Serum sclerostin levels in children with osteogenesis imperfecta

Abstract

Context

Sclerostin inhibits bone formation via the WNT/β-catenin pathway and is a therapeutic target in osteoporosis. Antibodies against sclerostin are currently under investigation for the treatment of osteogenesis imperfecta (OI), a rare, genetically and clinically heterogeneous bone fragility disorder. However, data on serum sclerostin levels in pediatric and adolescent patients with OI remain limited.

Design

This is a retrospective, cross-sectional analysis of serum sclerostin levels in a genetically heterogeneous cohort of children and adolescents with OI.

Patients and methods

80 serum samples from 74 OI patients (median age 8.9 years, range 0.1–20.7 years), classified by clinical severity and affected gene, were collected. Serum levels of sclerostin, osteoprotegerin (OPG), parathyroid hormone (PTH), alkaline phosphatase (AP) and 25-Hydroxy Vitamin D (25(OH)D) were measured and analyzed according to genotype and OI severity.

Results

The median serum sclerostin level in this cohort was 0.35 ng/mL (IQR 0.28–0.52). Disease severity showed an inverse correlation with serum sclerostin levels (Spearman ρ = −0.4547, 95% CI [−0.62, −0.25], P < 0.0001). Multivariable linear regression analysis revealed genotype-specific differences in sclerostin levels, particularly in patients with BMP1 or WNT1 mutations compared with other mutation subgroups. No significant correlations were found between sclerostin and OPG, PTH, 25(OH)D, or AP.

Conclusions

In this cohort of children and adolescents with OI, disease severity was inversely associated with serum sclerostin levels, and genotype-specific differences may reflect distinct pathophysiological mechanisms of bone metabolism. These findings may contribute identifying patient subgroups most likely to benefit from targeted anti-sclerostin therapies.

Keywords

Osteogenesis imperfecta WNT1 BMP1 osteoanabolic treatment sclerostin

Introduction

Elucidating the different modifying pathways of bone formation and resorption led to new treatment approaches. Sclerostin, a glycoprotein encoded by SOST gene, is a potent inhibitor and negative modifier of bone formation through the WNT/β-catenin pathway. Secreted primarily by osteocytes, sclerostin suppresses osteoblast differentiation, thereby reducing bone formation. Moreover, it plays a role in the adaptation of the bone structure according to biomechanical needs.^1–4 In murine models, overexpression of SOST has been associated with reduced bone mass, while genetic inactivation led to increased bone mass.^5,6 Consistent with these findings, individuals with SOST loss-of-function mutations exhibit a high bone mass phenotype and excessive bone overgrowth.^7,8 Although sclerostin is known to act primarily in a paracrine rather than an endocrine manner,^9,10 circulating serum sclerostin levels have been shown to correlate with bone marrow plasma concentrations, suggesting that they reflect local bone metabolism.¹¹ Counterintuitively to findings on SOST expression, a positive correlation between serum sclerostin levels and bone mineral density (BMD) was observed and has been interpreted in different ways—either as a counter-regulatory mechanism within bone homeostasis or as an indirect reflection of the total number of osteocytes, under the assumption that this number is proportional to overall bone mass.^12–14

Osteogenesis imperfecta (OI) is a rare genetic disease characterized by low bone mass, bone fragility, skeletal deformities and short stature. More than 90% of patients present with a mutation in one of the collagen genes COL1A1/A2 (collagen type I alpha 1 chain/collagen type I alpha 2 chain gene) leading to the altered bone structure. In the past decade, additional causative mutations have been identified in genes affecting diverse molecular pathways, including those influenced by sclerostin signaling.^15,16

Based on its role in bone remodeling, sclerostin has emerged as a potential target of interest for therapeutic intervention in patients at risk of fractures. Recently, Romosozumab—an anti-sclerostin antibody—has been approved for the treatment of postmenopausal osteoporosis.¹⁷ Furthermore, two phase 3 clinical trials are currently investigating anti-sclerostin antibodies (Romosozumab - NCT05972551, Setrusumab - NCT05768854) as potential treatments in children with OI. Preclinical studies in different murine OI models have shown promising results, including reduced fracture rates and increased cortical thickness and bone mass.^18–21 These findings are further supported by the results of the phase 2b study of Setrusumab in adults with OI (NCT03118570), which demonstrated a dose-dependent increase in areal BMD of the lumbar spine and total radial volumetric BMD after 12 months of treatment.²² Very recently, it was announced that NCT05768854 did not meet its primary endpoint of reducing the annualized clinical fracture rate, while improvements in bone mineral density were reported with strong statistical significance.²³

However, despite the growing interest in targeting sclerostin for OI treatment, data on circulating sclerostin levels in affected individuals—particularly children—remain scarce and inconsistent. In addition, previous studies that measured serum sclerostin levels in patients with OI^24–28 primarily categorized patients according to clinical OI types (I–IV), without stratification based on the underlying causative mutations beyond COL1A1 or COL1A2. Consequently, potential genotype-specific differences in sclerostin regulation remain largely unexplored. Moreover, the interaction between sclerostin, osteoprotegerin (OPG) and RANKL (receptor activator of nuclear factor kappa-B ligand) has not been systematically investigated in pediatric patients with OI.

The aim of this study was to evaluate serum sclerostin levels in children and adolescents with different types of OI. We hypothesized that different OI genotypes may be associated with distinct sclerostin profiles.

Methods

Patients and sample collection

A total of 80 blood samples from 74 patients with OI were analyzed. All included patients were seen consecutively in our specialized clinic for skeletal dysplasia and had a genetically confirmed diagnosis of OI. Patients were clinically classified as having mild, moderate, or severe OI based on morbidity, mobility, and phenotypic presentation. Mildly affected patients were independent of assistive mobility devices and experienced fewer than two fractures per year. Moderately affected patients required assistive devices for daily mobility and had undergone surgical interventions to correct skeletal deformities. Severely affected patients were wheelchair-dependent and continued to experience fractures and skeletal deformities despite surgical intervention.

Genetic categorization was based on the affected gene, resulting in 10 distinct groups: COL1A1 (MIM 120,150), COL1A2 (MIM 120,160), CRTAP (MIM 605,497), LEPRE1 (MIM 610,339), FKBP10 (MIM 607,063), BMP1 (MIM 112,264), TAPT1 (MIM 612,758), SERPINF1 (MIM 172,860), PLOD2 (MIM 601,865), and WNT1 (MIM 164,820). Written informed consent for participation, data collection, analysis and publication was obtained from all patients and their legal guardians. Ethical approval was obtained from the local ethics committee of the University (approval number: 13-202).

Blood samples were collected during routine clinical visits and standard blood draws between August 2011 and August 2014. Two patients had multiple samples taken at different time points (one patient with five and another with three samples, respectively).

Laboratory measurements

Serum levels of sclerostin, OPG, parathyroid hormone (PTH), alkaline phosphatase (AP) and 25-Hydroxy Vitamin D (25(OH)D) were measured. PTH, AP and 25(OH)D levels were determined using standard laboratory methods.

Blood samples for sclerostin and OPG were collected in serum tubes, allowed to clot, and centrifuged to obtain serum. Serum was aliquoted and stored at −80°C until analysis. All samples were visually inspected prior to analysis, and samples with evidence of hemolysis were excluded. According to the manufacturer, both assays are stable under storage at ≤ −80°C for >24 months, and up to three freeze–thaw cycles do not affect analyte stability. The median storage duration of the analyzed samples was 7.0 months (IQR 5.7–11.1 months), and no sample exceeded 24 months of storage prior to measurement.

Sclerostin concentrations were measured using a high-sensitivity human Sclerostin ELISA kit (Ref. TE1023HS; TECOmedical Group). The assay has a limit of detection (LOD) of 0.008 ng/mL and a limit of quantification (LOQ) of 0.01–3.34 ng/mL, with a calibration range of 0–3 ng/mL. OPG levels were determined with the MicroVue OPG ELISA kit (Ref. 8034; QUIDEL Corporation), with a limit of detection of 0.13 pmol/L and a calibration range of 1.875–60 pmol/L All measurements were performed according to the manufacturers’ instructions.

Statistical analysis

All statistical analyses were performed using only the first measurement per patient (n = 74), thereby ensuring a patient-level unit of analysis. The total of 80 measurements is reported only for descriptive purposes.

Statistical analyses were conducted using GraphPad Prism version 10.3 and R version 4.6.0. Normality was assessed using the Shapiro-Wilk test. Since sclerostin levels were not normally distributed, data are presented as median (interquartile range, IQR). Serum sclerostin concentrations were additionally transformed into age- and sex-adjusted z-scores based on reference percentiles published by Fischer et al.,²⁹ using the same assay kit. Median sclerostin concentrations and corresponding z-scores were analyzed using the Wilcoxon signed-rank test. Cohort-specific percentiles were used for descriptive visualization (Figure 3).

Group comparisons of sclerostin levels across disease severity groups (mild, moderate, severe) were performed using the Kruskal-Wallis test followed by pairwise Wilcoxon rank-sum (Mann-Whitney U) tests with FDR correction (Benjamini-Hochberg). Effect sizes were estimated as Hodges-Lehmann median differences with 95% confidence intervals.

Due to skewed distributions, both sclerostin concentrations and time since last antiresorptive treatment were log-transformed prior to regression analyses. Multivariable linear regression analyses were performed with log-transformed sclerostin concentrations as the dependent variable and age, sex, disease severity, mutation group, and log-transformed time since last antiresorptive treatment as independent variables. To account for multiple testing, false discovery rate (FDR) correction using the Benjamini-Hochberg procedure was applied and adjusted P-values (q-values) were used for statistical inference. A q-value <0.05 was considered statistically significant. Spearman correlation analyses were used to assess associations between sclerostin levels and disease severity, serum OPG, PTH, 25(OH)D and AP.

Results

Clinical data

A total of 74 children and adolescents with OI (29 females) were included in the study. Clinical data for these patients are presented in Table 1. The median age at the time of the first sclerostin measurement was 8.9 years with an age range of 0.1–20.7 years. Based on morbidity, mobility and phenotypical presentation (as described in the Methods section), 18 patients were classified as having mild OI, 17 as moderate, and 39 as severe. Genotype distribution across severity groups is shown in Table 1.

Table 1.

Clinical data of the cohort.


Patients (n)	74
Sex (female/male)	29/45
Age (in case of multiple measurements: Age of first measurement) (years)
median (IQR)	8.9 (5.7–13.0)
range	0.1–20.7
OI severity (n)
mild	18
moderate	17
severe	39
Affected gene (n)
COL1A1	34
COL1A2	20
CRTAP	2
LEPRE1	4
FKBP10	2
BMP1	3
TAPT1	1
SERPINF1	5
PLOD2	1
WNT1	2
OI severity (n)
mild	18
COL1A1	11
COL1A2	7
moderate	17
COL1A1	11
COL1A2	5
WNT1	1
severe	39
COL1A1	12
COL1A2	8
CRTAP	2
LEPRE1	4
FKBP10	2
BMP1	3
TAPT1	1
SERPINF1	5
PLOD2	1
WNT1	1
Previous antiresorptive treatment before sclerostin measurement (n)	71

Abbreviations: IQR – interquartile range.

OMIM numbers of the listed genes: COL1A1 (MIM 120,150), COL1A2 (MIM 120,160), CRTAP (MIM 605,497), LEPRE1 (MIM 610,339), FKBP10 (MIM 607,063), BMP1 (MIM 112,264), TAPT1 (MIM 612,758), SERPINF1 (MIM 172,860), PLOD2 (MIM 601,865), WNT1 (MIM 164,820).

The most frequently affected genes were COL1A1 (n = 34) and COL1A2 (n = 20). Other causative genes included CRTAP (n = 2), LEPRE1 (n = 4), FKBP10 (n = 2), BMP1 (n = 3), TAPT1 (n = 1), SERPINF1 (n = 5), PLOD2 (n = 1), and WNT1 (n = 2). Of the 74 participants, 71 had received antiresorptive treatment (denosumab or bisphosphonates) prior to sclerostin measurement; for 70 patients, the exact time since the last bisphosphonate administration was available. Therapy duration until measurement and time since last dose for all patients, as well as for the two patients with multiple measurements, are provided in Tables S1 and S2.

Serum sclerostin levels

Serum sclerostin levels were measured in all 74 patients. Two patients had multiple samples collected at different time points—one with a SERPINF1 mutation (five samples) and the other with a WNT1 mutation (three samples)—resulting in a total of 80 measurements (Table 2). For all statistical analyses, only the first measurement per patient was included.

Table 2.

Sclerostin and OPG concentrations (median [IQR] or individual values).

	Sclerostin, n ≥ 3: Median (IQR), n < 3: Individual values, [ng/ml]	n_measurements (n_patients)	OPG, n ≥ 3: Median (IQR), n < 3: Individual values, [pmol/l]	n_measurements (n_patients)
All measurements	0.36 (0.28–0.55)	80 (74)	4.7 (3.9–5.3)	58 (57)
Excluding patient(s) with multiple measurements	0.35 (0.28–0.53)	72	4.7 (3.9–5.2)	56
Sex
Female	0.33 (0.27–0.52)	31 (29)	4.7 (3.9–6.0)	23 (22)
Male	0.36 (0.28–0.58)	49 (45)	4.7 (3.9–5.2)	35
OI severity
Mild	0.53 (0.40–0.73)	18	4.3 (3.8–5.3)	15
Moderate	0.33 (0.21–0.44)	19 (17)	4.7 (4.1–5.7)	13 (12)
Severe	0.32 (0.24–0.46)	43 (39)	4.7 (3.9–5.1)	30
Affected genes
COL1A1	0.39 (0.29–0.56)	34	4.7 (3.7–5.4)	26
COL1A2	0.34 (0.28–0.49)	20	4.7 (3.9–5.0)	16
CRTAP	0.32 and 0.24	2	4.1 and 6.4	2
LEPRE1	0.28 (0.19–0.30)	4	4.9 (4.1–5.2)	4
FKBP10	0.46 and 0.70	2	2.7	1
BMP1	1.10 (0.81–1.70)	3	4.9 and 4.8	2
TAPT1	0.33	1	5.1	1
SERPINF1	0.48 (0.34–0.78)	9 (5)	3.0 (2.8–3.9)	3
Excluding patient with multiple measurements	0.40 (0.20–0.55)	4	3.0 (2.8–3.9)	3
PLOD2	0.58	1	3.9	1
WNT1	0.17 (0.14–0.18)	4 (2)	6.8 and 6.00	2 (1)
Excluding patient with multiple measurements	0.18	1		0

Abbreviations: OPG, osteoprotegerin; IQR, interquartile range.

Serum sclerostin levels were measured in all 74 patients (=n_patients). Two patients had multiple samples collected at different time points—one with a SERPINF1 mutation (five samples) and the other with a WNT1 mutation (three samples)—resulting in a total of 80 measurements (=n_measurements). OPG levels were measured at the same time point as sclerostin in 58 instances (=n_measurements) across 57 patients (=n_patients). One patient with a WNT1 mutation had two paired measurements of OPG and sclerostin. OMIM numbers of the listed genes: COL1A1 (MIM 120,150), COL1A2 (MIM 120,160), CRTAP (MIM 605,497), LEPRE1 (MIM 610,339), FKBP10 (MIM 607,063), BMP1 (MIM 112,264), TAPT1 (MIM 612,758), SERPINF1 (MIM 172,860), PLOD2 (MIM 601,865), WNT1 (MIM 164,820).

In this cohort, the median serum sclerostin level in children with OI (0.35 ng/mL [0.28–0.52], n = 74) was not significantly different from that observed in unaffected children (0.42 ng/mL [0.33–0.54], n = 424). Likewise, the median age-adjusted sclerostin z-score in this OI cohort (−0.33 [−0.94–0.58], n = 74) was not significantly different from the reference median of unaffected children (P = 0.0760).²⁹

Stratification by affected gene revealed varying sclerostin levels depending on the underlying mutation (Figure 1, Table 2). For example, in this cohort children with WNT1 mutations (n = 2) exhibited low sclerostin levels (below the 25th percentile of the overall cohort distribution), whereas children with BMP1 mutations (n = 3) showed elevated levels (above the 90th percentile of the overall cohort distribution).³⁰

Figure 1.

Serum sclerostin concentrations (ng/ml) stratified by causative gene.

Within this cohort, increasing severity of OI was strongly associated with lower serum sclerostin levels (Spearman ρ = −0.4547, 95% CI [−0.62, −0.25], P < 0.0001, n = 74). Specifically, children with severe OI had significantly lower median sclerostin concentrations compared with those with mild OI (median [Hodges-Lehmann estimate] difference: −0.22 ng/mL, 95% CI [−0.35, −0.12], P < 0.0001, q = 0.0001). Similarly, moderately affected children showed significantly lower median sclerostin concentrations compared with those with mild OI (median [Hodges-Lehmann estimate] difference: −0.18 ng/mL, 95% CI [−0.33, −0.03], P = 0.005, q = 0.007), while no significant difference was observed between moderate and severe OI (Figure 2, Table 2).

Figure 2.

Serum sclerostin concentrations (ng/ml), stratified by OI severity. n = number of patients, q = FDR-adjusted P-values, ns = not significant.

There was no correlation between sclerostin levels and age (n = 74), sex (n = 74), or time since the last antiresorptive treatment (n = 70; no exact time available for one patient, and three patients had not received prior antiresorptive treatment). Given that age- and sex-related differences in sclerostin levels of children, as well as effects of antiresorptive treatment on sclerostin levels have been previously reported,^31–33 we conducted a multivariable linear regression analysis to adjust for these variables, as well as for OI severity and genotype group. After adjustment, OI severity (mild versus moderate) remained significantly associated with serum sclerostin levels (q-value = <0.0017). In addition, several genotype groups showed significant differences in adjusted analyses (Table 3), particularly BMP1 and WNT1 compared with other mutation groups.

Table 3.

Multivariable linear regression analysis of serum sclerostin levels.

Variable	Estimate	95% CI	Raw P-value	Adjusted P (=q-value)	Significance (based on q-value)
Age (years)	−0.009	−0.03 to 0.02	0.4590	0.5852	ns
Sex (male vs. female)	−0.105	−0.35 to 0.14	0.3934	0.5423	ns
Time since last antiresorptive treatment to measurement	0.041	−0.08 to 0.16	0.5047	0.6129	ns
Severity: Mild vs. mod	−0.377	−0.69 to −0.06	0.0204	0.0650	ns
Severity: Mild vs. severe	−0.784	−1.10 to −0.47	<0.0001	<0.0017	**
Severity: mod vs. severe	−0.407	−0.73 to −0.09	0.0135	0.0553	ns
BMP1 vs. Col1A1	−1.537	−2.11 to −0.96	<0.0001	<0.0017	**
BMP1 vs. COL1A2	−1.706	−2.30 to −1.11	<0.0001	<0.0017	**
BMP1 vs. CRTAP	−1.458	−2.27 to −0.65	0.0007	0.0051	**
BMP1 vs. LEPRE1	−1.649	−2.33 to −0.97	<0.0001	<0.0017	**
BMP1 vs. FKBP10	−0.736	−1.52 to 0.05	0.0647	0.1709	ns
BMP1 vs. TAPT1	−1.265	−2.27 to −0.27	0.0141	0.0553	ns
BMP1 vs. SERPINF1	−1.322	−1.99 to −0.66	0.0002	0.0017	**
BMP1 vs. PLOD2	−0.801	−1.81 to 0.21	0.1173	0.2063	ns
BMP1 vs. WNT1	−2.265	−3.08 to −1.45	<0.0001	<0.0017	**
WNT1 vs. COL1A1	0.728	0.09 to 1.36	0.0255	0.0765	ns
WNT1 vs. COL1A2	0.559	−0.09 to 1.21	0.0887	0.1967	ns
WNT1 vs. CRTAP	0.807	−0.09 to 1.71	0.0773	0.1877	ns
WNT1 vs. LEPRE1	0.616	−0.14 to 1.38	0.1097	0.2063	ns
WNT1 vs. FKBP10	1.529	0.65 to 2.41	0.0009	0.0057	**
WNT1 vs. TAPT1	1.000	−0.07 to 2.07	0.0670	0.1709	ns
WNT1 vs. SERPINF1	0.943	0.21 to 1.67	0.0125	0.0553	ns
WNT1 vs. PLOD2	1.464	0.40 to 2.53	0.0080	0.0408	*
COL1A1 vs. COL1A2	−0.170	−0.42 to 0.08	0.1861	0.3164	ns
COL1A1 vs. CRTAP	0.079	−0.60 to 0.76	0.8176	0.8528	ns
COL1A1 vs. LEPRE1	−0.112	−0.60 to 0.37	0.6455	0.7316	ns
COL1A1 vs. FKBP10	0.801	0.15 to 1.46	0.0177	0.0622	ns
COL1A1 vs. TAPT1	0.271	−0.64 to 1.18	0.5525	0.6553	ns
COL1A1 vs. SERPINF1	0.214	−0.25 to 0.68	0.3604	0.5106	ns
COL1A1 vs. PLOD2	0.736	−0.15 to 1.62	0.1018	0.2063	ns
COL1A2 vs. CRTAP	0.249	−0.46 to 0.95	0.4821	0.5997	ns
COL1A2 vs. LEPRE1	0.057	−0.44 to 0.56	0.8194	0.8528	ns
COL1A2 vs. FKBP10	0.971	0.30 to 1.64	0.0054	0.0306	*
COL1A2 vs. TAPT1	0.441	−0.48 to 1.36	0.3420	0.4983	ns
COL1A2 vs. SERPINF1	0.384	−0.09 to 0.86	0.1077	0.2063	ns
COL1A2 vs. PLOD2	0.905	0.01 to 1.80	0.0473	0.1340	ns
CRTAP vs. LEPRE1	−0.191	−0.96 to 0.57	0.6180	0.7163	ns
CRTAP vs. FKBP10	0.722	−0.16 to 1.60	0.1053	0.2063	ns
CRTAP vs. SERPINF1	0.135	−0.63 to 0.90	0.7245	0.7862	ns
CRTAP vs. TAPT1	0.192	−0.86 to 1.25	0.7161	0.7862	ns
CRTAP vs. PLOD2	0.657	−0.42 to 1.73	0.2267	0.3730	ns
LEPRE1 vs. FKBP10	0.913	0.16 to 1.67	0.0183	0.0622	ns
LEPRE1 vs. TAPT1	0.384	−0.59 to 1.36	0.4337	0.5821	ns
LEPRE1 vs. SERPINF1	0.327	−0.26 to 0.92	0.2710	0.4306	ns
LEPRE1 vs. PLOD2	0.848	−0.11 to 1.81	0.0819	0.1899	ns
FKBP10 vs. TAPT1	−0.530	−1.59 to 0.53	0.3201	0.4802	ns
FKBP10 vs. SERPINF1	−0.587	−1.32 to 0.15	0.1157	0.2063	ns
FKBP10 vs. PLOD2	−0.065	−1.12 to 0.99	0.9021	0.9056	ns
TAPT1 vs. SERPINF1	−0.057	−1.02 to 0.90	0.9056	0.9056	ns
TAPT1 vs. PLOD2	0.464	−0.77 to 1.70	0.4552	0.5852	ns
SERPINF1 vs. PLOD2	0.521	−0.43 to 1.48	0.2786	0.4306	ns

Abbreviations: versus – versus, 95% CI–95% confidence interval.

Multivariable linear regression analyses were performed with log-transformed sclerostin concentrations as the dependent variable and age, sex, disease severity, mutation group, and log-transformed time since last antiresorptive treatment as independent variables. To account for multiple testing, false discovery rate (FDR) correction using the Benjamini-Hochberg procedure was applied.

Significance codes refer to adjusted P-values (=q-values): *q < 0.05, **q < 0.01; ns = not significant.

Serum OPG levels

OPG levels were measured at the same time point as sclerostin in 58 instances across 57 patients (Figure 3(a), Table 2). One patient with a WNT1 mutation had two paired measurements of OPG and sclerostin.

Figure 3.

Serum OPG concentrations (pmol/l) (3A) and sclerostin/OPG ratio (3B) stratified by causative gene.

There was no correlation between serum sclerostin and OPG levels (Spearman ρ = −0.049, 95% CI [−0.31, 0.22], P = 0.7160, n = 57), although some mutation groups (FKBP10, PLOD2, WNT1) showed opposing trends with increased sclerostin-to-OPG ratios in FKBP10- and PLOD2-related OI and decreased ratios in WNT1-related OI (Figure 3(b)).

No correlation with PTH, 25(OH)D or AP

As observed for OPG, no significant correlations were observed between serum sclerostin levels and PTH, 25(OH)D or AP levels in this cohort.

Discussion

In this study, we evaluated serum sclerostin levels in children and adolescents with OI with regard to disease severity and genotype. To the best of our knowledge, this has not been previously investigated in a pediatric OI population.

The median serum sclerostin levels in this cohort were not significantly altered to those of unaffected children. Likewise, the median age-adjusted sclerostin z-score was not significantly different from the reference median of unaffected children.²⁹ Previous studies have reported conflicting results regarding serum sclerostin levels in patients with OI. Chen et al.²⁴ observed higher sclerostin concentrations in a mixed cohort of children and adults with OI, while studies in adult patients with OI (Nicol et al.²⁶ and Kocijan et al.²⁵) reported lower sclerostin concentrations. In contrast, our findings align with previous pediatric studies by Palomo et al.²⁷ and Brunetti et al.,²⁸ which also reported no significant differences in serum sclerostin levels between children with OI and healthy controls. Notably, all of these studies categorized patients based on clinical OI type (I–IV), without differentiation based on causative mutations other than those affecting COL1A1 or COL1A2.

Considering the genetic heterogeneity of OI and the involvement of multiple molecular pathways, it is conceivable that sclerostin regulation varies depending on the specific mutation. Our findings support this notion: while most genotype-specific differences did not reach statistical significance in a multivariable regression model, likely due to small subgroup sizes, a significant difference in serum sclerostin levels was observed mainly in patients with BMP1 and WNT1 mutations in comparison to other mutation groups after adjustment for age, sex, disease severity and time since last antiresorptive treatment. Notably, patients with BMP1 mutations showed elevated sclerostin levels, while those with WNT1 mutations exhibited markedly low levels: BMP1 mutations impair procollagen processing and lead to increased collagen mineralization, resulting in a high bone mass OI phenotype.³⁴ The higher bone mass may reduce overall mechanical strains, which are key negative regulators of sclerostin expression by osteocytes.^2,3 Consequently, reduced mechanical stimulation in BMP1-mutated patients may contribute to increased sclerostin production. In contrast, WNT1 mutations disrupt WNT/β-catenin signaling and suppress osteoblast-specific gene expression, resulting in impaired bone formation despite normal mineralization.³⁴ The markedly low sclerostin levels observed in these patients may reflect a compensatory mechanism to enhance osteoblast activity and promote bone formation. Our result is further supported by transcriptomic and proteomic data of patients with WNT1-related osteogenesis imperfecta which showed significantly downregulated expression of SOST compared to controls with COL1A1 mutation.³⁵ However, given the small genotype subgroup sizes (BMP1 n = 3 patients; WNT1 n = 2 patients), these findings, as well as their interpretation, should be considered exploratory. Additionally, as circulating sclerostin levels are also influenced by bone mass and osteocyte number,³⁶ the observed differences between BMP1 and WNT1 groups and other mutation groups may also be driven by individual variation in skeletal mass and osteocyte number rather than genotype-specific effects. Due to the retrospective nature of this study, there were no sufficient BMD data available to investigate this potential confounding factor.

We also found a significant inverse association between OI severity and serum sclerostin levels. Children with more severe forms of the disease had lower sclerostin concentrations compared to those with milder forms. This finding was stable after adjusting for age, sex, genotype and time since last antiresorptive treatment. While this may reflect a physiological adaptation aimed at enhancing bone formation in the context of severe skeletal fragility, as discussed above, it may also reflect differences in skeletal mass rather than a direct pathogenic mechanism. Two studies in adult patients with OI reported lower sclerostin levels in individuals with clinical OI type I compared to those with types III/IV, although the differences were not statistically significant.^25,26 To the best of our knowledge, no other studies have specifically investigated the relationship between sclerostin levels and OI severity.

Regarding other bone markers, serum sclerostin levels did not correlate with OPG concentrations in this cohort. OPG, a glycoprotein secreted by osteoblasts and osteocytes, acts as a decoy receptor for RANKL, thereby inhibiting osteoclast differentiation and activity, and ultimately reducing bone resorption.³⁷ OPG is a WNT/β-catenin target gene and it has been shown that sclerostin, as an inhibitor of the WNT/β-catenin-pathway, suppresses OPG expression.^38,39 In contrast to these expected mechanisms, a previous study in children with OI reported a positive correlation between serum sclerostin and OPG levels.²⁸ Interestingly, although no overall correlation between serum sclerostin and OPG levels was found in this cohort, mutation-specific trends emerged in our data. In selected subgroups, OPG levels showed opposing patterns to sclerostin concentrations, potentially reflecting distinct mutation-specific mechanisms of bone remodeling: The elevated OPG levels observed in patients with WNT1 mutation may seem surprising, considering that OPG is a target gene of WNT/β-catenin signaling. However, WNT1 is only one of several WNT ligands,⁴⁰ and there are also other factors such as estrogens that promote OPG gene expression independently of the WNT/β-catenin pathway.^41–43 In the context of the concurrently observed low sclerostin levels—and thus reduced inhibition of the WNT/β-catenin pathway—along with the possibility that OPG expression can occur independently of WNT1, the elevated OPG levels likely represent a counter-regulatory mechanism to reduce osteoclast activity in WNT1-related OI. In contrast, patients with FKBP10- and PLOD2-related OI showed increased sclerostin-to-OPG ratios. Both mutations are known to impair collagen crosslinking,³⁴ leading to a structurally unstable extracellular matrix. Such matrix alterations may affect mechanotransduction in bone cells resulting in decreased OPG expression.⁴⁴ These mutation-specific changes may thus represent a compensatory response to increase bone stability.

Finally, we found no significant correlations between sclerostin levels and other bone metabolism markers, including PTH, 25(OH)D and AP. These findings appear to contrast with the known inverse regulation of SOST expression by PTH signaling,⁴⁵ as well as in studies in adults reporting negative correlations between serum sclerostin and circulating PTH levels.^11,46–50 However, two studies in children—one in unaffected individuals,³¹ and another in children with OI following bisphosphonate treatment²⁷—similarly reported no correlation between sclerostin and PTH, supporting our findings in this cohort.

We also did not observe a correlation between serum sclerostin and 25(OH)D or AP. Reported associations in previous studies are inconsistent: some describe positive correlations in healthy children and in those with OI receiving bisphosphonate therapy, while others report no such correlation.^27,31,51

In adults, a negative correlation between AP and sclerostin has been observed in elderly women but not in men.⁵² These discrepancies may reflect age-related and/or sex-related differences as well as the influence of small sample sizes.

This study has several limitations. Due to the retrospective nature of the analysis, it was not possible to collect additional relevant clinical data, such as fracture rates or BMD—the latter known to be positively correlated with sclerostin levels.⁵¹ Given rarity of OI, subgroup sizes—particularly when stratified by genotype—were small and the observed genotype-specific differences may therefore be individual-specific and related to clinical factors rather than disease biology. In particular, the BMP1 and WNT1 findings are based on very limited numbers of samples and should be interpreted with caution. Accordingly, these results should be considered exploratory. In nearly all cases, only a single measurement of serum sclerostin was available, limiting the assessment of intra-individual variability. This is particularly relevant given that sclerostin levels are regulated by multiple factors.³⁶ Another limitation is that, in the assay, only the standards and controls were performed in duplicates, but not the samples. Therefore, no intra- and inter-assay coefficients of variation could be calculated. However, all samples in the sclerostin assay and OPG assay respectively were run in one batch. Nonetheless, the study has notable strengths. Most importantly, it addresses a highly relevant clinical question in light of two ongoing phase 3 clinical trials investigating anti-sclerostin antibodies in children with OI (NCT05972551, NCT05768854). As these therapies may soon become part of the clinical care, baseline data on sclerostin variability across OI genotypes and severity levels are important. Our findings may contribute to identifying patients—based on their genotype—who are most likely to benefit from targeted anti-sclerostin treatment and may help guide future research into potential predictive markers of therapeutic response. This is of particular interest, as two recent studies in OI (including early results from NCT05768854) have shown that increases in bone mass do not necessarily correlate with a reduction in fracture rates, highlighting the need for genotype-based patient selection and individualized therapeutic strategies.^23,53

Supplemental material

Supplemental material - Serum sclerostin levels in children with osteogenesis imperfecta

Supplemental material for Serum sclerostin levels in children with osteogenesis imperfecta by Susanna Reincke, Mirko Rehberg, Stefanie Stasek, Shino Junghänel-Welzing, Vera Fuessgen, Oliver Semler, and Heike Hoyer-Kuhn in Annals of Clinical Biochemistry

Footnotes

Acknowledgements

We sincerely thank all the patients and their families for their participation in the measurements, making this analysis possible. We also thank the DFG research unit FOR2722 (SE2373/1-2, Project: 384170921), which supported several of our authors (OS, MR, SS, HHK).

ORCID iDs

Susanna Reincke

Vera Fuessgen

Oliver Semler

Consent to participate

Written informed consent was obtained from all patients and their legal guardians. No animal experiments were conducted in this study.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the “Ein Herz für Kinder” Foundation (MED-0054132).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

Ethical approval was obtained from the local ethics committee of the University (approval number: 13-202).

Contributionship

SR wrote the first draft of the paper. OS and HHK designed the study. All data were collected during clinical visits with the support of MR, SJW and HHK. Data analysis was done by SR and revised by OS, SS and HHK. Revision and final approval of the manuscript were done by all authors. All authors agree to be accountable for the work and to ensure that any questions relating to the accuracy and integrity of the paper are investigated and properly resolved.

Guarantor

PD Dr. med. Heike Hoyer-Kuhn.

Supplemental material

Supplemental material for this article is available online.

References

Zhang

Kang

, et al. Sclerostin binds to LRP5/6 and antagonizes canonical wnt signaling. J Biol Chem 2005; 280: 19883–19887. https://doi.org/10.1074/jbc.M413274200

Robling

Niziolek

Baldridge

, et al. Mechanical stimulation of bone in vivo reduces osteocyte expression of sost/Sclerostin. J Biol Chem 2008; 283: 5866–5875. https://doi.org/10.1074/jbc.M705092200

Rhee

Condon

, et al. Sost downregulation and local wnt signaling are required for the osteogenic response to mechanical loading. Bone 2011; 50: 209–217. https://doi.org/10.1016/J.BONE.2011.10.025

Baron

Kneissel

. WNT signaling in bone homeostasis and disease: from human mutations to treatments. Nat Med 2013; 19: 179–192. https://doi.org/10.1038/nm.3074

Marulanda

Tauer

Boraschi-Diaz

, et al. Effect of sclerostin inactivation in a mouse model of severe dominant osteogenesis imperfecta. Sci Rep 2023; 13: 5010. https://doi.org/10.1038/S41598-023-32221-3

Kramer

Loots

Studer

, et al. Parathyroid hormone (PTH)-induced bone gain is blunted in SOST overexpressing and deficient mice. J Bone Miner Res 2010; 25: 178–189. https://doi.org/10.1359/jbmr.090730

Balemans

Ebeling

Patel

, et al. Increased bone density in sclerosteosis is due to the deficiency of a novel secreted protein (SOST). Hum Mol Genet 2001; 10: 537–543. https://doi.org/10.1093/hmg/10.5.537

Brunkow

Gardner

Van Ness

, et al. Bone dysplasia sclerosteosis results from loss of the SOST gene product, a novel cystine knot-containing protein. Am J Hum Genet 2001; 68: 577–589. https://doi.org/10.1086/318811

Aryana

IGPS

Rini

Soejono

. Importance of sclerostin as bone-muscle mediator crosstalk. Ann Geriatr Med Res 2022; 26: 72–82. https://doi.org/10.4235/AGMR.22.0036

10.

Han

You

Xing

, et al. Paracrine and endocrine actions of bone—The functions of secretory proteins from osteoblasts, osteocytes, and osteoclasts. Bone Res 2018; 6: 16. https://doi.org/10.1038/S41413-018-0019-6

11.

Drake

Srinivasan

Mödder

, et al. Effects of parathyroid hormone treatment on circulating sclerostin levels in postmenopausal women. J Clin Endocrinol Metab 2010; 95: 5056–5062. https://doi.org/10.1210/JC.2010-0720

12.

Garnero

Sornay-Rendu

Munoz

, et al. Association of serum sclerostin with bone mineral density, bone turnover, steroid and parathyroid hormones, and fracture risk in postmenopausal women: the OFELY study. Osteoporos Int 2013; 24: 489–494. https://doi.org/10.1007/s00198-012-1978-x

13.

Hampson

Edwards

Conroy

, et al. The relationship between inhibitors of the wnt signalling pathway (Dickkopf-1(DKK1) and sclerostin), bone mineral density, vascular calcification and arterial stiffness in post-menopausal women. Bone 2013; 56: 42–47. https://doi.org/10.1016/J.BONE.2013.05.010

14.

Arasu

Cawthon

Lui

, et al. Serum sclerostin and risk of hip fracture in older caucasian women. J Clin Endocrinol Metab 2012; 97: 2027–2032. https://doi.org/10.1210/JC.2011-3419

15.

Marini

Forlino

Bächinger

, et al. 2017 Osteogenesis imperfecta. Nat Rev Dis Primers 2017; 3(1 3): 1–19. https://doi.org/10.1038/nrdp.2017.52

16.

Stasek

Zaucke

Hoyer-Kuhn

, et al. Osteogenesis imperfecta: shifting paradigms in pathophysiology and care in children. J Pediatr Endocrinol Metab 2024; 38: 1–15. https://doi.org/10.1515/JPEM-2024-0512

17.

Cosman

Crittenden

Adachi

, et al. Romosozumab treatment in postmenopausal women with osteoporosis. N Engl J Med 2016; 375: 1532–1543. https://doi.org/10.1056/NEJMoa1607948

18.

Sinder

Salemi

Ominsky

, et al. Rapidly growing Brtl/+ mouse model of osteogenesis imperfecta improves bone mass and strength with sclerostin antibody treatment. Bone 2015; 71: 115–123. https://doi.org/10.1016/J.BONE.2014.10.012

19.

Grafe

Alexander

Yang

, et al. Sclerostin antibody treatment improves the bone phenotype of Crtap(-/-) mice, a model of recessive osteogenesis imperfecta. J Bone Miner Res 2016; 31: 1030–1040. https://doi.org/10.1002/JBMR.2776

20.

Jacobsen

Barber

Ayturk

, et al. Targeting the LRP5 pathway improves bone properties in a mouse model of osteogenesis imperfecta. J Bone Miner Res 2014; 29: 2297–2306. https://doi.org/10.1002/jbmr.2198

21.

Cardinal

Dessain

Roels

, et al. Sclerostin-antibody treatment decreases fracture rates in axial skeleton and improves the skeletal phenotype in growing oim/oim mice. Calcif Tissue Int 2020; 106: 494–508. https://doi.org/10.1007/S00223-019-00655-5

22.

Glorieux

Langdahl

Chapurlat

, et al. Setrusumab for the treatment of osteogenesis imperfecta: 12-month results from the phase 2b asteroid study. J Bone Miner Res 2024; 39: 1215–1228. https://doi.org/10.1093/JBMR/ZJAE112

23.

2025) Ultragenyx announces phase 3 orbit and cosmic results for setrusumab (UX143) in osteogenesis imperfecta. https://ir.ultragenyx.com/news-releases/news-release-details/ultragenyx-announces-phase-3-orbit-and-cosmic-results-setrusumab. Accessed 26 Jan 2026.

24.

Chen

Gao

Shen

, et al. Serum sclerostin and its association with bone turnover marker in metabolic bone diseases. Dis Markers 2022; 2022: 7902046–7902048. https://doi.org/10.1155/2022/7902046

25.

Kocijan

Muschitz

Fahrleitner-Pammer

, et al. Serum sclerostin levels are decreased in adult patients with different types of osteogenesis imperfecta. J Clin Endocrinol Metab 2014; 99: E311–E319. https://doi.org/10.1210/jc.2013-2244

26.

Nicol

Wang

Smith

, et al. Serum sclerostin levels in adults with osteogenesis imperfecta: comparison with normal individuals and response to teriparatide therapy. J Bone Miner Res 2018; 33: 307–315. https://doi.org/10.1002/JBMR.3312

27.

Palomo

Glorieux

Rauch

. Circulating sclerostin in children and young adults with heritable bone disorders. J Clin Endocrinol Metab 2014; 99: E920–E925. https://doi.org/10.1210/jc.2013-3852

28.

Brunetti

Papadia

Tummolo

, et al. Impaired bone remodeling in children with osteogenesis imperfecta treated and untreated with bisphosphonates: the role of DKK1, RANKL, and TNF-α. Osteoporos Int 2016; 27: 2355–2365. https://doi.org/10.1007/s00198-016-3501-2

29.

Fischer

Mischek

Wolf

, et al. Paediatric reference values for the C-terminal fragment of fibroblast-growth factor-23, sclerostin, bone-specific alkaline phosphatase and isoform 5b of tartrate-resistant acid phosphatase. Ann Clin Biochem 2012; 49: 546–553. https://doi.org/10.1258/acb.2012.011274

30.

Tietze

Pott

Kanzelmeyer

, et al. LMS-based continuous pediatric reference values for soluble receptor activator of nuclear factor kappa B ligand (sRANKL) and osteoprotegerin (OPG) in the HARP cohort. Osteoporos Int 2023; 35: 533–542. https://doi.org/10.1007/S00198-023-06959-5

31.

Kirmani

Amin

McCready

, et al. Sclerostin levels during growth in children. Osteoporos Int 2012; 23: 1123–1130. https://doi.org/10.1007/s00198-011-1669-z

32.

Gatti

Viapiana

Adami

, et al. Bisphosphonate treatment of postmenopausal osteoporosis is associated with a dose dependent increase in serum sclerostin. Bone 2012; 50: 739–742. https://doi.org/10.1016/j.bone.2011.11.028

33.

Catalano

Morabito

Basile

, et al. Zoledronic acid acutely increases sclerostin serum levels in women with postmenopausal osteoporosis. J Clin Endocrinol Metab 2013; 98: 1911–1915. https://doi.org/10.1210/jc.2012-4039

34.

Etich

Leßmeier

Rehberg

, et al. Osteogenesis imperfecta—Pathophysiology and therapeutic options. Molecul Cellul Pediatr 2020; 7(17): 1–9. https://doi.org/10.1186/S40348-020-00101-9

35.

Tan

Chen

Zhang

, et al. Multi-omics analyses reveal aberrant differentiation trajectory with WNT1 loss-of-function in type XV osteogenesis imperfecta. J Bone Miner Res 2024; 39: 1253–1267. https://doi.org/10.1093/JBMR/ZJAE123

36.

Drake

Khosla

. Hormonal and systemic regulation of sclerostin. Bone 2016; 96: 8–17. https://doi.org/10.1016/J.BONE.2016.12.004

37.

Boyce

Xing

. Functions of RANKL/RANK/OPG in bone modeling and remodeling. Arch Biochem Biophys 2008; 473: 139–146. https://doi.org/10.1016/j.abb.2008.03.018

38.

Delgado-Calle

Condon

, et al. Osteocytes mediate the anabolic actions of canonical Wnt/β-catenin signaling in bone. Proc Natl Acad Sci USA 2015; 112: E478–E486. https://doi.org/10.1073/pnas.1409857112

39.

Wijenayaka

Kogawa

Lim

, et al. Sclerostin stimulates osteocyte support of osteoclast activity by a RANKL-dependent pathway. PLoS One 2011; 6: e25900. https://doi.org/10.1371/JOURNAL.PONE.0025900

40.

Qin

Fan

, et al. Canonical and noncanonical wnt signaling: multilayered mediators, signaling mechanisms and major signaling crosstalk. Genes Dis 2024; 11: 103–134. https://doi.org/10.1016/J.GENDIS.2023.01.030

41.

Bord

Ireland

Beavan

, et al. The effects of estrogen on osteoprotegerin, RANKL, and estrogen receptor expression in human osteoblasts. Bone 2003; 32: 136–141. https://doi.org/10.1016/S8756-3282(02)00953-5

42.

Jia

Zhou

Zeng

, et al. Estrogen stimulates osteoprotegerin expression via the suppression of miR-145 expression in MG-63 cells. Mol Med Rep 2017; 15: 1539–1546. https://doi.org/10.3892/MMR.2017.6168

43.

Shi

Morgan

. Estrogen and estrogen receptors mediate the mechanobiology of bone disease and repair. Bone 2024; 188: 117220. https://doi.org/10.1016/J.BONE.2024.117220

44.

Sanchez

Gabay

Salvat

, et al. Mechanical loading highly increases IL-6 production and decreases OPG expression by osteoblasts. Osteoarthr Cartil 2009; 17: 473–481. https://doi.org/10.1016/J.JOCA.2008.09.007

45.

Bellido

Ali

Gubrij

, et al. Chronic elevation of parathyroid hormone in mice reduces expression of sclerostin by osteocytes: a novel mechanism for hormonal control of osteoblastogenesis. Endocrinology 2005; 146(11): 4577. https://doi.org/10.1210/en.2005-0239

46.

Van Lierop

Witteveen

Hamdy

NAT

, et al. Patients with primary hyperparathyroidism have lower circulating sclerostin levels than euparathyroid controls. Eur J Endocrinol 2010; 163: 833–837. https://doi.org/10.1530/EJE-10-0699

47.

Costa

Cremers

Rubin

, et al. Circulating sclerostin in disorders of parathyroid gland function. J Clin Endocrinol Metab 2011; 96: 3804–3810. https://doi.org/10.1210/jc.2011-0566

48.

Kaji

Imanishi

Sugimoto

, et al. Comparisons of serum sclerostin levels among patients with postmenopausal osteoporosis, primary hyperparathyroidism and osteomalacia. Exp Clin Endocrinol Diabet 2011; 119: 440–444. https://doi.org/10.1055/s-0031-1275661

49.

Piemonte

Romagnoli

Bratengeier

, et al. Serum sclerostin levels decline in post-menopausal women with osteoporosis following treatment with intermittent parathyroid hormone. J Endocrinol Investig 2012; 35: 866–868. https://doi.org/10.3275/8522

50.

Kumbhani

Siwila-Sackman

, et al. Acute decline in serum sclerostin in response to PTH infusion in healthy men. J Clin Endocrinol Metab 2011; 96: E1848–E1851. https://doi.org/10.1210/JC.2011-1534

51.

Kim

Choi

Ahn

, et al. Associations between sclerostin and anthropometric and metabolic parameters in children and adolescents. Children 2021; 8: 788. https://doi.org/10.3390/children8090788

52.

Mödder

Hoey

Amin

, et al. Relation of age, gender, and bone mass to circulating sclerostin levels in women and men. J Bone Miner Res 2010; 26: 373–379. https://doi.org/10.1002/JBMR.217

53.

Hald

Weir

Keerie

, et al. Baseline characteristics of the TOPaZ study: randomised trial of teriparatide and zoledronic acid compared with standard care in adults with osteogenesis imperfecta. Calcif Tissue Int 2025; 116: 136. https://doi.org/10.1007/s00223-025-01440-3