Association of APOE4 genotype and treatment with cognitive outcomes in breast cancer survivors over time

This prospective longitudinal study of breast cancer survivors (n = 167) examined the association of apolipoprotein ε4 (APOE ε4) genotype with cognition and interactions with chemotherapy or endocrine therapy up to 6 years after treatment. In general, we found no effects of ε4 across timepoints and treatment exposures; post hoc analysis at 3–6 years suggested a trend towards worse cognition in the domains of attention and learning among ε4 carriers exposed to endocrine therapy. Further study is needed.

One-Carbon Metabolites, B Vitamin Intake, Apolipoprotein E Genotype, and Their Interactive Effects on Cognitive Performance: Secondary Outcomes of the REACH Cohort

Objectives To investigate associations between one-carbon metabolites and cognitive performance in older adults, and to examine the interactive effects of B vitamin intake, apolipoprotein E genotype, and one-carbon metabolites on cognition. Methods Three hundred and thirteen healthy older men and women (65–74 years, 65% female) were included in this secondary analysis of the REACH cohort. Cognitive performance was assessed by the Computerised Mental Performance Assessment System (COMPASS). Fasting plasma one-carbon metabolites (betaine, choline, cysteine, dimethylglycine, glycine, homocysteine, methionine, S-adenosylmethionine, serine) were quantified by ultra high performance liquid chromatography with tandem mass spectrometry, and four-day food records were analyzed for nutrient intake. Presence of the apolipoprotein E ε4 allele was measured by polymerase chain reaction amplification. Linear regression models were adjusted for age, sex, batch effects, education level, physical activity, energy intake and supplement use. Interaction terms were fit between continuous (metabolites) and categorical (quartiles of B vitamin intake or metabolites not fit as the main independent variable, presence of apolipoprotein ε4 allele) variables. Results Higher glycine concentrations were associated with better global cognitive performance (β = 1.340, P = 0.017), episodic memory (β = 1.396, P = 0.016) and location learning (β = 1.394, P = 0.027) in linear regression models, although this relationship was not apparent in participants with higher choline concentrations or the apolipoprotein ε4 genotype (interaction, P < 0.05). Conversely, the apolipoprotein ε4 genotype and higher vitamin B12 intake both attenuated the inverse association between Hcy and cognition across several domains of cognitive performance (interaction, P < 0.05). Conclusions The relationship between cognitive performance and one-carbon metabolites, notably glycine and homocysteine, is modified by vitamin B12 intake, apolipoprotein E genotype, and status of inter-connected metabolites. These findings point towards the need for a personalized approach to dietary interventions which protect against age-related cognitive decline. Funding Sources This work was supported by the Health Research Council of New Zealand and AgResearch Ltd.

NATURALISTIC DRIVING BEHAVIOR AS A NEUROBEHAVIORAL MARKER OF PRECLINICAL ALZHEIMER’S DISEASE

Decline in driving skills begins in preclinical AD, when an older adult remains cognitively normal, but the underlying disease process has begun. Preclinical AD is detectable among cognitively normal individuals using molecular biomarkers: positron emission tomography (PET) imaging and cerebrospinal fluid (CSF). The aim of this prospective, longitudinal study is to determine whether naturalistic driving behavior using in-vehicle dataloggers can distinguish older adults with (n=36) and without preclinical AD (n=134). Driving data was calculated as mean/month for several variables (number of trips/day, trip length, trip time, speeding, and hard-braking) for participants followed between one to 46 months. Using stepwise logistic regression, the area under the receiver operating curve (AUC) and 95% confidence interval for these five variables was 0.73 (0.63-0.79) in distinguishing those with and without preclinical AD via amyloid imaging. When age, gender, race, and education were added, the model improved: 0.80 (0.72-0.88). Finally, when apolipoprotein ε4 allele (APOε4), obtained via blood or saliva, was added to the model, accuracy improved: 0.84 (0.77-0.89). Similar results were found using CSF biomarker tau/Aβ42: AUCs (95% CI) were 0.68 (0.58-0.79) for driving variables alone, 0.77 (0.69-0.86) for driving variables and demographics, and 0.87 (0.80-0.94) driving variables, demographics, and apolipoprotein ε4 allele. These promising findings suggest that naturalistic driving behavior can predict those with and without preclinical AD. The AUC is further improved with demographics and APOε4, an easily obtainable genetic biomarker. This model may be used in clinical/research settings as a screen or adjunct for diagnostics and prognostics purposes.