HEALTHCARE PROBLEM
Intracranial aneurysms (IAs)—balloon-like bulges in the walls of brain arteries—are common, affecting approximately 3% of the population, or 10 million adults in the EU. While most aneurysms remain asymptomatic, rupture can cause aneurysmal subarachnoid hemorrhage (ASAH), a severe form of stroke. One-third of ASAH patients die, while another third suffer severe disability, requiring lifelong care. ASAH typically occurs around age 50—significantly younger than most other cardiovascular diseases—impacting patients during their working years and placing a heavy burden on families and healthcare systems.
Early detection of aneurysms is possible through imaging, which can enable preventive treatment via endovascular intervention or surgery. However, current screening strategies are inadequate due to a limited understanding of who is at risk. The pathogenesis of IAs and ASAH remains poorly understood, as only a few genetic, environmental, and imaging risk factors have been identified. Furthermore, the interplay between these factors remains unclear, partly due to the lack of large, well-characterized cohorts for adequately powered studies. Notably, the disease exhibits significant heterogeneity, with sex-related differences being particularly striking—two-thirds of ASAH patients are women. However, the reasons behind this increased risk in women remain unknown.
OVERALL OBJECTIVE
The PRYSM (Early recognition of intracranial aneurysms to prevent aneurysmal subarachnoid hemorrhage) project aimed to enhance disease understanding to enable early recognition and prevention of ASAH.
APPROACH
PRYSM introduced a novel, integrated approach combining genetic, clinical, and imaging risk factors, assessed using machine learning. Machine learning—an AI-driven computational method—analyzes data patterns to improve predictive accuracy. Through this approach, imaging markers, genetic variants, including those with roles in arterial wall structure and function, and environmental risk factors were identified. Additionally, interactions between female sex and specific risk factors contributing to the disease were uncovered, and sex-specific prediction models improved predictive accuracy for disease prediction.
CONCLUSION
The combined results of PRYSM advance our understanding of IA development and represent an important step toward individualized risk prediction and precision medicine in early IA detection. By improving early identification of high-risk individuals, these findings contribute to reducing the burden of ASAH and improving patient outcomes.
