Critical-size craniofacial bone defects heal poorly, and pharmacological strategies capable of accelerating bone repair remain limited. Soluble epoxide hydrolase (sEH) degrades epoxyeicosatrienoic acids (EETs), lipid mediators associated with inflammation resolution; however, the osteoblastic effectors that translate sEH inhibition into matrix mineralisation are unknown. We investigated whether pharmacological sEH inhibition accelerates craniofacial bone repair, and whether ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) a candidate regulator of pyrophosphate metabolism participates in this response. In vivo, bilateral 5-mm critical-size calvarial defects were created in male Wistar rats (n = 6 per group) treated with the sEH inhibitor TPPU (10 mg/kg/day, p.o.) or vehicle for 14 or 28 days. Repair was assessed by micro-computed tomography, histomorphometry, sEH immunohistochemistry and multiplex spatial RNA in situ hybridisation (RNAscope) of a predefined six-gene osteogenic panel (Spp1, Runx2, Col1a1, Alpl, Bglap, Enpp1). In vitro, human SAOS-2 osteoblastic cells were treated with TPPU and/or the ENPP1 inhibitor STF-1084 to assess viability, osteogenic gene expression, ENPP1 expression and enzymatic activity, alkaline phosphatase (ALP) activity and mineralised matrix formation. At 14 days, TPPU increased BV/TV, BS/TV and trabecular number and reduced trabecular separation and total porosity (p < 0.05); these differences were no longer detected at 28 days, whereas histomorphometry showed a greater newly formed bone area in TPPU-treated animals at both time points. sEH immunolabelling was reduced by TPPU at 14 and 28 days (p < 0.001), confirming target engagement. Spatially, Enpp1-positive cells were more abundant in TPPU-treated tissue at 14 days, the only panel gene reaching significance at the per-cell level. In vitro, TPPU increased ENPP1 mRNA and protein. STF-1084 dose-dependently reduced ENPP1 activity, ALP activity and mineralised matrix deposition without affecting viability, and co-treatment with TPPU did not restore mineralisation despite preserved or increased OSX, RUNX2 and BSP expression. Pharmacological sEH inhibition accelerates the early phase of craniofacial bone repair and is associated with increased ENPP1 expression in regenerating tissue. In human osteoblastic-SAOS cells, ENPP1 enzymatic activity is necessary for mineralisation and for the pro-mineralising effect of sEH inhibition, identifying ENPP1 as a candidate downstream effector of the sEH/EpFA axis and a target for further genetic validation.

Osteoporosis is a major cause of morbidity and fragility fractures in older adults, and increasing evidence suggests that antihypertensive therapies may influence skeletal health beyond blood pressure control. Although angiotensin receptor blockers (ARBs) have been associated with favorable skeletal outcomes, whether clinically relevant heterogeneity exists among individual ARBs remains unclear. Telmisartan possesses distinctive pharmacological properties, including partial peroxisome proliferator-activated receptor-γ (PPAR-γ) agonistic activity, which may confer additional effects on bone metabolism. This study evaluated whether skeletal outcomes differ between telmisartan and other ARBs using an active-comparator new-user design. To evaluate whether clinically relevant differences in osteoporosis and osteoporotic fracture risk exist between telmisartan and other ARBs using Taiwan's Health and Welfare Data Science Center (HWDC) Database. Using Longitudinal Generation Tracking Database (LGTD) housed within the Health and Welfare Data Science Center (HWDC), we conducted a nationwide active-comparator new-user cohort study. Hypertensive patients aged 51-80 years who newly initiated telmisartan or another ARB between 2002 and 2021 were identified. Propensity score matching (1:4) was performed to balance baseline characteristics. The primary outcome was a composite of incident osteoporosis or osteoporotic fracture and time to first osteoporotic fracture was additionally evaluated as a secondary outcome. Adjusted hazard ratios (aHRs) and 95% confidence intervals (CIs) were estimated using Cox proportional hazards models. During a mean follow-up of 4.35 ± 4.12 years, telmisartan use was associated with a lower risk of osteoporosis or osteoporotic fracture compared with other ARBs (adjusted HR, 0.65; 95% CI, 0.53-0.79). In analyses of individual outcomes, telmisartan use was associated with lower risks of osteoporosis (adjusted HR, 0.60; 95% CI, 0.43-0.84) and hip fracture (adjusted HR, 0.37; 95% CI, 0.19-0.70). In the secondary fracture-only analysis, telmisartan use was associated with a lower risk of first osteoporotic fracture (adjusted HR, 0.71; 95% CI, 0.57-0.90). Compared with other ARBs, telmisartan use was associated with a lower risk of osteoporosis and osteoporotic fractures in this nationwide cohort. These findings suggest that clinically relevant heterogeneity may exist within the ARB class and that antihypertensive drug selection may influence skeletal outcomes beyond blood pressure control. Further prospective studies are warranted to confirm these findings.

Vitamin C may be important for maintaining skeletal muscle health and preventing sarcopenia, but most previous epidemiological studies have investigated dietary intakes rather than vitamin C status. To address this gap, this study investigated cross-sectional associations between both plasma and dietary vitamin C and muscle strength and physical performance. This study analysed data from over 5000 participants of the European Prospective Investigation into Cancer and Nutrition (EPIC)-Norfolk cohort (aged 48-92 years). Multivariable linear regression models, adjusted for lifestyle, biological, medical, and dietary covariates, were used to investigate associations between dietary (categorised as quintiles, Qs) and plasma vitamin C (low < 25μmol/L, suboptimal 25 to < 50μmol/L, optimal ≥ 50μmol/L), and handgrip strength (adjusted for height: HGSht), gait speed, and 5-chair-stand time. Least square means for outcome measures were calculated from regression models. Women with optimal versus low plasma vitamin C had higher HGSht (+ 0.54 kg/m[2], p = 0.009), and faster gait speed and chair-stand time (+ 0.11m/s, p < 0.001; - 1.03s, p = 0.028, respectively). Men with optimal versus low plasma vitamin C had faster gait speed and chair-stand time (+ 0.05m/s, p = 0.01; - 0.71s, p = 0.026, respectively). Compared with Q1 of dietary intake, women in Q2 and Q5 had faster gait speed (+ 0.04m/s, p < 0.01 for both); women in Q4 and Q5 had faster chair-stand times (- 0.67s, p = 0.002; - 0.57s, p = 0.013, respectively). No associations were found for dietary vitamin C in men. These findings suggest that maintaining optimal plasma vitamin C status could be important for maintaining physical function. Longitudinal and intervention studies are needed to support our findings.

To investigate whether latent epigenetic ageing is causally associated with bone mineral density using Mendelian randomization and to identify monocyte-specific genes potentially mediating this relationship. A latent epigenetic ageing factor was constructed from four DNA methylation clocks (GrimAge, Hannum, PhenoAge, and intrinsic epigenetic age acceleration) using Genomic structural equation modelling. Causal effects on heel estimated bone mineral density (eBMD; N = 426,824) were tested by two-sample Mendelian randomization (MR) and the pleiotropy-robust CAUSE framework. Monocyte-specific mediating genes were identified through single-cell PrediXcan with OneK1K prediction models, Bayesian colocalization, SMR-HEIDI, and causal transcriptome-wide association study fine-mapping. Functional consequences of prioritized variants were characterized using AlphaGenome, and key candidates were examined at the protein level in RANKL-induced THP-1 osteoclast differentiation as a first step toward validation in primary human monocyte-derived osteoclasts. Forward MR provided evidence consistent with a potential causal association between the latent ageing factor and eBMD (inverse-variance weighted β = 0.067, P = 3.6 × 10[-4]), with concordant sensitivity and CAUSE estimates; reverse MR was null. Single-cell PrediXcan revealed monocyte enrichment and an 11-gene pleiotropic core set. Colocalization and SMR-HEIDI converged on IFT80 (posterior probability of hypothesis 4 = 0.769) and TRIM59 (posterior probability of hypothesis 4 = 0.886) at chromosome 3. Causal transcriptome-wide association study fine-mapping nominated 45 high-confidence eBMD genes, independently confirming both IFT80 (PIP = 0.981) and TRIM59 (PIP = 0.936) among the top-ranked candidates. AlphaGenome highlighted regulatory perturbations at two chr3 variants. Western blot confirmed significant IFT80 downregulation and TRIM59 upregulation during osteoclast differentiation, consistent with computational predictions. Multi-layered genomic evidence provides convergent support for a putative causal association between latent epigenetic ageing and bone mineral density through monocyte-specific regulatory mechanisms, identifying the TRIM59-IFT80 locus and 45 candidate genes as priority targets for mechanistic validation.

Systemic sclerosis (SSc) is associated with increased bone fragility, yet early, compartment-specific alterations and their osteoimmunological correlates remain incompletely characterized. This study aimed to assess bone status in SSc using dual X-ray absorptiometry (DXA) complemented by peripheral quantitative computed tomography (pQCT), and to evaluate their associations with a targeted osteoimmunological biomarker panel and clinical risk scores. This prospective cross-sectional study included 40 women with SSc (disease duration ≤ 7 years from diagnosis) and 45 non-immune female controls, frequency-matched by age. Bone status was assessed by DXA (lumbar spine, femoral neck) and distal radius pQCT. Serum levels of routine bone and mineral metabolism markers, alongside a targeted panel of osteoimmunological mediators (OPG, OPN, PDGF-BB, ACP5, RANKL, TNF-α, IL-6, IL-1β, DKK-1) using multiplex immunoassay were measured. FRAX scores were calculated to contextualize clinical fracture risk. With broadly comparable DXA-defined bone status, pQCT revealed significantly lower trabecular volumetric bone mineral density (vBMD) in SSc, while total and cortical+subcortical vBMD were comparable. Routine biochemical markers showed substantial overlap between groups, whereas SSc patients exhibited a distinct osteoimmune profile with significantly elevated OPG, ACP5, RANKL, PDGF-BB, IL-6, IL-1β, DKK-1, and RANKL/OPG after FDR correction. Exploratory biomarker-bone associations did not remain significant after correction for multiple testing. In women with relatively early SSc, pQCT identified compartment-specific differences, particularly within the trabecular compartment, despite comparable DXA-defined bone status. A distinct osteoimmunological biomarker profile was also observed in SSc, providing complementary information on disease-related skeletal alterations, while specific biomarker-bone associations require further confirmation.

Bone mechanical competence depends not only on bone mass but also on the anisotropic orientation of the collagen/biological apatite (BAp) composite, yet its relationship to the local mechanical environment in human vertebrae remains unclear. Here, we examined whether region-specific BAp c-axis orientation corresponds to predominant principal stress direction estimated by µCT-based finite element analysis. The anterior region showed predominantly uniaxial alignment along the craniocaudal axis, corresponding to a predominant principal stress direction oriented primarily along the craniocaudal axis. In contrast, the cranial and caudal subchondral regions exhibited a two-dimensional in-plane orientation pattern corresponding to predominant principal stress directions within the anteroposterior-mediolateral plane, consistent with load transfer through the intervertebral disc, which imposes transverse in-plane loading on the endplate region. At the corticotrabecular junction, the preferred BAp orientation changed markedly within a spatial range of 100 μm or less. Overall, these findings demonstrate region-specific stress-orientation correspondence in human vertebrae and support crystallographic orientation as a quantitative bone-quality indicator of vertebral mechanical adaptation.

Methylmalonic acid (MMA) is a functional marker of vitamin B12 insufficiency and an emerging metabolic factor associated with skeletal aging and osteoporosis. This review synthesizes epidemiological, experimental, and mechanistic evidence linking the vitamin B12-MMA axis to impaired bone homeostasis. Population studies associate elevated circulating MMA with lower bone mineral density, higher odds of osteoporosis, and lower knee extensor strength, while disturbed vitamin B12 metabolism has also been linked to altered bone turnover. Experimental MMA exposure and models of methylmalonyl-CoA mutase (MMUT) deficiency reveal context-dependent abnormalities in mitochondrial respiration, energy metabolism, redox balance, and quality control. These findings provide a rationale for investigating skeletal effects, but the broader consequences of vitamin B12 deficiency and inherited metabolic defects should be distinguished from direct MMA actions in bone cells. Concurrent homocysteine accumulation and reduced S-adenosylmethionine availability may further impair collagen cross-linking, epigenetic regulation, and skeletal remodeling. These pathways support a metabolic-mitochondrial framework in which functional vitamin B12 deficiency contributes to bone loss through coordinated effects on bioenergetics, redox balance, cell fate, and extracellular matrix quality. MMA may therefore complement vitamin B12, homocysteine, renal function, and conventional bone markers in the assessment of skeletal metabolic risk. Prospective human studies, bone-specific mechanistic experiments, and intervention trials are needed to establish causality, clinically useful thresholds, and the skeletal benefits of correcting vitamin B12 deficiency or targeting mitochondrial dysfunction.

Osteoporosis is a systemic disease characterized by decreased bone density, and deterioration in the micro-architectural remodelling of bone tissue, leading to an increased susceptibility to bone fractures. One molecule implicated in bone metabolism is Programmed Death Ligand 1 (PD-L1), a pharmacologically targetable immune-checkpoint molecule encoded by the gene Cd274. The aim of this study was to investigate the bone phenotype of Cd274 knockout (KO) mice at 26 weeks of age compared to C57BL/6J control wildtype mice on radiological, histological and biomechanical level. The cortical wall thickness and cortical volume increased, and the trabecular space decreased significantly in both femur and tibia of Cd274 KO mice. Moreover, there was a significant increase in BV/TV and trabecular volume of lumbar vertebrae of Cd274 KO mice. Yield load, stiffness, anorganic amount increased and organic amount decreased significantly in femoral bones of Cd274 KO mice. Additionally, there was a significantly higher number of osteocytes and lower number of osteoclasts in both humerus and lumbar vertebra of Cd274 KO mice. The findings demonstrate that constitutive knockout of Cd274 was associated with significantly increased bone mass and strength. These results-obtained in healthy, skeletally mature constitutive knockout mice-indicate an improved bone phenotype in the absence of Cd274 and thus contrast with previous studies that associate disruption of the PD-1/PD-L1 axis, including immune checkpoint inhibitor therapy, with bone loss and increased fracture risk. Further studies, particularly in dedicated osteoporosis models, are required to elucidate the influence of CD274 on bone metabolism before translational conclusions can be drawn.

Hip structural analysis parameters (HSAP) are associated with hip fracture risk, but associations with multiple fragility fracture phenotypes remain poorly characterised. We analysed 11 DXA-derived HSAP in patients referred between June 2004 and November 2019. Five mutually exclusive fracture phenotypes were defined: no fracture, single hip fracture, single non-hip fracture, multiple non-hip fractures, and multiple fractures including at least one hip fracture. Separate multinomial logistic regression models were fitted for each HSAP reported as odds ratio (OR) per standard deviation increase, adjusted for FRAX covariates and left femoral BMD. Benjamini-Hochberg false discovery rate correction was applied across all tests. A right-hip sensitivity analysis assessed bilateral consistency. 35,647 patients were included (mean age 64.5 years, 82% female). Of 13,777 (38.6%) with a fragility fracture, 1044 (2.9%) had a single hip fracture, 9723 (27.3%) a single non-hip fracture, 2268 (6.4%) multiple non-hip fractures, and 742 (2.1%) multiple fractures including at least one hip fracture. After multiple testing correction, no meaningful associations were observed for isolated hip or non-hip fracture phenotypes. Multiple fractures including at least one hip fracture were independently associated with greater hip axis length (OR 1.44, 95% CI 1.28 to 1.63), neck diameter (OR 1.24, 95% CI 1.11 to 1.38), and centre of mass to upper neck margin distance (OR 1.35, 95% CI 1.21 to 1.51). HSAP were significantly associated with multiple fractures including at least one hip fracture rather than isolated fracture phenotypes, potentially reflecting local structural vulnerability and broader biomechanical instability.

Computer vision is a rapidly evolving field within computer science that focuses on extracting meaningful information from digital images. In medical imaging, computer vision techniques such as object detection, classification, and segmentation are essential for analysing complex anatomical structures. This primer provides an overview of computer vision concepts and methods, highlighting both traditional approaches and modern machine learning techniques. Landmark detection and shape modelling are also discussed as these are particularly relevant for musculoskeletal imaging. Challenges associated with applying computer vision in clinical practice are examined. This includes considerations such as data limitations, domain shift, and the need for robust validation. Finally, future directions for integrating computer vision into osteoporosis research and clinical workflows are explored, emphasising the importance of explainability, fairness, and adherence to guiding principles for trustworthy clinical tools.

Chronic hypoparathyroidism (HypoPT) may increase cardiovascular risk, yet current care rarely includes structured cardiovascular assessment. This systematic review and meta‑analysis aimed to quantify the association of chronic HypoPT with major cardiovascular events. Electronic databases and registries were searched from their inception through 10 June 2026 to identify controlled cohort studies of adults with chronic HypoPT and non-HypoPT comparators that reported cardiovascular outcomes. Random‑effects meta‑analyses (for hazard ratios [HRs]) were conducted for the primary and secondary outcomes using R. Data from nine cohort studies (N=134, 598; HypoPT: 28, 625; controls: 105, 973) were analyzed. HypoPT was associated with higher risk of myocardial infarction (MI) (HR1.19, 95%CI1.06, 1.35; I[2]=0%). HypoPT was associated with higher occurrence of heart failure (HF) [HR 1.46(95%CI: 1.08, 1.96); I[2]=83.5%; P=0.012] which lost its significance on prediction-interval analysis (PIA) (95%CI: 0.62, 3.42). HypoPT had higher occurrence of arrhythmias [HR1.45(95%CI: 1.19, 1.76); I[2]=64.9%; P=0.0002] which lost its significance on PIA (95%CI: 0.87, 2.43). HypoPT had higher occurrence of stroke [HR 1.24(95% CI: 1.05, 1.46); I[2]=46.9%; P =0.009] which lost its significance on PIA (95%CI: 0.85, 1.81). HypoPT had higher all-cause mortality [HR1.40(95%CI: 1.07, 1.84); I[2]=80.9%; P = 0.015] which lost its significance on PIA (95% CI: 0.62, 3.15). Chronic HypoPT was associated with a higher pooled hazard of MI. Pooled associations with HF, arrhythmias, stroke, composite cardiovascular disease, CKD, and all-cause mortality were imprecise and heterogeneous, with prediction-intervals that included the null.