Based on our current work, we have found aberrations of both cellular immunity and humoral immunity in patients with Alzheimer's disease (AD), suggesting that the body's immune system plays an important role in the etiology and/or pathophysiology of at least one subset of AD (we consider AD as a "syndrome," not a disease). Depending upon the nature of the immune deficits and patients' responsiveness to appropriate irnmunomodulate therapy, we have thus far differentiated four subsets of AD patients: one subset with defect of a specific T cell (e.g., membrane flexibility) and response to therapy with pyrrolidone compounds; a second subset with autoantibodies to neuron-axon filament proteinsthese patients show clinical improvement after therapy with dialyzable leucocyte extract (DLE) containing transfer factor (TF); a third subset with antibodies to brain antigens (autoimmune) for which therapy is not yet developed; and a fourth subset with none of the abnormalities mentioned above, probably heterogeneous due to one or another biochemical abnormality. We believe that different therapeutic modalities will be necessary for different subsets, much like the situation with other "diseases" such as anemia and diabetes. Our therapeutic results provide additional evidence that AD is a syndrome, not a single disease. Moreover, the clinical improvement demonstrates that the defective function in AD is not due to death of neuronal cells, but rather atrophy or physiologic "suppression."