The use of biomarkers is growing in the early detection of Alzheimer's disease (AD). Although some biomarkers such as medial temporal lobe volumetry, amyloid Positron Emision Tomography (PET), and Aβ42 in CSF are being widely used, there is no clear consensus about the best biomarker to be used in each phase of the disease. Magnetic Resonance Spectroscopy (MRS) of the brain is less known as biomarker but has proven useful according to cross-sectional and longitudinal studies. This technique measures metabolite levels that reflect the degree of pathology in the brain. N-acetyl aspartate (NAA), a marker of neuronal density, decreases and Myo-inositol, a marker of glial proliferation, increases as the disease progresses. Decreased NAA levels have been detected in the prodromal phases of AD and even in presymptomatic stages in carriers of tau and amyloid protein mutations. Longitudinal studies have demonstrated good correlation between NAA levels and progression of AD, even in spite of treatment with cholinesterase inhibitors. From clinical trials we have learned that the current therapies have a modest effect on AD progression and that they do not have neuroprotective effects. This modest effect is reflected in the modest or null changes in metabolite levels in clinical trials using MRS as biomarker.