Integrative genomics identifies putative causal genes and therapeutic targets in thoracic aortic disease

Background Thoracic aortic dissection (TAD) is a life-threatening condition with high mortality if left untreated. A key risk factor for TAD is thoracic aortic aneurysm (TAA), defined by pathological dilation of the thoracic aorta. Current clinical guidelines rely on imaging and recommend prophylactic surgery once a critical aortic diameter is reached. However, no pharmacological therapies are available to slow TAA progression or prevent TAD. Improving risk stratification and enabling therapeutic development requires a deeper understanding of the genetic and molecular mechanisms underlying both TAD and TAA (TAA/D). Purpose We aim to improve the identification of candidate genes underlying TAA/D by integrating a large-scale GWAS meta-analysis with multi-tissue molecular trait data, and subsequently evaluate the druggability of the candidate genes to find potential therapeutic targets. Methods We conducted a GWAS meta-analysis of five independent TAA/D cohorts (14,409 cases). Putative causal genes were identified using multi-group causal transcriptome-wide association studies (M-cTWAS), integrating GWAS results with multi-omic data (gene expression, splicing, and protein abundance) from vascular tissues and blood plasma. To assess directionality and robustness of gene-trait associations, we performed Mendelian randomization (MR) analyses. Therapeutic potential of genes was evaluated using DrugnomeAI, a machine learning framework that estimates druggability, and the Open Targets Platform for clinical level evidence. Results By integrating GWAS and molecular trait data, we have identified 28 putative causal genes. Of these genes, 15 lacked a genome-wide significant variant within ±500kb of the gene in the individual cohort GWAS summary statistics. The identified genes included both established risk genes (e.g. TSHD4, COL6A3, CDH13) and novel candidates (e.g. MRC2, CRIM1). MR analyses supported the directionality of effect. Notably, increased expression of novel candidate gene MRC2, a regulator of collagen degradation, was associated with higher TAA/D risk. Drug-target profiling identified several genes with high druggability scores, including LRP1 and COL6A3, which are in advanced clinical development or have an approved drug, respectively. Conclusions Integration of large-scale GWAS data and multi-tissue molecular traits identified both established and novel candidate genes for TAA/D, and demonstrated the added value of molecular trait integration. By evaluating directionality and druggability, we highlight potential therapeutic targets and provide a framework for future functional studies.