HFE Genotype and Neurodegeneration: An Updated Systematic Review and Structured Evidence Synthesis with Four-Stratum Architecture

Background: Common variants in the homeostatic iron regulator gene HFE, principally p.Cys282Tyr (C282Y; rs1800562) and p.His63Asp (H63D; rs1799945), alter systemic iron handling and may influence neurodegeneration through disrupted brain iron homeostasis. Prior meta-analyses produced heterogeneous effect estimates and did not systematically examine sex as an effect modifier. Objectives: To synthesize evidence across four strata defined by design class and inferential scope: Stratum 1A, Alzheimer disease case-control studies; Stratum 1B, Parkinson disease case-control studies; Stratum 2, prospective cohort studies; and Stratum 3, genetic causal inference via GWAS and Mendelian randomization. Quantitative Updating Scope. No stratum contains an executed updated pooled meta-analysis. Stratum 1A (Alzheimer disease) is presented as structured narrative integration because primary study-level data for the foundational Lin 2011 corpus are not available for re-analysis. Stratum 1B (Parkinson disease) was planned for formal updated pooling, but the eligible meta-analyses used incompatible genetic models across partially overlapping study pools and the largest genetic study did not provide a poolable odds ratio; Stratum 1B is therefore also a structured evidence summary. Stratum 2 is not pooled due to cohort overlap. Stratum 3 is narrative. Methods: Systematic search of PubMed, Embase, and Web of Science (inception to February 2026). Eligible studies reported HFE genotype in relation to neurodegenerative outcomes in adults of European ancestry. All four strata were synthesized as structured evidence summaries; no formal updated pooled meta-analysis was executed because compatible primary study-level data were not available in any stratum. Results: Eleven reports were included across four strata. In Stratum 1A, the Lin 2011 synthesis (22 studies; 4,365 cases / 8,652 controls) found no C282Y association and a modest protective H63D signal (allele contrast OR=0.902, 95% CI 0.819-0.994). Tisato 2018 (276 AD cases / 1,086 controls) reported a borderline protective C282Y direction (unadjusted OR=0.41; adjusted OR=0.48, 95% CI 0.22-1.04). These sources are not combined. In Stratum 1B, two 2015 meta-analyses produced discordant C282Y findings and the largest genetic study (Saini 2021; 14,671 PD cases) found no HFE association with PD. In Stratum 2, male C282Y homozygotes in the UK Biobank showed elevated incident dementia (Atkins 2021: HR=1.83, 95% CI 1.23-2.72) and excess morbidity and mortality (Lucas 2024). Male H63D homozygotes in the ASPREE trial showed elevated incident dementia (Yu 2025: HR=2.39, 95% CI 1.25-4.57). Associations were absent in female homozygotes across all analyses. In Stratum 3, Mendelian randomization supported causal brain iron effects on non-Alzheimer dementia and Parkinson disease but not Alzheimer disease (Casanova 2024). Conclusions: Prospective cohort evidence supports a sex-specific elevated dementia risk among male HFE homozygotes. These findings are hypothesis-generating and do not support modification of clinical screening or treatment guidelines. Any consideration of iron depletion for neurodegeneration prevention would require independent replication of the male-specific signal, dementia subtype resolution, and demonstration of mechanistic modifiability.