Romosozumab versus Teriparatide for the Treatment of Postmenopausal Osteoporosis: A Systematic Review and Meta‐analysis through a Grade Analysis of Evidence

Abstract

Objective

To provide a systematic review about the efficacy and safety of romosozumab and teriparatide for the treatment of postmenopausal osteoporosis.

Method

Randomized controlled trials (RCTs) were searched from electronic databases, including PubMed (1996 to June 2019), Embase (1980 to June 2019), Cochrane Library (CENTRAL, June 2019), Web of Science (1998 to June 2019), and others. The primary outcomes included the following: the percentage change in bone mineral density of lumbar spine and total hip from baseline at month 6 and month 12 in each group. The secondary outcomes included the following: the percentage change in bone mineral density of femoral neck from baseline at month 6 and month 12 in each group and the incidence of adverse events at month 12 in each group.

Results

Four studies containing 1304 patients met our selection criteria. The result of our analysis indicated that romosozumab showed better effects in improving BMD of lumbar spine (month 6: MD = 3.54, 95% CI [3.13, 3.94], P<0.001; month 12: MD = 4.93, 95% CI [4.21, 5.64], P<0.001), total hip (month 6: MD = 2.27, 95% CI [0.62, 3.91], P = 0.007; month 12: MD = 3.17, 95% CI [2.68, 3.65], P<0.001), and femoral neck (month 6: MD = 2.30, 95% CI [0.51, 4.08], P = 0.01; month 12: MD = 3.04, 95% CI [2.29, 3.78], P<0.001). Also, the injection‐site reaction was less (month 12: RR = 2.84, 95% CI [1.22, 6.59], P = 0.02), but there were no significant difference in the incidence of serious adverse events (month 12: RR = 0.78, 95% CI [0.46, 1.33], P = 0.37) and death (month 12: RR = 0.61, 95% CI [0.08, 4.62], P = 0.63).

Conclusion

Based on the available studies, our current results demonstrate that romosozumab was better than teriparatide both in terms of efficacy and side effects.

To provide a systematic review about the efficacy and safety of romosozumab and teriparatide for the treatment of postmenopausal osteoporosis, Randomized controlled trials were searched from electronic database and then articles were screened according to inclusion and exclusion criteria. Risk of bias were assessmented, and statistical analysis and data synthesis were conducted with GRADE system to evaluate the level of the evidence. Our results demonstrate that romosozumab was better than teriparatide both in terms of efficacy and side effects.

Introduction

Postmenopausal osteoporosis is identified as a systemic skeletal disorder characterized by low bone mineral density (BMD) and qualitative changes in microarchitecture of bone tissue, with a consequent increase in bone fragility and susceptibility to fracture ¹ . In elderly patients, osteoporotic fracture (fragility fracture) is a catastrophic complication, which causes substantial morbidity and mortality ² . This fracture often occurs in the spine, hip, and wrist, but also affects other bones, such as the humerus and radius ³ . Drugs for postmenopausal osteoporosis fall into two major categories, antiresorptive drugs and osteoanabolic drugs. Antiresorptive drugs for postmenopausal osteoporosis increase those for bone mineral density and prevent the progression of structural damage but may not restore bone structure ⁴ . However, osteoanabolic drugs can reverse microarchitectural deterioration of bone tissue and seem to be better. For postmenopausal osteoporosis treatment, the classic drug bisphosphonate represents the vast majority of prescriptions, and is a conventional drug. However, long‐term use of bisphosphonates may cause atypical fractures, and intravenous use of bisphosphonates may cause osteonecrosis of the jaw. Teriparatide (brand name FOTTEOTM), an N‐terminal (1–34) fragment of human parathyroid hormone, was the first osteoanabolic drug approved by the Food and Drug Administration in 2003 ⁹ , ¹⁰ . It can significantly improve BMD. However, patients must inject this drug once each day in their thigh or abdomen. Besides, after teriparatide is discontinued, its benefits are quickly lost ¹¹ . What's worse, a study of the Forteo Patient Registry (FPR) anticipated that they will be able to detect a fourfold increase in the risk of osteosarcoma if one exists by 2024 ¹² .

Sclerostin, a glycoprotein produced primarily by osteocytes, is encoded by the SOST gene, which can specifically block the canonical Wnt signaling ¹³ , ¹⁴ . This pathway plays a pivotal role in promoting bone formation and regulating bone homeostasis ¹⁵ . Sclerostin increases the expression of RANKL and decreases that of OPG, resulting in bone absorption ¹⁶ , ¹⁷ . Romosozumab, a humanized monoclonal anti‐sclerostin antibody, is a new osteoanabolic drug that inhibits sclerostin with a dual effect on bone, increasing bone formation and decreasing bone resorption ⁴ .

However, the efficacy and safety of this new drug are not well‐documented. Therefore, we conducted a systematic review and meta‐analysis of randomized controlled trials (RCTs) of romosozumab and teriparatide to fully evaluate their effects in postmenopausal osteoporosis patients.

Materials and Methods

Search Strategy

The electronic databases PubMed, Embase, the Cochrane Library, Web of Science, and the Cochrane Controlled Trials Register were searched up to June 2019. The search terms were as follows: ([AMG 785 OR evenity OR romosozumab OT CDP 7851] AND hPTH [1–34] OR Human Parathyroid Hormone [1–34] OR Parathar OR Forteo) AND (postmenopausal osteoporosis OR Postmenopausal Bone Loss). The Flow chart of the trial selection process was presented in Fig. 1. We also used the PRISMA guidelines ¹⁸ , GRADE system ¹⁹ , and Cochrane Handbook ²⁰ to assess the quality of the included studies to make sure the data were reliable and veritable.

Fig. 1.

Flow chart of database searching, records screening, assessment of full‐text articles, and article inclusion.

Selection Criteria

Trials were included on conditions that they met the PICOS (population, intervention, comparator, outcome, study design) criteria.

• (i) Population: Female patients with postmenopausal osteoporosis.
• (ii) Intervention: Romosozumab.
• (iii) Comparator: Teriparatide.
• (iv) Outcomes: The primary outcomes included the following: the percentage change in bone mineral density of lumbar spine and total hip from baseline at month 6 and month 12 in each group. The secondary outcomes contained the following: the percentage change in bone mineral density of femoral neck from baseline at month 6 and month 12 in each group and the incidence of adverse events at month 12 in each group.
• (v) Study design: RCT.

Exclusion criteria were as follows: (i) non‐RCTs carried out in individuals with other disorders likely to affect bone and calcium metabolism (such as chronic kidney disease, pregnancy, and glucocorticoid use) or conducted in specific populations that might have a different risk of cardiovascular (CV) events (patients with cancer, transplant, or human immunodeficiency virus infection, or children); (ii) studies with duration less than 6 months; (iii) studies with zero CV events or without safety data published.

Data Extraction

A standard data extraction form was used to collect the relevant data from included studies. Two reviewers collected available data from included studies independently, and any disagreement between the two reviewers was judged by a third reviewer. The relevant data included authors, published dates, intervention types, age, sample size, outcomes, duration of follow‐up, and reference type. Baseline characteristics of included trials were presented in Table 1. Data on BMD (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) were obtained from the data presented in tables or figures if no direct data were available from the article text.

TABLE 1.

Baseline characteristics of included randomized controlled trials

Study (year)	Intervention	Age (years, mean ± SD)	Number of patients with LS BMD	Number of patients with TH BMD	Number of patients with FN BMD	Outcomes	Follow‐up (months)	Reference type
Genant et al. 2017 21	Romosozumab 210 mg per month Teriparatide 20 μg per day	64.3 ± 4.7 65.8 ± 5.7	24 30	9 19	— —	BMD changes at LS, TH, FN; incidence of AEs	12	RCT
Keaveny et al. 2017 22	Romosozumab 210 mg per month Teriparatide 20 μg per day	64.3 ± 4.7 65.8 ± 5.7	24 28	9 19	— —	BMD changes at LS, TH, FN; incidence of AEs	12	RCT
Langdahl et al. 2017 23	Romosozumab 210 mg per month Teriparatide 20 μg per day	71.8 ± 7.4 71.2 ± 7.7	206 209	206 209	206 209	BMD changes at LS, TH, FN; incidence of AEs	12	RCT
McClung et al. 2014 4	Romosozumab 210 mg per month Teriparatide 20 μg per day	66.3 ± 6.5 66.8 ± 5.7	49 46	49 46	49 46	BMD changes at LS, TH, FN; incidence of AEs	12	RCT

AEs, Adverse effect; BMD, Bone mineral density; FN, Femoral neck; LS, Lumber spine; RCT, randomized controlled trial; SD, Standard Deviation; TH, Total hip.

Risk of Bias Assessment

According to the Cochrane Handbook for Systematic Reviews of Interventions ²⁰ , the methodological quality and basis of the included literature were assessed as follows: randomization, allocation concealment, blind method, selective reporting, incomplete outcome data, and other bias (Figs 2 and 3).

Fig. 2.

Risk of bias graph according to the Cochrane Handbook for Systematic Reviews of Interventions.

Fig. 3.

Risk of bias summary according to the Cochrane Handbook for Systematic Reviews of Interventions.

Grading Quality of Evidence

We used the GRADE system to evaluate the level of the evidence and strength of recommendations for included outcomes. GRADE software was used to evaluate the evidence of included outcomes. Initially, RCTs were considered as high confidence in an estimate of effect and cohort studies were considered as low confidence in an estimate of effect. Reasons that might decrease the level of confidence include limitations, inconsistency, indirectness, imprecision, and publication bias. Reasons that might raise the level of confidence include large effect, plausible confounding, dose‐response. The GRADE evidence was divided into the following categories: (i) high‐quality evidence, which indicated that further research was unlikely to change the confidence in an estimate of effect; (ii) moderate‐quality evidence, which indicated that further research was likely to have an important impact on confidence in an estimate of effect and may change the estimate; (iii) low‐quality evidence, which indicated that further research was likely to have an important impact on confidence in an estimate of effect and was likely to change the estimate; and (v) very low‐quality evidence, which indicated that we were very uncertain about the results. The results of the GRADE analysis were presented in Table 2.

TABLE 2.

The GRADE evidence quality for each outcome

No of studies	Design	Decrease quality of evidence					Increase quality of evidence			Quality	Importance
		Limitations	Inconsistency	Indirectness	Imprecision	Publication bias	Large effect	Plausible confounding	Does‐response
Lumbar spine month 6	RCT	No	Serious	No	No	Likely	Large	No	No	Moderate	Critical
Lumbar spine month 12	RCT	No	Serious	No	No	Likely	Large	No	No	Moderate	Critical
Total hip month 6	RCT	No	Very serious	No	No	Likely	Large	No	No	Low	Important
Total hip month 12	RCT	No	Serious	No	No	Likely	Large	No	No	Moderate	Critical
Femoral neck month 6	RCT	No	Very serious	No	No	Likely	Large	No	No	Low	Important
Femoral neck month 12	RCT	No	Very serious	No	No	Likely	Large	No	No	Low	Important
Incidence of SAEs	RCT	No	No	No	No	Likely	No	No	No	Moderate	Critical
Incidence of death	RCT	No	No	No	No	Likely	No	No	No	Moderate	Critical
Incidence of Injection‐site reaction	RCT	No	No	No	No	Likely	Large	No	No	High	Critical

RCT, randomized controlled trial; High quality—further research is very unlikely to change our confidence in the estimate of effect; Moderate quality—further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate; Low quality—further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate; Very low quality—we are very uncertain about the estimate.

Statistical Analysis and Data Synthesis

Meta‐analyses were performed with Review Manager Software for Windows (version 5.3; Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014). The mean difference (MD) was used to assess continuous outcomes in month 6 and 12, such as BMD of different parts, with a 95% confidence interval (CI). Relative risks (RR) with a 95% CI were used to assess dichotomous outcomes, such as AEs. The inverse variance and Mantel–Haenszel methods were used to combine separate statistics. If P values were < 0.05, the results were considered statistically significant.

Statistical heterogeneity of the included studies was evaluated using the chi‐square test in accordance with the values of P and I ² . If the values of I ²  < 50%, the heterogeneity might not be important. A fixed‐effects model was used to assess these outcomes. If I ² was between 50% and 100%, it could represent substantial heterogeneity. We used random effects model to evaluate these outcomes. Thresholds for the interpretation of I ² can be misleading, since the importance of inconsistency depends on several factors. Therefore, subgroup analysis or sensitivity analysis was performed to interpret the potential source of heterogeneity. Because only four studies were included, publication bias test were not necessary.

Results

Characteristics of Included Studies

All patients were aged over 60 years. All follow‐up periods were 1 year. All postmenopausal women had a T score of −2.0 to −3.5 at the total hip or femoral neck. Patients were randomly assigned to receive subcutaneous injections of romosozumab (at a dose of 210 mg daily) or teriparatide (20 μg once daily) monthly for 12 months; thereafter, patients in each group received denosumab for 12 months, at a dose of 60 mg, administered subcutaneously every 6 months. The end points were the cumulative incidences of new vertebral fractures at 12 months and 24 months. Secondary end points included clinical (a composite of non‐vertebral and symptomatic vertebral) and non‐vertebral fractures.

Search Results

Initially, 198 citations were identified from electronic databases, of which 169 records were excluded by primary screening. After reading the full text of all remaining 29 studies in detail, 25 studies were also excluded according to the inclusion and exclusion criteria. Finally, four RCTs ⁴ , ²¹ , ²² , ²³ were included. But only two studies ⁴ , ²³ had the data measured in month 6 and the data of femoral neck BMD. The characteristics of the included studies were summarized in Table 1.

Primary Outcome

The BMD of lumbar spine and total hip were the primary outcome in our meta‐analysis, which were used to evaluate the therapeutic effect of postmenopausal osteoporosis. The treatment period was divided into two subgroups (month 6 and 12).

BMD of Lumbar Spine

Four studies assessed lumbar spine BMD of 620 patients through month 12 (I²  = 75%, P = 0.04) and two studies assessed 514 patients through month 6 (I²  = 86%, P < 0.001). Data were pooled according to the random effects model because of high heterogeneity. Compared with teriparatide, romosozumab significantly improved the BMD (month 6: MD = 3.54, 95% CI [3.13, 3.94], P<0.001; month 12: MD = 4.93, 95% CI [4.21, 5.64], P<0.001; Fig. 4).

Fig. 4.

A forest plot diagram showing lumbar spine bone mineral density at month 12 and month 6.

BMD of Total Hip

Four studies assessed total hip BMD of 570 patients through month 12 (I² = 84%, P < 0.001) and 514 patients through month 6 (I²  = 99%, P < 0.001). Data were pooled according to the random effects model because of high heterogeneity. Compared with teriparatide, romosozumab significantly improved the BMD (month 6: MD = 2.27, 95% CI [0.62, 3.91], P = 0.007; month 12: MD = 3.17, 95% CI [2.68, 3.65], P<0.001; Fig. 5).

Fig. 5.

A forest plot diagram showing total hip bone mineral density at month 12 and month 6.

Secondary Outcome

BMD of Femoral Neck

The BMD of femoral neck was reported in two studies ⁴ , ²³ , including 514 patients in month 6 (I²  = 99%, P < 0.001) and the same sample size in month 12 (I ²  = 94%, P < 0.001). Data were pooled according to the random effects model because of high heterogeneity. The percentage change from baseline in the romosozumab group was significantly improved (month 6: MD = 2.30, 95% CI [0.51, 4.08], P = 0.01; month 12: MD = 3.04, 95% CI [2.29, 3.78], P<.001; Fig. 6).

Fig. 6.

A forest plot diagram showing femoral neck bone mineral density at month 12 and month 6.

Adverse Events

There were also two studies ⁴ , ²³ evaluating the incidence of adverse events. The common points of interest were serious adverse event (I²  = 0%, P = 0.58), death (I²  = 0%, P = 0.63), and injection‐site reaction (I²  = 0%, P = 0.91). No significant differences were found between the two groups in the incidence of serious adverse events (month 12: RR = 0.78, 95% CI [0.46, 1.33], P = 0.37) and death (month 12: RR = 0.61, 95% CI [0.08, 4.62], P = 0.63). However, romosozumab could significantly alleviate the local response (month 12: RR = 2.84, 95% CI [1.22, 6.59], P = 0.02; Fig. 7).

Fig. 7.

A forest plot diagram showing the incidence of adverse events in month 12.

Discussion

With an aging population, postmenopausal osteoporosis, especially the most common postmenopausal osteoporosis has brought great economic burden to global public health, and also seriously affected the quality of life of patients themselves ²⁴ , ²⁵ . In America, age‐related fractures are projected to increase nationally to over 3 mn fractures in 2025 ²⁶ . The process of aging in women is associated with an increase in the rate of bone remodeling in both cancellous and cortical bone, combined with a negative remodeling balance, resulting in bone loss and disruption of bone microarchitecture ¹ . It is generally believed that the key to the treatment of postmenopausal osteoporosis is to restore the dynamic balance of bone metabolism, and the signal pathway between cells has become a key to research ²⁷ . With the emergence of new signaling pathways, new avenues targeting them are also emerging, such as melatonin ²⁸ .

The real‐life challenge, however, is rooted in the long therapeutic procedure for postmenopausal osteoporosis. Today, generally speaking, women still have a long life expectancy, possibly 30 years or more after menopause, and their fracture risk increases exponentially with age. There are few clinical extension trials for over 10 years for the treatment of postmenopausal osteoporosis, especially those of antiresorptive therapies. Additionally, serious adverse effects such as osteonecrosis of the jaw and atypical femoral fractures have been related to extended antiresorptive therapy, raising concerns of increased risks due to continuous inhibition of bone resorption. Osteoanabolic therapy is currently limited to 24 months of teriparatide treatment ² .

There are two major categories of drugs for postmenopausal osteoporosis, antiresorptive drugs, and osteoanabolic drugs. The former inhibit the recruitment and activity of osteoclasts, and probably do not fully correct the negative remodeling balance. The latter have anabolic skeletal effects, which can be achieved through changes in bone remodeling, bone modeling, or a combination of the two ¹ . Except teriparatide, there is another osteoanabolic drug, abaloparatide (brand name TymlosTM). It was approved by the FDA on 28 April 2017. Abaloparatide is a synthetic analogue of PTHrP, which can increase bone mass in animals ²⁹ and in humans ³⁰ . But patients still need daily subcutaneous injections like teriparatide. For a long time, abaloparatide injected subcutaneously in rats resulted in dose‐ and time‐dependent formation of osteosarcomas, with a comparable response to h‐PTH (1‐34) at similar exposure ³¹ . Although abaloparatide can reduce hypercalcaemia, its registration was denied in Europe on the grounds of concerns about its effectiveness in reducing nonvertebral fractures, and increases in heart rate and palpitations ¹ . The EMA's Committee for Medicinal Products for Human Use (CHMP) thought some data from study sites of abaloparatide were not reliable and had to be excluded as the study had not been conducted in compliance with “good clinical practice” (GCP) at those sites ³² . After a number of clinical trials in postmenopausal women with osteoporosis ²³ , ³³ , ³⁴ , ³⁵ , ³⁶ , ³⁷ , romosozumab (EVENITYTM) has been proved to be safe and effective, and was approved by the FDA in 2019 ³⁸ .

There is a new avenue for the treatment of postmenopausal osteoporosis. Dating back to 1979, Frost first proposed the concept of sequential therapy for postmenopausal osteoporosis ³⁹ . However, relevant DATA‐Switch studies have not attracted enough attention until now, and have shown better outcomes than monotherapy ⁴⁰ , ⁴¹ , ⁴² , ⁴³ . Similarly, there are few similar studies on romosozumab. Compared with monotherapy, the transition from romosozumab to other antiresorptive drugs may further increase BMD in postmenopausal osteoporotic women. Through bone‐targeting systems to deliver siRNA are also a new method for postmenopausal osteoporosis, this has already been examined in a preclinical study ⁴⁴ .

In summary, all the current drugs for postmenopausal osteoporosis more or less have some side effects or lack efficacy, and a very ideal postmenopausal osteoporosis therapy has not yet been developed. Other than drugs, good nutrition, regular physical activity, avoiding harmful lifestyle habits, and fall prevention are recommended for all patients at risk of postmenopausal osteoporosis and should be considered of equal value as medical treatment ⁴⁵ .

In our study, compared with teriparatide, romosozumab had better effectiveness for the treatment of postmenopausal osteoporosis, especially in increasing the BMD of lumbar spine, total hip, and femoral neck, decreasing the incidence of injection‐site reaction. But on the grounds of the concerns about small sample size, incomplete data, and heterogeneity for RCTs included, further studies are required to demonstrate our results.

Limitations

Our meta‐analysis has several limitations: (i) there were only four RCTs in our meta‐analysis, the sample size of included studies was small (N = 1304); (ii) in regard to the significant heterogeneity of LS BMD (I²  = 86%) and TH BMD (I²  = 84%), although we used a random effects model, we tried to find the source of heterogeneity. When we excluded the study of Keaveny et al., the heterogeneity of LS BMD (I²  = 71%) reduced at a level; when we excluded the study of Langdahl et al. and McClung et al., the heterogeneity of TH BMD (I²  = 0) reduced significantly. Therefore, we thought those excluded studies might be the source. The heterogeneity of FN BMD (I² = 99% in month 6, I² = 94% in month 12) could not be analyzed, because there were only two RCTs included; (iii) some original data could not be directly acquired; (iv) the follow‐up period was too short in included trials, so some AEs might not be revealed. The effectiveness and safety needed longer follow‐up time to be confirmed; and (v) only English publications were included in our meta‐analysis.

Conclusions

According to our study, romosozumab can significantly increase the BMD of lumbar spine, total hip, and femoral neck, with a lower incidence of injection‐site reaction. And those differences all have statistical significance (P < 0.05). There is no difference in the incidence of death and other serious adverse events. Fewer adverse events (P < 0.05) and longer half‐life may improve the compliance of patients and reduce the loss of benifits.

Author Contributions

Xinlong Ma and Jianxiong Ma contributed to the conception of the study. Shan Zhu contributed significantly to analysis and manuscript preparation. Aixian Tian and Haobo Jia performed the data analyses and wrote the manuscript. Bin Lu and Yan Li helped perform the analysis with constructive discussions.