Common Immune Gene Variant Cuts Male Longevity Odds by 36 Percent
A multi-cohort study finds HLA-DRB1*15:01 significantly reduces longevity likelihood in northern European men, possibly via Alzheimer's and immune reactivity to LDL.
Summary
Researchers studying the genetics of extreme old age discovered that a common immune gene variant, HLA-DRB1*15:01, is associated with a 36% lower likelihood of living to 95 or beyond — but only in men. The study compared German centenarians and long-lived individuals against younger controls, then replicated findings in Danish and UK Biobank cohorts. Men carrying this HLA allele were underrepresented among the long-lived: 10% of male cases versus 15% of controls carried it. The team also found computational evidence that this allele triggers immune responses against a component of LDL cholesterol (APOB-100), potentially linking immune over-reactivity to neurodegeneration and shortened lifespan. The sex-specific effect may relate to known male-specific associations between this allele and Alzheimer's disease.
Detailed Summary
Human longevity is partially heritable, but the genetic variants robustly linked to exceptional lifespan remain few. Beyond APOE and FOXO3, immune gene variants have been underexplored despite immunity's well-established role in aging. The highly polymorphic HLA region — encoding proteins central to pathogen recognition and immune activation — has rarely been analyzed at high resolution in longevity cohorts. This study systematically addressed that gap using imputed 2-field HLA alleles across three independent northern European populations.
The primary analysis enrolled 1,463 German long-lived individuals (LLI; mean age 99.0 years, range 94–110) against 6,464 geographically matched younger controls (mean age 57.2 years). HLA alleles were imputed from Immunochip genotype data using HLA-TAPAS with the 1000 Genomes reference panel. Only alleles with imputation r² > 0.8 and frequency > 1% were retained, yielding 146 class I and II alleles. Sex-stratified logistic regression with Bonferroni correction for multiple testing was performed, adjusting for the top three principal components to control for population stratification.
The primary discovery was a male-specific negative association of HLA-DRB1*15:01:01 with longevity (adjusted P = 2.80 × 10⁻², OR = 0.64, 95% CI: 0.48–0.82). Among male cases, only 10% carried this allele versus 15% of male controls — a substantial depletion. Female cases showed no meaningful difference (14%), making the effect distinctly sex-specific. Replication in the UK Biobank, using offspring of long-lived parents (fathers >95 years, n = 4,747 cases vs. 318,421 controls; mothers >95 years, n = 11,941 vs. 242,100), confirmed the male-specific association. The independent Danish cohort, analyzed via a tag-SNP (rs3135388) for DRB1*15:01, showed consistent directional results. Imputation accuracy was validated both by allele frequency comparison against known CEU frequencies and by targeted HLA sequencing of 200 randomly selected LLI.
To explore biological mechanisms, the researchers performed epitope binding and immunogenicity predictions for all HLA-DRB1 alleles against apolipoprotein B-100 (APOB-100), a major protein component of LDL particles. DRB1*15:01 was predicted to present an APOB-100 epitope with higher immunogenicity than most other DRB1 alleles. Critically, across all DRB1 alleles, higher predicted APOB-100 immunogenicity was significantly associated with reduced longevity (estimate −0.11, SE = 0.03, P = 0.005), suggesting that immune reactivity against LDL may be a general anti-longevity mechanism, not unique to DRB1*15:01 alone. Pairwise HLA interaction testing and APOE epistasis analyses in the UK Biobank found no significant interactions between DRB1*15:01 and APOE ε4 or ε2.
The proposed mechanistic pathway is that DRB1*15:01 promotes T-cell-mediated immune responses against APOB-100, contributing to neuroinflammation and accelerated Alzheimer's disease (AD) pathology — a disease already known to have a male-specific association with this same allele. This provides a plausible immunological bridge between HLA genotype, LDL immunity, neurodegeneration, and reduced male longevity. Caveats include the use of parental longevity as a proxy in the UK Biobank (which introduces noise), the observational design, and the limitation to populations of northern European ancestry, meaning these findings may not generalize to other ethnic groups.
Key Findings
- HLA-DRB1*15:01:01 was significantly associated with reduced male longevity across three cohorts (adjusted P = 2.80 × 10⁻², OR = 0.64, 95% CI: 0.48–0.82)
- Male long-lived individuals (mean age 99.0 years) carried DRB1*15:01 at 10% frequency vs. 15% in male controls — a 33% relative depletion
- Female long-lived individuals showed no significant difference in DRB1*15:01 frequency (~14%), confirming the effect is strictly male-specific
- Replication in UK Biobank used 4,747 cases (long-lived fathers) vs. 318,421 controls and confirmed the male-specific association directionally
- Higher predicted APOB-100 immunogenicity across all HLA-DRB1 alleles was significantly associated with reduced longevity (estimate −0.11, SE = 0.03, P = 0.005)
- No significant epistasis was detected between DRB1*15:01 and APOE ε4 or APOE ε2 in UK Biobank interaction analyses
- Imputation accuracy for DRB1*15:01:01 was validated by targeted HLA sequencing of 200 selected LLI and by concordance with CEU population allele frequencies (r² = 0.97 with tag-SNP rs3135388)
Methodology
Case-control design with 1,463 German LLI (mean age 99.0 years) versus 6,464 younger controls; HLA alleles imputed from Immunochip data using HLA-TAPAS (r² > 0.8 threshold, 146 alleles retained). Sex-stratified logistic regression with Bonferroni correction and principal component adjustment for population stratification. Findings replicated in a Danish cohort (800 cases, 1,148 controls) using a tag-SNP and in the UK Biobank (up to 11,941 parental longevity cases vs. 242,100 controls) with sex-stratified logistic regression.
Study Limitations
The study is limited to populations of northern European ancestry, restricting generalizability to other ethnic groups. The UK Biobank replication relies on parental longevity as a proxy rather than direct LLI phenotyping, which introduces phenotypic noise. Computational immunogenicity predictions, while directionally consistent, require experimental validation to confirm a causal mechanistic link. No conflicts of interest were reported by the authors.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.