Our data show that previously unrecognized monogenic bone disorders play an important role in PAO. Pregnancy should be considered a skeletal risk factor, which can promote the initial clinical onset of such skeletal disorders. The underlying increased calcium demand is essential in terms of prophylactic and therapeutic measures, which are especially required in individuals with a genetically determined low bone mass. The implementation of this knowledge in clinical practice can enable the partial recovery of the skeleton. Consistent genetic studies are needed to analyze the frequency of pathogenic variants in women with PAO.

Reduced bone mineral density (BMD; ie, Z-score ≤−2.0) occurring at a young age (ie, premenopausal women and men <50 years) in the absence of secondary osteoporosis is considered early-onset osteoporosis (EOOP). Mutations affecting the WNT signaling pathway are of special interest because of their key role in bone mass regulation. Here, we analyzed the effects of relevant LRP5 and LRP6 variants on the clinical phenotype, bone turnover, BMD, and bone microarchitecture. After exclusion of secondary osteoporosis, EOOP patients (n = 372) were genotyped by gene panel sequencing, and segregation analysis of variants in LRP5/LRP6 was performed. The clinical assessment included the evaluation of bone turnover parameters, BMD by dual-energy X-ray absorptiometry, and microarchitecture via high-resolution peripheral quantitative computed tomography (HR-pQCT). In 50 individuals (31 EOOP index patients, 19 family members), relevant variants affecting LRP5 or LRP6 were detected (42 LRP5 and 8 LRP6 variants), including 10 novel variants. Seventeen variants were classified as disease causing, 14 were variants of unknown significance, and 19 were BMDassociated single-nucleotide polymorphisms (SNPs). One patient harbored compound heterozygous LRP5 mutations causing osteoporosis-pseudoglioma syndrome. Fractures were reported in 37 of 50 individuals, consisting of vertebral (18 of 50) and peripheral (29 of 50) fractures. Low bone formation was revealed in all individuals. A Z-score ≤−2.0 was detected in 31 of 50 individuals, and values at the spine were significantly lower than those at the hip (−2.1 AE 1.3 versus −1.6 AE 0.8; p = .003). HR-pQCT analysis (n = 34) showed impaired microarchitecture in trabecular and cortical compartments. Significant differences regarding the clinical phenotype were detectable between index patients and family members but not between different variant classes. Relevant variants in LRP5 and LRP6 contribute to EOOP in a substantial number of individuals, leading to a high number of fractures, low bone formation, reduced Zscores, and impaired microarchitecture. This detailed skeletal characterization improves the interpretation of known and novel LRP5 and LRP6 variants.

No abstract