In Alzheimer's disease (AD) the accumulation of pathological forms of the beta-amyloid (Abeta) peptide are believed to be causal factors in the neurodegeneration that results in the loss of cognitive function in patients. Anti-Abeta antibodies have been shown to reduce Abeta levels in transgenic mouse models of AD and in AN-1792 clinical trial on AD patients; however, the clinical trial was halted when some patients developed meningoencephalitis. Theories on the cause of the adverse events include proinflammatory "primed patients," a Th1-inducing adjuvant, and Abeta autoreactive T cells. New immunotherapy approaches are being developed to eliminate these putative risk factors. Mannan, which is recognized by pattern recognition receptors of the innate immune system, can be utilized as a molecular adjuvant to promote a Th2-mediated immune response to conjugated B cell epitopes. The N-terminus of Abeta was conjugated to mannan, and used to immunize mice with low concentrations of immunoconjugate, without a conventional adjuvant. Mannan induced a significant and highly polarized toward Th2 phenotype anti-Abeta antibody response not only in BALB/c, but also in B6SJL F1 mice. New preclinical trials in AD mouse models may help to develop novel immunogen-adjuvant configurations with the potential to avoid the adverse immune response that occurred in the first clinical trial.