Alzheimer's disease (AD) is a complex neurodegenerative dementing illness. Over the past few years, however, remarkable advances have taken place in understanding both the genetic and molecular biology with the intracellular processing of amyloid and tau and the changes leading to the pathologic formation of extracellular amyloid plaques and the intraneuronal aggregation of hyperphosphorylated tau into neurofibrillary tangles. This progress in our understanding of the molecular pathology has set the stage for clinically meaningful advances in the development of biomarkers. Emerging diagnostic methods that are based on biochemical and imaging biomarkers of disease specific pathology hold the potential to provide effective measures of natural history (marker of disease that is predictive of outcome), biological activity (such as magnitude and frequency of response correlating with drug potency) and markers of surrogate endpoints (single or composite marker that accounts for clinical benefit of the therapy). Markers of biological activity should be also evaluated regarding their value to reflect disease progression, heterogeneity of the clinical population, for early decision making and characterization of new treatments. We focussed on the current status of core analytes which provide reasonable evidence for association with key mechanisms of pathogenesis or neurodegeneration in AD. In addition, feasibility was important, such as availability of a validated assay for the biological measure in question, with properties that included high precision and reliability of measurement, reagents and standards well described. On this basis we reviewed the body of literature that has examined CSF total tau (t-tau) and beta-amyloid 1-42 (Abeta(1-42)), phosphorylated tau (p-tau) and beta-amyloid-antibodies as diagnostic tests for AD versus clinically representative comparison groups. Measurement of t-tau and Abeta(1-42) in the CSF seems useful to discriminate early and incipient AD from age-associated memory-impairment, depression, and some secondary dementias. First studies showed that measurement of p-tau proteins significantly improves early and differential diagnosis, as well as disease prediction in subjects at risk for AD and comes closest to fulfilling proposed criteria of a biological marker for AD. However, the nature of the majority of reported findings are still preliminary and retrospective. General issues for biomarkers have to be adequately addressed, such as sensitivity of the method, frequency of assessments, stability of the method, standardization of methods and dynamic range. There is still a partial lack of comparison patient populations that must be addressed in future studies. International dementia networks have been recently established to advance the establishment of core biomarker candidates of AD as potential surrogate endpoints for clinical trials and their clinical use for predictive and diagnostic purposes.