A Haptoglobin (HP) Exon Deletion Polymorphism Alters the Effect of APOE Alleles on Alzheimer’s Disease in European‐Descent People with APOEε4

Background Haptoglobin (HP) is an antioxidant of apolipoprotein E (APOE) – the strongest risk gene for sporadic Alzheimer’s disease (AD). The HP gene has two functional alleles, HP2 and HP1 , which contains a two‐exon deletion that changes its protein structure conformation. We hypothesize that this structural variation is associated with AD. Method To investigate this, we imputed the HP genotypes for 12,403 cases and 11,699 normal controls from the Alzheimer’s Disease Genetics Consortium (ADGC), respectively within each cohort of 18 European‐descent cohorts, using a custom reference panel. We then evaluated the association between HP genotype and the AD status using logistic regression, adjusting for sex, age, and the first 3 principal components. Result The imputed HP1 frequency is 0.376. We found HP influences AD given an APOEε4 genetic context. In individuals with ≥1 APOEε4 allele (n=10,054), HP1 homozygotes attenuate the effect of the second APOE allele; both the protective effect of the ε2 allele and the risk effect of the ε4 allele are dramatically reduced (odds ratio, OR=0.92, p=0.415 of ε2 versus ε3 , OR=1.09, p=0.415 of ε4 versus ε3 ). However, the HP2 allele accentuates the effect of the second APOE allele, both the protective effect of the ε2 allele and the risk effect of the ε4 allele (OR=0.80, p= 0.003 of ε2 versus ε3 , OR=1.25, p=0.003 of ε4 versus ε3 ). This effect is even stronger in HP2 homozygotes (OR=0.64, p=0.003 of ε2 versus ε3 , OR=1.55, p=0.003 of ε4 versus ε3 ). In addition, we analyzed time to AD diagnosis with a cox proportional hazard regression model within the more consistently examined National Alzheimer's Coordinating Center (NACC) cohorts (n=7,448). A cox model shows that HP2 alleles additively decrease the hazard of AD (hazard ratio=0.94, p=6.85e‐3). Conclusion HP impacts AD risk in people with APOEε4 . Furthermore, this risk is different in people with APOEε24 , APOEε34 , and APOEε44 genotypes. Though the mechanism still awaits investigation, based on the current knowledge of HP function our results suggest a possible differential anti‐oxidative ability of HP proteins for distinct APOE proteins. As such, the HP protein may be an imminently translatable therapeutic target for AD in APOEε4 carriers.