Psychotic symptoms occurring in Alzheimer's disease (AD + psychosis, AD + P) are a marker for a more rapidly deteriorating phenotype. We have developed a polygenic model of AD + P risk, conditioned on the presence of AD. Whether risk genes for AD itself contribute to AD + P risk is not established, although our model predicts they will not. The most important identified genetic determinant of sporadic, late-onset AD is the apolipoprotein E epsilon 4 allele (APOE4). The effect of APOE4 on AD phenotype is to reduce the age of onset of AD. Prior studies examining the association of APOE4 with AD + P have reported conflicting results. However, no prior studies have examined if APOE4 reduces time to onset of psychosis in AD. The objective of this study was to examine the effect of APOE4 and alpha1-antichymotrypsin/AA (ACT/AA) genotypes on time to psychosis onset in subjects with AD. A longitudinal study of psychosis incidence in 316 subjects with AD with no history of current or prior psychotic symptoms at entry was undertaken. APOE and ACT genotyping was conducted per established protocols. Data were analyzed by survival analysis and Cox proportional hazards models. There were no significant associations of APOE or ACT genotypes with time to psychosis onset and no significant interaction of these genotypes with time to psychosis onset. There remained no significant associations after covarying for age, age of AD onset, degree of cognitive impairment, gender, race, and education. This is the first study to examine the genetic prediction of psychosis onset in AD. The findings support the hypothesis that these two genetic determinants of AD risk do not contribute to the risk of development of psychotic symptoms in AD.