Bone and Mineral Metabolism

MON-748 - Title: Data-Driven Genetic Clustering of Bone Mineral Density Loci Reveals Heterogeneous Mechanistic Pathways Associated with Osteoporosis

Monday, June 15, 2026

9:00 AM - 2:00 PM CT

Location: ENDOExpo; Poster Floor

Abstract Poster Presenter(s)

Nipith Charoenngam, MD

Watertown, MA, United States

Co-Author(s)

KS

Kirk Smith, MS

Department of Medicine, Harvard Medical School
Canbridge, Massachusetts, United States
MS

Magdalena Sevilla-Gonzalez, PhD

Massachusetts General Hospital
Boston, Massachusetts, United States
SH

Sarah Hsu, MS

Diabetes Unit and Center for Genomic Medicine, Massachusetts General Hospital
Boston, Massachusetts, United States
HL

Haojie Lu, n/a

Department of Internal Medicine, Erasmus MC University Medical Center
Rotterdam, Zuid-Holland, Netherlands
TL

Tianyuan Lu, PhD

Department of Population Health Sciences, School of Medicine and Public Health, University of Wisconsin-Madison
Madison, Wisconsin, United States
JK

John P. Kemp, PhD

Group Leader
Mater Research Institute, The University of Queensland, Translational Research Institute, Woolloongabba, Queensland, Australia
Woolloongabba, Queensland, Australia
DK

Douglas P. Kiel, MD

Hinda and Arthur Marcus Institute for Aging Research, Hebrew SeniorLife; Beth Israel Deaconess Medical Center and Harvard Medical School
Boston, Massachusetts, United States
Gc

GEFOS-CHARGE consortium, n/a

http://www.gefos.org/
Rotterdam, Zuid-Holland, Netherlands

Senior Author/Primary Investigator(s)

MU

Miriam S. Udler, MD, PhD

Massachusetts General Hospital
Boston, Massachusetts, United States
Michael Mannstadt, MD (he/him/his)

Chief, Endocrine Unit
Massachusetts General Hospital and Harvard Medical School
Boston, Massachusetts, United States

Disclosure(s):

Nipith Charoenngam, MD: No financial relationships to disclose

Michael Mannstadt, MD: Alexion Pharmaceutical: Advisory Board (), Clinical Trial Investigator (); BridgeBio: Advisory Board (), Clinical Trial Investigator (); MBX: Advisory Board (); septerna: Consultant (); Shire Pharmaceuticals/Takeda: Advisory Board (), Clinical Trial Investigator (), Consultant ()

Genome-wide association studies (GWAS) have established bone mineral density (BMD) as a highly polygenic trait, but the mechanisms underlying most loci remain unclear. To identify distinct mechanisms, we clustered BMD loci based on their associations with other phenotypes.

We extracted genome-wide significant, LD-independent lead loci for lumbar spine and femoral neck BMD from an unpublished European-ancestry GEFOS/UK Biobank (UKB) DXA-BMD meta-analysis (N = 157,441). We aligned these loci with UKB GWAS summary statistics for diverse traits, including body composition and laboratory measures. We constructed a 347-variant × 88-trait association matrix and applied Bayesian non-negative matrix factorization (bNMF) to derive genetic clusters of BMD loci. To quantify cluster-phenotype associations, we constructed cluster-specific partitioned polygenic scores (pPS) and tested each pPS for association with phenotypes in multi-ancestry samples from UKB and Mass General Brigham Biobank (MGBB). We applied epigenetic annotations to connect clusters to specific tissues.

Using bNMF, we identified 10 clusters of low-BMD loci, with all variants aligned to the BMD-lowering allele. They included clusters related to calcium metabolism, body mass index (BMI)/adrenal androgen, adiposity, sex hormone-binding globulin, hematopoiesis and five others with unclear underlying mechanisms. Notably, pPS of the calcium metabolism cluster containing 7 loci was associated with higher serum calcium, lower phosphate and higher parathyroid hormone. Annotation with parathyroid Hi-C chromatin interaction data showed that the top two loci mapped to enhancer elements of the calcium-sensing receptor (CASR) gene that interact with the CASR promoter. Another locus mapped to the DGKD gene, linked to kidney stone risk by affecting CASR signaling. These findings suggest a pathway through which CASR-related common regulatory variants influence BMD.

When assessing all low-BMD loci in aggregate, their combined genetic risk score was associated with lower BMI in UKB, consistent with low BMI as an osteoporosis risk. In contrast, a cluster of 29 low-BMD loci showed a surprising association with higher BMI and multiple adiposity measures in UKB, and higher BMI and type 2 diabetes risk in MGBB. This cluster includes loci near metabolic genes such as PPARG, ADCY3, RSPO3 and CCND2. ChromHMM analysis showed preferential enrichment of these loci in genic, enhancer and promoter regions in adipocytes and adipose-related mesenchymal stem cells.

In summary, using a data-driven genetic approach, we identified 10 clusters of BMD loci with shared multi-trait association patterns suggesting distinct underlying physiological processes. Integration of epigenetic annotations demonstrated tissue specificity, supporting the interpretations of genetic mechanisms and highlighting calcium and adipose-related biology.

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*