Abstract
Background
Pregnancy and lactation-associated osteoporosis (PLO) is a rare disorder causing rapid bone loss and severe fragility fractures. While familial clusterings suggest genetic susceptibility, the specific clinical features, genetic landscape, and skeletal recovery trajectory of PLO remain poorly characterized in non-Caucasian populations. This study aimed to evaluate a Taiwanese PLO cohort to optimize diagnostic awareness.
Methods
Eleven postpartum women with PLO were evaluated using bone densitometry and spinal imaging. Whole-exome sequencing was performed utilizing a targeted 398-gene skeletal disease virtual panel. Patients were followed longitudinally for a median of 25 months during lactation cessation and/or targeted bone therapy.
Results
Seven women developed symptoms after their first pregnancy and four after their second. Vertebral fractures occurred in 91% (10/11) of patients, 80% (8/10) of whom had multi-level involvement predominantly localized to the thoracolumbar spine (T3-L5, peaking at T11-L2). Baseline lumbar bone mineral density (BMD) was significantly lower than hip sites (P = 0.018). Genetic variants were identified in 10 of 11 patients (91%), with six individuals harboring pathogenic variants in WNT1, LRP5, COL10A1, LTBP3, and CALCR. Patients with pathogenic variants exhibited significantly lower lumbar spine BMD than those without (Z-scores: -3.7 vs. -3.0; P = 0.043). Post-treatment, the cohort’s median Z-score improved from -3.5 to -2.3 (P = 0.005), though none achieved a score above -1.0.
Conclusions
Our findings suggest a significant genetic burden associated with PLO, potentially revealed by pregnancy and lactation stressors. The predominance of vertebral bone loss highlights acute postpartum back pain as a notable clinical sign warranting further investigation. Optimizing long-term maternal outcomes may require rapid diagnostic evaluation, prompt lactation cessation, and the future integration of genotype-informed management strategies.
Keywords: bone mineral density (BMD), genetic predisposition, postpartum back pain, pregnancy and lactation-associated osteoporosis (PLO), vertebral fragility fractures, whole-exome sequencing (WES)
Introduction
Pregnancy and lactation-associated osteoporosis (PLO) is a rare skeletal disorder characterized by early-onset osteoporosis, often manifesting as vertebral compression fractures or transient bone loss of the hip during late pregnancy or breastfeeding (1–4). Epidemiological data on the incidence of PLO are limited, but estimates vary, including 6.8 cases per 100,000 pregnancies in the UK (Edinburgh) (5) and 460 cases per million deliveries in Japan (6). The exact causes of PLO remain unclear, and the condition is often associated with significant morbidity. It is believed that PLO is related to hormonal changes during pregnancy and lactation. Nevertheless, PLO is extremely rare in normal pregnancies, implying that additional factors may contribute to its development. Peris et al. (7) reported that 53% of first-degree relatives of PLO patients had osteoporosis, compared to 15% in matched controls, with multiple affected members within the same families, suggesting an underlying genetic predisposition to low peak bone mass. A German study identified risk variants in bone-related genes such as LRP5, WNT1 and COL1A1/COL1A2 in nearly 50% of PLO cases (8), and a recent exome-sequencing study of 110 North American women with PLO reported rare predicted deleterious variants in 30%, most frequently in WNT-pathway genes (LRP5, LRP6 and WNT1) (9). The broader clinical phenotype, genetic architecture, and longitudinal skeletal recovery remain poorly characterized in Asian populations. This data gap often results in diagnostic delays and missed interventions for women presenting with debilitating postpartum back pain. This study aimed to characterize the clinical features and skeletal recovery of a Taiwanese PLO cohort, utilizing whole-exome sequencing (WES) to identify underlying genetic susceptibilities and improve diagnostic awareness for this severe maternal condition.
Methods
Participants and ethical approval
PLO is clinically characterized by vertebral compression fractures or significant hip bone loss during pregnancy or within 6 months postpartum (1-3). We retrospectively enrolled all women diagnosed with PLO between 2017 and 2026 at our institution’s Endocrinology Department. Prior to diagnosis, secondary causes of osteoporosis were excluded, including celiac disease, idiopathic hypercalciuria, chronic kidney disease (estimated glomerular filtration rate less than 30 ml/min), primary hyperparathyroidism, and chronic corticosteroid use. This study was conducted in accordance with the Declaration of Helsinki and received approval from our hospital’s Research Ethics Committee (reference number: 201508107RIND). All participants provided written informed consent for the evaluation of osteoporosis and its underlying genetic basis after a comprehensive explanation of the study objectives and procedures.
Radiological assessment
Bone mineral density (BMD) of the hip and lumbar spine was assessed using dual-energy X-ray absorptiometry (DXA). In accordance with international guidelines (10) for premenopausal women, Z-scores were utilized instead of T-scores to ensure an appropriate age-matched reference for evaluating bone mass in younger populations. X-ray and magnetic resonance imaging (MRI) were employed to identify the locations of fractures, with a primary focus on the thoracic and lumbar spine, as well as areas of reported bone pain. Vertebral fractures on X-ray were evaluated using the semi-quantitative method proposed by Genant et al. (11). DXA scans and X-rays were obtained at baseline and during follow-up, with MRI reserved for patients presenting with multiple fractures or when the underlying condition could not be determined.
Whole-exome sequencing and bioinformatics analysis
Genomic DNA from all patients was subjected to WES. Exonic regions were enriched using the Agilent SureSelect Human All Exon V6 Kit following the manufacturer’s protocol. Sequencing was carried out on the Illumina NovaSeq 6000 platform to generate 150 bp paired-end reads. Raw sequencing data were processed using the Illumina DRAGEN Bio-IT Platform (v4.2; Illumina Inc., San Diego, CA, USA). Reads were aligned to the human reference genome (GRCh38), and variant calling was performed according to the DRAGEN (12) pipeline best practices. Annotation of genetic variants was carried out using the Ensembl Variant Effect Predictor (VEP v.111) (13).
A custom-designed gene panel consisting of 398 genes associated with bone mass regulation, skeletal dysplasias, dysostoses, and connective tissue disorders, as listed in Supplementary Table 1, was applied for focused analysis. The complete filtering and interpretation workflow is summarized in Supplementary Figure 1. Variants were retained for prioritization only if they carried a “PASS” filter status, a minor allele frequency (MAF) of 0.01 or less in the Genome Aggregation Database (gnomAD), a read depth (DP) of at least 5, and a variant allele fraction (VAF) of at least 20%. Retained variants were then triaged into three classes: (i) loss-of-function variants, comprising frameshift, stop-gain, and canonical splice-site variants; (ii) nonsynonymous variants, comprising missense, in-frame indel, stop-loss, and initiation-codon variants, predicted to be protein-damaging by Combined Annotation Dependent Depletion (CADD) (14) at a phred score of 25 or above, or by Deleterious Annotation of genetic variants using Neural Networks (DANN) at a score of 0.93 or above where a CADD score was unavailable; and (iii) splicing-region variants with a predicted splicing impact (SpliceAI (15) delta score of 0.5 or above). The resulting candidate set was restricted to the 398-gene virtual panel and was screened in parallel for variants or genes recurring across samples. Panel-restricted candidates were classified according to the five-tier system recommended by the American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) guidelines (16) by manual review, supported by an in-house developed workflow, the Gendiseak (GDK) platform (17, 18), and cross-checked against ClinVar (January 2025 release) (19), VarSome (June 2025) (20), and GeneBe (June 2025) (21). Allele frequencies in the gnomAD East Asian subset and in the Taiwan Biobank (TWB) were additionally retrieved and are reported for all disease-causing variants.
In parallel with the prioritization pathway described above, and independently of the allele frequency filter, variants already reported as pathogenic or likely pathogenic in ClinVar, confirmed in MITOMAP, or pathogenic/likely pathogenic in a gene on the ACMG secondary findings list were also reviewed (Supplementary Figure 1, pathway 1). This step was performed as a supplementary check of the annotation pipeline and did not form part of the primary analysis; no variant relevant to PLO and no reportable secondary finding was identified through it, and its output is therefore not included in the variant counts reported in the Results.
For downstream analyses, variants were categorized as follows. Disease-causing variants (DCVs) were defined as heterozygous variants classified as pathogenic or likely pathogenic (P/LP) by manual review, VarSome, or GeneBe, in genes with autosomal dominant (AD) or combined autosomal dominant/autosomal recessive (AD/AR) inheritance. Relevant variants (RVs) were a collection of genetic variants potentially associated with bone fragility and skeletal disorders, including DCVs, heterozygous P/LP variants in AR inheritance pattern, as well as variants of uncertain significance (VUS) regardless of inheritance mode. Gene-disease inheritance information was obtained from the Gene Curation Coalition (GenCC) database.
Statistical methods
Continuous variables were presented as median (Q1, Q3), whereas categorical variables were expressed as frequencies (percentages). The rates of all outcomes were calculated with 95% confidence intervals (CIs). BMD Z-scores were compared across anatomical regions, as well as at two time points (diagnosis and 2–3 years post-treatment), using the Wilcoxon signed-rank test. To compare subgroup differences based on genetic variants, the Wilcoxon rank-sum test for continuous data and the Fisher’s exact test for categorical data were used. A P-value less than or equal to 0.05 was considered statistically significant. All statistical analyses were conducted using Stata (version 12.1; StataCorp, College Station, TX, USA).
Results
Clinical and radiographic features of PLO
The study cohort comprised 11 women with clinically diagnosed PLO, with a median age of 35 years (34, 37) (Table 1). These patients had the following demographic characteristics: median height of 160 cm (155, 165), weight of 51 kg (43, 63), and body mass index of 21.0 kg/m² (17.9, 24.6). All patients developed symptoms during lactation, with onset after the first delivery in 7 cases and after the second delivery in 4 cases. Symptom occurred at a median of 3 months postpartum (1, 4). Seven women presented within the first 3 months, and four between 4 and 6 months. Six of the patients experienced height loss, with a median height loss of 4.5 cm among these patients (3, 9), while the median height reduction for all 11 patients was 3.0 cm (0, 5). Back pain was the predominant presentation in all patients (11 cases).
Table 1.
Clinical and radiographic characteristics of PLO patients with and without disease-causing variants (DCVs).
| Characteristics | Total (n = 11) |
Without DCVs (n = 5) |
With DCVs (n = 6) |
P-value* |
|---|---|---|---|---|
| Age at diagnosis, years | 35 (34, 37) | 37 (35, 38) | 34 (30, 36) | 0.054 |
| Height, cm | 160 (155, 165) | 159 (155, 163) | 161 (156, 165) | 0.715 |
| Weight, kg | 51 (43, 63) | 49 (46, 65) | 53 (42, 58) | 0.584 |
| Body mass index, kg/m2 | 21.0 (17.9, 24.6) | 22.5 (18.4, 25.7) | 19.7 (17.1, 23.8) | 0.361 |
| Height decrease, cm | 3 (0, 5) | 0 (0, 4) | 3 (0, 5) | 0.630 |
| Parity | ||||
| 1 | 7 (63.6%) | 2 (40%) | 5 (83%) | 0.242 |
| 2 | 4 (36.4%) | 3 (60%) | 1 (17%) | 0.242 |
| Disease onset during lactation | ||||
| In the first 3 months after childbirth | 7 (63.6%) | 3 (60%) | 4 (67%) | 1.000 |
| In the 4–6 month postpartum period | 4 (36.4%) | 2 (40%) | 2 (33%) | 1.000 |
| Interval between delivery and disease onset, months | 3 (1, 4) | 3 (1, 4) | 2.5 (1, 4) | 0.925 |
| Breastfeeding, months | 6 (4, 8) | 6 (4, 8) | 5.5 (4, 6) | 0.782 |
| Lumbar spine BMD, Z-score | -3.5 (-3.8, -3.0) | -3.0 (-3.3, -2.5) | -3.7 (-4.1, -3.5) | 0.043 |
| Femoral neck BMD, Z-score | -1.3 (-2.0, -1.1) | -1.1 (-2.0, -0.4) | -1.65 (-2.25, -1.2) | 0.280 |
| Total hip BMD, Z-score | -1.6 (-2.4, -0.7) | -0.7 (-3.1, 0.1) | -1.95 (-2.35, -1.2) | 0.480 |
| Vertebral fractures | 10 (91%) | 4 (80%) | 6 (100%) | 0.455 |
| Vertebral fractures, multiple (at least 2 vertebrae) | 8 (73%) | 2 (40%) | 6 (100%) | 0.061 |
| Vertebral fracture number per case | 4 (1, 8) | 1 (1, 7) | 4 (3, 8) | 0.312 |
Continuous variables are presented as median (Q1, Q3), and categorical variables are expressed in frequency (percentage).
*P-values were calculated to compare characteristics between patients with and without DCVs, using the Wilcoxon rank-sum test for continuous variables and Fisher’s exact test for categorical variables.
Bold values indicate P < 0.05.
At diagnosis, the median BMD Z-scores were significantly reduced: -3.5 (-3.8, -3.0) at the lumbar spine, -1.3 (-2.0, -1.1) at the femoral neck, and -1.6 (-2.4, -0.7) at the total hip. Furthermore, the lumbar spine demonstrated a significantly lower BMD compared to the femoral neck (-3.5 vs. -1.3, P = 0.018) and the total hip (-3.5 vs. -1.6, P = 0.018) region of interest, with statistically significant differences confirmed by Wilcoxon signed-rank tests.
Among the 11 PLO women, nine experienced only vertebral compression fractures. One patient had a combination of vertebral and left pubic bone fractures. The remaining patient had fractures at the left sacral ala and bilateral pubic bones. Among the ten patients with vertebral fractures, a total of 58 fractured vertebrae were identified, yielding a median of 4 fractures per patient (1, 8). These spinal fractures occurred between T3 and L5, particularly at T11–L2 (Figure 1). Eight patients sustained multiple spinal fractures (≥2 sites), and the most severe case involved 14 levels. Review of pre-pregnancy medical histories revealed no prior fractures in any of the patients.
Figure 1.
Distribution of Fractured Vertebral Levels in PLO. Ten out of 11 PLO women had vertebral compression fractures, affecting a total of 58 vertebrae. These fractures predominantly occurred in the thoracolumbar region, specifically between T3 and L5, with a high frequency observed at the levels of T11 to L2.
Genetic analysis results: 91% of PLO patients had relevant variants, and 55% carried disease-causing variants
Whole-exome sequencing using a customized 398-gene skeletal disease virtual panel was performed on 11 women with PLO, identifying heterozygous variants in 10 cases (Figure 2). Based on the prior PLO study’s classification framework (8), relevant variants (RVs) were defined as genetic variants in skeletal disorder-related genes meeting ACMG/AMP categories II–V criteria. Our analysis revealed that 91% of PLO patients carried RVs (Table 2). Disease-causing variants (DCVs), a subset of RVs classified as ACMG/AMP categories IV and V, were identified in six PLO patients (55%). Seven DCVs were detected in these six women, affecting five genes: WNT1 (n = 2), LRP5 (n = 2), LTBP3 (n = 1), COL10A1 (n = 1) and CALCR (n = 1); one woman (Case 8) carried two DCVs, one in COL10A1 and one in LRP5, and is counted once in the cohort-level proportion (Table 3). All identified DCVs were visually confirmed by manual review of read alignments in the Integrative Genomics Viewer.
Figure 2.
Genetic landscape and cohort distribution based on variant status. The pie chart classifies the 11 women by variant profile: those with at least one disease-causing variant (DCV; with or without concurrent relevant variants [RVs]), those with RVs but no DCV, and those with neither. The adjacent panel lists the genes affected by the seven DCVs identified in the six DCV-positive women. The “COL10A1/LRP5” entry denotes the single woman (Case 8) who carried two DCVs, one in each of these genes; she is counted once in the pie chart. Gene-level counts in this panel correspond to the DCV rows of Table 2 and to Table 3.
Table 2.
Case-level summary of disease-causing variants (DCVs) and relevant variants (RVs) identified by virtual panel analysis in the PLO cohort.
| Case | Category | Gene | HGVSc | HGVSp | ACMG classes | VARSOME (2025 June) | Genebe (2025 June) | ClinVar (2025_Jan) | MOI (GenCC) |
|---|---|---|---|---|---|---|---|---|---|
| 1 | DCV | WNT1 | NM_005430.4:c.677C>T | p.(Ser226Leu) | LP | LP | LP | Not reported | AD/AR |
| RV | IDUA | NM_000203.5:c.1597C>T | p.(Pro533Ser) | LP | LP | VUS | P(1) VUS(1) | AR | |
| RV | FGFR2 | NM_000141.5:c.253A>G | p.(Thr85Ala) | VUS | VUS | VUS | Not reported | AD | |
| RV | P3H1 | NM_022356.4:c.326T>G | p.(Phe109Cys) | VUS | VUS | VUS | VUS (1) | AD/AR | |
| 2 | RV | COL27A1 | NM_032888.4:c.3786T>A | p.(Tyr1262*) | P | LP | P | Not reported | AR |
| RV | MCM5 | NM_006739.4:c.1976-9T>A | VUS | VUS | VUS | Not reported | Unknown | ||
| 3 | RV | PIK3C2A | NM_002645.4:c.1352T>C | p.(Ile451Thr) | VUS | VUS | VUS | VUS (1) | AR |
| DCV | WNT1 | NM_005430.4:c.291C>A | p.(Phe97Leu) | VUS | LP | VUS | Not reported | AD/AR | |
| 4 | RV | FAM20C | NM_020223.4:c.956 + 12_956 + 13insGTCCTCCCTCCATCCGCGACAGGTGAGCCCTTCC | VUS | VUS | VUS | Not reported | AR | |
| 5 | not found | ||||||||
| 6 | DCV | LTBP3 | NM_001130144.3:c.1936G>A | p.(Gly646Ser) | VUS | VUS | LP | Not reported | AD/AR |
| RV | PAPSS2 | NM_001015880.2:c.397C>T | p.(Arg133Cys) | VUS | VUS | VUS | VUS (1) | AR | |
| 7 | RV | COL5A2 | NM_000393.5:c.4352C>T | p.(Ser1451Phe) | VUS | VUS | VUS | Not reported | AD |
| RV | EVC | NM_153717.3:c.769C>A | p.(Leu257Ile) | VUS | VUS | VUS | Not reported | AD/AR | |
| RV | VPS33A | NM_022916.6:c.1302 + 5G>A | VUS | VUS | VUS | Not reported | AR | ||
| 8 | DCV | COL10A1 | NM_000493.4:c.1228G>C | p.(Gly410Arg) | VUS | VUS | LP | Not reported | AD |
| RV | GHSR | NM_198407.2:c.107_109delAGC | p.(Gln36del) | VUS | VUS | VUS | Not reported | AD/AR | |
| DCV | LRP5 | NM_002335.4:c.3608C>T | p.(Ala1203Val) | VUS | VUS | LP | Not reported | AD/AR | |
| 9 | RV | ACAN | NM_001369268.1:c.7369A>G | p.(Met2457Val) | VUS | VUS | VUS | Not reported | AD/AR |
| DCV | CALCR | NM_001164737.3:c.520A>T | p.(Arg174*) | VUS | LP | VUS | Not reported | Unknown | |
| RV | CFAP410 | NM_004928.3:c.674G>C | p.(Arg225Pro) | VUS | VUS | VUS | Not reported | AD | |
| RV | MEOX1 | NM_004527.4:c.593G>A | p.(Arg198Gln) | VUS | VUS | VUS | VUS (1) | AD/AR | |
| 10 |
RV | CRTAP | NM_006371.5:c.143A>T | p.(Tyr48Phe) | VUS | VUS | VUS | VUS (2) | AD/AR |
| RV | CWC27 | NM_005869.4:c.427C>T | p.(Arg143*) | P | P | P | P(3) | AR | |
| RV | DSPP | NM_014208.3:c.2024_2025insGGACAGCAGCAGTAGCAG | p.(Ser674_Ser675insArgAspSerSerSerSer) | VUS | VUS | VUS | Not reported | AD | |
| 11 | DCV | LRP5 | NM_002335.4:c.350G>A | p.(Trp117*) | P | LP | P | Not reported | AD |
| RV | ASCC1 | NM_001198800.3:c.317dup | p.(Gly107TrpfsTer15) | P | LP | P | Not reported | AR | |
| RV | FKBP10 | NM_021939.4:c.745C>T | p.(Gln249*) | P | LP | P | Not reported | AR | |
| RV | LRRK1 | NM_024652.6:c.3439G>T | p.(Ala1147Ser) | VUS | VUS | LP | VUS (1) | AR |
All variants were heterozygous. Annotations included manual ACMG classes, VarSome (June 2025), GeneBe (June 2025), ClinVar (Jan 2025), and mode of inheritance (MOI; GenCC). Abbreviations: AD, autosomal dominant; AR, autosomal recessive; HGVS, Human Genome Variation Society nomenclature (coding level); HGVSp, HGVS description at the protein level; LP, likely pathogenic; MOI, mode of inheritance; P, pathogenic; VUS, variant of uncertain significance. Case 5 carried no variant meeting the RV criteria. Case 8 carried two DCVs, one in COL10A1 and one in LRP5; the seven DCVs listed here were therefore identified in six women, and Case 8 is counted once in the cohort-level DCV proportion of 6/11 (55%). This is the basis for the “COL10A1/LRP5” category shown in Figure 2. Note: for variant NM_001164737.3 (CALCR):c.520A>T the variant allele fraction (VAF) was 0.29.
Table 3.
The seven disease-causing variants (DCVs) identified by virtual panel analysis in six of the 11 women with PLO (55%).
| Gene | HGVSc | Protein level | ACMG classes | AF in gnomAD v4.1.0 |
AF in gnomAD v4.1.0 East-Asian |
AF in TWB | ClinVar (2025_Jan) | CADD | SIFT | Polyphen-2 | Bayes Del_addAF |
|---|---|---|---|---|---|---|---|---|---|---|---|
| WNT1 | NM_005430.4:c.677C>T | p.(Ser226Leu) | LP | 0.000001240 | 0.00004468 | 0 | Not reported | 30 | Damaging | Damaging | Damaging |
| WNT1 | NM_005430.4:c.291C>A | p.(Phe97Leu) | LP | 0.000004960 | 0.0001783 | 0.0005 | Not reported | 27 | Damaging | Damaging | Damaging |
| LTBP3 | NM_001130144.3:c.1936G>A | p.(Gly646Ser) | LP | Not found | Not found | 0 | Not reported | 30 | Damaging | Damaging | Damaging |
| COL10A1* | NM_000493.4:c.1228G>C | p.(Gly410Arg) | LP | 0.000001239 | 0.00004458 | 0 | Not reported | 25 | Damaging | Damaging | Damaging |
| LRP5* | NM_002335.4:c.3608C>T | p.(Ala1203Val) | LP | 0.000 | 0.000 | 0 | Not reported | 21 | Damaging | Damaging | Damaging |
| LRP5 | NM_002335.4:c.350G>A | p.(Trp117*) | P | 0.000 | 0.000 | 0 | Not reported | 40 | n/a | n/a | Damaging |
| CALCR | NM_001164737.3:c.520A>T | p.(Arg174*) | LP | Not found | Not found | 0 | Not reported | 29 | Damaging | n/a | Tolerant |
Annotations include manual ACMG classes, ClinVar (Jan 2025) and in silico predictors. Allele frequencies (AF) were obtained from gnomAD v4.1.0, its East Asian subset, and the Taiwan Biobank (TWB). Variants marked with an asterisk (*) co-occurred in the same patient (Case 8). All variants were heterozygous. In silico predictions were used only as supporting evidence within the ACMG/AMP framework; none of these variants underwent functional or segregation validation. Abbreviations: BayesDel_addAF, BayesDel with allele frequency; CADD, Combined Annotation Dependent Depletion; LP, likely pathogenic; P, pathogenic; PolyPhen-2, Polymorphism Phenotyping v2; SIFT, Sorting Intolerant From Tolerant; n/a, not applicable. Note: for variant NM_001164737.3 (CALCR):c.520A>T the variant allele fraction (VAF) was 0.29.
Subgroup differences based on the genetic variants
To evaluate the impact of genetic status on clinical presentation of PLO, we compared clinical and radiographic features between patients with and without DCVs (Table 1). No significant differences were found in anthropometrics, vertebral fracture occurrence, or hip BMD Z-scores. In contrast, lumbar spine BMD Z-scores were significantly lower in patients with DCVs compared to those without (-3.7 vs. -3.0, P = 0.043, Wilcoxon rank-sum test). Statistical analysis revealed a trend toward significance for a younger age at diagnosis (34 vs. 37, P = 0.054) and a higher proportion of multiple vertebral fractures (100% vs. 40%, P = 0.061) in patients with DCVs (Table 1).
Treatments and follow-up
The lumbar spine BMD Z-scores at diagnosis and after a median follow-up of 25 months (23.5, 31) are illustrated in Figure 3. Upon PLO diagnosis, all 11 patients discontinued breastfeeding and initiated standard combined supplementation consisting of calcium (600–800 mg/day) and vitamin D (cholecalciferol 1500–2000 IU/day or alfacalcidol 0.5 μg/day). Due to severe back pain, a profound sense of maternal inadequacy, and significant psychological distress, three patients received additional bone-active therapies. Specifically, Case 1 (a WNT1 pathogenic carrier) achieved pain relief and an improved femoral neck Z-score from -2.5 to -1.9 after 3 years of denosumab. Case 3 (another WNT1 pathogenic variant carrier) attained complete pain resolution and a lumbar spine Z-score increase from -3.3 to -1.3 over 2 years via sequential therapy consisting of 8 months of teriparatide followed by three doses of denosumab. Similarly, Case 11 (an LRP5 pathogenic variant carrier) underwent a 2-year regimen of romosozumab (12 months) and denosumab (three doses), resulting in complete pain relief and a lumbar spine Z-score increase from -3.5 to -2.4. Notably, Case 11 was the only patient with pre-conception densitometry data, having recorded a lumbar Z-score of -2.9 four years prior to pregnancy. Within the cohort, the eight patients managed solely with lactation cessation and supplementation demonstrated a significant lumbar spine Z-score improvement from -3.45 to -2.3 over the follow-up period (P = 0.012). When stratified by variant status, patients without DCVs exhibited a significant improvement from -3.0 to -2.0 (P = 0.043), while those with DCVs showed a non-significant change from -3.8 to -3.3 (P = 0.109). Furthermore, no statistically significant differences were observed between the conservative treatment and bone-active therapy groups regarding lumbar spine BMD Z-scores at baseline (-3.45 vs. -3.5, P = 0.538), at final follow-up (-2.3 vs. -1.85, P = 0.792), or in the magnitude of increase (0.7 vs. 1.55, P = 0.296). Overall, the entire cohort achieved complete back pain resolution and a statistically significant increase in median lumbar spine Z-scores (from -3.5 to -2.3, P = 0.005) at the median 25-month follow-up. Despite these improvements, skeletal recovery remained incomplete, with no individual achieving a Z-score above -1.0. During the follow-up period, no subsequent pregnancies or fracture recurrences were observed within the cohort.
Figure 3.
Clinical management and longitudinal skeletal outcomes of the PLO cohort. Asterisks (*) denote carriers of pathogenic DCVs. All individuals discontinued breastfeeding and initiated calcium and vitamin D supplementation upon diagnosis. Driven by severe back pain and secondary psychological distress, three patients (Cases 1, 3, and 11) received adjunctive bone-active therapies. At the 2-3-year follow-up, the entire cohort achieved complete back pain resolution and significantly improved bone mineral density (median lumbar spine Z-score increased from -3.5 to -2.3; P = 0.005), though individual Z-scores consistently remained below -1.0. Note: Lumbar spine densitometry was not assessed for Case 1; this individual exhibited an improved femoral neck Z-score from -2.5 to -1.9 at year three (data not shown).
Discussion
PLO is a rare and heterogeneous disorder, with accumulating evidence implicating genetic susceptibility in its pathogenesis (7). Previous exome-sequencing studies have reported a high prevalence of skeletal disorder-associated variants in women with PLO, ranging from 30% in a North American cohort (9) and 50% in a German cohort (8) to 78.6% in a Chinese cohort (22), although these frequencies were derived using different analytical and classification frameworks. In our cohort, RVs were identified in 10 of 11 patients (91%), and six patients (55%) carried variants classified as DCVs under our predefined framework. Patients with DCVs exhibited significantly lower lumbar spine BMD, together with a trend toward both a younger age at PLO diagnosis and a higher proportion of multiple vertebral fractures than those without, suggesting a genetic contribution to disease severity. While the high variant frequency in our cohort suggests a significant genetic component in PLO, these findings warrant caution due to the small sample size. This observation creates an opportunity for pre-pregnancy genetic testing in women with existing fractures or confirmed family history of monogenic bone disorders, potentially shifting the focus from broad prediction to diagnostic clarification and personalized monitoring. In light of the absence of current guidelines for targeted screening in high-risk women, ensuring adequate calcium and vitamin D supplementation before and during pregnancy, and avoiding breastfeeding postpartum, could represent a preliminary preventative strategy for high-risk women with pathogenic variants pending further validation. Furthermore, our observations are consistent with previous exome-sequencing studies (8, 9, 22) showing that variants in the WNT/β-catenin signaling cascade predispose women to pregnancy- and lactation-associated bone loss. This finding also aligns with genome-wide association studies and gene-based analyses on BMD and osteoporosis in the Taiwanese population, which have highlighted that WNT-mediated signaling is one of the pivotal pathways in bone health (23).
The WNT1 protein, a central regulator of the WNT/β-catenin pathway, is essential for osteoblast differentiation and bone formation (24). Variants in the WNT1 gene have been widely linked to early-onset osteoporosis and osteogenesis imperfecta (25–27). We identified a recurrent heterozygous variant, NM_005430.4(WNT1):c.677C>T p.(Ser226Leu), previously shown to impair WNT/β-catenin signaling and to downregulate BMP-2 expression in preosteoblasts (28, 29), and a second missense variant, NM_005430.4(WNT1):c.291C>A(p.Phe97Leu), which is absent from the ClinVar database (January 2025 release) and requires functional characterization. LRP5 has been linked to both primary osteoporosis and PLO (30–32). Previous reports have described heterozygous and compound heterozygous LRP5 variants that contribute to the clinical manifestation of PLO (33, 34). Here we identified a rare nonsense variant, NM_002335.4(LRP5):c.350G>A p.(Trp117*), which was not reported in ClinVar at the time of annotation and is predicted to truncate the LRP5 protein within its first β-propeller domain and thereby to impair WNT co-receptor function. Additionally, one patient in our cohort carried a heterozygous variant in NM_002335.4(LRP5):c.3608C>T p.(Ala1203Val) alongside a heterozygous variant in NM_000493.4(COL10A1):c.1228G>C p.(Gly410Arg). These two missense variants are previously unreported and require further cellular functional verification. COL10A1 encodes type X collagen, which is crucial for endochondral ossification, and has been linked to both metaphyseal chondrodysplasia and osteoporosis (35). The co-occurrence of these variants suggests a possible polygenic contribution to skeletal fragility in this population. We also identified rare variants in CALCR and LTBP3 genes implicated in bone remodeling. CALCR encodes the calcitonin receptor regulating osteoclast-mediated bone resorption (36). The nonsense variant NM_001164737.3(CALCR):c.520A>T p.(Arg174*) truncates the receptor immediately before the position at which the two human calcitonin receptor isoforms differ by a 16-amino-acid insertion within the first intracellular loop (37, 38), and would be expected to abolish receptor signaling. Of note, this variant was detected at a variant allele fraction of 0.29, and although it was confirmed on manual review of the read alignments, somatic mosaicism or allelic imbalance in the capture data cannot be excluded without an independent orthogonal assay. LTBP3 gene encodes latent transforming growth factor-beta binding protein 3 (LTBP3), a critical regulator of transforming growth factor-beta (TGF-β) bioavailability. Pathogenic variants in this gene are associated with skeletal dysplasia and low BMD (39). Mechanistically, LTBP3 deficiency may decrease active TGF-β in bone and cartilage, potentially impairing osteoblast and osteoclast function. Further research is warranted to validate the NM_001130144.3(LTBP3):c.1936G>A p.(Gly646Ser) variant identified here in the pathogenesis of osteoporosis and fragility fractures. During pregnancy, bone turnover markers (formation and resorption) are elevated with consistently high parathyroid hormone-related protein driving calcium mobilization from maternal bone (1). While the specific interplay in human gestation remains uncharacterized, animal models suggest that CALCR gene expression appears upregulated (2- to 3-fold) in both cortical and trabecular bones at the end of pregnancy and lactation compared to virgin controls, expressed specifically by osteoclasts (40), with the potential to restrict excessive bone resorption. In addition, studies in osteoporotic rats indicate that calcitonin receptor activation could upregulate WNT ligands (41), suggesting a potential synergistic mechanism for preserving skeletal integrity under physiological stress. However, direct evidence for such a CALCR-WNT/LRP5 interplay during human pregnancy is currently lacking and warrants further investigation. Furthermore, during lactation, animal models have demonstrated that osteoclast-mediated bone resorption (mainly in trabecular bone) and osteocytic osteolysis (in trabecular and cortical bones) mobilize maternal bone minerals (42–45), while hypothalamic cellular communication network factor 3 (CCN3) signaling drives compensatory bone formation (46). This delicate balance involves LTBP3 and COL10A1 during local matrix remodeling, incorporates CALCR signaling to restrict excessive resorption, and features CCN3 cross-talk with the WNT1/LRP5 signaling axis to stimulate osteoblastic recovery throughout the breastfeeding period. Together, these findings support that PLO is not just a “mechanical” result of pregnancy/lactation, but often has a pre-existing genetic susceptibility.
During lactation, the physiological mobilization of calcium occurs preferentially from trabecular bone (42–45), a pattern reflected in our cohort by significantly lower lumbar spine Z-scores compared to the hip. Consistent with previous reports (6), vertebral fractures were the predominant manifestation of PLO, often involving multiple thoracolumbar levels, though sacral alae and pubic bone fractures were also observed. Beyond physical disability, these fractures caused debilitating pain that severely impaired essential newborn care, such as lifting or bathing infants. This functional incapacity, compounded by the risk of fracture recurrence, often led to a profound sense of maternal inadequacy and significant psychological distress. Consequently, the management of PLO must integrate both skeletal stabilization and comprehensive psychosocial support for the mother.
In the absence of controlled trials, PLO screening and management remain empirical and lack established guidelines. A thorough investigation is necessary to identify potential contributing factors, such as underweight, advanced maternal age, nutritional deficiencies (calcium and vitamin D), sedentary lifestyle, smoking, chronic diseases, and previous fragility fractures (6, 47–50). Early lactation cessation and calcium/vitamin D supplementation constitute the mainstay of management (6, 47–49). Current evidence has not yet established a definitive role for bone formation or anti-resorptive agents in the management of PLO (51). In our cohort, patients managed solely with this conservative approach demonstrated significant lumbar BMD recovery over the follow-up period. While these preliminary observations suggest that conservative management could serve as a foundational strategy, our study lacked the statistical power to conclusively evaluate the therapeutic efficacy of adjunctive anabolic/anti-resorptive agents. Consequently, whether these therapies could alter the trajectory of BMD accrual, accelerate back pain resolution, or expedite the restoration of maternal function remains to be elucidated. Larger, multi-center datasets are required to fully clarify these therapeutic implications for PLO patients both with and without specific genetic risk variants.
Epidemiological studies indicate that lactation generally does not alter peak bone mass or long-term fracture risk because BMD typically recovers within 6–12 months after weaning (42). In our cohort, all 11 patients showed BMD improvement 25 months after weaning and treatment. However, none reached a Z-score above -1.0. This finding aligns with a recent systematic review (52) of 451 PLO patients (1988-2022). While that review noted significant overall BMD gains across treatments, evaluation of the compiled individual case profiles reveals that post-treatment lumbar Z-scores predominantly remained below -1.0. Although a Z-score above -2.0 formally defines bone density as “within the expected range for age” (10), a Z-score below -1.0 corresponds to the 16th percentile or lower of an age-matched population. This outcome may reflect low pre-pregnancy bone mass or incomplete skeletal recovery. While our cohort had no history of pre-pregnancy fractures, Case 11 was the sole patient with a documented baseline BMD. Her pre-pregnancy lumbar spine Z-score of -2.9 explicitly demonstrates that her bone density was compromised below the expected range for age before her pregnancy. Moreover, previous reports suggest that PLO may be a predictor of persistent osteopenia (53) and future fractures (54). Furthermore, high-resolution peripheral quantitative computed tomography (HR-pQCT) studies in lactating women have revealed persistent impairments in trabecular and cortical parameters. These findings suggest potential permanent alterations in bone microarchitecture (55, 56). While most women regain bone mass after weaning, our results and prior HR-pQCT data indicate that recovery may be incomplete in individuals with underlying genetic defects. This underscores the clinical value of pre-pregnancy BMD screening for identifying women at high risk for fragility fractures and highlights the necessity of long-term clinical monitoring.
Study limitations
Our study has several limitations that warrant consideration. First, the sample size is relatively small (n = 11), which is a common constraint when investigating rare conditions such as PLO. The high detection rate of genetically relevant variants (91%) and pathogenic defects (55%) observed in our cohort suggests a potential genetic contribution to PLO, highlighting the need for larger studies. Second, no functional or segregation studies were performed. Variant classification relied on computational prediction and database evidence, and describes pathogenicity under the ACMG/AMP framework rather than a demonstrated contribution to the phenotype. Variants without prior PLO-specific functional or segregation evidence require functional validation and, where possible, family segregation analysis before their contribution to the PLO phenotype can be established. Genetic variants alone cannot establish causality in an observational cohort of this size. Third, our cohort was drawn entirely from an Asian population at a single tertiary medical center; consequently, these findings may not fully represent the genetic architecture or clinical progression of PLO in ethnically diverse populations. Fourth, because pre-pregnancy BMD data were unavailable for most participants, we could not definitively establish baseline skeletal health prior to the physiological stress of childbearing. Finally, while we observed significant BMD improvement over a 25-month follow-up period, longer-term longitudinal tracking is required to determine the lifetime fracture risk and the permanent trajectory of skeletal recovery in women harboring these specific genetic variants.
Conclusion
Our study demonstrates that PLO is characterized by severe, multi-level vertebral fractures and a significant genetic burden. The high prevalence of genetic variants observed in this cohort suggests a potential genetic contribution to PLO, likely precipitated by the physiological stressors of pregnancy and lactation. Specific pathogenic variants indentified (WNT1, LRP5, COL10A1, LTBP3, CALCR) appear associated with phenotypic severity, offering preliminary insights into the connection between molecular discoveries and clinical presentation. The clinical hallmark of PLO is preferential trabecular bone loss, leading to a high frequency of vertebral fractures that cause debilitating pain and severely impair newborn care. Although timely lactation cessation and therapeutic interventions significantly improved lumbar BMD at a median 25-month longitudinal follow-up, skeletal recovery remains incomplete, with no patient achieving a Z-score above -1.0. Therefore, acute postpartum back pain warrants immediate investigation. Recognizing this underlying genetic burden warrants consideration of early intervention, long-term bone health monitoring, and future development of genotype-informed management strategies to potentially improve maternal outcomes.
Acknowledgments
We are grateful to Ms. Shu-Chen Chung for her assistance in data collection, and to the National Center for High-performance Computing (NCHC) of National Applied Research Laboratories (NARLabs) of Taiwan for providing computational and storage resources.
Funding Statement
The author(s) declared that financial support was received for this work and/or its publication. This study was funded by the National Science and Technology Council, Taiwan (NSTC- 112-2314-B-002-154), the Wong-Yuan Endocrine Fund, the Osteoporosis and Metabolic Bone Diseases Research Fund (MS042), and research grants from the Department of Medical Research in the authors’ hospital (105-N3322). The funders had no role in study design, data collection, analysis, interpretation, manuscript preparation, or the decision to submit the work for publication.
Footnotes
Edited by: Vito Guarnieri, Home for Relief of Suffering (IRCCS), Italy
Reviewed by: Nataliya Gak, Imperial College London, United Kingdom
Yusen Qiao, The First Affiliated Hospital of Soochow University, China
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.
Ethics statement
The studies involving humans were approved by the Research Ethics Committee at National Taiwan University Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
W-YC: Visualization, Formal analysis, Data curation, Writing – original draft, Methodology, Conceptualization, Investigation, Writing – review & editing. Y-AC: Conceptualization, Writing – original draft, Software, Methodology, Visualization, Writing – review & editing. S-RS: Methodology, Writing – review & editing, Conceptualization, Investigation, Writing – original draft. R-SY: Conceptualization, Investigation, Methodology, Writing – review & editing, Writing – original draft. M-HH: Writing – review & editing, Investigation, Conceptualization, Methodology, Writing – original draft. W-PY: Methodology, Conceptualization, Writing – review & editing, Investigation, Writing – original draft. P-LC: Visualization, Conceptualization, Methodology, Supervision, Writing – review & editing, Software, Writing – original draft. K-ST: Investigation, Supervision, Writing – review & editing, Methodology, Conceptualization, Data curation, Writing – original draft. WSY: Methodology, Supervision, Conceptualization, Funding acquisition, Visualization, Writing – review & editing, Writing – original draft.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was used in the creation of this manuscript. During the preparation of this work, the authors used Claude from Anthropic to improve the language, grammar, and spelling. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fendo.2026.1930247/full#supplementary-material
References
- 1. Gak N, Abbara A, Dhillo WS, Keen R, Comninos AN. Current and future perspectives on pregnancy and lactation-associated osteoporosis. Front Endocrinol (Lausanne). (2024) 15:1494965. doi: 10.3389/fendo.2024.1494965 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Hardcastle SA. Pregnancy and lactation associated osteoporosis. Calcif Tissue Int. (2022) 110:531–45. doi: 10.1007/s00223-021-00815-6 [DOI] [PubMed] [Google Scholar]
- 3. Qian Y, Wang L, Yu L, Huang W. Pregnancy- and lactation-associated osteoporosis with vertebral fractures: a systematic review. BMC Musculoskelet Disord. (2021) 22:926. doi: 10.1186/s12891-021-04776-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Terzi R, Terzi H, Özer T, Kale A. A rare cause of postpartum low back pain: pregnancy- and lactation-associated osteoporosis. BioMed Res Int. (2014) 2014:287832. doi: 10.1155/2014/287832 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Orhadje E, Berg K, Hauser B, Ralston SH. Clinical features, incidence and treatment outcome in pregnancy-associated osteoporosis: a single-centre experience over two decades. Calcif Tissue Int. (2023) 113:591–6. doi: 10.1007/s00223-023-01139-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kasahara K, Tanaka-Mizuno S, Tsuji S, Ohashi M, Kasahara M, Kawasaki T, et al. Pregnancy and lactation-associated osteoporosis as a major type of premenopausal osteoporosis: a retrospective cohort study based on real-world data. BMC Pregnancy Childbirth. (2024) 24:301. doi: 10.1186/s12884-024-06520-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Peris P, Guañabens N, Monegal A, Pons F, Martínez de Osaba MJ, Ros I, et al. Pregnancy associated osteoporosis: the familial effect. Clin Exp Rheumatol. (2002) 20:697–700. [PubMed] [Google Scholar]
- 8. Butscheidt S, Tsourdi E, Rolvien T, Delsmann A, Stürznickel J, Barvencik F, et al. Relevant genetic variants are common in women with pregnancy and lactation-associated osteoporosis (PLO) and predispose to more severe clinical manifestations. Bone. (2021) 147:115911. doi: 10.1016/j.bone.2021.115911 [DOI] [PubMed] [Google Scholar]
- 9. Lynch L, Shea PR, Vena N, Chung WK, Shane E, Kamanda-Kosseh M, et al. Genetic contributors to osteoporosis in pregnancy and lactation associated osteoporosis (PLO). Bone. (2026) 204:117744. doi: 10.1016/j.bone.2025.117744 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Shuhart C, Cheung A, Gill R, Gani L, Goel H, Szalat A. Executive summary of the 2023 adult position development conference of the International Society for Clinical Densitometry: dxa reporting, follow-up bmd testing and trabecular bone score application and reporting. J Clin Densitom. (2024) 27:101435. doi: 10.1016/j.jocd.2023.101435 [DOI] [PubMed] [Google Scholar]
- 11. Genant HK, Wu CY, van Kuijk C, Nevitt MC. Vertebral fracture assessment using a semiquantitative technique. J Bone Miner Res. (1993) 8:1137–48. doi: 10.1002/jbmr.5650080915 [DOI] [PubMed] [Google Scholar]
- 12. Behera S, Catreux S, Rossi M, Truong S, Huang Z, Ruehle M, et al. Comprehensive genome analysis and variant detection at scale using dragen. Nat Biotechnol. (2025) 43:1177–91. doi: 10.1038/s41587-024-02382-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. McLaren W, Gil L, Hunt SE, Riat HS, Ritchie GR, Thormann A, et al. The ensembl variant effect predictor. Genome Biol. (2016) 17:122. doi: 10.1186/s13059-016-0974-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kircher M, Witten DM, Jain P, O'Roak BJ, Cooper GM, Shendure J. A general framework for estimating the relative pathogenicity of human genetic variants. Nat Genet. (2014) 46:310–5. doi: 10.1038/ng.2892 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Jaganathan K, Kyriazopoulou Panagiotopoulou S, McRae JF, Darbandi SF, Knowles D, Li YI, et al. Predicting splicing from primary sequence with deep learning. Cell. (2019) 176:535–48:e24. doi: 10.1016/j.cell.2018.12.015 [DOI] [PubMed] [Google Scholar]
- 16. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. (2015) 17:405–24. doi: 10.1038/gim.2015.30 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Taiwan Ai Labs. Taigenomics. Gendiseak Platform . Taiwan Ai Labs, Taigenomics, Gendiseak Platform. Available online at: https://www.tw23.org/en/service/gendiseak/ (Accessed September 30, 2026).
- 18. Chen YA, Yuan TH, Huang JH, Wang YB, Hung TM, Chen CY, et al. Toward streamline variant classification: discrepancies in variant nomenclature and syntax for clinvar pathogenic variants across annotation tools. Hum Genomics. (2025) 19:70. doi: 10.1186/s40246-025-00778-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Landrum MJ, Lee JM, Benson M, Brown GR, Chao C, Chitipiralla S, et al. Clinvar: improving access to variant interpretations and supporting evidence. Nucleic Acids Res. (2018) 46:D1062–7. doi: 10.1093/nar/gkx1153 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Kopanos C, Tsiolkas V, Kouris A, Chapple CE, Albarca Aguilera M, Meyer R, et al. Varsome: the human genomic variant search engine. Bioinformatics. (2019) 35:1978–80. doi: 10.1093/bioinformatics/bty897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Stawinski P, Ploski R. Genebe.Net: implementation and validation of an automatic acmg variant pathogenicity criteria assignment. Clin Genet. (2024) 106:119–26. doi: 10.1111/cge.14516 [DOI] [PubMed] [Google Scholar]
- 22. Fu Z, Ni X, Han S, Li M, Wang O, Xing X, et al. Gene variants and bisphosphonates treatment in pregnancy and lactation-associated osteoporosis (PLO): a retrospective study of 22 Chinese patients. Calcif Tissue Int. (2025) 116:70. doi: 10.1007/s00223-025-01381-x [DOI] [PubMed] [Google Scholar]
- 23. Lin YJ, Liang WM, Chiou JS, Chou CH, Liu TY, Yang JS, et al. Genetic predisposition to bone mineral density and their health conditions in East Asians. J Bone Miner Res. (2024) 39:929–41. doi: 10.1093/jbmr/zjae078 [DOI] [PubMed] [Google Scholar]
- 24. Baron R, Kneissel M. Wnt signaling in bone homeostasis and disease: from human mutations to treatments. Nat Med. (2013) 19:179–92. doi: 10.1038/nm.3074 [DOI] [PubMed] [Google Scholar]
- 25. Laine CM, Joeng KS, Campeau PM, Kiviranta R, Tarkkonen K, Grover M, et al. Wnt1 mutations in early-onset osteoporosis and osteogenesis imperfecta. N Engl J Med. (2013) 368:1809–16. doi: 10.1056/NEJMoa1215458 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Keupp K, Beleggia F, Kayserili H, Barnes AM, Steiner M, Semler O, et al. Mutations in Wnt1 cause different forms of bone fragility. Am J Hum Genet. (2013) 92:565–74. doi: 10.1016/j.ajhg.2013.02.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Wang Q, Liu M, Cao BY, Su C, Meng X, Ding Y, et al. Osteoporosis caused by monoallelic variant of Wnt1 gene in four pediatric patients. Am J Med Genet A. (2025) 197:e63987. doi: 10.1002/ajmg.a.63987 [DOI] [PubMed] [Google Scholar]
- 28. Zhang B, Li R, Wang W, Zhou X, Luo B, Zhu Z, et al. The role of Wnt1 mutant variant (Wnt1 c.677 C>T) in osteogenesis imperfecta. Ann Hum Genet. (2020) 84:447–55. doi: 10.1111/ahg.12399 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Hu J, Lin X, Gao P, Zhang Q, Zhou B, Wang O, et al. Genotypic and phenotypic spectrum and pathogenesis of Wnt1 variants in a large cohort of patients with OI/osteoporosis. J Clin Endocrinol Metab. (2023) 108:1776–86. doi: 10.1210/clinem/dgac752 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Hartikka H, Makitie O, Mannikko M, Doria AS, Daneman A, Cole WG, et al. Heterozygous mutations in the LDL receptor-related protein 5 (LRP5) gene are associated with primary osteoporosis in children. J Bone Miner Res. (2005) 20:783–9. doi: 10.1359/JBMR.050101 [DOI] [PubMed] [Google Scholar]
- 31. Saarinen A, Saukkonen T, Kivela T, Lahtinen U, Laine C, Somer M, et al. Low density lipoprotein receptor-related protein 5 (LRP5) mutations and osteoporosis, impaired glucose metabolism and hypercholesterolaemia. Clin Endocrinol (Oxf). (2010) 72:481–8. doi: 10.1111/j.1365-2265.2009.03680.x [DOI] [PubMed] [Google Scholar]
- 32. Ferrari SL, Deutsch S, Baudoin C, Cohen-Solal M, Ostertag A, Antonarakis SE, et al. LRP5 gene polymorphisms and idiopathic osteoporosis in men. Bone. (2005) 37:770–5. doi: 10.1016/j.bone.2005.06.017 [DOI] [PubMed] [Google Scholar]
- 33. Butscheidt S, Delsmann A, Rolvien T, Barvencik F, Al-Bughaili M, Mundlos S, et al. Mutational analysis uncovers monogenic bone disorders in women with pregnancy-associated osteoporosis: three novel mutations in LRP5, COL1A1, and COL1A2. Osteoporos Int. (2018) 29:1643–51. doi: 10.1007/s00198-018-4499-4 [DOI] [PubMed] [Google Scholar]
- 34. Campos-Obando N, Oei L, Hoefsloot LH, Kiewiet RM, Klaver CC, Simon ME, et al. Osteoporotic vertebral fractures during pregnancy: be aware of a potential underlying genetic cause. J Clin Endocrinol Metab. (2014) 99:1107–11. doi: 10.1210/jc.2013-3238 [DOI] [PubMed] [Google Scholar]
- 35. Lu Y, Wang W, Yang B, Cao G, Du Y, Liu J. Screening and analysis of core genes for osteoporosis based on bioinformatics analysis and machine learning algorithms. Indian J Orthop. (2024) 58:944–54. doi: 10.1007/s43465-024-01152-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Masi L, Becherini L, Gennari L, Colli E, Mansani R, Falchetti A, et al. Allelic variants of human calcitonin receptor: distribution and association with bone mass in postmenopausal Italian women. Biochem Biophys Res Commun. (1998) 245:622–6. doi: 10.1006/bbrc.1998.8445 [DOI] [PubMed] [Google Scholar]
- 37. Gorn AH, Lin HY, Yamin M, Auron PE, Flannery MR, Tapp DR, et al. Cloning, characterization, and expression of a human calcitonin receptor from an ovarian carcinoma cell line. J Clin Invest. (1992) 90:1726–35. doi: 10.1172/JCI116046 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Moore EE, Kuestner RE, Stroop SD, Grant FJ, Matthewes SL, Brady CL, et al. Functionally different isoforms of the human calcitonin receptor result from alternative splicing of the gene transcript. Mol Endocrinol. (1995) 9:959–68. doi: 10.1210/mend.9.8.7476993 [DOI] [PubMed] [Google Scholar]
- 39. Duncan EL, Danoy P, Kemp JP, Leo PJ, McCloskey E, Nicholson GC, et al. Genome-wide association study using extreme truncate selection identifies novel genes affecting bone mineral density and fracture risk. PloS Genet. (2011) 7:e1001372. doi: 10.1371/journal.pgen.1001372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Davey RA, Clarke MV, Golub SB, Russell PK, Zajac JD. The calcitonin receptor regulates osteocyte lacunae acidity during lactation in mice. J Endocrinol. (2021) 249:31–41. doi: 10.1530/JOE-20-0599 [DOI] [PubMed] [Google Scholar]
- 41. Hsiao CY, Chen TH, Chu TH, Ting YN, Tsai PJ, Shyu JF. Calcitonin induces bone formation by increasing expression of Wnt10b in osteoclasts in ovariectomy-induced osteoporotic rats. Front Endocrinol (Lausanne). (2020) 11:613. doi: 10.3389/fendo.2020.00613 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Kovacs CS. Maternal mineral and bone metabolism during pregnancy, lactation, and post-weaning recovery. Physiol Rev. (2016) 96:449–547. doi: 10.1152/physrev.00027.2015 [DOI] [PubMed] [Google Scholar]
- 43. Athonvarangkul D, Wysolmerski JJ. Crosstalk within a brain-breast-bone axis regulates mineral and skeletal metabolism during lactation. Front Physiol. (2023) 14:1121579. doi: 10.3389/fphys.2023.1121579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Wysolmerski JJ. Osteocytes remove and replace perilacunar mineral during reproductive cycles. Bone. (2013) 54:230–6. doi: 10.1016/j.bone.2013.01.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Cohen A. Bone metabolism, bone mass, and bone structure during pregnancy and lactation: normal physiology and pregnancy and lactation-associated osteoporosis. Endocrinol Metab Clin North Am. (2024) 53:453–70. doi: 10.1016/j.ecl.2024.05.003 [DOI] [PubMed] [Google Scholar]
- 46. Babey ME, Krause WC, Chen K, Herber CB, Torok Z, Nikkanen J, et al. A maternal brain hormone that builds bone. Nature. (2024) 632:357–65. doi: 10.1038/s41586-024-07634-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Dunne F, Walters B, Marshall T, Heath DA. Pregnancy associated osteoporosis. Clin Endocrinol (Oxf). (1993) 39:487–90. doi: 10.1111/j.1365-2265.1993.tb02398.x [DOI] [PubMed] [Google Scholar]
- 48. O'Sullivan SM, Grey AB, Singh R, Reid IR. Bisphosphonates in pregnancy and lactation-associated osteoporosis. Osteoporos Int. (2006) 17:1008–12. doi: 10.1007/s00198-006-0112-3 [DOI] [PubMed] [Google Scholar]
- 49. Hadji P, Boekhoff J, Hahn M, Hellmeyer L, Hars O, Kyvernitakis I. Pregnancy-associated osteoporosis: a case-control study. Osteoporos Int. (2017) 28:1393–9. doi: 10.1007/s00198-016-3897-8 [DOI] [PubMed] [Google Scholar]
- 50. Grassi G, Zampogna M, Atlasova A, Rondinella S, Ghielmetti A, Siracusano G, et al. Identifying risk factors for pregnancy and lactation-associated osteoporosis: insights from an Italian survey in PLAO patients and controls. Front Endocrinol (Lausanne). (2025) 16:1612188. doi: 10.3389/fendo.2025.1612188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Hadji P, Athanasiadis A, Brandi ML, Chakhtoura M, Chandran M, Chervenak F, et al. Pregnancy- and lactation-associated osteoporosis: A position statement of the IAPM, IOF, ECTS, ESCEO, IMS, and EMAS. Int J Gynaecol Obstet. (2026). doi: 10.1002/ijgo.71203 [DOI] [PubMed] [Google Scholar]
- 52. Anagnostis P, Lampropoulou-Adamidou K, Bosdou JK, Trovas G, Galanis P, Chronopoulos E, et al. Comparative effectiveness of therapeutic interventions in pregnancy and lactation-associated osteoporosis: A systematic review and meta-analysis. J Clin Endocrinol Metab. (2024) 109:879–901. doi: 10.1210/clinem/dgad548 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53. Kurabayashi T, Nagata H, Takeyama N, Matsushita H, Tanaka K. Bone mineral density measurement in puerperal women as a predictor of persistent osteopenia. J Bone Miner Metab. (2009) 27:205–12. doi: 10.1007/s00774-008-0028-4 [DOI] [PubMed] [Google Scholar]
- 54. Kyvernitakis I, Reuter TC, Hellmeyer L, Hars O, Hadji P. Subsequent fracture risk of women with pregnancy and lactation-associated osteoporosis after a median of 6 years of follow-up. Osteoporos Int. (2018) 29:135–42. doi: 10.1007/s00198-017-4239-1 [DOI] [PubMed] [Google Scholar]
- 55. Bjornerem A, Ghasem-Zadeh A, Wang X, Bui M, Walker SP, Zebaze R, et al. Irreversible deterioration of cortical and trabecular microstructure associated with breastfeeding. J Bone Miner Res. (2017) 32:681–7. doi: 10.1002/jbmr.3018 [DOI] [PubMed] [Google Scholar]
- 56. Brembeck P, Lorentzon M, Ohlsson C, Winkvist A, Augustin H. Changes in cortical volumetric bone mineral density and thickness, and trabecular thickness in lactating women postpartum. J Clin Endocrinol Metab. (2015) 100:535–43. doi: 10.1210/jc.2014-2825 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.