Study Design and Data Sources
We applied a 2-sample MR analysis to evaluate the causal effect of insulin resistance on the risk of small vessel stroke and Alzheimer Disease (Figure 1). MR design is based on the theory that genotypes are randomly assorted at meiosis and independent of confounding factors, so that potential confounders and reverse causation can be controlled and more reliable causal inferences can be obtained. MR relies on three assumptions: 1) the instrument is associated with the exposure; 2) the instrument influences the outcome only through the exposure; 3) the instrument is not associated with other confounders.
Single nucleotide polymorphisms (SNPs) that are associated with insulin resistance and satisfy with the MR assumptions were extracted as instrumental variable. SNPs for insulin resistance based on fasting insulin were obtained from the Meta-Analyses of Glucose and Insulin-related traits Consortium (MAGIC). SNPs for insulin resistance based on gold standard measures were obtained from the GENEticS of Insulin Sensitivity consortium (GENESIS). Data on association of SNPs with small vessel disease were obtained from the Multiancestry Genome-Wide Association Study of Stroke consortium (MEGASTROKE). Data on association of SNPs with Alzheimer’s Disease were obtained from the Psychiatric Genomics Consortium-Alzheimer Disease Workgroup (PGC-ALZ). All the consortia are based upon genome-wide association studies (GWASs) of individual of European ancestry. Characteristics of the GWAS studies used in this study are presented in Table 1.
Selection of Genetic Variants
We used two methods to measure insulin resistance: 1) a common proxy of insulin resistance measured by fasting insulin, and 2) gold standard measures of insulin resistance based on the euglycemic-hyperinsulinemic clamp and the insulin suppression test. We selected SNPs associated with insulin resistance from two GWAS consortia: the MAGIC and the GENESIS. The MAGIC assessed potential SNPs associated with fasting insulin in 108,557 non-diabetic individuals of European ancestry. The consortia using the Illumina CardioMetabochip containing ~66000 SNPs for a range of cardiovascular and metabolic traits, contributed ~1000 SNPs for fasting insulin. The estimates of associations were adjusted for body mass index (BMI) and the fasting insulin was natural logarithm transformed. They identified 12 loci that achieved significance (P < 5×10-8) for the insulin resistance. We extracted independent genetic variants without linkage disequilibrium with other SNPs for insulin resistance. The GENSIS analyzed GWAS data of 2764 individual and replication in 2860 individuals of European ancestry with direct, standard measures of insulin resistance from four cohort GWASs (the RISC consortium, Uppsala Longitudinal Study of Adult Men, the EUGENE2 consortium, and the Stanford Insulin Suppression Test). They identified 5 SNPs that reached significance (P < 6×10−6) for the insulin resistance. The association of the 17 individual SNPs with the insulin resistance is presented in Table 2. These insulin resistance-associated SNPs were at different loci and without any linkage disequilibrium (r2 < 0.2). Furthermore, they were not associated with other potential risk factors related to CSVD or Alzheimer Disease at a genome-wide significance level (P < 5×10−8) after performing a search in the PhenoScanner database.
Summary statistics for the associations of individual SNP with small vessel stroke were acquired from previously published GWAS of Multiancestry Genome-Wide Association Study of Stroke (MEGASTROKE) consortium. The MEGASTROKE consortium is a multi-ancestry genome-wide association meta-analysis on stroke and stroke subtypes that tests ~8 million SNPs and indels with minor allele frequency ≥ 0.01 in 521,612 individuals (67,162 stoke cases and 454,450 controls) for association with stroke. Analysis in European individuals involved 40,585 stoke cases and 406,111 controls (Methods in the online-only Data Supplement).
Summary statistics for the associations of individual SNPs with Alzheimer Disease were acquired from the previously published GWAS of the Psychiatric Genomics Consortium-Alzheimer’s Disease Workgroup (PGC-ALZ). The PGC-ALZ is a three-pahase genome-wide meta-analysis involving 455,258 individuals (71,880 cases and 383,378 controls) of European ancestry. The genome-wide meta-analysis of clinically diagnosed Alzheimer Disease case–control status was based on four independent consortia: the PGC-ALZ, the International Genomics of Alzheimer’s Project (IGAP), Alzheimer’s Disease Sequencing Project (ADSP), and UK Biobank. The diagnosis of Alzheimer Disease in PGC-ALZ is in accordance with the recommendations of National Institute on Aging-Alzheimer’s Association, National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) and International Classification of Diseases Tenth Revision (ICD-10) research criteria. Alzheimer Disease in the IGAP and ADSP is autopsy-confirmed or clinically diagnosed with NINCDS-ADRDA criteria, Diagnostic and Statistical Manual of Mental Disorders (DSM) Fourth Edition) criteria, Alzheimer’s Disease Diagnostic and Treatment Centers State of California criteria, or DSM Third Edition Revised criteria. UK Biobank constructs an Alzheimer Disease-by-proxy status as part of self-report questionnaire based on parental Alzheimer Disease information weighted by age during the in-person interview, which is additionally confirmed with ICD-10 codes (G30, F00) in medical records. All individuals were entered into a meta-analysis of clinical Alzheimer Disease GWAS and the Alzheimer Disease-by-proxy GWAS (Methods in the online-only Data Supplement). The associations of individual SNPs for insulin resistance with small vessel stroke and Alzheimer Disease are presented in Table 3.
We performed 2-sample MR analyses on summary statistics to evaluate the impact of insulin resistance-associated genetic variants on small vessel stroke and Alzheimer Disease. The SNP-insulin resistance and SNP-outcome associations were used to calculate estimates of insulin resistance-outcome (small vessel stroke and Alzheimer Disease) associations using inverse-variance weighted (IVW) MR analysis. In sensitivity analyses, we complementarily performed analyses using robust IVW, MR-Egger, simple median, weighted median, weighted mode-based estimator (MBE), and MR pleiotropy residual sum and outlier (MR-PRESSO) methods. These methods include pleiotropic or invalid instruments that are more robust to potential violations of the standard instrumental variable assumption. The MR-Egger method can identify and control for bias due to directional pleiotropy (SNPs influence the outcome through different biological pathways other than exposure). The weighted median method allows stronger SNPs to contribute more toward the estimate and, therefore, allows them to contribute more toward stronger causal inferences. The MR-PRESSO method was applied to identify and correct potential outliers in multi-instrument summary-level MR testing. Potential pleiotropic effects of these SNPs were evaluated via the MR-Egger regression, in which the slope represents causal estimates corrected by pleiotropy and the intercept represents the average pleiotropic effects of all SNPs. Heterogeneity of SNPs was estimated by Cochran Q statistic. If there is heterogeneity, random-effects IVW models are applied; otherwise, the fixed-effect IVW model is applied. In order to estimate the influence of outlying or pleiotropic genetic variants, we conducted a leave-one-out analysis in which we re-estimated the effect by sequentially dropping 1 SNP at a time.
The odds ratios (ORs) with their 95% confidence interval (CI) per 1-SD log-transformed genetically predicted increase in insulin resistance were used to represent the association between insulin resistance and outcomes (small vessel stroke and Alzheimer Disease). We additionally plotted the association of each genetic variant with insulin resistance against its effect on the outcomes. The Bonferroni-adjusted significance for small vessel stroke and Alzheimer Disease was calculated as P < 0.025 (0.05/2 = 0.025) to ensure the validity of our conclusions. In the present analyses, 2-sided P < 0.05 was considered statistically significant for a potential, but yet to be confirmed, causal association, and a 2-sided P < 0.025 was considered to be statistically significant for a causal association. All analyses were conducted with R 3.5.3 (R Development Core Team).