Alzheimer’s disease (AD) is currently the leading cause of dementia, being expected to double in the coming decades. Increasing evidence suggests the pathological mechanisms of AD become active several years before neurons start dying and dementia manifests. During this prodromal stage, referred to as Mild Cognitive Impairment (MCI), effective treatments would have the greatest impact because cognitive function is still relatively preserved. Hence, it is crucial to identify these subjects at increased risk of developing AD at the earliest possible stage. Several neuroimaging, cerebrospinal fluid and neuropsychological biomarkers for early detection of AD have been proposed. However, none of these biomarkers has been established as an ideal diagnostic or prognostic indicator. Considering the complexity of the AD process and the multiple dynamic pathological changes that occur during the course of the disease it is rather unlikely that a single ideal biomarker even exists. Hence a combination of both already established and new sensitive biomarkers has the potential of significantly improving the accuracy of diagnosis compared to each of them alone. In the current project we will combine several multimodal biomarkers from neuroimaging, cerebrospinal fluid, genetics and cognitive tests in multivariate analyses in order to determine which marker or combination of markers is optimal for early diagnosis of AD. We will create a discriminant pattern of disease for the diagnosis of AD, prediction of conversion of MCI to AD and discrimination of MCI due to AD pathology and MCI associated with Parkinson’s disease. These patterns will be based on information from well-established measures that have shown value as AD biomarkers, in addition to new measures suggested as potentially new markers of AD. This project will benefit the Work Program by providing tools able to diagnose and discriminate early prodromal stages of AD that can be potentially implemented in clinical practice.