Alzheimer's disease (AD) is the most common form of dementia. Its most important pathological hallmarks are profound neuronal loss, presence of intracellular neurofibrillary tangles, and extracellular deposition of beta-amyloid protein (Aβ) as beta-amyloid plaques. These latter aggregations result in neuronal degeneration in brain regions important not only for memory, but also for other cognitive functions. One of the most important risk factors for AD is age and with the increase of life-expectancy AD has thus become the most common form of dementia. It is now formally recognized by several new diagnostic criteria that AD is not a homogeneous disease. The current "Holy Grail" is to be able to diagnose variants of AD before they manifest clinically and before irreparable brain damage is done. To achieve this goal, robust and reliable biomarkers that reflect the pathogenesis of AD have to be implemented. This is of paramount importance because such biomarkers may provide clues to pathways that can be targeted for interventions aimed at disease prevention or improvement. Although much attention has focused on Aβ as a major component of AD, Aβ may be a less attractive target since an increasing amount of data has raised concerns about its causative role in AD. This review will be in two parts, this first part will deal with the current clinical knowledge and the questions raised by the Aβ cascade hypothesis in the pathogenesis of AD and the second part will aim to synthesize our current knowledge and to discuss new data that suggest how immune alterations may contribute to the development of AD and may therefore provide beneficial targets in novel approaches for the treatment of AD.