A Multivariate Method for Meta-Analysis of Multiple Outcomes in Genetic Association StudiesIn this work we present a simple, yet powerful approach for performing multivariate meta-analysis of genetic association studies when multiple outcomes are assessed. The key element of our approach is the analytical calculation of the within-studies covariances. We propose a model based on summary data, uniformly defined for both discrete and continuous outcomes (using log odds-ratios or mean differences). The within-studies covariances can be calculated using the cross-classification of the genotypes in both outcomes, which are retrieved using a log-linear model using the iterative proportional fitting algorithm under the assumption of no three-way interaction. As an example, we examine the association of GNB3 C825T polymorphism with two non-exclusive dichotomous outcomes (Type 2 Diabetes Mellitus and Essential Hypertension). We also present an application using continuous outcomes (diastolic and systolic blood pressure). We show the applicability and the generality of the method performing the analysis assuming the genetic model beforehand or following a genetic model-free approach. The method is simple and fast, it can be extended for several outcomes and can be fitted in nearly all statistical packages. There is no need for individual patient data or the simultaneous evaluation of both outcomes in all studies. We conclude that the proposed method constitute a useful framework for performing meta-analysis for multiple outcomes within the context of genetic association studies. Connections to other similar models presented in the literature, are discussed, as well as potential applications of the method to other areas.Apolipoprotein E polymorphism and left ventricular failure in beta-thalassemia: A meta-analysisApolipoprotein E (ApoE) acts as a scavenger of free radicals and as ApoE4 allele decreases antioxidant and iron-binding activity, may serve as a genetic risk factor for the development of left ventricular failure (LVF), the main cause of death in beta-thalassemia homozygotes. In our setting, studies report on more than one outcome, as patients are classified into three groups according to the severity of the symptoms and a particular polymorphism (ApoE) with multiple haplotype variants emerging thus the utilisation of a multivariate random effects meta-analysis that takes into account that the log odds ratios (logORs) are correlated. Afterwards, formal overall Wald tests can be constructed as well as a trend analysis to test different aspects of the shape of the function of the ORs for each outcome vs. the reference category within the different allelic contrasts. In total, 4 individual studies were recruited with 613 beta-thalassemic patients and 664 healthy controls. The multivariate meta-analysis approach, using E3 as a reference haplotype variant, revealed a highly statistically significant overall association (Wald test p-value=0.009). This was mainly attributed to the contrast of E4 vs. E3 allele which yields an OR of 2.32 (95% CI: 1.19, 4.53) for Group II and 3.34 (95% CI: 1.78, 6.26) for Group III. The trend analysis revealed a marginally non significant finding (χ2(2)=5.55, p-value=0.062), indicating a possible underlying linear trend of the 3 ORs for each outcome vs. the reference category within the 2 allelic contrasts. Further studies are needed in order to enlighten the mechanism of this association as well as to study its effects on larger populations. We conclude that the multivariate approach presented here can be applied in several fields of research, since it is the first method that addresses simultaneously multiple exposures (i.e. alleles) and multiple comparison groups in genetic association studies.International Journal Publications
Software
Posters and oral presentations in International and National Conferences
|
IntDaMuS > Work Packages >