There is increasing support for the role of neuroinflammation and aberrant immune regulation in the pathophysiology of Alzheimer's disease (AD). The immunoregulatory human leukocyte antigen (HLA) complex has been linked to susceptibility for a number of neurodegenerative diseases including AD, however, studies to date have failed to consistently identify a risk HLA haplotype for AD. Contributing to this difficulty are the complex genetic organization of the HLA region, differences in sequencing and allelic imputation methods, and diversity across ethnic populations. Building on prior work linking the HLA to AD, we used a robust imputation method on two case-control cohorts to examine the relationship between HLA haplotypes and AD risk in 309 individuals from UCSF and 11,381 individuals from the Alzheimer's Disease Genetics Consortium (ADGC), a NIA-funded national data repository. The strongest association with AD risk occurred with MHC haplotype A*03:01∼B*07:02∼DRB1*15:01∼DQA1*01:02∼DQB1*06:02, p=9.6x10-4, odds ratio (95% confidence interval) = 1.2 (1.1–1.4) in the combined UCSF + ADGC cohort. Secondary analysis suggested that this effect may be driven primarily by individuals who are negative for established AD genetic risk factor, APOE E4, and may be of particular relevance in men more than women. Separate analyses of class I and II haplotypes further supported the role of class I haplotype A*03:01∼B*07:02 and class II haplotype DRB1*15:01- DQA1*01:02- DQB1*06:02 (DR15) as risk factors for AD. We also examined cerebrospinal fluid (CSF) biomarker measures and longitudinal cognitive data from the Alzheimer's Disease Neuroimaging Initiative to assess clinical relevance of identified risk haplotypes and found A*03:01∼B*07:02 was associated with higher CSF amyloid levels while DR15 haplotype was associated with greater rates of cognitive decline in a dose-dependent manner. We provide evidence that variation in the HLA—including risk haplotype DR15—contributes to AD risk. DR15 has also been associated with risk for multiple sclerosis, and its component alleles have been implicated in Parkinson's disease and narcolepsy. Our findings thus raise the possibility that DR15-associated mechanisms may contribute to pan-neuronal disease vulnerability. Results must be considered in the context of only representing European-ancestry, clinically diagnosed individuals, and utilizing imputed genotypes for a subset of HLA genes.