The Modulatory Effect of Sex on the Association between APOE and Neuropsychiatric Symptom Burden in Older Adults at Risk for Alzheimer's Disease.

Abstract

<h3>Introduction</h3> Apolipoprotein (APOE) ε4 is one of the strongest genetic biomarkers for Alzheimer's disease (AD) however, the relationship between APOE and Neuropsychiatric Symptoms (NPS) remains unclear. Our recent analyses suggested that female sex modulated the association of APOE ε4 on specific NPS domains such as psychosis¹ and nighttime behaviours² in an autosomal recessive manner. The goal of this new analysis was to investigate the associations between APOE and sex on overall NPS burden in a cohort of participants with a condition putting them at risk for AD: Mild cognitive impairment (MCI) and/or a history of Major Depressive Disorder (MDD). <h3>Methods</h3> Baseline Neuropsychiatric Inventory Questionnaire (NPI-Q), APOE and demographic data were obtained in 181 participants from an ongoing clinical trial, "Preventing Alzheimer's dementia with cognitive remediation plus transcranial direct current stimulation in mild cognitive impairment and depression (PACt-MD)". NPS burden was measured with the NPI-Q Total score and NPI-Severity Total. One extreme outlier was omitted from analysis (>7 inter quartile ranges (IQR) from the median NPI-Severity and >2 IQR from the next nearest data point). We tested whether APOE status, sex, or an interaction of the two were associated with our two outcomes using Ordinal Least Square (OLS) regression. Subsequent models were fit adjusting for the effects of age, education and diagnosis group (i.e., MCI or MDD (with/without MCI)). APOE ε3/ε3, the most common isoform, was used as the baseline category in regression analysis. All findings reported are compared to this baseliine category. <h3>Results</h3> When analysing the effects of APOE and sex alone, we only found a significant inverse association between NPI-Q Total and APOE ε3/ε4 carriers compared to APOE ε3/ε3 carriers (n = 38, <i>b</i> = -1.04, <i>t</i>(170) = -2.28, <i>p</i> = 0.024). When analysing our interaction terms APOE ε4/ε4 females (n = 6) were significantly associated with higher NPI-Q total scores (<i>b = 1.95, t(</i>171) = 2.14, <i>p</i> = 0.034). We also found that APOE ε2/ε3 females (n = 10) were associated with higher NPI-Q total scores <i>(b =</i> 1.55<i>, t(</i>171) = 2.04, <i>p</i> = 0.044). Results were consistent after adjusting for covariates. Sex or APOE status alone were not associated with NPI-Q Total severity scores. When analysing the interaction terms, APOE ε2/ε3 females showed higher NPI-Q Total severity scores than APOE ε3/ε3 carriers (<i>b = 3.57, t(</i>171) = 2.45, <i>p</i> = 0.015). After adjusting for covariates, being APOE ε2/ε3 females (<i>b</i> = 3.3, <i>t</i>(162) = 1.49, <i>p</i> = 0.027) or APOE ε4/ε4 females (<i>b</i> = 3.8, <i>t</i>(162) = 1.78 <i>p</i> = 0.033) was associated with greater NPI-Severity. <h3>Conclusions</h3> In older adults at risk for progression to AD, sex may play a modulatory role in the association between APOE and NPS burden. Consistent with recent findings¹^,², females who are homozygous for APOE ε4 may be susceptible to the highest NPS burden. Further investigation of the interactive effects of sex and APOE ε4 on NPS burden with larger samples are needed. <h3>Funding</h3> This Project has been made possible by Brain Canada through the Canada Brain Research Fund, with the financial support of Health Canada and the Chagnon Family.