# Relationship between bone mineral density and strength derived from 3D-shaper and HR-pQCT in patients with X-linked osteoporosis related to PLS3

**Authors:** Z. Zervou, R. Mabrouk, E. F. S. van Velsen, M. S. A. M. Bevers, E. Alizadeh, J. P. van den Bergh, M. C. Zillikens

PMC · DOI: 10.1007/s11657-026-01656-2 · 2026-01-21

## TL;DR

This study compares bone density and strength measurements in patients with a rare genetic bone disease using two imaging techniques, finding that trabecular bone is more affected than cortical bone.

## Contribution

The study is the first to compare HR-pQCT and 3D-DXA in X-linked osteoporosis caused by PLS3 variants.

## Key findings

- 3D-DXA and HR-pQCT both showed greater trabecular than cortical bone deficits in PLS3 patients.
- 3D-DXA parameters were highly correlated with each other and with aBMD, but not with HR-pQCT parameters.
- Lumbar spine aBMD was lower than total hip aBMD in PLS3 patients.

## Abstract

In individuals with PLS3 genetic variants, HR-pQCT and 3D-DXA analysis indicate a larger trabecular than cortical deficit. This is consistent with the lower aBMD from DXA at the lumbar spine than at the total hip.

X-linked osteoporosis due to PLS3 genetic variants is a rare disease, clinically affecting men more than women. This study aimed to explore relationships between parameters obtained from DXA, high-resolution peripheral quantitative computed tomography (HR-pQCT), and 3D-DXA in thirteen adults with PLS3 variants.

The parameters were for 3D-DXA: trabecular volumetric bone mineral density (Tb.BMD), cortical surface BMD (Ct.sBMD), and bone strength at the total hip using 3D-DXA-based finite element models (3D Shaper® software, v2.14.0), and for HR-pQCT: Tb.BMD, cortical volumetric BMD (Ct.BMD), and bone strength at the distal radius and tibia using micro-finite element analysis. Results were compared with normative data of the scanner system (DXA) or from literature (3D-DXA, HR-pQCT) and expressed as median Z-scores. Spearman correlations were determined between the different parameters.

aBMD at the lumbar spine was numerically lower (median Z-score − 1.7) than at the total hip (Z −1.3). 3D-DXA revealed a low Tb.BMD (Z −1.9), while Ct.sBMD was less affected (Z −1.0). HR-pQCT also showed a lower Tb.BMD (Z −1.9 at the radius and −2.5 at the tibia) than Ct.BMD (Z −0.4 and −1.3, respectively). 3D-DXA parameters were highly correlated with each other (p < 0.001) and with aBMD. No significant correlations were found between any of the 3D-DXA and HR-pQCT parameters.

Both HR-pQCT and 3D-DXA showed lower trabecular than cortical BMD in individuals with PLS3 variants, in line with the lower aBMD at the lumbar spine than at the total hip. Studies in larger cohorts and other bone disorders are needed to examine relationships between imaging modalities and their ability to predict bone fragility.

The online version contains supplementary material available at 10.1007/s11657-026-01656-2.

## Linked entities

- **Genes:** PLS3 (plastin 3) [NCBI Gene 5358]

## Full-text entities

- **Genes:** COL1A2 (collagen type I alpha 2 chain) [NCBI Gene 1278] {aka EDSARTH2, EDSCV, OI4}, PLS3 (plastin 3) [NCBI Gene 5358] {aka BMND18, DIH5, T-plastin}, COL1A1 (collagen type I alpha 1 chain) [NCBI Gene 1277] {aka CAFYD, EDSARTH1, EDSC, OI1, OI2, OI3}
- **Diseases:** Osteoporosis (MESH:D010024), bone fragility (MESH:C536063), peripheral fractures (MESH:D010523), Vertebral fractures (MESH:C535781), hypercalciuria (MESH:D053565), fracture (MESH:D050723), bone failure (MESH:D000080983), pain (MESH:D010146), vertebral compression fractures (MESH:D050815), osteogenesis imperfecta (MESH:D010013), physical disability (MESH:D059445), diabetes mellitus (MESH:D003920), age-related disorder (MESH:D008569), bone diseases (MESH:D001847)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Figures

1 figure with captions in the complete paper: https://tomesphere.com/paper/PMC12823622/full.md

---
Source: https://tomesphere.com/paper/PMC12823622