Sex differences in the risk of cataract associated with type 2 diabetes: a Mendelian randomization study

Haoyang Zhang, Xuehao Xiu, Yuedong Yang, Yuanhao Yang, Huiying ZhaomedRxiv 2021 · DOI

Abstract

Type 2 diabetes (T2D) is a well-established risk factor for the development of cataracts. However, whether the genetic architecture and causal relationship between these two conditions differ by sex remains poorly understood. This study evaluated sex-specific genetic correlation ($r_g$) and putative genetic causality between T2D and cataracts using linkage disequilibrium score regression (LDSC) and six Mendelian randomization (MR) approaches. Leveraging large-scale genome-wide association studies (GWAS) from East Asian and European populations, the study found a significant genetic correlation and a consistently stronger causal effect of T2D on cataracts in East Asian males compared to females. In contrast, while European populations showed significant genetic correlation in both sexes, the evidence for a causal relationship was weak. These findings provide the first evidence of sex-specific causal differences in East Asians and suggest genetic heterogeneity in the shared mechanisms of T2D and cataracts across different ethnicities.

Introduction

Clinicians have long recognized Type 2 Diabetes (T2D) as a major driver of cataract formation. However, observational studies—which analyze the co-occurrence of these diseases—have yielded conflicting results regarding sex-based risk. Some data suggest women are more vulnerable, while other studies find no significant difference.

The primary challenge in observational research is confounding: variables such as BMI, lifestyle, and blood pressure can obscure whether T2D directly causes cataracts or if both conditions simply share environmental triggers. This study addresses these limitations by utilizing Mendelian randomization (MR), a method that uses genetic variants as instrumental variables to establish causal links, effectively bypassing the "noise" of traditional observational data.

Methodology

The researchers employed a robust suite of statistical genetics tools to analyze summary statistics from large-scale GWAS datasets:

  • Linkage Disequilibrium Score Regression (LDSC): This was used to estimate the genetic correlation ($r_g$) between T2D and cataracts, determining the extent to which the two diseases share common genetic variants.
  • Mendelian Randomization (MR): Acting as a "natural randomized controlled trial," MR utilized genetic instruments to test the causal flow from T2D to cataracts. Six different models, including Inverse Variance Weighted (IVW), MR-Egger, and CAUSE, were used to ensure the consistency of the results.
  • Summary-data-based Mendelian Randomization (SMR): This tool helped identify specific functional genes that may serve as the biological bridge between the two conditions.

Datasets

The study utilized massive, sex-stratified datasets across two major populations:

  1. East Asian Population: Data derived from the Asian Genetic Epidemiology Network (AGEN) and BioBank Japan (BBJ).
  2. European Population: Data from the UK Biobank (UKB), involving over 500,000 participants.
  3. Reference Genomes: The 1000 Genomes project served as the genetic baseline for both populations.

Key Results

The East Asian Sex Disparity

The most significant finding was a clear disparity in East Asian populations. LDSC identified a significant genetic correlation in East Asian males ($r_g = 0.68, p = 8.60 \times 10^{-3}$) but a non-significant correlation in females ($r_g = 0.25, p = 8.38 \times 10^{-2}$).

MR analyses further confirmed that the causal effect of T2D on cataracts is significantly stronger in males. The liability odds ratio (OR) for males ranged from 1.20 to 1.41, while for females, it was notably lower, ranging from 1.12 to 1.21. A paired t-test confirmed this difference was statistically significant ($|t|=5.87, p=2.04 \times 10^{-3}$).

Ethnic Heterogeneity

Interestingly, the strong causal link observed in East Asians was not mirrored in the European cohort. While European males and females both showed significant genetic correlations ($r_g \approx 0.20$ and $0.17$, respectively), the MR analysis provided only weak evidence for a direct causal relationship in either sex. This suggests that the genetic drivers linking T2D to cataracts may vary significantly between ethnic groups.

Functional Genetic Bridges

The study identified specific genes that may drive these risks:

  • KLHL42 and BRD3: Identified as male-specific functional genes in the T2D-cataract pathway for East Asians.
  • KCNJ11: Linked to risk in both sexes but exhibiting different functional patterns, suggesting it plays a complex role in disease progression.

Discussion and Clinical Implications

This research provides the first robust genetic evidence that the causal link between T2D and cataracts is more pronounced in East Asian men. The identification of male-specific genes like KLHL42 and BRD3 offers a starting point for understanding the biological mechanisms—potentially involving sex hormones or sex-specific environmental interactions—that accelerate cataract formation in diabetic patients.

Limitations

The authors noted several limitations, including potential sample-overlap bias in the European (UK Biobank) data and the need for loosened significance thresholds in certain sub-analyses to maintain statistical power. Furthermore, while the study establishes a causal link, the exact biological pathway through which these genes operate remains to be fully elucidated.

Conclusion

These findings support a shift toward more personalized, sex-specific strategies in diabetic eye care. For male T2D patients of East Asian descent, early and frequent cataract screening may be particularly critical. By uncovering the shared genetic architecture of these diseases, this study lays the theoretical foundation for targeted prevention and potential therapeutic interventions.

Summary updated

Abstract

Abstract Type 2 diabetes (T2D) is a recognized risk factor for developing cataract. However, it is unclear if the shared genetic variance and potential genetic causal relationship between T2D and cataract are different for males and females. We evaluated sex-specific genetic correlation ( r g ) and putative genetic causality between the two diseases by using linkage disequilibrium score regression (LDSC) and six Mendelian randomization (MR) approaches after leveraging large-scale population-based genome-wide association studies (GWAS) summary of T2D and cataract. Application of LDSC found a significant genetic correlation between T2D and cataract in East Asian males ( r g =0.68, 95% confident interval [CI]=0.17 to 1, p-value=8.60×10 −3 ) but a non-significant genetic correlation in East Asian females ( r g =0.25, CI= -0.02 to 0.52, p-value=8.38×10 −2 ). MR analyses indicated a consistently stronger (paired t-test |t|=5.87, p-value=2.04×10 −3 ) causal effect of T2D on cataract in East Asian males (liability OR=1.20 to 1.41, p-value=5.86×10 −27 to 6.60×10 −6 ) than in females (liability OR=1.12 to 1.21, p-value=2.02×10 −14 to 1.82×10 −2 ). In Europeans, the LDSC analysis suggested a close significant genetic correlation between the two diseases in males ( r g =0.20, 95% confident interval [CI]=0.08 to 0.32, p-value=7.00×10 −4 ) and females ( r g =0.17, CI= 0.05 to 0.29, p-value=4.90×10 −3 ); but the MR analyses provided weak evidences on a causal relationship between the two diseases in both sexes. These results presented the first evidence on sex difference of the casual relationship between cataract and T2D in East Asians, and supported a potential genetic heterogeneity of the shared genetics underlying T2D and cataract between East Asians and Europeans in both sexes.
BibTeX
@misc{zhang2021sex,
  author = {Haoyang Zhang and Xuehao Xiu and Yuedong Yang and Yuanhao Yang and Huiying Zhao},
  title = {{Sex differences in the risk of cataract associated with type 2 diabetes: a Mendelian randomization study}},
  year = {2021},
  doi = {10.1101/2021.08.05.21261684},
  howpublished = {medRxiv},
}
Figure