|Molecular Vision 2000;
Received 9 October 2000 | Accepted 29 December 2000 | Published 31 December 2000
Association of the Apolipoprotein E gene with age-related macular degeneration: Possible effect modification by family history, age, and gender
Silke Schmidt,1,2 Ann
M. Saunders,1,3 Monica A. De La Paz,1,4
Eric A. Postel,1,4 Ruth M.
Heinis,1,4 Anita Agarwal,5 William K. Scott,1,2 John
R. Gilbert,1,2 Julie G. McDowell,1,2 Amy Bazyk,6 J.
Donald M. Gass,5 Jonathan
L. Haines,6 Margaret A.
1Department of Medicine, 2Center for Human Genetics, 3Division of Neurology, and 4Duke University Eye Center, Duke University Medical Center, Durham, NC; 5Department of Ophthalmology and Visual Sciences and 6Program in Human Genetics, Vanderbilt University Medical Center, Nashville, TN
Correspondence to: Margaret Pericak-Vance, Ph.D., Center for Human Genetics, Department of Medicine, Duke University Medical Center, Box 3445, Durham, NC, 27710; Phone: (919) 684-3422; FAX: (919) 684-2275; email: email@example.com
Purpose: Age-related macular degeneration (AMD) is a complex disorder affecting older adults in which genetic factors are likely to play a role. It has been previously suggested that the e4 allele of the apolipoprotein E (APOE) gene may have a protective effect on AMD risk and that the e2 allele may increase disease risk. The purpose of our study was to examine whether an independent data set would support the proposed role of APOE in AMD etiology.
Methods: We compared AMD cases (n=230) to controls (n=372) with respect to APOE genotypes using c2 tests and logistic regression analysis. We also conducted separate analyses for familial (n=129) and sporadic (n=101) AMD cases since these groups may have a different disease etiology.
Results: We did not find evidence for the risk-increasing effect attributed to the e2 allele in either familial or sporadic AMD. No evidence for a protective effect of the e4 allele was obtained for sporadic AMD. The age- and sex-adjusted odds ratio (OR) for e4 carriers among familial AMD cases compared to controls was 0.66 (95% confidence interval: 0.38-1.12, p=0.13). In the subgroup of individuals younger than 70 years of age, an OR of 0.24 (95% confidence interval: 0.08-0.72, p=0.004) was obtained.
Conclusions: Our data modestly support a protective effect of the APOE-e4 allele on AMD risk, but emphasize the need to investigate more thoroughly whether the effect could be restricted to cases with a family history of AMD and whether it varies across age and sex groups.
Age-related macular degeneration (AMD) is the leading cause of irreversible vision loss in older adults, affecting at least 15 million people in the United States. The prevalence of AMD increases with age, affecting approximately 9% of the population over the age of 65 and approximately 28% of individuals over the age of 75. AMD is a degenerative disorder of the retina and retinal pigment epithelium (RPE) primarily affecting the macular region. Two clinical subtypes of advanced AMD have been described and are referred to as an exudative ("wet") form characterized by choroidal neovascularization (CNV) and a "dry" form characterized by geographic atrophy. Vision loss may be severe with either form. Soft drusen are the earliest sign of AMD and confer an increased risk of geographic atrophy and CNV . Both the incidence and prevalence of soft drusen are age-related. There is no reliable treatment for dry AMD, and only 5-15% of patients with the wet subtype are candidates for laser coagulation therapy, which has a variable clinical outcome. The advanced forms of AMD occur in approximately 7% of the population over the age of 75 years and are more frequent in females (7.8%) than males (5.6%) .
AMD is a multifactorial disorder. Dietary and environmental risk factors for AMD have been suggested in numerous studies, and the adverse effects of smoking and high cholesterol levels are well established. Genetic studies are complicated by the late onset, clinical heterogeneity, and complex etiology of AMD. However, the substantial role of genetic factors has been supported by epidemiologic studies comparing the family history of cases and controls, by twin studies, and by segregation analyses . Recently, interesting hypotheses about the involvement of specific genotypes in AMD have been raised. Allikmets et al.  reported an increased prevalence of several variants of the Stargardt disease gene (ABCR) in AMD patients compared to controls. Subsequent studies indicated both presence [5-8] and absence [9-11] of associations of ABCR gene variants and AMD, leaving a resolution of the conflicting findings open for further research.
The apolipoprotein E (APOE) gene has also been implicated in AMD. Klaver et al.  and Souied et al.  have demonstrated a statistically significant protective effect of the APOE-e4 allele on AMD risk. The odds ratio (OR) associated with carrying an APOE-e4 allele was estimated as 0.43 (95% confidence interval: 0.21-0.88)  and 0.34 (95% confidence interval: 0.17-0.68) , respectively. One study  could not confirm these findings in Chinese AMD patients. Klaver et al.  also found some evidence for an increased risk of AMD conferred by the e2 allele of APOE. Soft drusen, the earliest sign of AMD, are characterized by protein and lipid deposits within Bruch's membrane . Thus, the APOE protein is an intuitive candidate to study because of the central role it plays in lipid metabolism. Additionally, interest in the APOE gene is supported by its known role in other neuro-degenerative diseases, most notably Alzheimer's disease, where the e4 allele increases risk in a dose-dependent manner [16-18].
In this report, we examine the effect of APOE on AMD risk in a series of patients ascertained as part of a large genetic study on AMD. Patients included in the analysis represent both sporadic and familial AMD cases. Their APOE genotype and allele frequencies are compared to a control group composed of previously genotyped spouses of non-AMD patients from the same clinical population base and spouses of AMD patients who were examined for evidence of AMD and found to be unaffected.
All AMD cases in this study were Caucasian and were ascertained in the southeastern United States at either Duke University Medical Center (DUMC) or Vanderbilt University Medical Center (VUMC). Probands were identified within the clinic population or through referral to the study site from local ophthalmologists. The study was conducted in accordance with the appropriate Institutional Review Board guidelines, and informed consent was obtained from all study participants. All cases were either evaluated at DUMC or VUMC, or medical records and fundus photographs were obtained from referring ophthalmologists for review. Grading of severity of macular disease based on fundus photographs was performed according to an internationally accepted classification system , as described previously . Following Bressler and Rosberger , we considered the presence of soft drusen, which is likely to lead to the subsequent development of RPE abnormalities, geographic atrophy, and CNV, sufficient to diagnose AMD at an early stage. Advanced stages of AMD include geographic atrophy ("dry" AMD) and exudative lesions ("wet" AMD). In our grading scheme, individuals were considered unaffected if they were classified as grade 1, no or small (<63 mm) drusen, or grade 2, non-extensive intermediate (>63 mm) drusen. Individuals with extensive intermediate or large (>125 mm) soft drusen, with or without RPE detachment, were classified as grade 3 and considered affected. Similarly, those with the more advanced findings of geographic atrophy (grade 4) and exudative lesions (grade 5) were considered to suffer from AMD.
All probands were administered a structured family history questionnaire in order to classify them as a familial or sporadic case. Confirmation of diagnosis for individuals identified as AMD suspects and examination of unaffected individuals (based on family history information supplied by the probands) were undertaken whenever possible. Emphasis in ascertainment was placed on the probands' parents (when available) and siblings. Our study population for the analysis presented here consists of 61 independent cases from multiple families (2 or more confirmed affected family members), 101 sporadic cases (no known family history), and 68 cases from families where additional affected family members were reported by the patient but were unavailable for examination. The total number of cases with known age and gender who were genotyped for APOE was 230, which is the largest number of AMD cases studied to date for an association with APOE.
For controls, we analyzed 333 spouses of patients ascertained through the Joseph and Kathleen Bryan Alzheimer's Disease Research Center (ADRC) and previously genotyped for APOE . These data included spouses of Alzheimer's disease as well as non-Alzheimer's dementia patients. This control group was ascertained from the same clinical population base (DUMC) as the AMD patients. The spouses were questioned about their ocular disease history and were free of obvious signs of advanced visual impairment at the time their blood sample was obtained. However, they were not specifically examined for early signs of AMD (see discussion). In addition to the ADRC spouse controls, we included, as controls, 39 spouses of AMD patients who had an eye examination at DUMC or VUMC and were found to be unaffected. Basic characteristics of the case and control population are summarized in Table 1. Since the ocular changes associated with AMD reflect a clinical continuum, an unambiguous age of onset for AMD is almost impossible to obtain with any accuracy. Thus, the age at examination for both cases and controls was used in the analysis. No controls younger than the youngest AMD case (41 years) were included.
For APOE genotyping of cases and controls, genomic DNA was extracted from whole blood after isolation of blood lymphocytes according to established protocols . APOE genotyping was performed as previously described .
To test whether APOE genotypes were in Hardy-Weinberg equilibrium (HWE) among cases and controls, c2 tests comparing observed proportions of genotypes with those expected under HWE were computed. Differences of genotype and allele frequency distributions between AMD cases and controls were also assessed by c2 statistics. For contingency tables where the reliability of p values based on c2 approximations was questionable, we report p values for Fisher's exact test (SAS version 6.12, SAS Institute, Cary, NC). For exact tests involving the e4 allele of APOE, we report a left sided p value, which corresponds to a pre-specified alternative hypothesis based on previous studies [12,13] proposing that the proportion of APOE-e4 carriers among cases is smaller than among controls.
To investigate the influence of family history on the proposed association of APOE and AMD, subgroup analyses were performed for the group of cases with confirmed or historical evidence of familial AMD (n=129) and for the group of sporadic cases (n=101). Since there was a significant difference between the case and control population with respect to age at examination and sex (Table 1), we computed stratified and Mantel-Haenszel odds ratios (OR) for age/sex groups chosen to have appropriate sample sizes for analysis. We also estimated age/sex-adjusted ORs for carriers of the e2 and e4 allele via logistic regression analysis. In this model, the group with e2/e4 genotype was included in both the e2 and the e4 terms and the ancestral e3/e3 genotype was the referent group. After verifying that a linear relationship of age and logit(AMD) was reasonable to assume (data not shown), age was incorporated in the logistic regression model as a continuous variable to use the most information available.
Overall APOE genotype and allele frequency distributions for cases and controls are shown in Table 2. Since the test of Hardy-Weinberg equilibrium (HWE) among controls was non-significant (c2=1.90, df=3, p=0.59), the comparison of allele frequencies, which assumes independence of alleles within genotypes, is valid . APOE genotypes of cases were also in HWE (c2=4.47, df=3, p=0.21). The overall unadjusted genotype and allele frequency distributions were not significantly different from each other (p=0.68 and p=0.35, respectively), although the frequency of the e4 allele was lower among cases (11.7%) than controls (14.6%).
As mentioned earlier, significant differences in the age and sex distribution of cases and controls were found, necessitating appropriate analysis methods to control for the potential confounding effects of these variables. No effect of APOE genotype on the risk for AMD was found in the logistic regression model for all cases and controls when adjusting for age and sex (OR for e4 carriers 0.88, 95% confidence interval: 0.58-1.35, p=0.56; OR for e2 carriers 0.99, 95% confidence interval: 0.59-1.66, p=0.97). In the subgroup analysis of sporadic AMD cases (n=101) and controls (n=372), we again did not find a significant effect of APOE (OR for e4 carriers 1.32, 95% confidence interval: 0.76-2.30, p=0.32; OR for e2 carriers 1.14, 95% confidence interval: 0.58-2.22, p=0.71). The comparison of familial AMD cases (n=129) with controls pointed toward a possible protective effect of the e4 allele among younger cases. In Table 3, we present genotype and allele frequencies for familial cases and controls stratified by age (three groups) and sex. For both women and men in the youngest age group (41-69 years), the frequency of genotypes with the e4 allele was lower among cases (3.6% for women, 27.3% for men) than controls (30.9% for women, 31.4% for men). In the older age groups, the frequencies were very similar for female cases and controls. Among males, the oldest age group (>79 years) actually had a higher proportion of cases (16.7%) than controls (4.6%) with e4 genotypes, but these frequencies were based on small sample sizes. Note that there were no familial cases with the e4/e4 genotype, whereas 2.7% of controls were homozygous for e4.
In Table 4, odds ratio estimates for e4 carriers, relative to e3/e3 genotypes, are shown separately for the six age/sex groups. In the youngest age group, the OR for e4 carriers was 0.24 (95% confidence interval: 0.08-0.72, p=0.004, left sided Fisher's exact test). The ORs for the other age groups were not significantly different from one, with the OR for the oldest age group having a particularly wide confidence interval due to the limited sample size in that strata. The overall Mantel-Haenszel OR for e4 carriers and its p value (OR 0.66, 95% confidence interval: 0.39-1.13, p=0.13) were practically identical to the values obtained from the age- and sex-adjusted logistic regression analysis (OR 0.66, 95% confidence interval: 0.38-1.12, p=0.13), also shown in Table 4. These results indicate that there is no statistically significant evidence for an association of the APOE-e4 allele and AMD when comparing the combined group of familial cases to controls, after adjusting for age and sex. However, the Breslow-Day test of homogeneity of odds ratios across age strata was significant (p=0.04), supporting the possibility that a protective effect of e4 may exist in the subgroup of younger cases with familial AMD. Sex-specific odds ratios were not computed in the Mantel-Haenszel analysis due to limited subgroup sample sizes, however, it can be seen from Table 3 that most of the evidence for a protective effect of e4 in the youngest age group came from the female cases and controls. Table 3 also shows that the e4 allele was less frequent in male compared to female controls in all but the youngest age group (see discussion), whereas among cases, we observed a higher frequency of e4 in males than females in the youngest and oldest age groups.
For the e2 allele of APOE, we did not detect any evidence of association in either the stratified Mantel-Haenszel analysis (Mantel-Haenszel OR 0.78, 95% confidence interval: 0.42-1.43, p=0.42, with no evidence for heterogeneity of ORs across age groups) or the logistic regression model (OR 0.82, 95% confidence interval: 0.43-1.54, p=0.53) when comparing familial AMD cases with controls (Table 4).
This study provides modest evidence to support the protective effect of the APOE-e4 allele on the risk of AMD that was suggested previously by Klaver et al.  and Souied et al. . However, our analyses indicate that the protective effect may be restricted to familial AMD cases and that, within this subgroup, it may be more prominent in younger cases. A possible gender difference remains to be investigated as our sample size was insufficient to allow for finer stratification. We did not formally adjust for the multiple tests performed on the data since the subgroup analyses were of a more exploratory nature and clearly require replication on larger data sets to be confirmed. An increased disease risk associated with the APOE-e2 allele was not supported by our data in any of the subgroups we considered.
We realize that the control group available to us for comparison of APOE genotype/allele frequencies is less than ideal since only a fraction of them (39/372) have been specifically examined for AMD. However, there are no significant differences (p>0.50) in APOE genotype or allele distribution between our controls (n=372) and those used by Klaver et al.  (n=901), which were randomly selected study subjects without atrophic or neovascular AMD based on fundus photography. Similarly, our control genotype/allele distributions are not significantly different (p>0.55, Fisher's exact test) from those of the subgroup of controls with self-reported or clinically confirmed normal vision (n=91) used by Souied et al. . Note that the overall frequency of the e4 allele in our controls (14.6%) is lower than in the Klaver et al.  controls (15.6%) and in the normal-vision Souied et al.  controls (17.6%). This means that our potential bias due to including unexamined controls is conservative.
It is quite plausible that the disease etiology may differ for familial and sporadic AMD cases. Unfortunately, the previous studies [12,13] did not report whether their AMD patients were familial or sporadic, making an exact comparison of previous and current results difficult. The power of this study to detect a true odds ratio of 0.5 (close to the value estimated by Klaver ), assuming that the population frequency of e4 carriers is 0.27, is 95.5% for the data set of all AMD cases and controls. It decreases to 86.6% when only the sample size for familial AMD cases is considered. Thus, our data set should have sufficient power to detect the postulated effect if there were no additional complexities introduced by confounding variables. However, a drawback of our study is the fact that cases and controls were not matched a priori for age and sex. To adjust for these important variables in an unmatched study, we incorporated them in an unconditional logistic regression model and we computed stratum-specific odds ratio estimates for reasonably large and more homogeneous age/sex strata. This, however, clearly limits the sample size in these subgroups, and thus our power to detect an APOE effect.
The power of our study (all AMD cases and controls) to detect an effect of the much rarer e2 allele, assuming a population frequency of e2 carriers of 0.15 and the OR of 1.5 estimated by Klaver et al. , is only 45.5%. The OR in the previous study  was not significantly different from one (95% confidence interval: 0.80-2.82). The OR in this study tends in the other direction from unity without reaching statistical significance. More data are needed to establish whether or not there is any association of the e2 allele with AMD, and if so, whether its presence leads to an increase or decrease in disease risk.
As for the APOE-e4 allele, it will be necessary to analyze a larger number of AMD patients, taking into account family history, age, and gender, in order to investigate more thoroughly if and how these factors influence the postulated protective effect. If such a study was carried out with age- and sex-matched controls, the main effect of these confounder variables could be eliminated, yet the assessment of potential interactions between these factors and APOE would still be possible. It has been observed previously [25,26] that APOE genotype and allele frequencies in control groups may differ by age and sex. As discussed by Bickeboeller et al.  in the context of an Alzheimer's disease study, the sex difference in APOE genotypes might be due to an early selection against male e4 carriers since this allele increases risk of ischemic heart disease in middle-aged and elderly men. For Alzheimer's disease, there is a well-documented higher disease prevalence among females compared to males. Some studies have suggested that this could be explained by a larger increase in Alzheimer's disease risk associated with the APOE-e4 allele for women compared to men [27-29], but the absence of a gender difference in risk has been reported as well [25,26]. Alternative explanations for the higher Alzheimer's disease prevalence in women may be (i) the lower e4 frequency in elderly male controls due to the early selection process described above , or (ii) the longer survival of affected women compared to affected men , both possibilities being compatible with a similar e4 associated Alzheimer's disease risk for males and females. These observations underscore the need to compare APOE frequencies of cases and controls only within the same age/sex group to eliminate the confounding association of the genotype distribution with age and sex and thus account for the known effect of APOE on longevity. In the study by Souied et al. , the AMD cases and controls were age- and sex-matched, but possible interaction effects of age and sex with APOE were not studied. Klaver et al.  noted significant age differences between their AMD patients and the control group, but reported that the frequency of the e4 allele was lower among cases than controls in all three age groups they considered (55-75, 76-85, and >85 years) and that there were no significant differences in allele frequencies between these groups. They did not present stratified and Mantel-Haenszel ORs for the three age groups. They did not comment on possible sex differences, although they did adjust for sex in their logistic regression analysis.
Our data provide suggestive evidence that the protective effect of APOE-e4 may be restricted to younger cases with familial AMD. Whether or not genetic factors in general play a more prominent role in disease etiology for the subgroup of patients with earlier age of onset has not been established unequivocally . If indeed one or more susceptibility gene(s) increase disease risk for the early-onset form of the disorder, it is conceivable that the e4 allele of APOE might reduce the risk conferred by the other gene(s) through some form of biological interaction. However, this possibility would clearly have to be investigated more thoroughly on a molecular basis once the role of APOE has been established unequivocally and other susceptibility genes for AMD have been identified from genome screens  or from candidate gene association studies.
If the proposed statistical association of the APOE-e4 allele and AMD is real, there are two possible mechanisms that could explain this observation. The first is linkage disequilibrium between the APOE gene and a gene whose product directly influences AMD susceptibility such that the e4 allele is associated with a particular mutation in this gene. The second explanation is a direct role of APOE in AMD pathogenesis. A major role of APOE in the CNS is to bind free cholesterol and lipids released from cell membrane degeneration and distribute them for use in the renewal of the membranes [32,33]. A true protective effect of APOE-e4 in AMD pathogenesis would be particularly interesting since it would be the exact opposite of the effect found in other neurodegenerative disorders, most notably Alzheimer's disease, where the e4 allele increases disease risk. Both Klaver et al.  and Souied et al.  have proposed hypothetical biological mechanisms by which APOE-e4 might reduce the risk of AMD.
In conclusion, we have demonstrated that the previously proposed protective effect of the APOE-e4 allele and AMD may be limited to the subgroup of familial AMD patients. Furthermore, we have emphasized the need to carefully adjust for the confounding effect of age and sex due to the association of these factors with the APOE genotype distribution in the general population. Studies with larger sample sizes are required to investigate more carefully whether the effect of APOE on AMD is indeed modified by family history, age, and gender.
We thank all of the families whose participation made this project possible. This research was supported in part by NIH grants EY12118 (M.A.P.-V.) and NS23360 (M.A.P.-V.). We thank the personnel at the Center for Human Genetics of Duke University Medical Center for their assistance in data management and in preparation of this manuscript.
1. Klein R, Klein BE, Jensen SC, Meuer SM. The five-year incidence and progression of age-related maculopathy: the Beaver Dam Eye Study. Ophthalmology 1997; 104:7-21.
2. Klein R, Klein BE, Linton KL. Prevalence of age-related maculopathy. The Beaver Dam Eye Study. Ophthalmology 1992; 99:933-43.
3. Yates JR, Moore AT. Genetic susceptibility to age related macular degeneration. J Med Genet 2000; 37:83-7.
4. Allikmets R, Shroyer NF, Singh N, Seddon JM, Lewis RA, Bernstein PS, Peiffer A, Zabriskie NA, Li Y, Hutchinson A, Dean M, Lupski JR, Leppert M. Mutation of the Stargardt disease gene (ABCR) in age-related macular degeneration. Science 1997; 277:1805-7.
5. Dean M, Allikmets R, Shroyer NF, Lupski JR, Lewis RA, Leppert M, Bernstein PS, Seddon JM. ABCR gene and age-related macular degeneration. Science [Online] 1998; 279:1107a.
6. Lewis RA, Shroyer NF, Singh N, Allikmets R, Hutchinson A, Li Y, Lupski JR, Leppert M, Dean M. Genotype/Phenotype analysis of a photoreceptor-specific ATP-binding cassette transporter gene, ABCR, in Stargardt disease. Am J Hum Genet 1999; 64:422-34.
7. Shroyer NF, Lewis RA, Allikmets R, Singh N, Dean M, Leppert M, Lupski JR. The rod photoreceptor ATP-binding cassette transporter gene, ABCR, and retinal disease: from monogenic to multifactorial. Vision Res 1999; 39:2537-44.
8. Allikmets R. Further evidence for an association of ABCR alleles with age-related macular degeneration. The International ABCR Screening Consortium. Am J Hum Genet 2000; 67:487-91.
9. Stone EM, Webster AR, Vandenburgh K, Streb LM, Hockey RR, Lotery AJ, Sheffield VC. Allelic variation in ABCR associated with Stargardt disease but not age-related macular degeneration. Nat Genet 1998; 20:328-329.
10. Kuroiwa S, Kojima H, Kikuchi T, Yoshimura N. ATP binding cassette transporter retina genotypes and age related macular degeneration: an analysis on exudative non-familial Japanese patients. Br J Ophthalmol 1999; 83:613-5.
11. De La Paz MA, Guy VK, Abou-Donia S, Heinis R, Bracken B, Vance JM, Gilbert JR, Gass JD, Haines JL, Pericak-Vance MA. Analysis of the Stargardt disease gene (ABCR) in age-related macular degeneration. Ophthalmology 1999; 106:1531-6.
12. Klaver CC, Kliffen M, van Duijn CM, Hofman A, Cruts M, Grobbee DE, van Broeckhoven C, de Jong PT. Genetic association of apolipoprotein E with age-related macular degeneration. Am J Hum Genet 1998; 63:200-6.
13. Souied EH, Benlian P, Amouyel P, Feingold J, Lagarde JP, Munnich A, Kaplan J, Coscas G, Soubrane G. The epsilon4 allele of the apolipoprotein E gene as a potential protective factor for exudative age-related macular degeneration. Am J Ophthalmol 1998; 125:353-9.
14. Leung YF, Fan DSP, Chan WM, Baum L, Pang CP, Lam DSC. Apolipoprotein E alleles in age-related macular degeneration. Invest Ophthalmol Vis Sci 1999; 40:S920.
15. Pauleikhoff D, Barondes MJ, Minassian D, Chisholm IH, Bird AC. Drusen as risk factors in age-related macular disease. Am J Ophthalmol 1990; 109:38-43.
16. Strittmatter WJ, Saunders AM, Schmechel D, Pericak-Vance M, Enghild J, Salvesen GS, Roses AD. Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci U S A 1993; 90:1977-81.
17. Saunders AM, Strittmatter WJ, Schmechel D, George-Hyslop PH, Pericak-Vance MA, Joo SH, Rosi BL, Gusella JF, Crapper-MacLachlan DR, Alberts MJ, Hulette C, Crain B, Goldgaber D, Roses AD. Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease. Neurology 1993; 43:1467-72.
18. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak-Vance MA. Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science 1993; 261:921-3.
19. Bird AC, Bressler NM, Bressler SB, Chisholm IH, Coscas G, Davis MD, de Jong PT, Klaver CC, Klein BE, Klein R, Mitchell P, Sarks SH, Soubrane G, Taylor HR, Vingerling JR. An international classification and grading system for age-related maculopathy and age-related macular degeneration. The International ARM Epidemiological Study Group. Surv Ophthalmol 1995; 39:367-74.
20. Bressler SB, Rosberger D. Nonneovascular (nonexudative) age-related macular degeneration. In: Guyer DR, Yannuzzi L, Chang S, Shields JA, Green WR, editors. Retina-Vitreous-Macula. Philadelphia: WB Saunders and Company; 1999. p. 79-93.
21. Scott WK, Saunders AM, Gaskell PC, Locke PA, Growdon JH, Farrer LA, Auerbach SA, Roses AD, Haines JL, Pericak-Vance MA. Apolipoprotein E epsilon2 does not increase risk of early-onset sporadic Alzheimer's disease. Ann Neurol 1997; 42:376-8.
22. Vance JM. The Collection of Biological Samples for DNA Analysis. In: Haines JL, Pericak-Vance MA, editors. Approaches to Gene Mapping in Complex Human Diseases. New York: Wiley-Liss; 1998. p. 201-211.
23. Grubber JM, Saunders AM, Crane-Gatherum AR, Scott WK, Martin ER, Haynes CS, Conneally PM, Small GW, Roses AD, Haines JL, Pericak-Vance MA. Analysis of association between Alzheimer disease and the K variant of butyrylcholinesterase (BCHE-K). Neurosci Lett 1999; 269:115-9.
24. Schaid DJ, Jacobsen SJ. Biased tests of association: comparisons of allele frequencies when departing from Hardy-Weinberg proportions. Am J Epidemiol 1999; 149:706-11.
25. Bickeboller H, Campion D, Brice A, Amouyel P, Hannequin D, Didierjean O, Penet C, Martin C, Perez-Tur J, Michon A, Dubois B, Ledoze F, Thomas-Anterion C, Pasquier F, Puel M, Demonet JF, Moreaud O, Babron MC, Meulien D, Guez D, Chartier-Harlin MC, Frebourg T, Agid Y, Martinez M, Clerget-Darpoux F. Apolipoprotein E and Alzheimer disease: genotype-specific risks by age and sex. Am J Hum Genet 1997; 60:439-46.
26. Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC Jr, Rimmler JB, Locke PA, Conneally PM, Schmader KE, Tanzi RE, Small GW, Roses AD, Pericak-Vance MA, Haines JL. Apolipoprotein E, survival in Alzheimer's disease patients, and the competing risks of death and Alzheimer's disease. Neurology 1995; 45:1323-8.
27. Payami H, Montee KR, Kaye JA, Bird TD, Yu CE, Wijsman EM, Schellenberg GD. Alzheimer's disease, apolipoprotein E4, and gender. JAMA 1994; 271:1316-7.
28. Payami H, Zareparsi S, Montee KR, Sexton GJ, Kaye JA, Bird TD, Yu CE, Wijsman EM, Heston LL, Litt M, Schellenberg GD. Gender difference in apolipoprotein E-associated risk for familial Alzheimer disease: a possible clue to the higher incidence of Alzheimer disease in women. Am J Hum Genet 1996; 58:803-11.
29. Bretsky PM, Buckwalter JG, Seeman TE, Miller CA, Poirier J, Schellenberg GD, Finch CE, Henderson VW. Evidence for an interaction between apolipoprotein E genotype, gender, and Alzheimer disease. Alzheimer Dis Assoc Disord 1999; 13:216-21.
30. Gorin MB, Breitner JC, De Jong PT, Hageman GS, Klaver CC, Kuehn MH, Seddon JM. The genetics of age-related macular degeneration. Mol Vis 1999; 5:29 <http://www.molvis.org/molvis/v5/a29/>.
31. Weeks DE, Conley YP, Mah TS, Paul TO, Morse L, Ngo-Chang J, Dailey JP, Ferrell RE, Gorin MB. A full genome scan for age-related maculopathy. Hum Mol Genet 2000; 9:1329-49.
32. Poirier J, Baccichet A, Dea D, Gauthier S. Cholesterol synthesis and lipoprotein reuptake during synaptic remodelling in hippocampus in adult rats. Neuroscience 1993; 55:81-90.
33. Poirier J, Minnich A, Davignon J. Apolipoprotein E, synaptic plasticity and Alzheimer's disease. Ann Med 1995; 27:663-70.