Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Alzheimers Dement. Author manuscript; available in PMC 2017 April 1.
Published in final edited form as:
PMCID: PMC4841700

Individual estimates of age at detectable amyloid onset for risk factor assessment

Murat Bilgel, B.S.,a,b,* Yang An, M.S.,a Yun Zhou, Ph.D.,c Dean F. Wong, M.D., Ph.D.,c Jerry L. Prince, Ph.D.,b,c,d Luigi Ferrucci, M.D., Ph.D.,e and Susan M. Resnick, Ph.D.a



Individualized estimates of age at detectable amyloid-beta (Aβ) accumulation, distinct from amyloid positivity, allow for analysis of onset age of Aβ accumulation as an outcome measure to understand risk factors.


Using longitudinal Pittsburgh compound B (PiB) PET data from Baltimore Longitudinal Study of Aging, we estimated the age at which each PiB+ individual began accumulating Aβ. We used survival analysis methods to quantify risk of accumulating Aβ and differences in onset age of Aβ accumulation in relation to APOE ε4 status and sex among 36 APOE ε4 carriers and 83 non-carriers.


Age at onset of Aβ accumulation for the APOE ε4− and ε4+ groups was 73.1 and 60.7, respectively. APOE ε4 positivity conferred a 3-fold risk of accumulating Aβ after adjusting for sex and education.


Estimation of onset age of amyloid accumulation may help gauge treatment efficacy in interventions to delay symptom onset in Alzheimer’s disease.

Keywords: APOE, Pittsburgh compound B, PET, amyloid imaging, age at onset


Cerebral amyloid-beta (Aβ) deposition is the defining characteristic of preclinical stages of Alzheimer’s disease (AD) [1] and begins years before cognitive symptoms are evident [2]. The apolipoprotein E (APOE) ε4 allele is the most influential known genetic risk factor for late-onset AD [3,4], and is associated with higher cortical Aβ burden in cognitively normal individuals [523].

Age at onset of AD is a phenotype of great clinical interest, as delaying onset will decrease population prevalence and disease burden [24]. Furthermore, finding characteristics associated with onset age might provide clues about disease mechanisms. Among AD risk genes, APOE exhibits the strongest relationship with the age at onset of AD [25,26], with the presence of one or more ε4 alleles shifting the onset to earlier ages [27]. However, studies of APOE in relation to onset age of AD-related neuropathology are limited. Two cross-sectional studies [7,22] and a recent meta-analysis [23] have shown that Aβ deposition is more likely at younger ages in groups of APOE ε4 carriers compared with non-carriers, suggesting an earlier age at onset of Aβ accumulation. However, the effect of APOE genotype on the age at onset of Aβ accumulation at the individual level has not been investigated. Individualized estimates of the age at onset of Aβ accumulation will allow for the analysis of onset age as an outcome measure to understand the effects of risk factors.

Here, we use longitudinal Pittsburgh compound B (PiB) positron emission tomography (PET) scans from the Baltimore Longitudinal Study of Aging to estimate the age at which each individual with Aβ pathology began exhibiting increases in cortical fibrillar Aβ, a measure distinct from the age of PiB positivity. We investigate both the risk of accumulating Aβ and the age at which this began in relation to APOE genotype and sex.



Longitudinal positron emission tomography (PET) data were acquired using the radiotracer Pittsburgh compound B (PiB) for 132 participants, aged 55 and older, from the Baltimore Longitudinal Study of Aging (BLSA) neuroimaging substudy [28]. Research protocols were approved by local institutional review boards, and all participants gave written informed consent at each visit. At enrollment into neuroimaging substudy, participants were free of central nervous system disease (dementia, stroke, bipolar illness, epilepsy), severe cardiac disease (myocardial infarction, coronary artery disease requiring angioplasty or coronary artery bypass surgery), severe pulmonary disease, or metastatic cancer. A subset of the BLSA neuroimaging participants received PiB-PET scans beginning in 2005. At baseline PiB-PET, 3 of the 132 participants had diagnoses of mild cognitive impairment (MCI) and 2 had diagnoses of AD. Remaining individuals were cognitively normal at baseline. At last PiB-PET, 3 participants had diagnoses of MCI, 4 had AD, and one had dementia of unspecified etiology. Diagnoses of dementia and Alzheimer’s disease (AD) were based on DSM-III-R [29] and NINCDS-ADRDA criteria [30], respectively. Mean (SD) age at baseline PiB-PET imaging was 77.1 (7.8). Participants were studied at 1 to 2-year intervals with up to 8 (mean 2.5, SD 1.8) repeated scans over up to 9 years. Participant demographics for the whole sample and for each APOE group are presented in Table 1. Of the 36 ε4 carriers, 3 were ε2/ε4, 29 were ε3/ε4, and 4 were ε4/ε4. Of the 83 non-carriers, 16 were ε2/ε3 and 67 were ε3/ε3.

Table 1
Participant characteristics overall and by APOE ε4 status.

Mean (SD, range) age at baseline for males (n=71) and females (n=61) was 79.0 (6.8, 60.9–92.3) and 74.9 (8.5, 55.7–93.4) years, respectively. Sixteen (25.4%) of 63 males and 20 (35.7%) of 56 females with known APOE status were ε4 carriers.

Continuous variables were compared between APOE ε4 groups, as well as between males and females, using Wilcoxon rank sum tests. Proportions were compared using Fisher’s exact test.

Image acquisition

Dynamic PiB-PET scans were acquired in 3D mode on a GE Advance scanner immediately following an intravenous bolus injection of a mean (SD) 14.7 (0.8) mCi of [11C]-PiB. Participants were fitted with a thermoplastic head mask to minimize motion during scanning. PET data were acquired according to the following protocol for the duration of the frames: 4×0.25, 8×0.5, 9×1, 2×3, and 10×5 min (70 minutes total, 33 frames). Dynamic images were reconstructed using filtered back-projection with a ramp filter, yielding a spatial resolution of approximately 4.5 mm full width at half max at the center of the field of view (image matrix = 128×128, 35 slices, pixel size = 2×2mm, slice thickness = 4.25mm).

Magnetization-prepared rapid gradient echo (MPRAGE) images acquired on a 3T scanner (Philips Achieva, repetition time [TR]=6.8ms, echo time [TE]=3.2ms, flip angle=8°, image matrix=256×256, 170 slices, pixel size=1×1mm, slice thickness=1.2mm) were used as the MRIs whenever available; however, these images were unavailable for 19 of the 132 participants due to scanner changes over time. Six participants underwent an MPRAGE acquisition sequence on a 1.5T scanner (Philips Intera, TR=6.8ms, TE=3.3ms, flip angle=8°, image matrix=256×256, 124 slices, pixel size=0.94×0.94 mm, slice thickness=1.5 mm). Remaining 13 participants underwent a spoiled gradient-recalled (SPGR) acquisition sequence on a 1.5T scanner (GE Signa, TR=35ms, TE=5ms, flip angle=45°, image matrix=256×256, 124 slices, pixel size=0.94×0.94mm, slice thickness=1.5mm). A baseline MRI was matched with its concurrent or closest-in-time PiB-PET image for each participant. Seventeen participants did not have an MRI concurrent with any of their PiB-PET scans. The interval between the MRI and closest-in-time PiB-PET for these participants was 2.5 years (SD 1.1, range 1.0–4.6), excluding two participants who were claustrophobic and had a single MRI 10.5 and 14.7 years prior to PiB-PET.

Image processing

Time frames of each dynamic PET scan were aligned to the average of the first two minutes to correct for motion [31]. For registration purposes, we obtained static images by averaging the first 20 minutes of each dynamic PET scan. Follow-up PET scans were rigidly registered onto the baseline PET within each participant using 20-minute average images. Baseline MRIs were rigidly registered onto their corresponding 20-minute average. We generated a study-specific template using baseline 3T MPRAGE images in a diffeomorphic registration framework [32]. Corresponding 1.5T MPRAGE and 1.5T SPGR templates were generated from the 3T template using a patch-based image synthesis method [33]. 3T MPRAGE template was segmented using FreeSurfer (version 5.1, [34]. The appropriate MRI template was registered onto each subject’s baseline MRI using diffeomorphic registration [35]. The FreeSurfer segmentation in template space was transformed accordingly onto the baseline PET scan, and then rigidly transformed onto each longitudinal PET scan using the within-subject transformations. Consecutive transformations were concatenated so that FreeSurfer segmentations were interpolated only once. Distribution volume ratio (DVR) images were calculated in the native space of each PET image using the simplified reference tissue model with the cerebellar gray matter as reference tissue [36]. Mean cortical DVR (cDVR) was calculated as the average of the DVR values in cingulate, frontal, parietal (including precuneus), lateral temporal, and lateral occipital cortical regions, excluding the sensorimotor strip.

Estimation of mean cDVR threshold

A two-class Gaussian mixture model was fitted to baseline mean cDVR data. The DVR value corresponding to the intersection of the probability density functions of the two classes was used to classify participants into PiB− and PiB+ groups based on their last visit. The longitudinal mean cDVR data are presented in Figure 1.

Figure 1
Longitudinal mean cortical DVR versus age

APOE genotype

APOE genotype was determined using standard procedures [37]. Thirteen participants with missing APOE genotype information were excluded in APOE analyses. APOE ε4 status was coded for each participant as 1 if at least one ε4 allele was present, and as 0 otherwise.

Nonlinear mixed effects model for estimating the age of onset of amyloid deposition

We estimated the age at which each PiB+ individual began exhibiting increases in mean cDVR using a nonlinear mixed effects model. This age reflects the age at earliest detectable amyloid accumulation, prior to surpassing the threshold for PiB positivity. The model included an intercept and a slope associated with age for the PiB− group. Subjects who were PiB+ at last visit were modeled using a two-part piecewise linear trajectory: they were assumed to follow the trajectory associated with the PiB− group until they reached detectable amyloid levels, beyond which they were modeled with a separate trajectory that had a different slope associated with age. Random effects included intercept and age at detectable amyloid accumulation, allowing for the estimation of these values for each individual. The rate of change in mean cDVR was assumed to be the same across individuals within each of the PiB groups. This assumption reduced the variability in the estimated parameters and allowed for the estimation of the age at detectable amyloid accumulation for individuals with a single PiB-PET scan. See Appendix for details. Once amyloid levels are detectable, individuals need to accumulate enough amyloid to surpass the mean cDVR threshold to be considered PiB+. We also calculated the age at which each individual became PiB+ based on the fitted model.

Survival analyses

The nonlinear mixed effects model allows for the estimation of the age at detectable amyloid accumulation only for the individuals who developed elevated levels of cortical Aβ over the course of the study. In order to accurately assess the risk of APOE genotype and sex on developing elevated levels of Aβ, we have to account for the possibility that individuals who remained PiB− during the study might have begun accumulating Aβ after their last visit. We therefore used survival analysis methods to examine the effects of APOE ε4 status and sex on the age at detectable amyloid accumulation, adjusting for level of education. Age at detectable amyloid accumulation served as the time of event for the PiB+ individuals; those who had not developed cortical Aβ pathology by their last visit were right-censored. First, we compared the Kaplan-Meier curves for APOE ε4 groups and for males and females. To study the effect of APOE ε4 status and sex on the risk of reaching detectable amyloid levels, we used the Cox proportional hazards model. The Cox model allows for the quantification of risk, but does not inform about the differences in the age at onset across groups. To directly study the effect of APOE status and sex on the age at detectable amyloid accumulation, we used the accelerated failure time (AFT) model. The AFT model is given by logTi=xiTβ+σεi, where Ti is time to event for ith subject, xi is a vector of independent variables, εi is random noise, and β and σ are parameters to be estimated. The Akaike Information Criterion (AIC) was used to select the best distribution (among exponential, log-logistic, Weibull, log-normal, and gamma distributions) for Ti in the AFT model. In survival analyses, sex was coded as 1 for males and 0 for females. All statistical analyses were conducted in R, version 3.0.2 (


Distribution of baseline age (p=0.19), proportion of females (p=0.24), years of education (p=0.54), number of PiB-PET scans (p=0.28) and follow-up duration (p=0.29) were comparable between the APOE groups (Table 1). Proportion of APOE ε4 carriers (p=0.24 restricting the comparison to individuals with known APOE status), years of education (p=0.63), number of PiB-PET scans (p=0.79) and follow-up duration (p=0.93) were comparable between males and females. Females were younger than males at baseline (p=0.004).

The Gaussian mixture model yielded a mean cDVR threshold of 1.066 for classifying individuals into PiB− and PiB+ groups (Appendix, Figure A.1). Of 83 ε4− and 36 ε4+ individuals, 25 (30.1%) and 18 (50.0%) were PiB+ at last visit, respectively (Table 1). Of 71 males and 61 females, 30 (42.3%) and 17 (27.9%) were PiB+ at last visit, respectively. Proportion of PiB+ individuals at last visit was higher in the ε4+ group compared to ε4− as well as among males compared to females, but these differences did not reach statistical significance (p=0.06 and 0.10, respectively). ε4 carriers had significantly higher mean cDVR at last visit compared to non-carriers (p=0.02). Mean cDVR at last visit was 1.13 (SD 0.20) for males and 1.07 (SD 0.16) for females, but this difference was not significant (p=0.18).

The results of the nonlinear mixed effects model are presented in Table A.1, and predicted mean cDVR values for each individual are shown in Figure A.2. We estimated that individuals in our sample began accumulating cortical fibrillar amyloid as measured by PiB-PET at age 69.3 (SD 8.1) on average. Estimated ages at detectable amyloid accumulation ranged from 54.0–88.8 years. After reaching detectable amyloid levels, individuals exhibited increases in mean cDVR by 0.024/year. There was a slight downward slope in the PiB− group showing decreases by 0.001/year in mean cDVR.

Using individual-level estimates, we calculated that individuals surpass the cDVR threshold at age 72.5 (SD 8.3, range 56.7–92.8) on average. Average within-subject interval between age at detectable amyloid accumulation and age at which the individual became PiB+ was 3.2 years (SD 0.4, range 1.5–4.0).

Kaplan-Meier curves (Figure 2) were statistically different between APOE ε4 carrier and non-carriers (log-rank test p=0.0009) but not between males and females (p=0.26). The Cox model showed that APOE ε4 carrier status was significantly associated with reaching detectable amyloid levels (chi-squared test p=0.0006) after adjusting for sex and years of education (Table 2). APOE ε4+ individuals had a 3-fold increase in risk of reaching detectable amyloid levels compared to the ε4− individuals (hazard ratio [HR]=3.05, 95% CI 1.61–5.78). Neither sex nor level of education was significant (p=0.12 and 0.77, respectively).

Figure 2
Kaplan-Meier curves
Table 2
Cox model results.

The log-normal distribution yielded the smallest AIC, and therefore was selected as the distribution function for the AFT model. The effect of APOE ε4 status on age at detectable amyloid accumulation was significant in the AFT analysis (chi-squared test p=0.0005), with ε4+ individuals reaching detectable amyloid levels 16.9% (95% CI 7.4–26.4) earlier on average compared to ε4− individuals after adjusting for sex and years of education (Table 3). Neither sex nor level of education was significant (p=0.25 and 0.84, respectively). Using the result of the AFT model in conjunction with the mixed effects model finding that the population-average age at detectable amyloid accumulation is 69.3, we approximated that the age at detectable amyloid accumulation is 73.1 (95% CI 68.2–76.1) and 60.7 (95% CI 53.6–65.1) years for the APOE ε4− and ε4+ groups, respectively.

Table 3
Accelerated failure time (AFT) model results.

Analyses using precuneus DVR values instead of mean cDVR yielded similar results, with ε4 carrier status associated with a more than 3-fold risk of reaching detectable amyloid levels as well as an 18.2% earlier onset of amyloid accumulation (Appendix, Tables A.8 and A.9).


We estimated the age at onset of fibrillar cortical Aβ pathology using PiB-PET imaging in a longitudinal sample, and investigated its associations with APOE genotype, adjusting for sex and levels of education. On average, Aβ accumulation began at age 69.3 in our sample. APOE ε4 carrier status not only increased the risk of developing Aβ pathology, i.e., PiB positivity, but was also associated with an earlier onset of Aβ accumulation. There was a more than 12-year difference in the initiation of Aβ accumulation between APOE ε4 carriers and non-carriers, with ε4+ and ε4− individuals beginning to accumulate at age 60.7 and 73.1 years, respectively. We found no significant effects of sex or education on level or timing of onset of Aβ accumulation.

Our finding that APOE ε4 carrier status is associated with a higher risk of developing cortical fibrillar Aβ agrees with previous reports of in vivo amyloid imaging in cognitively normal healthy individuals showing higher cortical amyloid levels and a greater proportion of amyloid positive individuals in the APOE ε4+ group [6,9,11,23,38,39]. In this paper, we extend these findings by introducing a novel approach to estimate the onset age of Aβ accumulation to allow investigation of factors that modify the initiation of Aβ deposition at the individual level. The earlier onset of Aβ in APOE ε4 positive individuals is closely linked to the finding that the presence of the APOE ε4 allele shifts the age at diagnosis of AD by about 10 to 15 years [3,27,40]. Our finding suggests that the onset of preclinical AD, which is characterized by asymptomatic amyloidosis [1], is also earlier in APOE ε4 carriers, supporting the hypothesis that APOE ε4 positivity shifts the entire disease timeline rather than accelerating disease progression. We and others have shown that after adjusting for age, APOE ε4 status does not significantly affect rates of Aβ accumulation, measured by PiB-PET [2,5,21,41] or cerebrospinal fluid Aβ1-42 [21], suggesting that once the neuropathological changes are underway, the effects of APOE status on disease progression are less pronounced.

Our analyses using longitudinal data corroborate the findings in two cross-sectional samples that APOE ε4 positivity is associated with an earlier onset of Aβ accumulation [7,22]. Fleisher et al. found that Florbetapir PET signal began showing increases at age 58, with positivity occurring at age 71 in cognitively normal adults [7]. APOE ε4 carriers were found to become amyloid positive at age 56 on average, whereas non-carriers at age 76 [7]. In a study of cognitively normal adults, Jack et al. found that the ages at which 10% of individuals would be considered PiB-PET positive is 57 for APOE ε4 carriers and 64 for non-carriers [22]. Our study is distinguished from these studies in several aspects: Our analyses were based on longitudinal rather than cross-sectional data, we characterized the age at onset of amyloid accumulation rather than age at amyloid positivity, and we estimated an onset age for each individual in addition to reporting averages by APOE group. We estimated that PiB signal began showing increases at age 69, with positivity occurring at age 72 in our overall sample. We further estimated that ages at detectable amyloid accumulation were 61 for APOE ε4 carriers and 73 for non-carriers. Based on our finding that there is approximately a 3-year period between detectable amyloid accumulation and amyloid positivity, our results indicate that amyloid positivity occurs around age 64 for ε4 carriers and 76 for non-carriers. The different findings across these studies might be due to the difference in radiotracers, definitions of amyloid positivity, and statistical approaches, the use of standardized uptake volume ratio versus DVR to quantify the amyloid levels, and the cross-sectional versus longitudinal nature of the datasets. Furthermore, our sample included individuals 55 and older, but the samples used in Fleisher et al. [7] and Jack et al. [22] had participants as young as 18 and 30, respectively. Inclusion of younger adults might have benefited the estimation of subtle trends at earlier ages that we were unable to extract in our sample. In addition, our initial sample of BLSA participants was selected for health, as they had remained healthy enough to participate over an 11-year follow-up at the initial PiB studies. BLSA participants also have high levels of education, which might have a protective effect against amyloid deposition although this was not evident in our sample [42].

Sex differences in cortical Aβ levels or the proportion of amyloid positive individuals have previously not been observed among healthy individuals in large studies of PET imaging, including the BLSA, although a small study of 26 individuals found higher cortical amyloid burden in males [20]. The proportion of PiB+ individuals was not significantly different among males and females in our sample, and males and females did not differ significantly in terms of their risk or age at onset of Aβ accumulation, which agrees with the finding of Jack et al. [22]. The proportion of APOE ε4+ individuals was higher among females than among males in our sample. This difference in older adults may in part reflect an earlier dropout of males due to the association between the ε4 genotype and risk for cardiovascular disease [43], which occurs earlier in men than women.

Limitations of our study include the limited number of longitudinal follow-ups due to the relatively recent development of amyloid imaging. The majority of the estimated ages of detectable amyloid accumulation from the nonlinear mixed effects model are based on relatively few direct observations since only a few individuals went from being PiB− to PiB+ during the course of the study. Estimation of these ages for individuals who were PiB+ at their initial scan was based on the assumption that the rate of DVR change is constant across PiB+ individuals. Since PiB-PET measures fibrillar amyloid, our nonlinear mixed effects model estimates reflect the ages at which we start to measure a discernible change in PiB signal, not necessarily the age at which amyloid begins to accumulate. Our PiB-PET images were not corrected for partial volume effects, and therefore the mean cDVR measures may be confounded by effects of atrophy. Due to sample size limitations, we were unable to investigate different APOE allele combinations or the interaction of APOE status and sex in relation to Aβ. High education levels of the BLSA cohort may limit generalizability of our findings to individuals with lower levels of education.

In conclusion, individualized estimates of the onset age for detectable amyloid accumulation in conjunction with survival analysis methods allow for novel analyses using longitudinal amyloid data. Determination of onset age of amyloid accumulation is of critical importance for evaluation of treatments to prevent or delay AD and can inform the optimal time window for anti-amyloid interventions.

Research in Context

1. Systematic Review

We searched PubMed for “amyloid”, “APOE”, “PiB”, and “Alzheimer”. Our review indicated that the age at onset of amyloid accumulation has not been explored at the individual level, and no longitudinal study to date has investigated the associations between the onset age of amyloid accumulation and APOE genotype.

2. Interpretation

APOE ε4 positivity confers a 3-fold risk of accumulating Aβ after adjusting for sex and education. The onset of Aβ accumulation occurs more than 12 years earlier for APOE ε4+ individuals compared to ε4−. Understanding the factors that influence the onset age of amyloid accumulation may help gauge treatment efficacy and inform the optimal treatment window in interventions to prevent or delay symptom onset in Alzheimer’s disease.

3. Future directions

Future work in larger samples with more follow-up visits will also evaluate additional risk factors that might modulate the age at onset of amyloid accumulation.

Supplementary Material



This research was supported in part by the Intramural Research Program of the National Institute on Aging, National Institutes of Health. We are grateful to the BLSA participants and staff for their dedication to these studies and the staff of the Johns Hopkins PET facility for their assistance. We thank Brieana Viscomi and Wendy Elkins for assistance with coordination and data collection for the PET studies.


Alzheimer’s disease
Baltimore Longitudinal Study of Aging
Cortical DVR
Distribution volume ratio
Hazard ratio
Mild cognitive impairment
Pittsburgh compound B


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, et al. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s Dement. 2011;7:280–92. doi: 10.1016/j.jalz.2011.03.003. [PMC free article] [PubMed] [Cross Ref]
2. Villemagne VL, Burnham S, Bourgeat P, Brown B, Ellis KA, Salvado O, et al. Amyloid β deposition, neurodegeneration, and cognitive decline in sporadic Alzheimer’s disease: a prospective cohort study. Lancet Neurol. 2013;12:357–67. doi: 10.1016/S1474-4422(13)70044-9. [PubMed] [Cross Ref]
3. Corder EH, Lannfelt L, Bogdanovic N, Fratiglioni L, Mori H. The role of APOE polymorphisms in late-onset dementias. Cell Mol Life Sci. 1998;54:928–34. doi: 10.1007/s000180050223. [PubMed] [Cross Ref]
4. Liu C-C, Kanekiyo T, Xu H, Bu G. Apolipoprotein E and Alzheimer disease: risk, mechanisms and therapy. Nat Rev Neurol. 2013;9:106–18. doi: 10.1038/nrneurol.2012.263. [PMC free article] [PubMed] [Cross Ref]
5. Vlassenko AG, Mintun MA, Xiong C, Sheline YI, Goate AM, Benzinger TLS, et al. Amyloid-beta plaque growth in cognitively normal adults: longitudinal [11C]Pittsburgh compound B data. Ann Neurol. 2011;70:857–61. doi: 10.1002/ana.22608. [PMC free article] [PubMed] [Cross Ref]
6. Reiman EM, Chen K, Liu X, Bandy D, Yu M, Lee W, et al. Fibrillar amyloid-β burden in cognitively normal people at 3 levels of genetic risk for Alzheimer’s disease. PNAS. 2009;106:6820–5. doi: 10.1073/pnas.0900345106. [PubMed] [Cross Ref]
7. Fleisher AS, Chen K, Liu X, Ayutyanont N, Roontiva A, Thiyyagura P, et al. Apolipoprotein E ε4 and age effects on florbetapir positron emission tomography in healthy aging and Alzheimer disease. Neurobiol Aging. 2013;34:1–12. doi: 10.1016/j.neurobiolaging.2012.04.017. [PubMed] [Cross Ref]
8. Sojkova J, Zhou Y, An Y, Kraut MA, Ferrucci L, Wong DF, et al. Longitudinal patterns of β-amyloid deposition in nondemented older adults. Arch Neurol. 2011;68:644–9. [PMC free article] [PubMed]
9. Villemagne VL, Pike KE, Chételat G, Ellis KA, Mulligan RS, Bourgeat P, et al. Longitudinal assessment of Aβ and cognition in aging and Alzheimer disease. Ann Neurol. 2011;69:181–92. doi: 10.1002/ana.22248. [PMC free article] [PubMed] [Cross Ref]
10. Vemuri P, Wiste HJ, Weigand SD, Knopman DS, Shaw LM, Trojanowski JQ, et al. Effect of apolipoprotein E on biomarkers of amyloid load and neuronal pathology in Alzheimer disease. Ann Neurol. 2010;67:308–16. doi: 10.1002/ana.21953. [PMC free article] [PubMed] [Cross Ref]
11. Morris JC, Roe CM, Xiong C, Fagan AM, Goate AM, Holtzman DM, et al. APOE predicts amyloid-beta but not tau Alzheimer pathology in cognitively normal aging. Ann Neurol. 2010;67:122–31. doi: 10.1002/ana.21843. [PMC free article] [PubMed] [Cross Ref]
12. Fouquet M, Besson FL, Gonneaud J, La Joie R, Chételat G. Imaging brain effects of APOE4 in cognitively normal individuals across the lifespan. Neuropsychol Rev. 2014;24:290–9. doi: 10.1007/s11065-014-9263-8. [PubMed] [Cross Ref]
13. Jagust WJ, Landau SM. Apolipoprotein E, not fibrillar β-amyloid, reduces cerebral glucose metabolism in normal aging. J Neurosci. 2012;32:18227–33. doi: 10.1523/JNEUROSCI.3266-12.2012. [PMC free article] [PubMed] [Cross Ref]
14. Murphy KR, Landau SM, Choudhury KR, Hostage CA, Shpanskaya KS, Sair HI, et al. Mapping the effects of ApoE4, age and cognitive status on 18F-florbetapir PET measured regional cortical patterns of beta-amyloid density and growth. Neuroimage. 2013;78:474–80. doi: 10.1016/j.neuroimage.2013.04.048. [PMC free article] [PubMed] [Cross Ref]
15. Mathis CA, Kuller LH, Klunk WE, Snitz BE, Price JC, Weissfeld LA, et al. In vivo assessment of amyloid-β deposition in nondemented very elderly subjects. Ann Neurol. 2013;73:751–61. doi: 10.1002/ana.23797. [PMC free article] [PubMed] [Cross Ref]
16. Hinrichs AL, Mintun MA, Head D, Fagan AM, Holtzman DM, Morris JC, et al. Cortical binding of Pittsburgh compound B, an endophenotype for genetic studies of Alzheimer’s disease. Biol Psychiatry. 2010;67:581–3. doi: 10.1016/j.biopsych.2009.09.012. [PMC free article] [PubMed] [Cross Ref]
17. Kantarci K, Lowe V, Przybelski SA, Weigand SD, Senjem ML, Ivnik RJ, et al. APOE modifies the association between Aβ load and cognition in cognitively normal older adults. Neurology. 2012;78:232–40. [PMC free article] [PubMed]
18. Roe C, Fagan A, Grant E, Hassenstab J, Moulder K, Dreyfus DM, et al. Amyloid imaging and CSF biomarkers in predicting cognitive impairment up to 7. 5 years later. Neurology. 2013;80:1784–91. doi: 10.1016/j.jalz.2013.05.1448. [PMC free article] [PubMed] [Cross Ref]
19. Mormino EC, Betensky RA, Hedden T, Schultz AP, Ward A, Huijbers W, et al. Amyloid and APOE ε4 interact to influence short-term decline in preclinical Alzheimer disease. Neurology. 2014;82:1760–7. doi: 10.1212/WNL.0000000000000431. [PMC free article] [PubMed] [Cross Ref]
20. Scheinin NM, Wikman K, Jula A, Perola M, Vahlberg T, Rokka J, et al. Cortical 11C-PIB uptake is associated with age, APOE genotype, and gender in “healthy aging. J Alzheimer’s Dis. 2014;41:193–202. doi: 10.3233/JAD-132783. [PubMed] [Cross Ref]
21. Resnick SM, Bilgel M, Moghekar A, An Y, Cai Q, Wang M-C, et al. Longitudinal changes in Aβ biomarkers and associations with APOE genotype in two longitudinal cohorts. Neurobiol Aging. 2015 In press. [PMC free article] [PubMed]
22. Jack CR, Wiste HJ, Weigand SD, Knopman DS, Vemuri P, Mielke MM, et al. Age, sex, and APOE ε4 effects on memory, brain structure, and β-amyloid across the adult life span. JAMA Neurol. 2015;72:511–9. doi: 10.1001/jamaneurol.2014.4821. [PMC free article] [PubMed] [Cross Ref]
23. Jansen WJ, Ossenkoppele R, Knol DL, Tijms BM, Scheltens P, Verhey FRJ, et al. Prevalence of cerebral amyloid pathology in persons without dementia. JAMA. 2015;313:1924–38. doi: 10.1001/jama.2015.4668. [PMC free article] [PubMed] [Cross Ref]
24. Brookmeyer R, Gray S, Kawas C. Projections of Alzheimer’s disease in the United States and the public health impact of delaying disease onset. Am J Public Heal. 1998;88:1337–42. doi: 10.2105/AJPH.88.9.1337. [PubMed] [Cross Ref]
25. Thambisetty M, An Y, Tanaka T. Alzheimer’s disease risk genes and the age-at-onset phenotype. Neurobiol Aging. 2013;34:2696.e1–5. doi: 10.1016/j.neurobiolaging.2013.05.028. [PMC free article] [PubMed] [Cross Ref]
26. Naj AC, Jun G, Reitz C, Kunkle BW, Perry W, Park YS, et al. Effects of multiple genetic loci on age at onset in late-onset Alzheimer disease: a genome-wide association study. JAMA Neurol. 2014;71:1394–404. doi: 10.1001/jamaneurol.2014.1491. [PMC free article] [PubMed] [Cross Ref]
27. Khachaturian AS, Corcoran CD, Mayer LS, Zandi PP, Breitner JCS. Apolipoprotein E ε4 count affects age at onset of Alzheimer disease, but not lifetime susceptibility. Arch Gen Psychiatry. 2004;61:518–24. [PubMed]
28. Resnick SM, Goldszal AF, Davatzikos C, Golski S, Kraut MA, Metter EJ, et al. One-year age changes in MRI brain volumes in older adults. Cereb Cortex. 2000;10:464–72. [PubMed]
29. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-III-R. 4. Washington, DC: 1987.
30. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34:939–44. [PubMed]
31. Jenkinson M, Bannister P, Brady M, Smith S. Improved optimization for the robust and accurate linear registration and motion correction of brain images. Neuroimage. 2002;17:825–41. doi: 10.1006/nimg.2002.1132. [PubMed] [Cross Ref]
32. Avants BB, Yushkevich P, Pluta J, Minkoff D, Korczykowski M, Detre J, et al. The optimal template effect in hippocampus studies of diseased populations. Neuroimage. 2010;49:2457–66. doi: 10.1016/j.neuroimage.2009.09.062. [PMC free article] [PubMed] [Cross Ref]
33. Jog A, Roy S, Carass A, Prince JL. Magnetic resonance image synthesis through patch regression. IEEE 10th Int. Symp. Biomed. Imaging; San Francisco, CA, USA. 2013; pp. 350–3. [PMC free article] [PubMed]
34. Desikan RS, Ségonne F, Fischl B, Quinn BT, Dickerson BC, Blacker D, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage. 2006;31:968–80. doi: 10.1016/j.neuroimage.2006.01.021. [PubMed] [Cross Ref]
35. Avants BB, Epstein CL, Grossman M, Gee JC. Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain. Med Image Anal. 2008;12:26–41. doi: 10.1016/ [PMC free article] [PubMed] [Cross Ref]
36. Zhou Y, Resnick SM, Ye W, Fan H, Holt DP, Klunk WE, et al. Using a reference tissue model with spatial constraint to quantify [11C]Pittsburgh compound B PET for early diagnosis of Alzheimer’s disease. Neuroimage. 2007;36:298–312. doi: 10.1016/j.neuroimage.2007.03.004. [PMC free article] [PubMed] [Cross Ref]
37. Hixson JE, Vernier DT. Restriction isotyping of human apolipoprotein E by gene amplification and cleavage with with HhaI. J Lipid Res. 1990;31:545–8. [PubMed]
38. Rodrigue KM, Kennedy KM, Devous MD, Sr, Rieck JR, Hebrank AC, Diaz-Arrastia R, et al. β-amyloid burden in healthy aging. Neurology. 2012;78:387–95. [PMC free article] [PubMed]
39. Johnson KA, Sperling RA, Gidicsin CM, Carmasin JS, Maye JE, Coleman RE, et al. Florbetapir (F18-AV-45) PET to assess amyloid burden in Alzheimer’s disease dementia, mild cognitive impairment, and normal aging. Alzheimer’s Dement. 2013;9:S72–83. doi: 10.1016/j.jalz.2012.10.007. [PMC free article] [PubMed] [Cross Ref]
40. Meyer MR, Tschanz JT, Norton MC, Welsh-Bohmer KA, Steffens DC, Wyse BW, et al. APOE genotype predicts when -not whether- one is predisposed to develop Alzheimer disease. Nat Genet. 1998;19:321–2. [PubMed]
41. Jack CR, Wiste HJ, Lesnick TG, Weigand SD, Knopman DS, Pankratz VS, et al. Brain beta-amyloid load approaches a plateau. Neurology. 2013;80:890–6. [PMC free article] [PubMed]
42. Yasuno F, Kazui H, Morita N, Kajimoto K, Ihara M, Taguchi A, et al. Low amyloid-β deposition correlates with high education in cognitively normal older adults: a pilot study. Int J Geriatr Psychiatry. 2014 Nov 26; doi: 10.1002/gps.4235. In press. [PubMed] [Cross Ref]
43. Eichner JE, Dunn ST, Perveen G, Thompson DM, Stewart KE, Stroehla BC. Apolipoprotein E polymorphism and cardiovascular disease: A HuGE review. Am J Epidemiol. 2002;155:487–95. [PubMed]