Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 1.
Published in final edited form as: Neurobiol Aging. 2015 Jan 14;36(4):1751–1756. doi: 10.1016/j.neurobiolaging.2015.01.008

Apolipoprotein E ε4 genotype and the temporal relationship between depression and dementia

Ida K Karlsson a, Anna M Bennet a, Alexander Ploner a, Therese M-L Andersson a, Chandra A Reynolds b, Margaret Gatzc c,a, Nancy L Pedersen a,c,*
PMCID: PMC4380668  NIHMSID: NIHMS655951  PMID: 25670333

Abstract

To investigate how apolipoprotein E (APOE) affects the temporal relationship between depression and dementia, we conducted a nested case-control study with longitudinal depression and dementia evaluations from several population studies, using 804 dementia cases and 1600 matched controls, totaling 1519 unique individuals.

Depression within ten years of dementia onset was strongly associated with dementia diagnosis regardless of APOE status (IRR 5.25, 95%CI 3.32-8.31 for ε4 carriers, IRR 4.40, 95%CI 3.23-5.99 for non-carriers). However, we found a significant interaction between depression more than ten years prior to dementia onset and APOE (p=0.01), with depression more distal to dementia being a risk factor only in ε4 carriers (IRR 3.39, 95%CI 1.69-6.78 for carriers, IRR 1.01, 95%CI 0.60-1.70 for non-carriers). Thus, depression with onset close in time to dementia onset is associated with disease irrespective of APOE genotype, while depression more distal to dementia onset is a risk factor only in ε4-carriers. This is the first study to show the interaction between APOE and depression to be dependent on timing of depression onset.

Keywords: dementia, Alzheimer's disease, depression, apolipoprotein E

1. Introduction

More than 24 million individuals are estimated to suffer from dementia worldwide, the most common form being Alzheimer's disease (AD), and the number is expected to double every 20 years (Ferri, et al., 2005). The strongest and most well-established genetic risk factor for AD is the apolipoprotein E (APOE) ε4 allele (Huang, 2010). Some of the other known risk factors are low education, hypertension, diabetes and depression (Barnes and Yaffe, 2011). The focus of the present inquiry was the interaction between APOE and depression in relation to dementia and AD.

Depression has long been recognized as a risk factor for dementia (Jorm, 2001). However, the mechanism behind the association remains unclear, and there is an ongoing debate as to whether depression is a risk factor for dementia or rather a prodrome of the disease. A study from a sample of Swedish twins showed that depression with a first onset within ten years of dementia was associated with disease, while depression with onset more than ten years before dementia was not (Brommelhoff, et al., 2009). Similarly, Li et al. (Li, et al., 2011) found depression with a first onset after the age of 50 to be associated with dementia, while depression with onset earlier in life was not. Others have found both midlife and late-life, as well as recurrent depression, to be associated with dementia (Barnes, et al., 2012), although the association was strongest for late-life and recurrent depression.

Taking this literature together, depression seems to increase the risk of dementia and to present most consistently during the prodromal stage, although the interpretation is far from clear. It has been suggested that APOE might play a role in this complex association. Four previous studies have analyzed the risk for dementia of having either only APOE ε4, only depression, or both risk factors (Irie, et al., 2008,Kim, et al., 2010,Meng and D'Arcy, 2012,Steffens, et al., 1997). Two found a significant interaction between APOE and late-life depression in men. Only individuals with both risk factors were at increased risk of dementia (Irie, et al., 2008,Kim, et al., 2010). The other two studies did not find a significant interaction although the pattern of results was similar (Meng and D'Arcy, 2012,Steffens, et al., 1997). Only one of the studies, using a sample of 142 twins, addressed the temporal relationship between depression and dementia, finding that APOE and late-onset depression were independent risk factors for AD and that the association between depression and dementia was stronger when depression onset was closer in time to dementia (Steffens, et al., 1997).

In the present study, we investigate the association between depression and dementia and how this association is affected by the presence of APOE ε4, while considering differences between depression in the preclinical phase of dementia versus more distal to dementia onset, and depression with its onset in mid- versus late-life.

2. Methods

2.1 Study population

The Swedish Twin Registry (STR) was established in the late 1950s and now contains more than 194,000 twins born in Sweden between 1886 and 2000 (Magnusson, et al., 2012). The participants in the present study include members from the STR who were ascertained similarly for dementia in three longitudinal and one cross-sectional study of aging: The Swedish Adoption/Twin Study of Aging (SATSA) (Finkel and Pedersen, 2004), Origins of Variance in the Oldest Old: Octogenarian Twins (OCTO-Twin) (McClearn, et al., 1997), Aging in Women and Men (GENDER) (Gold, et al., 2002), and The Study of Dementia in Swedish Twins (HARMONY) (Gatz, et al., 2005). All studies have been described in detail previously. Briefly, SATSA is an ongoing longitudinal study of twin pairs separated before the age of 11, matched with a sample of twins reared together. The study includes a face-to-face examination on a three-year rolling schedule. OCTO-Twin is a completed longitudinal study of 351 same-sex twin-pairs who were at least age 80 at baseline. All participated in at least one of five face-to-face examinations that occurred every two years. GENDER is a completed longitudinal study of 249 unlike-sex twin pairs, consisting of three face-to-face examinations every four years Participants in any of these longitudinal studies who at any wave showed indications of cognitive dysfunction were referred for a complete dementia evaluation. HARMONY is a cross-sectional study that started with a telephone screening for cognitive dysfunction of all twins in the STR aged 65 or older. All individuals who screened positive for cognitive dysfunction, their co-twins, and a control sample, were invited to participate in a clinical phase with physical and cognitive examination. In total, 2884 individuals received a complete dementia work-up through one of the four studies.

In the current study, we used a nested case-control design. For each dementia case we randomly selected two controls originating from the same study, matched on year of birth within two years and sex. Matching was within study so that cases and controls would be comparable in any procedural differences across studies. Cases and controls were not allowed to be co-twins, and controls had to still be participating in the study and be cognitively intact at age of dementia diagnosis in the case. The sample for analyses included 804 dementia cases and 1600 matched controls. To obtain a representative sample of exposure status in the population and person-time at risk we used incidence density sampling (Greenland and Thomas, 1982). Participants were hence allowed to serve as controls for more than one dementia case and cases were allowed to serve as controls until onset of dementia; the final study sample contains 1519 unique individuals (the number of cases and unique individuals from each study can be found in table S1).

All participants provided informed consent and this study was approved by both the Regional Ethics Board in Stockholm and the Institutional Review Board at the University of Southern California.

2.2 Assessment of dementia

Dementia ascertainment was performed similarly in all four studies (Gatz, et al., 1997). In brief, a cognitive screening was performed using either the Mini-Mental State Examination (MMSE) (Folstein, et al., 1975) or telephone screening with the TELE (Gatz, et al., 2002) with individuals and informants. Cutoffs were set to maximize sensitivity. Every twin in the STR aged 65 or older was invited for screening on at least one occasion. Screening was followed by a clinical work-up of suspected cases and their co-twins, including cognitive testing, physical and neurological examinations, informant interviews, reviews of medical records and laboratory tests. Final dementia diagnosis was set at multidisciplinary consensus conferences, according to DSM-III-R (American Psychiatric Association, 1987, 1987) or DSM-IV (American Psychiatric Association, 1987, 1994) criteria. Dementia was further differentially diagnosed into Alzheimer's disease according to the NINCDS/ADRDA criteria (McKhann, et al., 1984). For prevalent dementia cases, age at onset was assessed by informant interviews and review of medical records.

2.3 Assessment of depression

Information about depression was available from four sources, namely from national registries, medical records, information about antidepressant medication, and the Center for Epidemiologic Studies Depression (CES-D) scale (Radloff, 1977).

Depression diagnoses were gathered from the National Patient Registry (NPR) through linking based on the national personal number assigned to all inhabitants of Sweden. The NPR contains information about all in-patient care at public hospitals in Sweden. Diseases are classified according to International Classification of Diseases (ICD) codes (Socialstyrelsen, 2000). ICD codes used to identify depression were ICD-7 code 314.99, ICD8 codes 296.00, 298.00, 300.40-41, 790.20, ICD-9 codes 296C/D/W, 298A, 300E, 309A/B, 311X, and ICD-10 codes F32, F33, F34.1, F41.2.

Medical records from both in-patient and out-patient care were collected and reviewed as part of each study. Use of antidepressant medication was also available, both self-reported and from medical records, and included as another criterion for depression.

The CES-D scale, a 20-item self-report scale developed for epidemiological studies of depressive symptoms in the general population (Radloff, 1977), was administered during every testing occasion in the longitudinal studies. A cut-off point of 20 was used to indicate depression in the present study (Himmelfarb and Murrell, 1983).

For all participants, any occurrence of depression prior to dementia onset in the case was recorded. Individuals with no depression information from medical records, the CES-D scale, nor antidepressant information were considered missing for depression (n=16). For depression onset, age at the earliest record of onset was used.

The different studies contain slightly different depression information. CES-D scores were not available from prevalent dementia cases, including dementia cases identified by HARMONY. Depression from medical records was not available for all twins from SATSA. Because of the differences in depression information, we matched cases and controls on study of origin and included sources of depression information as a covariate. We also performed sensitivity analyses modeling each source of depression separately.

Depression was further categorized based on the temporal relation to dementia onset. For both cases and their controls, depression was categorized into first identified onset occurring within ten years, or more than ten years prior to age at dementia onset in the case.

Finally, we categorized depression as late-life depression if first record of onset was at the age of 60 or later, or as midlife depression if first record of onset was before the age of 60. In both categorizations, cases and controls with first depression onset after the age at dementia onset in the case were considered non-depressed.

2.4 APOE genotype

APOE genotype was dichotomized into ε4 carriers and non-carriers, where carriers had genotypes 34 and 44. All other genotypes were considered non-carriers, including 24, which was considered neutral due to the protective effect of the ε2 allele. Information about APOE status was available for 1516 out of the 1519 unique individuals included in the study.

2.5 Covariates

Covariates included level of education, stroke prior to age at dementia onset in the case, and sources of depression data. Education was dichotomized into 0-7 years versus more than 7 years of education (information on education was missing for 7 individuals). Information about prior stroke was identified through the NPR (ICD-7 codes 331-332, ICD-8 and ICD-9 codes 431-434, and ICD-10 codes I61 and I63).

2.6 Statistical analyses

Three different analyses were performed. First, the effect of APOE ε4 on dementia and AD was investigated. Secondly, we investigated the effect of depression on dementia and AD, with both any history of depression as the exposure and with depression categorized relative to dementia onset and according to age of first record of depression. The association between depression and dementia was modeled with and without APOE status as a covariate. Lastly, we investigated the association between the different depression categories—timing and age of onset—and dementia after stratification on APOE status. The interaction between APOE and depression was evaluated by using an interaction term (APOE × depression) in the model. All analyses were adjusted for education and previous stroke. Analyses with depression included in the model were also adjusted for source of depression data. Sex and age were controlled for by design.

Conditional logistic regression was performed in SAS 9.3 using the logistic procedure. Given the use of a nested case-control design, the regression coefficients were interpreted as log-incidence rate ratios (Rodrigues and Kirkwood, 1990). Since the use of twins across matched risk sets violates the assumption of independence between strata, we provide inference based on bootstrapped standard errors, where re-sampling with replacement was performed on the population of matched risk sets. The resulting p-values and confidence intervals are conservatively corrected for potential correlation between risk sets, but suffer from lack of convergence and are provided only as auxiliary evidence in the supplement.

Additional sensitivity analyses are presented in the supplementary section with 1) each source of depression modeled separately as an exposure for dementia outcome, 2) each of the four studies modeled separately, 3) individuals with a prior stroke excluded, and 4) depression categorized based only on CES-D scores, medical records and diagnoses in the NPR, and further stratified on untreated or treated with antidepressant medication.

3. Results

3.1 Population characteristics

Of the 804 dementia cases, 469 had the differential diagnosis of AD. Among the 1519 unique individuals included in the study, 424 met the criteria for depression at some point. Of these, 332 were categorized as having had their first identified depression within ten years of dementia development and 92 as having had depression more than ten years prior to dementia. 388 individuals were categorized as having had late-life depression, and 36 as having had midlife depression. 520 individuals were carriers of the ε4 allele and 996 were non-carriers. Dementia and AD cases were significantly more likely to have low education, to have met criteria for depression, and to be carriers of the APOE ε4 allele (Table 1).

Table 1. Demographic characteristics of dementia cases and controls.

No dementiaa (n=715) All dementia (n=804) Alzheimer's disease (n=469)
Age at baseline, mean (SD) 78.9 (6.6) 80.1 (6.6)* 81.0 (6.4)
Age at death, mean (SD) 88.2 (6.0) 85.8 (5.8)* 86.4 (5.8)*
Female sex, n (%) 424 (59.3) 515 (64.1) 333 (71)*
Low education, n (%) 441 (61.7) 569 (71.3)* 341 (73.5)*
Prior stroke, n (%)b 34 (4.8) 70 (8.7)* 12 (2.6)
Depression, n (%)c 131 (18.6) 293 (36.7)* 155 (33.3)*
APOE ε4 carriers, n (%)d 168 (23.5) 352 (44.0)* 224 (48.0)*
Open in a new tab

Abbreviations: SD, standard deviation; APOE ε4, apolipoprotein E ε4.

Number of exposed individuals (percent of number of individuals with covariate data) for binary variables, and mean age (standard deviation) for continuous variables.

a

Unique individuals in the control group

b

For controls, stroke prior to dementia onset of the matched case with the highest age at dementia onset

c

Out of 1503 individuals with available depression data. For controls, depression prior to dementia onset of the matched case with the highest age at dementia onset.

d

Out of 1516 individuals with available APOE genotype. APOE ε4 carriers are defined as genotypes 34 and 44.

*

Significant difference compared to the no dementia group, p=0.05.

Mean age at dementia onset was 78 years. When stratified on APOE ε4 status, there was a significant difference in age at dementia onset between non-carriers and carriers of the risk-allele (p<0.0001, mean age at onset 79 and 77 respectively). Mean age at depression onset was 74 years. Among dementia cases, mean age of onset was 78 years (range 58-90) for depression within ten years of dementia onset, and 58 years (range 23-76) for depression more than ten years prior. For the controls in the study, the corresponding numbers were 77 years (range 60-89) and 63 years (range 38-79), respectively.

3.2 Association between APOE and dementia

As expected, carriers of the APOE ε4 allele had significantly increased rate of dementia development (IRR 1.96, 95% CI 1.64-2.35) as well as AD (IRR 2.64, 95% CI 2.06-3.39) after adjusting for education and prior stroke. These results did not change after bootstrapping (Table S2).

3.3 Association between depression and dementia

A significant association was found between depression and both dementia and AD after adjusting for education, prior stroke and sources of depression data (Table 2, column 1). This association remained significant after including APOE as a covariate (Table 2, column 2).

Table 2. The association between depression and dementia/Alzheimer's disease, and the interaction with APOE genotype.

Dementia Total sample Total sample APOE × depression interaction model
Adjusted for APOE ε4 non-carriersa ε4 carriersa p-value
n=2404 n=2404 n=1597 n=801
No depression 1 1 1 1
Any depression 3.41 (2.72-4.27) 3.51 (2.78-4.41) 3.07 (2.34-4.02) 4.68 (3.16-6.93) 0.07
Depression within 10 years of dementia 4.46 (3.44-5.76) 4.67 (3.59-6.08) 4.40 (3.23-5.99) 5.25 (3.32-8.31) 0.51
Depression >10 years before dementia 1.58 (1.07-2.34) 1.55 (1.04-2.31) 1.01 (0.60-1.70) 3.39 (1.69-6.78) 0.01
Late-life depression (≥60) 3.56 (2.81-4.51) 3.67 (2.88-4.68) 3.26 (2.46-4.33) 4.79 (3.16-7.26) 0.12
Midlife depression (<60) 2.43 (1.35-4.35) 2.44 (1.34-4.44) 1.74 (0.80-3.79) 4.21 (1.53-11.58) 0.17
Alzheimer's disease Total sample Total sample APOE × depression interaction model
Adjusted for APOE ε4 non-carriersa ε4 carriersa p-value
n=2404 n=2404 n=1597 n=801
No depression 1 1 1 1
Any depression 2.89 (2.09-4.00) 3.06 (2.18-4.30) 2.50 (1.69-3.70) 5.29 (2.75-10.18) 0.04
Depression within 10 years of dementia 3.45 (2.39-4.98) 3.76 (2.56-5.53) 3.57 (2.28-5.59) 4.49 (2.20-9.15) 0.58
Depression >10 years before dementia 1.75 (1.01-3.03) 1.69 (0.95-3.01) 0.90 (0.43-1.89) 9.60 (2.22-41.44) 0.01
Late-life depression (≥60) 2.93 (2.08-4.12) 3.09 (2.16-4.43) 2.57 (1.70-3.87) 5.08 (2.59-9.93) 0.07
Midlife depression (<60) 2.63 (1.13-6.09) 2.81 (1.17-6.76) 2.03 (0.74-5.52) 9.49 (0.98-92.06) 0.22
Open in a new tab

Abbreviations: APOE ε4, apolipoprotein E ε4

Hazard ratios (95% confidence intervals) of dementia/Alzheimer's disease in the presence of depression, for the total sample and stratified by APOE ε4 status.

The model is adjusted for education, prior stroke and sources of depression data available.

a

APOE ε4 carriers are defined as genotypes 34 and 44, all other genotypes are considered non-carriers.

The association with dementia and AD was much stronger for depression occurring within ten years of dementia onset in the case compared to depression more than ten years earlier (Table 2, column 1 rows 3-4 and 9-10). After including APOE as a covariate, depression within ten years of dementia remained a significant predictor of both dementia and AD; the effect of depression more than ten years prior to dementia onset was non-significant for AD.

When depression was categorized according to age at first onset, both late-life and midlife depression was shown to be associated with dementia and AD (Table 2, column 1 rows 5-6 and 11-12). This effect remained significant after including APOE status as a covariate.

After bootstrapping, the effect of depression within ten years of dementia and late-life depression remained significant. However, the effect of depression more than ten years prior to dementia onset and midlife depression were no longer significant (Table S3, column 1).

3.4 Interaction between APOE and depression on dementia outcome

There was a statistically significant interaction between overall depression and APOE on AD, but not dementia, although both carriers and non-carriers with depression showed an increased rate of dementia and AD compared to non-depressed individuals in the same genotype category (Table 2, column 3-4).

After categorizing depression into first onset occurring more or less than ten years prior to dementia onset, incidence rate of developing dementia and AD after a recent onset depression were significantly increased in both carriers and non-carriers of ε4 (Table 2, column 3-4 rows 3-4 and 9-10).

However, there was a statistically significant interaction between APOE and depression more than ten years before dementia onset for both dementia and AD. Carriers of the ε4 allele with a depression more than ten years before dementia onset had more than three times higher incidence of dementia and more than nine times higher incidence of AD, while the incidence for non-carriers of ε4 with a depression more than ten years before dementia development was not increased (Table 2, column 3-4 rows 3-4 and 9-10).

When the analyses were repeated with depression stratified into onset before or after the age of 60, the general pattern was similar, but the differences between groups less pronounced (Table 2, column 3-4 rows 5-6 and 11-12). Neither depression late in life nor midlife was found to significantly interact with APOE genotype for dementia or AD.

Bootstrapping results showed the same pattern of significances for the interaction model, except for midlife depression and depression more than 10 years before onset of AD, both of which displayed lack of convergence and extremely wide confidence intervals. (Table S3, column 3-5).

Sensitivity analyses with each source of depression modeled separately lacked power but generally showed the same pattern as the main results (Table S4). The exception was depression assessed by review of medical records, for which depression within ten years of dementia and in late life also significantly interacted with APOE. Furthermore, the estimates for the effect of depression assessed by the CES-D scale on dementia were low for all groups. Sensitivity analyses modelling each of the four studies separately generally showed a similar pattern for the point estimates, but suffered from low power with wide confidence intervals limiting the interpretation (Table S5). The SATSA study showed generally lower estimates as compared to the other studies.

Sensitivity analysis excluding stroke cases did not change the results or interpretations (Table S6). Sensitivity analyses modelling treated and untreated depression separately showed a statistically significant interaction between APOE and untreated, but not treated depression (table S5).

4. Discussion

Our results suggest that depression with onset close in time to dementia, and depression with first known onset late in life, increase the rate of dementia irrespective of APOE genotype, while depression more than ten years before dementia onset is a risk factor only in carriers of APOE ε4. These findings offer further evidence regarding the temporal relationship between depression and dementia, while also highlighting the complex nature of the association between the two diseases, especially the role of genetic risk.

While the association between late-life depression and dementia has been rather consistent across studies, there have been discrepancies between studies of the association between early or midlife depression and dementia (Byers and Yaffe, 2011). We were able to provide further support to the relationship between late-life depression and dementia by showing this association to be robust and unaffected by APOE status. In these individuals, there remains the possibility that depression is a prodrome of the dementing disease. We were also able to suggest a resolution for the inconsistencies between studies of the association between early or midlife depression and dementia, by offering APOE status as an explanatory factor.

The results also highlight the differences between depression in late life and that in midlife. Compared to depression earlier in life, late-onset depression is known to present with a different symptomology as well as a different etiology. Late onset depression is less likely to be associated with a family history of depression, and more likely to include cognitive deficits, white matter intensities and a vascular pathology (Fiske, et al., 2009).

There are several mechanisms by which depression could increase the risk of dementia. Because depression increases the risk of vascular disease, it could thereby increase the vascular risk of dementia (Byers and Yaffe, 2011). Alternatively, depression could increase stress-induced increase of glucocorticoid levels, which can lead to decreased neurogenesis, reduction of neurotrophic factors, hippocampal atrophy, and possibly also to increased levels of Aβ formation (Byers and Yaffe, 2011,Jorm, 2001). Inflammatory changes have also been shown in depression and are known to play a central role in both dementia and vascular disease (Byers and Yaffe, 2011,Hansson and Libby, 2006).

Eriksson et al. (Eriksson, et al., 2010) demonstrated that, among those with nonstroke cardiovascular disease, only carriers of APOE ε4 are at increased risk of developing AD. Similarly, our results indicate that only carriers of the APOE ε4 allele have increased vulnerability to dementia after suffering from depression earlier in life. APOE is associated with dementia through several mechanisms, including neurotoxicity and impaired neural plasticity (Vance and Hayashi, 2010). The latter could lead to decreased ability of the brain to cope with the detrimental effects of depression, thereby magnifying other underlying mechanisms for APOE ε4 carriers.

The results presented here are based on a large cohort of twins with long follow-up time and enough information available to distinguish between depression with onset within or more than ten years prior to dementia. The chief limitation is that ascertainment of depression comes from a variety of sources; the CES-D scale, medical records, antidepressant information, and diagnoses from the National Patient Registry. The latter only contains patients who receive a depression diagnosis whilst being admitted to a hospital, who are, naturally, more likely to have a severe depression. In general, geriatric depression has been shown to have a milder symptom load (Fiske, et al., 2009), and for those with only NPR as a source of depression information, some cases of geriatric depression may hence have been missed. Antidepressant medication is used not only in depression but also for other psychiatric disorders, and this source of depression may therefore include some other psychiatric disorders (Thronson and Pagalilauan, 2014). Prior depression could not be assessed in prevalent dementia cases through the CES-D scale, which was available prior to dementia onset only for participants in the longitudinal studies, and for a small proportion of participants medical records were not reviewed. To deal with these issues, we matched on study origin and included source of depression data as a covariate. Furthermore, although sensitivity analyses where each source of depression information was modeled separately lacked power, all sources showed the same general patterns as the main analyses.

Estimation of the age at depression onset may also be biased. For cases identified through the NPR, the dates of hospitalization were used for age of onset. For cases identified by medical records, it was not possible to go back to early midlife for all individuals. For cases identified from CES-D scores, the age of depression onset was determined by the age at the interview when the CES-D was completed. Due to the rolling scheme of testing occasions, it is possible that depressive episodes occurring between testing occasions that were missed. Furthermore, for individuals meeting the criterion for depression according to the CES-D scale at baseline but with no record of depression from other sources, age at onset may have been earlier than as assessed by age at interview. Information about depression before mid-life was only available from the NPR. The registry has a low sensitivity for depression, since not all depressed individuals are ever admitted to hospital. Further, the registry did not have national coverage until 1987. In general, then, the earliest age of onset is therefore probably earlier than identified by our data, and early-onset depression is likely under-ascertained in this study. Additionally, individuals categorized as having had their first depression onset in late life, or within ten years of dementia, may have had a prior episode of depression that would have been missed. However, misclassification of depression as late-life instead of midlife would drive the estimates toward the null-hypothesis for both the association between recent depression and dementia and for the interaction between APOE and depression more than ten years prior to dementia.

In conclusion, we provide new data in a large population-based sample about the interaction of depression and APOE as risk factors for dementia and AD. To the best of our knowledge, this is the first study to show APOE to affect the association between depression and dementia differently depending on the timing of depression onset in relation to dementia. Depression with onset close in time to dementia onset was strongly associated with disease and not affected by APOE genotype. However, the association between depression more distal to dementia onset and occurrence of dementia seems to be more complex, and was evident only in the presence of the APOE ε4 risk allele.

Supplementary Material

supplement

Table S1: Number of cases and unique individuals, and mean age at baseline for the four different studies

Table S2: Bootstrap analysis: The association between Apolipoprotein E genotype and dementia/Alzheimer's disease

Table S3: Bootstrap analysis: The association between depression and dementia/Alzheimer's disease and the interaction with Apolipoprotein E genotype

Table S4: The temporal association between depression and dementia, and the interaction with Apolipoprotein E genotype, for each source of depression separately

Table S5: The temporal association between depression and dementia, and the interaction with Apolipoprotein E genotype, for each study separately

Table S6: The association between depression and dementia, and the interaction with Apolipoprotein E genotype after excluding stroke cases

Table S7: The association between depression and dementia and the interaction with Apolipoprotein E genotype, stratified on treated or untreated depression

Highlights.

  • Nested case-control study of 804 dementia cases and 1600 matched controls

  • Investigated how APOE ε4 modifies the association between depression and dementia

  • We differentiated between depression close to and more distal to dementia onset

  • Depression close to dementia increase dementia rate irrespective of APOE genotype

  • Depression more distal to dementia is a risk factor only in APOE ε4 carriers

Acknowledgments

This work was supported by U.S. National Institutes of Health (Grants AG028555, AG08724, AG08861, AG10175, and AG04563), The MacArthur Foundation Research Network on Successful Aging, The Axel and Margaret Ax:son Johnsons Foundation, The Swedish Council for Social Research, The Swedish Foundation for Health Care Sciences and Allergy Research, The Strategic Research Program in Epidemiology at Karolinska Institutet, and the Eva and Oscar Ahrén Research Foundation.

Dr Pedersen had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Footnotes

Financial disclosures: The authors have no financial or personal relationships or affiliations that could inappropriately influence their decisions, work, or manuscript.

Author Contributions: IK performed the statistical analysis and wrote the manuscript. AP and TA contributed to the statistical analyses. AB, CR, MG, and NP were responsible for the conceptualization and implementation of the study. MG, CR, and NP were responsible for the acquisition of data. AB, AP, TA, CR, MG and NP critically appraised the draft.

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.

Contributor Information

Ida K. Karlsson, Email: Ida.Karlsson@ki.se.

Anna M. Bennet, Email: Anna.Bennet@ki.se.

Alexander Ploner, Email: Alexander.Ploner@ki.se.

Therese M.-L. Andersson, Email: Therese.M-L.Andersson@ki.se.

Chandra A. Reynolds, Email: Chandra.Reynolds@ucr.edu.

Margaret Gatzc, Email: Gatz@dornsife.usc.edu.

References

  1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders : DSM-III-R. Washington, DC: American Psychiatric Association; 1987. [Google Scholar]
  2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders : DSM-IV. Washington, DC: American Psychiatric Association; 1994. [Google Scholar]
  3. Barnes DE, Yaffe K. The projected effect of risk factor reduction on Alzheimer's disease prevalence. Lancet Neurol. 2011;10(9):819–28. doi: 10.1016/S1474-4422(11)70072-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barnes DE, Yaffe K, Byers AL, McCormick M, Schaefer C, Whitmer RA. Midlife vs late-life depressive symptoms and risk of dementia: differential effects for Alzheimer disease and vascular dementia. Arch Gen Psychiatry. 2012;69(5):493–8. doi: 10.1001/archgenpsychiatry.2011.1481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Brommelhoff JA, Gatz M, Johansson B, McArdle JJ, Fratiglioni L, Pedersen NL. Depression as a risk factor or prodromal feature for dementia? Findings in a population-based sample of Swedish twins. Psychol Aging. 2009;24(2):373–84. doi: 10.1037/a0015713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Byers AL, Yaffe K. Depression and risk of developing dementia. Nat Rev Neurol. 2011;7(6):323–31. doi: 10.1038/nrneurol.2011.60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Eriksson UK, Bennet AM, Gatz M, Dickman PW, Pedersen NL. Nonstroke cardiovascular disease and risk of Alzheimer disease and dementia. Alzheimer Dis Assoc Disord. 2010;24(3):213–9. doi: 10.1097/WAD.0b013e3181d1b99b. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M, Hall K, Hasegawa K, Hendrie H, Huang Y, Jorm A, Mathers C, Menezes PR, Rimmer E, Scazufca M. Global prevalence of dementia: a Delphi consensus study. Lancet. 2005;366(9503):2112–7. doi: 10.1016/S0140-6736(05)67889-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Finkel D, Pedersen N. Processing Speed and Longitudinal Trajectories of Change for Cognitive Abilities: The Swedish Adoption/Twin Study of Aging. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 2004;11(2):325–45. doi: 10.1080/13825580490511152. [DOI] [Google Scholar]
  10. Fiske A, Wetherell JL, Gatz M. Depression in older adults. Annu Rev Clin Psychol. 2009;5:363–89. doi: 10.1146/annurev.clinpsy.032408.153621. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatric Res. 1975;12(3):189–98. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  12. Gatz M, Fratiglioni L, Johansson B, Berg S, Mortimer JA, Reynolds CA, Fiske A, Pedersen NL. Complete ascertainment of dementia in the Swedish Twin Registry: the HARMONY study. Neurobiol Aging. 2005;26(4):439–47. doi: 10.1016/j.neurobiolaging.2004.04.004. [DOI] [PubMed] [Google Scholar]
  13. Gatz M, Pedersen NL, Berg S, Johansson B, Johansson K, Mortimer JA, Posner SF, Viitanen M, Winblad B, Ahlbom A. Heritability for Alzheimer's disease: the study of dementia in Swedish twins. J Gerontol A Biol Sci Med Sci. 1997;52(2):M117–25. doi: 10.1093/gerona/52a.2.m117. [DOI] [PubMed] [Google Scholar]
  14. Gatz M, Reynolds CA, John R, Johansson B, Mortimer JA, Pedersen NL. Telephone screening to identify potential dementia cases in a population-based sample of older adults. Int Psychogeriatr. 2002;14(3):273–89. doi: 10.1017/s1041610202008475. [DOI] [PubMed] [Google Scholar]
  15. Gold CH, Malmberg B, McClearn GE, Pedersen NL, Berg S. Gender and health: a study of older unlike-sex twins. J Gerontol B Psychol Sci Soc Sci. 2002;57(3):S168–76. doi: 10.1093/geronb/57.3.s168. [DOI] [PubMed] [Google Scholar]
  16. Greenland S, Thomas DC. On the need for the rare disease assumption in case-control studies. American journal of epidemiology. 1982;116(3):547–53. doi: 10.1093/oxfordjournals.aje.a113439. [DOI] [PubMed] [Google Scholar]
  17. Hansson GK, Libby P. The immune response in atherosclerosis: a double-edged sword. Nat Rev Immunol. 2006;6(7):508–19. doi: 10.1038/nri1882. [DOI] [PubMed] [Google Scholar]
  18. Himmelfarb S, Murrell SA. Reliability and validity of five mental health scales in older persons. J Gerontol. 1983;38(3):333–9. doi: 10.1093/geronj/38.3.333. [DOI] [PubMed] [Google Scholar]
  19. Huang Y. Mechanisms linking apolipoprotein E isoforms with cardiovascular and neurological diseases. Curr Opin Lipidol. 2010;21(4):337–45. doi: 10.1097/MOL.0b013e32833af368. [DOI] [PubMed] [Google Scholar]
  20. Irie F, Masaki KH, Petrovitch H, Abbott RD, Ross GW, Taaffe DR, Launer LJ, White LR. Apolipoprotein E epsilon4 allele genotype and the effect of depressive symptoms on the risk of dementia in men: the Honolulu-Asia Aging Study. Arch Gen Psychiatry. 2008;65(8):906–12. doi: 10.1001/archpsyc.65.8.906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Jorm AF. History of depression as a risk factor for dementia: an updated review. Aust N Z J Psychiatry. 2001;35(6):776–81. doi: 10.1046/j.1440-1614.2001.00967.x. [DOI] [PubMed] [Google Scholar]
  22. Kim JM, Kim SY, Bae KY, Kim SW, Shin IS, Yang SJ, Song YH, Yoon JS. Apolipoprotein e4 genotype and depressive symptoms as risk factors for dementia in an older korean population. Psychiatry Investig. 2010;7(2):135–40. doi: 10.4306/pi.2010.7.2.135. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Li G, Wang LY, Shofer JB, Thompson ML, Peskind ER, McCormick W, Bowen JD, Crane PK, Larson EB. Temporal relationship between depression and dementia: findings from a large community-based 15-year follow-up study. Arch Gen Psychiatry. 2011;68(9):970–7. doi: 10.1001/archgenpsychiatry.2011.86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Magnusson PK, Almqvist C, Rahman I, Ganna A, Viktorin A, Walum H, Halldner L, Lundstrom S, Ullen F, Langstrom N, Larsson H, Nyman A, Gumpert CH, Rastam M, Anckarsater H, Cnattingius S, Johannesson M, Ingelsson E, Klareskog L, de Faire U, Pedersen NL, Lichtenstein P. The Swedish Twin Registry: Establishment of a Biobank and Other Recent Developments. Twin Res Hum Genet. 2012;16(1):317–29. doi: 10.1017/thg.2012.104. [DOI] [PubMed] [Google Scholar]
  25. McClearn GE, Johansson B, Berg S, Pedersen NL, Ahern F, Petrill SA, Plomin R. Substantial genetic influence on cognitive abilities in twins 80 or more years old. Science. 1997;276(5318):1560–3. doi: 10.1126/science.276.5318.1560. [DOI] [PubMed] [Google Scholar]
  26. 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(7):939–44. doi: 10.1212/wnl.34.7.939. [DOI] [PubMed] [Google Scholar]
  27. Meng X, D'Arcy C. Apolipoprotein E gene, environmental risk factors, and their interactions in dementia among seniors. Int J Geriatr Psychiatry. 2012;28(10):1005–14. doi: 10.1002/gps.3918. [DOI] [PubMed] [Google Scholar]
  28. Radloff LS. The CES-D Scale: A Self-Report Depression Scale for Research in the General Population. Appl Psychol Meas. 1977;1(3):385–401. doi: 10.1177/014662167700100306. [DOI] [Google Scholar]
  29. Rodrigues L, Kirkwood BR. Case-control designs in the study of common diseases: updaes on the demise of the rare disease assumption and the choice of sampling scheme for controls. Int J Epidemiol. 1990;19(1):205–13. doi: 10.1093/ije/19.1.205. [DOI] [PubMed] [Google Scholar]
  30. Socialstyrelsen. [Cited 2014 Aug 28];Värdering av diagnoskvaliteten för akut hjärtinfarkt i patientregistret 1987 och 1995. 2000 Avaliable at: http://www.socialstyrelsen.se/publikationer2000/2000-0-100.
  31. Steffens DC, Plassman BL, Helms MJ, Welsh-Bohmer KA, Saunders AM, Breitner JC. A twin study of late-onset depression and apolipoprotein E epsilon 4 as risk factors for Alzheimer's disease. Biol Psychiatry. 1997;41(8):851–6. doi: 10.1016/S0006-3223(96)00247-8. [DOI] [PubMed] [Google Scholar]
  32. Thronson LR, Pagalilauan GL. Psychopharmacology. Med Clin North Am. 2014;98(5):927–58. doi: 10.1016/j.mcna.2014.06.001. [DOI] [PubMed] [Google Scholar]
  33. Vance JE, Hayashi H. Formation and function of apolipoprotein E-containing lipoproteins in the nervous system. Biochim Biophys Acta. 2010;1801(8):806–18. doi: 10.1016/j.bbalip.2010.02.007. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement

Table S1: Number of cases and unique individuals, and mean age at baseline for the four different studies

Table S2: Bootstrap analysis: The association between Apolipoprotein E genotype and dementia/Alzheimer's disease

Table S3: Bootstrap analysis: The association between depression and dementia/Alzheimer's disease and the interaction with Apolipoprotein E genotype

Table S4: The temporal association between depression and dementia, and the interaction with Apolipoprotein E genotype, for each source of depression separately

Table S5: The temporal association between depression and dementia, and the interaction with Apolipoprotein E genotype, for each study separately

Table S6: The association between depression and dementia, and the interaction with Apolipoprotein E genotype after excluding stroke cases

Table S7: The association between depression and dementia and the interaction with Apolipoprotein E genotype, stratified on treated or untreated depression

NIHMS655951-supplement.docx (46KB, docx)

RESOURCES