Nudelman, Kelly N. H., Jackson, Trever, Rumbaugh, Malia C., Eloyan, Ani, Pentchev, Julian V., Faber, Kelley M., Snoddy, Casey, Foroud, Tatiana M., Carrillo, Maria C., Dickerson, Brad C., Rabinovici, Gil D., and Apostolova, Liana G.
Background: Only a small percentage of early‐onset Alzheimer's disease (EOAD) participants are known to carry autosomal dominant pathogenic variants in the amyloid beta precursor protein (APP) gene or in presenilin 1 or 2 (PSEN1/2); more research is needed to identify additional causative genetic variants. This study presents results of the initial genetic analyses in the Longitudinal Early Onset Alzheimer's Disease Study (LEADS) and planned next steps. Methods: Whole exome sequencing (WES) is being performed for all cognitively impaired participants, including individuals with EOAD or frontotemporal dementia. Sequencing data is processed using GATK best practices and all variants for APP, PSEN1/2, progranulin (GRN), and microtubule associated protein tau (MAPT) are reviewed for previously reported pathogenic variants. C9ORF72 is also assessed for pathogenic repeat expansions. WES was analyzed with CANOES and CoNIFER for copy number variants (CNVs). WES data has been reviewed for reported pathogenic variants in other neurodegenerative disease‐associated genes. Gene burden testing is being done for the LEADS cases and age‐matched controls from the Parkinson's Progression Markers Initiative for rare variants in these genes. Results: To date, WES has been generated for 301 cases, and GWAS for 384 participants (290 cases, 89 controls). Reported pathogenic variants or repeat expansions in C9ORF72, PSEN1, GRN, and MAPT were identified in 2% of cases for the six screened genes. No CNVs were detected in APP, PSEN1/2, GRN, or MAPT. Review also identified individuals with pathogenic variants or CNVs in genes linked with diseases including Parkinson's disease and Niemann‐Pick disease, including reported or predicted pathogenic variants in GBA, LRRK2, and SMPD1. Surprisingly, results do not show enrichment of functional TREM2 variants compared to population frequency. Conclusions: Initial results indicate that LEADS is strongly enriched for novel genetic risk and/or causative factors, as only 2% have a reported pathogenic variant or repeat expansion in APP, PSEN1/2, GRN, MAPT, or C9ORF72. Review of reported pathogenic variants in other neurodegenerative disease‐associated genes indicates that while a small portion of disease etiology may be explained by these variants, more research is needed to identify additional causative and risk factors contributing to EOAD. [ABSTRACT FROM AUTHOR]