Multiple sclerosis is an idiopathic inflammatory disease of the central nervous system and the second most common cause of disability in young adults. Choosing an effective treatment is crucial to preventing disability. However, response to treatment varies greatly between patients. Because of this, accurate and timely detection of individual response to treatment is an essential requisite of efficient personalised multiple sclerosis therapy. Nowadays, there is a lack of comprehensive predictive models of response to individual treatment.This paper arises from the clinical need to improve this situation. To achieve it, all patient’s information was used to evaluate the effectiveness of demographic, clinical and paraclinical variables of individual response to fourteen disease-modifying therapies in MSBase, an international cohort. A personalized prediction model to three stages of disease, as a support tool in clinical decision making for each MS patient, was developed applying machine learning and Big Data techniques. These techniques were also used to reduce the data set and define a minimum set of characteristics for each patient. Best predictors for the response to treatment were identified to refine the predictive model. Fourteen relevant variables were selected. A web application was implemented to be used to support the specialist neurologist in real time. This tool provides a prediction of progression in EDSS from the last relapse of an individual patient, and a report for the medical expert.