An integrated approach to predict disease activity in the early phases of Multiple Sclerosis
In this project FRRB finances the italian Coordinator: IRCCS Ospedale San Raffaele, Milano. The Principal Investigator responsible of the project is Dr. Federica Esposito.
|
Pathology of interest: |
Multiple Sclerosis |
|
Area of research: |
Neurological diseases |
|
Start date: |
June 2019 |
|
End date: |
June 2022 |
|
Funding: |
€ 499.959,00 |
|
Project partners: |
IRCCS Ospedale San Raffele – leading partner Centre Hospitalier Universitaire de Toulouse (CHUT) National Research Council - Institute of Biomedical Technologies GeneXplain GmbH (GXP) |
PROJECT SUMMARY
Multiple Sclerosis (MS) is an autoimmune disorder of the central nervous system characterized by inflammation, demyelination and axonal degeneration. It is a disabling disorder affecting more than 2 million people worldwide with a high socio-economic impact. Given its marked heterogeneity, including clinical manifestations and individual treatment response, MS is a typical condition where a more personalized intervention would be highly beneficial.
Our hypothesis is that a comprehensive characterization of a large set of patients, integrating multi-layer data (clinical, molecular, environmental), could contribute to accelerate personalized medicine in MS through the identification of biomarkers of inflammatory activity, and the development of network-based and artificial intelligence approaches able to predict disease activity and to support treatment choice in the early phases of the disease.
Taking advantage of an already available well-characterized cohort of MS patients (>4,500), we will assess genetic and environmental factors associated with disease activity. A subset of 300 patients will be extensively studied at the molecular level by performing a comprehensive “omics” profiling covering transcriptome, miRNome and methylome, together with vitamin D measurements. Omics data will be integrated using genome-scale biological networks to yield modules of altered genes and unravel the molecular mechanisms underlying disease activity. We will then design a predictive algorithm of disease activity based on Deep Learning models suitable for personalized medicine applications in clinical practice.
The findings from the present project, besides clarifying the master-regulators of MS heterogeneity, should contribute to guide a more tailored use of currently available drugs. This more personalized MS management will allow to improve quality of life and to slow disability progression, in turn reducing health system costs.