Risk and Protective Factors for Cognitive Performance in Late Midlife

Introduction. Cognitive performance undergoes alterations already from middle adulthood, alterations which accelerate in later life, and may lead to dementia, such as Alzheimer’s Disease (AD) (Ferreira et al., 2015; Harada et al., 2013; Jack et al., 2018a). But the presence and extent of these cognitive changes are highly variable across cognitive domains and between individuals (Nyberg et al., 2012). Several risk and protective factors were suggested to explain this variability in age-related cognitive change, including modifiable factors (e.g., cognitive reserve, affective state, allostatic load, sleep quality), non-modifiable factors (sex, age, genetic predisposition), AD biomarkers (amyloid-beta, tau), and subjective cognitive decline (SCD) (Benedict et al., 2015; Da Silva et al., 2013; Hanseeuw et al., 2019; Jessen et al., 2020; Karlamangla et al., 2014; Livingston et al., 2017a; McFall et al., 2019; Salvato, 2015; Stern et al., 2018). However, most of the previous studies included only a handful of risk and protective factors, and rarely studied them together. Moreover, these factors were less often studied in late midlife (50-69 years), a time window associated with the first AD-related manifestations (Coupé et al., 2019).Objectives. Our main objective was to increase the theoretical and empirical knowledge about risk and protective factors for cognitive performance in late midlife. Specifically, we had three more specific objectives: (1) identify modifiable factors explaining the variability in global cognition in this age group; (2) verify if specific cognitive domains such as episodic memory, executive function and attention are related to modifiable and non-modifiable risk and protective factors; (3) investigate if subjective cognitive complaints are related to objective cognitive capacities, subclinical affective state, and amyloid-beta level, when they are considered together.Materials and Methods. We have recruited 101 healthy and cognitively normal individuals aged between 50 and 69 years, among which 68 women. We have realized extensive phenotyping of the participants, which included neuropsychological assessment, collection of blood and urinary samples for estimation of physiological measures and genotyping, structural Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET) for estimation of AD-related biomarkers (the proteins amyloid-beta and tau), actigraphy for rest-activity cycle evaluation, as well as self-assessment questionnaires for daily activities and lifestyle, affective state, and sleep quality. Results. Our results revealed that higher cognitive reserve and lower allostatic load are related to better global cognitive efficiency. Among the sub-factors of cognitive reserve, crystallized intelligence was the only one significantly related to global cognition, whereas among the sub-factors of allostatic load, only sympathetic nervous system functioning and lipid metabolism were related to global cognition. Moreover, higher cognitive reserve was the main correlate of better cognitive performance in episodic memory, and executive and attentional functioning. In addition, age was negatively associated with attentional functioning, whereas sex was a significant predictor for episodic memory, with women having better performance than men. Finally, we have demonstrated that higher level of cognitive complaints was significantly associated with lower episodic memory performance, worse affective state, and higher accumulation of amyloid-beta. Conclusion. The present results indicate that cognitive performance in late middle-aged population are associated with risk and protective factors. A combination of modifiable (cognitive reserve and allostatic load), non-modifiable (sex and age) factors, the presence of amyloid-beta burden, and cognitive complaints seem to predict global cognition and/or certain specific aspects related of cognitive functioning in late middle-aged population. While these results provide new interesting insights, we have confirmed only a part of our hypotheses. Future studies aiming to investigate risk and protective factors for cognitive performance should consider wider age range, differentially defined inclusion criteria, more comprehensive longitudinal assessment, and better refined measures, especially concerning sleep quality and the choice of PET radiotracer with higher specificity for tau.