people with osteoporosis. In 2002, the introduction of teriparatide, or recombinant parathyroid hormone (PTH [1-34]), opened a promising chapter in osteoporosis care.1 For the first time, there was an anabolic agent that significantly increased bone mineral density (BMD), reduced fracture risk, and restored bone architecture back to, or close to, normal. However, despite an impressive track record of both safety and efficacy, teriparatide has had a limited clinical reach as compared with other agents, largely owing to its requirement for daily subcutaneous injection, a black-box warning about osteosarcoma in rats, and its high cost. Since antiresorptive therapies do not restore bone architecture and have a number of other limitations, finding new treatments for osteoporosis has been a high priority.
The results of the study by McClung et al.2 now reported in the Journal represent a potential breakthrough in osteoporosis therapeutics. The study introduces romosozumab, a humanized monoclonal antibody directed against the osteocyte-derived glycoprotein known as sclerostin. Humans with genetic deficiencies of sclerostin and mice with knockout of the sclerostin gene (Sost) have high bone mass, increased bone strength, and resistance to fracture. Sclerostin works by inhibiting the Wnt and bone morphogenetic protein signaling pathways that are critical for osteoblast proliferation and activity. By inhibiting sclerostin, romosozumab should enhance osteoblastic function.
This phase 2 study was a randomized, placebo-controlled trial that included two comparator drugs. Participants were healthy postmenopausal women with osteopenia, randomly assigned to one of eight study groups - romosozumab administered subcutaneously either monthly or every 3 months at various doses; oral alendronate at a dose of 70 mg weekly; subcutaneous teriparatide at a dose of 20 μg daily; or placebo injections given monthly or every 3 months. Primary and secondary end points included changes in BMD as compared with placebo, changes in markers of bone metabolism, and comparisons of the study drug with alendronate and teriparatide.
The results were impressive. As compared with baseline, BMD was significantly improved for all doses of romosozumab and at all sites except at the distal third of the radius, which remained essentially unchanged. At the highest monthly dose of romosozumab, increases in BMD at the spine and hip were rapid and robust, surpassing the BMD values with alendronate and teriparatide at 6 months and remaining significantly higher than the BMD values with either comparator by the end of the trial.
If the changes in BMD for a presumed anabolic agent were predictable, the changes in bone-turnover markers were not. Levels of bone-formation markers increased rapidly after the first dose of romosozumab but then declined. By month 6, the bone-formation markers were nearly back to baseline levels, despite continued administration of the drug. In contrast and perhaps most surprising, markers of bone resorption declined in the first week and remained suppressed for the duration of the trial.
The pattern of brief anabolic stimulation coupled with chronic suppression of bone resorption seen with romosozumab is unprecedented among current therapies for osteoporosis. Potent antiresorptive agents such as bisphosphonates and denosumab suppress both bone-resorption and bone-formation markers. Teriparatide increases levels of bone-formation markers early on but, after a delay, stimulates bone-resorption markers as well. The so-called anabolic window opened by teriparatide may be prematurely closed by this effect on resorption, blunting the bone-strengthening properties of the drug. PTH-related protein, another potential anabolic agent, was recently shown to act similarly to PTH.3 Odanacatib, a cathepsin K inhibitor, initially suppresses both bone-formation and bone-resorption markers, but the levels of bone-formation markers increase back to baseline values by 1 year.4
Can we reproduce the effect of romosozumab on bone remodeling with existing osteoporosis therapies? Results from small studies suggest that perhaps we can. Combination therapy with teriparatide and potent but intermittently dosed antiresorptive agents such as zoledronic acid5 or denosumab6 administered one or two times per year shows promising results. Shorter courses of PTH followed sequentially or cyclically by oral bisphosphonates may increase bone formation without stimulating resorption.7,8
Many questions about romosozumab remain. Will changes in BMD translate into potent antifracture efficacy? Will it be safe over time? In the current study, there were no clinically significant adverse events other than injection-site irritation. Will longer administration (>1 year) cause bony complications such as cranial-nerve palsies or spinal stenosis? What duration of treatment is associated with the highest rate of response? Why did BMD not improve at the wrist? A phase 3 clinical trial of romosozumab is under way in a cohort of postmenopausal women with osteoporosis (ClinicalTrials.gov number, NCT01631214) and may answer some of these questions. For now, more than a decade after the introduction of teriparatide, we may at last have a sequel to the anabolic story.
Romosozumab in Postmenopausal Women with Low Bone Mineral Density
Michael R. McClung, M.D., Andreas Grauer, M.D., Steven Boonen, M.D., Ph.D., Michael A. Bolognese, M.D., Jacques P. Brown, M.D., Adolfo Diez-Perez, M.D., Ph.D., Bente L. Langdahl, Ph.D., D.M.Sc., Jean-Yves Reginster, M.D., Ph.D., Jose R. Zanchetta, M.D., Scott M. Wasserman, M.D., Leonid Katz, M.D., Judy Maddox, D.O., Yu-Ching Yang, Ph.D., Cesar Libanati, M.D., and Henry G. Bone, M.D.*
N Engl J Med 2014
Sclerostin is an osteocyte-derived inhibitor of osteoblast activity. The monoclonal antibody romosozumab binds to sclerostin and increases bone formation.
In a phase 2, multicenter, international, randomized, placebo-controlled, parallel-group, eight-group study, we evaluated the efficacy and safety of romosozumab over a 12-month period in 419 postmenopausal women, 55 to 85 years of age, who had low bone mineral density (a T score of -2.0 or less at the lumbar spine, total hip, or femoral neck and -3.5 or more at each of the three sites). Participants were randomly assigned 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 teriparatide (20 μg daily). The primary end point was the percentage change from baseline in bone mineral density at the lumbar spine at 12 months. Secondary end points included percentage changes in bone mineral density at other sites and in markers of bone turnover.
All dose levels of romosozumab were associated with significant increases in bone mineral density at the lumbar spine, including an increase of 11.3% with the 210-mg monthly dose, as compared with a decrease of 0.1% with placebo and increases of 4.1% with alendronate and 7.1% with teriparatide. Romosozumab was also associated with large increases in bone mineral density at the total hip and femoral neck, as well as transitory increases in bone-formation markers and sustained decreases in a bone-resorption marker. Except for mild, generally nonrecurring injection-site reactions with romosozumab, adverse events were similar among groups.
In postmenopausal women with low bone mass, romosozumab was associated with increased bone mineral density and bone formation and with decreased bone resorption. (Funded by Amgen and UCB Pharma; ClinicalTrials.gov number, NCT00896532.)
Osteoporosis is characterized by low bone mass and defects in microarchitecture that are responsible for decreased bone strength and increased risk of fracture.1Antiresorptive drugs for osteoporosis increase bone mineral density and prevent the progression of structural damage but may not restore bone structure. Stimulation of bone formation is necessary to achieve improvements in bone mass, architecture, and strength.
Sclerostin, encoded by the gene SOST, is an osteocyte-secreted glycoprotein that has been identified as a pivotal regulator of bone formation. By inhibiting the Wnt and bone morphogenetic protein signaling pathways, sclerostin impedes osteoblast proliferation and function, thereby decreasing bone formation. 2-4 The fact that SOST expression is limited to skeletal tissue makes the inhibition of sclerostin a particularly attractive target, because it would affect skeletal health but limit the risk of off-target effects.5
Studies of the molecular effects of sclerostin support the concept that blocking the action of sclerostin results in positive effects on the skeleton. Patients with a genetic deficiency of sclerostin have high bone mass, correspondingly increased bone strength, and resistance to fractures.6,7 Mice in which the sclerostin gene was deleted had increased bone formation and high bone mass and strength.8 In estrogen-deficient rat and monkey models of postmenopausal osteoporosis, treatment with antisclerostin antibodies restored bone mass and bone strength to levels higher than those in control animals.9,10
Furthermore, Wnt activation resulting from the inhibition of sclerostin has been associated with decreased bone resorption both in humans and in animal models, probably owing to direct or indirect actions on osteoclasts through the Wnt pathway.11,12
Romosozumab (formerly known as AMG 785/CDP7851, Amgen and UCB Pharma) is a humanized monoclonal anti-sclerostin antibody. In a phase 1 study, single injections of romosozumab stimulated bone formation, decreased bone resorption, and increased bone mineral density.13 Here we report the results of a 1-year, phase 2 study evaluating the efficacy and safety of romosozumab in postmenopausal women with low bone mass.
This study, which included 419 postmenopausal women with low bone mass, showed that inhibiting sclerostin with romosozumab, a humanized monoclonal antibody targeted to sclerostin, induced large increases in bone-formation markers, decreased a bone-resorption marker, and increased bone mineral density when the drug was administered by means of subcutaneous injection at 1-month or 3-month intervals. Although dose-dependency was not formally tested, the larger doses that were administered monthly (140 mg or 210 mg) appeared to produce greater changes than the other regimens. The response profiles of alendronate and teriparatide with respect to bone mineral density were as expected. The increase in bone mineral density at the lumbar spine and proximal femur was rapid and substantial with romosozumab by 3 months, and by 6 months the increase was greater with the 210-mg monthly dose of romosozumab than with either active comparator.
The effects of romosozumab on bone turnover reflect a rapid, marked, and transitory increase in bone formation and a moderate but more sustained decrease in bone resorption. These results confirm similar observations from the phase 1 clinical trial.13 The changes in bone-remodeling markers that were observed with romosozumab contrast with the effects of bisphosphonates and receptor activator of nuclear factor-κB ligand inhibitors, which reduce both bone-resorption and bone-formation markers. They also differ from the response to parathyroid hormone, with which the initial increase in bone-formation markers is followed by an increase in markers of bone resorption. The consequence of these divergent effects on bone formation and bone resorption with romosozumab is a strongly positive balance in bone turnover, accounting for the rapid and large increases in bone mineral density that we observed.
Circulating markers of bone formation increased rapidly with romosozumab but returned to baseline values despite continued administration, whereas a decrease in a circulating marker of bone resorption was maintained over the 12-month dosing period. The reason for the transitory nature of the effect on bone formation is unclear. Changes in counterregulatory signaling pathways would not be unexpected, given the complexity with which bone remodeling is controlled. No apparent relationship was observed between romosozumab antibodies and measures of efficacy. The implications of the transitory nature of the stimulation of bone formation on a dosing regimen for romosozumab in clinical practice will require more study. The specific mechanism responsible for the observed reduction in bone resorption with romosozumab is not understood.
Patients with lifelong homozygous or heterozygous genetic deficiency of sclerostin provide insight into the expected safety of inhibiting sclerostin signaling. Homozygous persons exhibit bony overgrowth and skeletal deformity, especially of the skull and face, which are observed during growth and which can result in symptoms related to compression of the VII and VIII cranial nerves.6,7 However, heterozygous carriers of the SOST mutation have increased bone density and modestly increased markers of bone formation but none of the sequelae of bony overgrowth.15,16 Therefore, potential complications due to bone overgrowth would not be expected with pharmacologic inhibition of sclerostin over a limited period of time in adults. Nevertheless, larger studies of romosozumab will be required in order to further evaluate the potential consequences of bone overgrowth.
The safety of drugs cannot be adequately evaluated in studies involving small treatment groups. In the current study, the overall incidence of adverse events was balanced between groups, with the exception of the increased frequency of injection-site reactions in the romosozumab groups as compared with the other groups. Numerical differences between the romosozumab groups and the control groups were noted with respect to the frequency of other reported adverse events, but some differences occurred in favor of romosozumab and others against romosozumab, and there was no relationship to the dose or to the timing of the event after administration of the drug.
In conclusion, romosozumab, administered subcutaneously at intervals of 1 month or 3 months over a period of 12 months, was associated with prompt, transitory increases in markers of bone formation; moderate, sustained decreases in markers of bone resorption; and rapid, large increases in bone mineral density in the spine and hip regions. The increases in bone mineral density were greater with romosozumab than with placebo, alendronate, or teriparatide. These results support further evaluation of romosozumab as a treatment for patients with osteoporosis.
A total of 419 participants were enrolled in the study and underwent randomization, and 383 (91%) completed the 12-month visit; 36 participants (9%) withdrew from the study (Fig. S1 in the Supplementary Appendix). The main reason reported for study discontinuation was withdrawal of consent (5% of participants). The demographic characteristics and key characteristics at baseline among the women enrolled in the study were balanced across the eight randomized groups (Table 1, and Table S1 in the Supplementary Appendix). The mean age of the participants was 67 years, 86% of the participants were white, and the mean T scores at the lumbar spine, total hip, and femoral neck were -2.29, -1.53, and -1.93, respectively.
Bone Mineral Density
At month 12, participants in the pooled romosozumab group, as compared with the pooled placebo group, had a significant increase in bone mineral density at the lumbar spine (primary end point; P<0.001), regardless of dose frequency (monthly or every 3 months) and dose level (140 mg or 210 mg) (Table S2 in the Supplementary Appendix). In addition, each of the five romosozumab groups, as compared with the pooled placebo group, had a significant increase in bone mineral density at the lumbar spine (Figure 2 (Figure 2, and Table S4 in the Supplementary Appendix) and femoral neck (Figure 2, and Table S5 in the Supplementary Appendix) (P<0.001 for all comparisons).
The largest gains were observed with the 210-mg monthly dose of romosozumab, with mean increases from baseline to 12 months of 11.3% at the lumbar spine, 4.1% at the total hip, and 3.7% at the femoral neck. These increases were significantly greater than those observed in the alendronate and teriparatide groups (P<0.001 for all three comparisons) (Figure 2, and Tables S3, S4, and S5 in the Supplementary Appendix). No noteworthy differences in bone mineral density at the distal third of the radius were observed at 12 months between any of the romosozumab groups and the pooled placebo group, the alendronate group, or the teriparatide group (Table S6 in the Supplementary Appendix).
The bone mineral density at the lumbar spine and total hip was also significantly increased at month 6 in all the romosozumab groups as compared with the pooled placebo group (P≤0.006) (Figure 2, and Tables S3 and S4 in the Supplementary Appendix). The increases in bone mineral density at the femoral neck were significantly greater at month 6 in the groups that received romosozumab in doses of 140 mg monthly, 210 mg monthly, and 210 mg every 3 months than in the pooled placebo group (P<0.02) (Figure 2, and Table S5 in the Supplementary Appendix). Increases in bone mineral density at the lumbar spine, total hip, and femoral neck at month 6 were also significantly greater in the groups that received the two highest doses of romosozumab (140 mg monthly and 210 mg monthly) than in the groups that received alendronate or teriparatide (P≤0.01).
Markers of Bone Turnover
In all the romosozumab groups, increases in bone-formation markers were transitory. Increases were noted 1 week after the initial dose was administered and were greatest at month 1. The levels returned to baseline values or fell below baseline values between months 2 and 9, depending on the dose and the marker (Figure 3A, and Tables S7, S8, and S9 in the Supplementary Appendix).
In all the romosozumab groups, the level of the bone-resorption marker serum β-CTX initially decreased from baseline, with the largest median decrease apparent in the first week. In the groups that received monthly doses of romosozumab, the levels of serum β-CTX remained below baseline values at month 12 (Figure 3B, and Table S10 in the Supplementary Appendix).
Treatment with romosozumab was associated with a decrease from baseline in the serum calcium level (Table S11 in the Supplementary Appendix). The mean nadir was observed by month 1, at which time the change from baseline ranged from -0.03 to -0.07 mmol per liter (-0.12 to -0.28 mg per deciliter), representing a decrease of 1.30 to 2.68% from baseline, following a dose-response pattern. The serum calcium levels returned to baseline values at subsequent visits. Compensatory increases from baseline in the level of parathyroid hormone were observed in the romosozumab groups in a dose-dependent manner (Table S12 in the Supplementary Appendix). No adverse events of hypocalcemia were reported.
Adverse Events and Safety
The proportions of participants reporting adverse events and serious adverse events were similar in the pooled placebo group and the romosozumab groups (Table 3, and Table S13 in the Supplementary Appendix). No apparent relationship between dose and adverse events was observed. Injection-site reactions were observed more frequently with romosozumab than with placebo, but no dose-response relationship was observed. These reactions were generally mild, did not lead to discontinuation of the study drug or withdrawal from the study, and did not generally recur with continued administration of romosozumab.
The incidence of serious adverse events was 14% in the placebo group (7 of 50 participants), 8% in the alendronate group (4 of 51), 9% in the teriparatide group (5 of 54), 10% in the group that received the 210-mg monthly dose of romosozumab (5 of 51), and 7% across all romosozumab groups (17 of 255). Among the 5 participants with serious adverse events in the group that received the 210-mg monthly dose of romosozumab, the following events were reported (in 1 participant each): breast cancer, chronic obstructive pulmonary disease, noncardiac chest pain, wrist fracture (radius and ulna), and renal oncocytoma (benign).
No serious adverse event was reported by more than 1 participant in any group, and none of the serious adverse events were considered by the investigator to be treatment-related. One death due to colon cancer was reported in a participant who received placebo, and one death associated with a postoperative ileus (after aortobifemoral bypass) was reported in the group that received the 70-mg monthly dose of romosozumab.
No notable changes from baseline in vital signs, laboratory values, or electrocardiographic variables were noted in any of the participants. All the participants who received placebo, alendronate, or teriparatide tested negative for binding antibodies. Among participants who received romosozumab, binding antibodies were identified in 20% and antibodies with in vitro neutralizing activity in 3%. The development of antibodies had no discernible effect on adverse events, pharmacokinetics, or pharmacodynamics.