Dissecting the genetic architecture of knee alignment reveals its contribution to osteoarthritis risk

To investigate the biological and clinical relevance of knee alignment in osteoarthritis by integrating population-scale imaging, genome-wide association, and functional genetic analyses.

Femorotibial angle was derived from dual-energy X-ray absorptiometry scans in UK Biobank using machine-learning methods. Associations with knee and hip osteoarthritis outcomes were assessed. A genome-wide association study of mean femorotibial angle was performed, followed by fine-mapping and pathway enrichment analyses. Mendelian randomization was used to explore potential causal relationships.

Varus alignment was strongly and progressively associated with knee pain, knee osteoarthritis, and total knee replacement (HR 3.42 [95% CI 2.92, 4.02]), with no association observed for hip osteoarthritis. GWAS identified 20 independent loci associated with femorotibial angle, enriched for pathways related to skeletal development, cartilage biology, and endochondral ossification. Post-GWAS analyses demonstrated regulatory effects across fetal and adult joint tissues, supporting life course influences on alignment. Genetic correlation analyses showed shared architecture between femorotibial angle and knee osteoarthritis. Causal analyses suggested that genetic liability to osteoarthritis reduces femorotibial angle (β -0.11 [-0.16, -0.06]), while evidence for an overall causal effect of femorotibial angle on osteoarthritis risk was limited (OR 0.93 [0.79, 1.10]).

Knee alignment and susceptibility to knee osteoarthritis are partially genetically determined. At the population level, these genetic determinants support a causal effect of osteoarthritis on knee alignment, whereas evidence for a causal effect of alignment on knee osteoarthritis was limited. Furthermore, this study identifies novel genetic loci linking knee alignment with pathways involved in skeletal development and cartilage biology relevant to osteoarthritis.