Table_1_Exploring Genetic Associations of Alzheimer’s Disease Loci With Mild Cognitive Impairment Neurocognitive Endophenotypes.DOCX EspinosaAna Hernández-OlasagarreBegoña Moreno-GrauSonia KleineidamLuca Heilmann-HeimbachStefanie HernándezIsabel WolfsgruberSteffen WagnerHolger Rosende-RocaMaitée MauleónAna VargasLiliana LafuenteAsunción Rodríguez-GómezOctavio AbdelnourCarla GilSilvia MarquiéMarta Santos-SantosMiguel A. SanabriaÁngela OrtegaGemma Monté-RubioGemma PérezAlba IbarriaMarta RuizSusana KornhuberJohannes PetersOliver FrölichLutz HüllMichael WiltfangJens LuckTobias Riedel-HellerSteffi MontrrealLaura CañabatePilar MorenoMariola PrecklerSilvia AguileraNuria de RojasItziar OrellanaAdelina AlegretMontserrat ValeroSergi M. NöthenMarkus WagnerMichael JessenFrank TárragaLluis BoadaMercè RamírezAlfredo RuizAgustín 2018 <p>The role of genetic risk markers for Alzheimer’s disease (AD) in mediating the neurocognitive endophenotypes (NEs) of subjects with mild cognitive impairment (MCI) has rarely been studied. The aim of the present study was to investigate the relationship between well-known AD-associated single-nucleotide polymorphisms (SNPs) and individual NEs routinely evaluated during diagnosis of MCI, AD, and other dementias. The Fundació ACE (ACE) dataset, comprising information from 1245 patients with MCI, was analyzed, including the total sample, amnestic MCI (aMCI) (n = 811), and non-amnestic MCI (naMCI) (n = 434). As probable-MCI (Pr-MCI) patients with memory impairment have a higher risk of AD, which could influence the statistical power to detect genetic associations, the MCI phenotype was also stratified into four related conditions: Pr-aMCI (n = 262), Pr-naMCI (n = 76), possible (Pss)-aMCI (n = 549), and Pss-naMCI (n = 358). Validation analyses were performed using data from the German study on Aging, Cognition, and Dementia in primary care patients (AgeCoDe), and the German Dementia Competence Network (DCN). SNP associations with NEs were calculated in PLINK using multivariate linear regression analysis adjusted for age, gender, and education. In the total MCI sample, APOE-ε4 was significantly associated with the memory function NEs “delayed recall (DR)” (β = -0.76, p = 4.1 × 10<sup>-10</sup>), “learning” (β = -1.35, p = 2.91 × 10<sup>-6</sup>), and “recognition memory” (β = -0.58, p = 9.67 × 10<sup>-5</sup>); and with “DR” in the aMCI group (β = -0.36, p = 2.96 × 10<sup>-5</sup>). These results were confirmed by validation in the AgeCoDe (n = 503) and DCN (n = 583) datasets. APOE-ε4 was also significantly associated with the NE “learning” in individuals classified as having Pss-aMCI (β = -1.37, p = 5.82 × 10<sup>-5</sup>). Moreover, there was a near study-wide significant association between the HS3ST1 locus (rs6448799) and the “backward digits” working memory NE (β = 0.52, p = 7.57 × 10<sup>-5</sup>) among individuals with Pr-aMCI, while the AP2A2 locus (rs10751667) was significantly associated with the language NE “repetition” (β = -0.19, p = 5.34 × 10<sup>-6</sup>). Overall, our findings support specific associations of established AD-associated SNPs with MCI NEs.</p>