Case-Case Genome-Wide Association Study of Age-Related Cancer and Alzheimer’s Disease

Background: Research over the past five years has strengthened in support of an inverse epidemiological correlation between Alzheimer’s disease (AD) and cancer--individuals with cancer are less likely to develop AD and those with AD have reduced cancer risk. Since cancer is characterized by uncontrolled cell division, and AD by neuronal death (and limited neuronal regeneration), this inverse relationship may point to dysregulation in some common underlying pathways. Here, we aim to investigate the genetic underpinnings of this unique relationship which have not been fully explored, using a unique case-case genome-wide association study (GWAS) design between an AD and cancer cohort. Hypothesis: We hypothesize that suggestive association signals will be observed when comparing the AD to cancer group, with the most interesting signals being those that are stronger when comparing cases-to-cases than when comparing cases-to-controls. Methods: Genome-wide SNP data for AD, Cancer, and Control groups were created using two publically available datasets: Breast Cancer (BrCa) and Prostate Cancer (PCa) Cohort Consortium and Alzheimer’s Disease Neuroimaging Initiative. Breast and prostate cancer were combined to form the Cancer group, which according to Cancer Research UK, are the most prevalent forms of adult and elderly cancers. All samples were typed with the Illumina Human610-Quad BeadChip. Rigorous data management and quality control measures were taken: group matching, updating map location, permutations test, sex check and filtering of low genotyping individuals and loci as well as loci with HWE issues. Three association analyses were performed: AD–Control, Cancer–Control, and AD–Cancer. Results: After matching for age, gender, and Caucasian ethnicity 492 individuals were included in the AD group (Avg age: 75 years, 37% female), 691 individuals in Cancer group (Avg age: 67.7 years, 37% female), and 1150 individuals in the combined Control group (avg age: 71 years, 37% female). Association analysis of the AD–Cancer study indicated one marker, rs2075650, as significant at p -8. Initial analysis also indicated possible clustering of significant SNPs on chromosomes 8 and 11. Conclusions: Case-case GWAS provides a novel means for identifying novel loci involved in the dichotomous relationship of risk of AD and risk of BrCa/PCa. These signals may point to critical genomic regions involved in age-related pathologies of cancer and AD.