Metabolomica: un nuovo approccio per lo studio della predittività di farmacoresistenza e di complicanze cardiovascolari in epilessia

About 30% of epileptic subjects is “drug resistant” (DR) to mono and multiple treatments. The risk of sudden unexpected death (SUDEP) is more than 20 times higher in epileptic patients than that in the general population; SUDEP represents the main cause of death in patients with refractory epilepsy. Several mechanisms probably exist, but most research has focused on seizure related cardiac arrhythmia. No evidence-based intervention to prevent SUDEP exists. Biological basis of DR are still unclear and applying innovative diagnostic techniques for prediction of outcome in the pharmacotherapy of epilepsy is a mission with important social and economical aspects. Genomic and proteomic are not able to depict the holistic approach that is fundamental for a complete description of phenotiping in human physiology. Metabolomic responses, as integration of genomic and proteomic expressions with the environmental solicitations, can give us the possibility to investigate about complex interactions in the metabolic networks induced by pathologies or by drugs. Data presented in this paper are 1H-NMR spectroscopy measurements performed by our team in our Clinical Metabolomic Labs, University of Cagliari, on blood samples of epileptic patients, previously classified by us as drug-responders or not. Using supervised models we are extracting a metabolic characterization, higher in specificity and sensibility with increasing of number of patient, to realize a profile of drug resistance in epilepsy. In this study we present a PLS_DA model. This is a preliminary model of metabolic alterations in our cohort of DR patients, which could represent a new and non-invasive method for early detection of epileptic patients at risk of sudden death and simultaneously help to clarify the pathophysiological mechanisms of DR and SUDEP.