Change in Cognitive Functioning Associated With ApoE Genotype in a Community Sample of Older Adults Scott M. Hofer Pennsylvania State University Helen Christensen Australian National University Andrew J. Mackinnon Mental Health Research Institute of Victoria and Monash University Ailsa E. Korten, Anthony F. Jorm, Alexander S. Henderson, and Simon Easteal Australian National University The influence of a genetic risk factor, apolipoprotein E (apoE) 4 variant, was assessed in older adults aged 70 to 94 on 3 occasions over 7 years. The results of latent growth curve analyses are reported for individuals genotyped for apoE at the 2nd measurement occasion (n = 601) and for a subsample of individuals without probable or definite dementia during the 1st or 2nd occasion (n = 434). ApoE-4 status was a significant predictor of level and change in memory performance and change in speed performance in the full sample, and of initial level and change in memory performance in the nonde- mented subsample. These results support previous findings that apoE-4 is associated with accelerated memory deterioration in individuals without clinical dementia. Apolipoprotein E (apoE) is a plasma protein involved in the metabolism of lipoproteins. There are three common allelic forms of apoE: 2, 3, and 4. he apoE-4 allele, which is present in approximately 15% of Caucasian populations, has been strongly identified as a risk factor for the development of Alzheimer’s disease (AD). There is an increased risk for AD in the “young-old” rather than in the “old-old” (aged 85 and older; Farrer et al., 1997), and there is evidence for increased progression of dementia in individuals homozygous for the 4 allele (Craft et al., 1998). The apoE-4 allele has also been associated with accelerated cognitive decline, decreased longevity, and increased cardiovascular disease (J. D. Smith, 2000). The apoE-2 allele has been found to be weakly protective of dementia and cognitive impairment. How- ever, there is some question (e.g., Small, Basun, & Backman, 1998) as to whether apoE-4 is a genetic risk factor for “normal” cognitive aging, or whether the apoE-4 association is due to preclinical dementia in older samples. Evaluation of this question requires relatively long-term studies that permit retrospective anal- ysis of change in functioning within individuals not diagnosed with dementia (Sliwinski, Lipton, Buschke, & Stewart, 1996). A recent review (Anstey & Christensen, 2000) identified studies that examined the relationship between cognitive change and apoE-4 and apoE-2 allelic variants. With respect to apoE-4, five studies found that apoE-4 predicted cognitive decline (Feskens et al., 1994; Haan, Shemanski, Jagust, Manolio, & Kuller, 1999; Hyman et al., 1996; Jonker, Schmand, Lindeboom, Havekes, & Launer, 1998; Yaffe, Cauley, Sands, & Browner, 1997), three studies reported that the effect was present but ob- served only on certain tests (Henderson, Easteal, et al., 1994; O’Hara, Yesavage, Kraemer, Mauricio, Friedman, & Murphy, 1998; Small et al., 1998), and two studies reported that the rate of change did not differ (Brayne et al., 1996; Helkala et al., 1996). Several additional apoE-4 studies have recently been reported. One study indicated greater memory decline in apoE-4 individ- uals (Carmelli et al., 2000), and another that individuals without the 4 allele maintained high cognitive status longer (Riley et al., 2000). Additionally, one study suggested that the cognitive decline/apoE-4 association is present only in those with Mini- Mental State Examination (MMSE; Folstein, Folstein & McHugh, 1975) scores below 27 (Dik et al., 2000), and another demonstrated no association in a very old sample (Juva et al., 2000). Staehelin, Perrig-Chiello, Mitrache, Miserez, and Perrig (1999) found that healthy older individuals over 65 years with either 4/4 or 3/4 genotypes performed lowest on tests of processing speed, delayed free recall, and semantic long-term memory but experienced no significant differences in change over 2 years. While the findings have been somewhat mixed, perhaps because of the relatively small effect size related to the apoE genotype, a majority of these findings support the view that the average rate of Scott M. Hofer, Department of Human Development and Family Stud- ies, Pennsylvania State University; Helen Christensen, Ailsa E. Korten, Anthony F. Jorm, and Alexander S. Henderson, Centre for Mental Health Research, Australian National University, Canberra, Australian Capital Territory, Australia; Andrew J. Mackinnon, Mental Health Research Insti- tute of Victoria, Parkville, Victoria, Australia and Department of Psycho- logical Medicine, Monash University, Melbourne, Victoria, Australia; Si- mon Easteal, Human Genetics Group, John Curtin School of Medical Research, Australian National University. This study was supported by National Health and Medical Research Council Grant 973302 and by the Australian Rotary Health Research Fund. This work was partly completed while Scott M. Hofer was a visiting research fellow (2000 –2002) at the Centre for Mental Health Research, Australian National University. Correspondence concerning this article should be addressed to Scott M. Hofer, Department of Human Development and Family Studies, 110 South Henderson Building, Pennsylvania State University, University Park, Pennsylvania 16802, or to Helen Christensen, Centre for Mental Health Research, Australian National University, Canberra, Australian Capital Territory, 0200 Australia. E-mail: smh21@psu.edu or helen.christensen@ anu.edu.au Psychology and Aging Copyright 2002 by the American Psychological Association, Inc. 2002, Vol. 17, No. 2, 194 –208 0882-7974/02/$5.00 DOI: 10.1037//0882-7974.17.2.194 194