Approximation of bias and mean-squared error in two-sample Mendelian randomization analyses

Mendelian randomization (MR) is a type of instrumental variable (IV) analysis that uses genetic variants as IVs for a risk factor to study its causal effect on an outcome. Extensive investigations on the performance of IV analysis procedures, such as the one based on the two-stage least squares (2SLS) procedure, have been conducted under the one-sample scenario, where measures on IVs, the risk factor, and the outcome are assumed to be available for each study participant. Recent MR analysis usually is performed with data from two independent or partially overlapping genetic association studies (two-sample setting), with one providing information on the association between the IVs and the outcome, and the other on the association between the IVs and the risk factor. We investigate the performance of 2SLS in the two-sample–based MR when the IVs are weakly associated with the risk factor. We derive closed form formulas for the bias and mean squared error of the 2SLS estimate and verify them with numeric simulations under realistic circumstances. Using these analytic formulas, we can study the pros and cons of conducting MR analysis under one-sample and two-sample settings and assess the impact of having overlapping samples. We also propose and validate a bias-corrected estimator for the causal effect.     

A semiparametric empirical likelihood method for biased sampling schemes with auxiliary covariates

We consider a semiparametric inference procedure for data from epidemiologic studies conducted with a two-component sampling scheme where both a simple random sample and multiple outcome- or outcome-/auxiliary-dependent samples are observed. This sampling scheme allows the investigators to oversample certain subpopulations believed to have more information about the regression model while still gaining insights about the underlying population through the simple random sample. We focus on settings where there is no additional information about the parent cohort and the sampling probability is nonidentifiable. We motivate our problem with an ongoing study to assess the association between the mutation level of epidermal growth factor receptor (EGFR) and the antitumor response to EGFR-targeted therapy among nonsmall cell lung cancer patients. The proposed method applies to both binary and multicategorical outcome data and allows an arbitrary link function in the framework of generalized linear models. Simulation studies show that the proposed estimator has nice small sample properties. The proposed method is illustrated with a data example.     

The association between smokeless tobacco use and pancreatic adenocarcinoma: A systematic review

BackgroundSmokeless tobacco is a possible risk factor for developing pancreatic adenocarcinoma. This systematic review addressed the question: Is there an association between smokeless tobacco use and pancreatic adenocarcinoma diagnosis?MethodsFive electronic databases, grey literature, and citations of relevant articles were searched to identify studies. Six researchers double-reviewed records for inclusion in the review. The information extracted from these studies was selected using criteria outlined in the Newcastle–Ottawa Quality Assessment Scale for observational studies. A qualitative synthesis of included studies was performed.ResultsThe search of electronic databases resulted in a total of 1747 citations. Eleven studies met the inclusion criteria for this review, including three cohort studies, seven case control studies and one study that pooled data from multiple case-control studies. Studies were heterogeneous in their assessment of exposure intensity and ascertainment of outcomes. Quality of the studies varied. Existing investigations of the association of interest appear to exhibit several types of biases including selection bias, information bias and bias in the analysis.ConclusionThe association between smokeless tobacco use and pancreatic adenocarcinoma is inconclusive. More definitive conclusions regarding this relationship await the results of more methodologically rigorous epidemiologic studies.     

Reduced folate carrier gene is a risk factor for neural tube defects in a Chinese population

BACKGROUND: There is a considerable body of data demonstrating that periconceptional supplementation of folic acid can prevent a significant proportion of neural tube defects (NTDs). At present, the mechanism by which folic acid exerts its beneficial effect remains unknown. Folate transporter genes, including the reduced folate carrier gene (RFC1), have been proposed as NTD risk factors. METHODS: The study population included 104 nuclear families with NTDs and 100 nonmalformed control families. We investigated the possible association between a common RFC1 polymorphism (A80G) and NTD risk among offspring, as well as potential gene-environment interactions between the infant RFC1 genotype and maternal periconceptional use of folic acid through a population-based case-control study. RESULTS: We observed that the infants of the GG genotype were associated with a 2.56-fold increased risk of NTDs when compared to the AA genotype (odds ratio [OR], 2.56; 95% confidence interval [CI], 1.04-6.36) in our study population. Among mothers who did not utilize folic acid supplements, the risk for having a child with an NTD was 3.30 (95% CI, 1.15-9.65) for offspring with the GG genotype, compared to the reference (AA) genotype. Children who had the GG genotype and whose mothers did not take folic acid had an elevated risk for NTDs (OR, 8.80; 95% CI, 2.83-28.69), compared to offspring with the AA and GA genotypes whose mothers utilized folic acid supplements. CONCLUSIONS: Our findings suggest that the RFC1 G allele is likely to be an important genetic factor in determining folate transport and subsequently may be a risk factor for NTDs in this Chinese population.     

Association between Stress Response Genes and Features of Diurnal Cortisol Curves in the Multi-Ethnic Study of Atherosclerosis: A New Multi-Phenotype Approach for Gene-Based Association Tests

The hormone cortisol is likely to be a key mediator of the stress response that influences multiple physiologic systems that are involved in common chronic disease, including the cardiovascular system, the immune system, and metabolism. In this paper, a candidate gene approach was used to investigate genetic contributions to variability in multiple correlated features of the daily cortisol profile in a sample of European Americans, African Americans, and Hispanic Americans from the Multi-Ethnic Study of Atherosclerosis (MESA). We proposed and applied a new gene-level multiple-phenotype analysis and carried out a meta-analysis to combine the ethnicity specific results. This new analysis, instead of a more routine single marker-single phenotype approach identified a significant association between one gene (ADRB2) and cortisol features (meta-analysis p-value=0.0025), which was not identified by three other commonly used existing analytic strategies: 1. Single marker association tests involving each single cortisol feature separately; 2. Single marker association tests jointly testing for multiple cortisol features; 3. Gene-level association tests separately carried out for each single cortisol feature. The analytic strategies presented consider different hypotheses regarding genotype-phenotype association and imply different costs of multiple testing. The proposed gene-level analysis integrating multiple cortisol features across multiple ethnic groups provides new insights into the gene-cortisol association.     

Factor analytic models of clustered multivariate data with informative censoring

This article describes a general class of factor analytic models for the analysis of clustered multivariate data in the presence of informative missingness. We assume that there are distinct sets of cluster-level latent variables related to the primary outcomes and to the censoring process, and we account for dependency between these latent variables through a hierarchical model. A linear model is used to relate covariates and latent variables to the primary outcomes for each subunit. A generalized linear model accounts for covariate and latent variable effects on the probability of censoring for subunits within each cluster. The model accounts for correlation within clusters and within subunits through a flexible factor analytic framework that allows multiple latent variables and covariate effects on the latent variables. The structure of the model facilitates implementation of Markov chain Monte Carlo methods for posterior estimation. Data from a spermatotoxicity study are analyzed to illustrate the proposed approach.     

Maternal PCMT1 gene polymorphisms and the risk of neural tube defects in a Chinese population of Lvliang high-risk area

Protein-L-isoaspartate (D-aspartate) O-methyltransferase 1 (PCMT1) gene encodes for the protein repair enzyme L-isoaspartate (D-aspartate) O-methyltransferase (PIMT), which is known to protect certain neural cells from Bax-induced apoptosis. Previous study has shown that PCMT1 polymorphisms rs4552 and rs4816 of infant are associated with spina bifida in the Californian population. The association between maternal polymorphism and neural tube defects is still uncovered. A case-control study was conducted to investigate a possible association between maternal PCMT1 and NTDs in Lvliang high-risk area of Shanxi Province in China, using a high-resolution DNA melting analysis genotyping method. We found that increased risk for anencephaly in isolated NTDs compared with the normal control group was observed for the G (vs. A) allele (p=0.034, OR=1.896, 95% CI, 1.04-3.45) and genotypes GG+GA (p=0.025, OR=2.237, 95% CI, 1.09-4.57). Although the significance was lost after multiple comparison correction, the results implied that maternal polymorphisms in PCMT1 might be a potential genetic risk factor for isolated anencephaly in this Chinese population.     

Obesity,a major risk factor for immunity and severe outcomes of COVID-19

An influenza-like virus named severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) is responsible for COVID-19 disease and spread worldwide within a short time. COVID-19 has now become a significant concern for public health. Obesity is highly prevalent worldwide and is considered a risk factor for impairing the adaptive immune system. Although diabetes, hypertension, cardiovascular disease (CVD), and renal failure are considered the risk factors for COVID-19, obesity is not yet well-considered. The present study approaches establishing a systemic association between the prevalence of obesity and its impact on immunity concerning the severe outcomes of COVID-19 utilizing existing knowledge. Overall study outcomes documented the worldwide prevalence of obesity, its effects on immunity, and a possible underlying mechanism covering obesity-related risk pathways for the severe outcomes of COVID-19. Overall understanding from the present study is that being an immune system impairing factor, the role of obesity in the severe outcomes of COVID-19 is worthy.     

Application of Multi-SNP Approaches Bayesian LASSO and AUC-RF to Detect Main Effects of Inflammatory-Gene Variants Associated with Bladder Cancer Risk

The relationship between inflammation and cancer is well established in several tumor types, including bladder cancer. We performed an association study between 886 inflammatory-gene variants and bladder cancer risk in 1,047 cases and 988 controls from the Spanish Bladder Cancer (SBC)/EPICURO Study. A preliminary exploration with the widely used univariate logistic regression approach did not identify any significant SNP after correcting for multiple testing. We further applied two more comprehensive methods to capture the complexity of bladder cancer genetic susceptibility: Bayesian Threshold LASSO (BTL), a regularized regression method, and AUC-Random Forest, a machine-learning algorithm. Both approaches explore the joint effect of markers. BTL analysis identified a signature of 37 SNPs in 34 genes showing an association with bladder cancer. AUC-RF detected an optimal predictive subset of 56 SNPs. 13 SNPs were identified by both methods in the total population. Using resources from the Texas Bladder Cancer study we were able to replicate 30% of the SNPs assessed. The associations between inflammatory SNPs and bladder cancer were reexamined among non-smokers to eliminate the effect of tobacco, one of the strongest and most prevalent environmental risk factor for this tumor. A 9 SNP-signature was detected by BTL. Here we report, for the first time, a set of SNP in inflammatory genes jointly associated with bladder cancer risk. These results highlight the importance of the complex structure of genetic susceptibility associated with cancer risk.     

Quantitative risk assessment for multivariate continuous outcomes with application to neurotoxicology: the bivariate case

The neurotoxic effects of chemical agents are often investigated in controlled studies on rodents, with multiple binary and continuous endpoints routinely collected. One goal is to conduct quantitative risk assessment to determine safe dose levels. Such studies face two major challenges for continuous outcomes. First, characterizing risk and defining a benchmark dose are difficult. Usually associated with an adverse binary event, risk is clearly definable in quantal settings as presence or absence of an event; finding a similar probability scale for continuous outcomes is less clear. Often, an adverse event is defined for continuous outcomes as any value below a specified cutoff level in a distribution assumed normal or log normal. Second, while continuous outcomes are traditionally analyzed separately for such studies, recent literature advocates also using multiple outcomes to assess risk. We propose a method for modeling and quantitative risk assessment for bivariate continuous outcomes that address both difficulties by extending existing percentile regression methods. The model is likelihood based; it allows separate dose-response models for each outcome while accounting for the bivariate correlation and overall characterization of risk. The approach to estimation of a benchmark dose is analogous to that for quantal data without the need to specify arbitrary cutoff values. We illustrate our methods with data from a neurotoxicity study of triethyl tin exposure in rats.     

Estimating the power of linkage analysis in hereditary breast cancer.

Because evidence from several sources suggests the existence of a major genetic factor contributing to the risk for breast cancer, there is current interest in searching for genetic linkage between the putative cancer susceptibility gene(s) and polymorphic DNA markers. However, because of high population rates of the nongenetic form of the condition, and because of the possibility of underlying genetic heterogeneity, it is expected that the number of informative families required for detection of linkage will be greater for breast cancer than for other conditions with major genetic components. Computer simulation approaches may be useful in estimating the power of proposed linkage studies. Here we have simulated multiple genotypes for a medium-sized breast-cancer family and have analyzed the generated pedigrees with a linkage program. The effects of possible phenocopies and of genetic heterogeneity were measured by comparing the results obtained when the parameters were varied in the models. When population-based rates for sporadic cases of breast cancer were incorporated, the number of families required to detect linkage was approximately twice that expected in the absence of phenocopies. Incorrect specification of the probability of phenocopies in the analytic model did not materially alter the power of the proposed study, although significant effects on the estimated recombination fractions were noted.     

The Association between Proximity to Animal Feeding Operations and Community Health: A Systematic Review

Background

A systematic review was conducted for the association between animal feeding operations (AFOs) and the health of individuals living near AFOs.

Methodology/Principal Findings

The review was restricted to studies reporting respiratory, gastrointestinal and mental health outcomes in individuals living near AFOs in North America, European Union, United Kingdom, and Scandinavia. From June to September 2008 searches were conducted in PUBMED, CAB, Web-of-Science, and Agricola with no restrictions. Hand searching of narrative reviews was also used. Two reviewers independently evaluated the role of chance, confounding, information, selection and analytic bias on the study outcome. Nine relevant studies were identified. The studies were heterogeneous with respect to outcomes and exposures assessed. Few studies reported an association between surrogate clinical outcomes and AFO proximity. A negative association was reported when odor was the measure of exposure to AFOs and self-reported disease, the measure of outcome. There was evidence of an association between self-reported disease and proximity to AFO in individuals annoyed by AFO odor.

Conclusions/Significance

There was inconsistent evidence of a weak association between self-reported disease in people with allergies or familial history of allergies. No consistent dose response relationship between exposure and disease was observable.     

Light at night co-distributes with incident breast but not lung cancer in the female population of Israel

Recent studies of shift-working women have reported that excessive exposure to light at night (LAN) may be a risk factor for breast cancer. However, no studies have yet attempted to examine the co-distribution of LAN and breast cancer incidence on a population level with the goal to assess the coherence of these earlier findings with population trends. Coherence is one of Hill's "criteria" (actually, viewpoints) for an inference of causality. Nighttime satellite images were used to estimate LAN levels in 147 communities in Israel. Multiple regression analysis was performed to investigate the association between LAN and breast cancer incidence rates and, as a test of the specificity of our method, lung cancer incidence rates in women across localities under the prediction of a link with breast cancer but not lung cancer. After adjusting for several variables available on a population level, such as ethnic makeup, birth rate, population density, and local income level, a strong positive association between LAN intensity and breast cancer rate was revealed (p<0.05), and this association strengthened (p<0.01) when only statistically significant factors were filtered out by stepwise regression analysis. Concurrently, no association was found between LAN intensity and lung cancer rate. These results provide coherence of the previously reported case-control and cohort studies with the co-distribution of LAN and breast cancer on a population basis. The analysis yielded an estimated 73% higher breast cancer incidence in the highest LAN exposed communities compared to the lowest LAN exposed communities.     

Mortality and occupational exposure to radiation: first analysis of the National Registry for Radiation Workers.

OBJECTIVE--To study cause specific mortality of radiation workers with particular reference to associations between fatal neoplasms and level of exposure to radiation. DESIGN--Cohort study. SETTING--United Kingdom. SUBJECTS--95,217 radiation workers at major sites of the nuclear industry. MAIN OUTCOME MEASURE--Cause of death. RESULTS--Most standardised mortality ratios were below 100: 83 unlagged, 85 with a 10 year lag for all causes; 84 unlagged, 86 lagged for all cancers; and 80 for all known other causes, indicating a "healthy worker effect." The deficit of lung cancer (75 unlagged, 76 lagged) was significant at the 0.1% level. Standardised mortality ratios were significantly raised (214 unlagged, 303 lagged) for thyroid cancer, but there was no evidence for any trend with external recorded radiation dose. Dose of external radiation and mortality from all cancers were weakly correlated (p = 0.10), and multiple myeloma was more strongly correlated (p = 0.06); for leukaemia, excluding chronic lymphatic, the trend was significant (p = 0.03; all tests one tailed). The central estimates of lifetime risk derived from these data were 10.0% per Sv (90% confidence interval less than 0 to 24%) for all cancers and 0.76% per Sv (0.07 to 2.4%) for leukaemia (excluding chronic lymphatic leukaemia). These are, respectively, 2.5 times and 1.9 times the risk estimates recommended by the International Commission on Radiological Protection, but 90% confidence intervals are large and the commission''s risk factors fall well within the range. The positive trend with dose for all cancers, from which the risk estimate was derived, was not significant. The positive association between leukaemia (except chronic lymphatic leukaemia) was significant and robust in subsidiary analyses. This study showed no association between radiation exposure and prostatic cancer. CONCLUSION--There is evidence for an association between radiation exposure and mortality from cancer, in particular leukaemia (excluding chronic lymphatic leukaemia) and multiple myeloma, although mortality from these diseases in the study population overall was below that in the general population. The central estimates of risk from this study lie above the most recent estimates of the International Commission on Radiological Protection for leukaemia (excluding chronic lymphatic leukaemia) and for all malignancies. However, the commission''s risk estimates are well within the 90% confidence intervals from this study. Analysis of combined cohorts of radiation workers in the United States indicated lower risk estimates than the commission recommends, and when the American data are combined with our analysis the overall risks are close to those estimated by the commission. This first analysis of the National Registry for Radiation Workers does not provide sufficient evidence to justify a revision in risk estimates for radiological protection purposes.     

Internalizing psychopathology and all‐cause mortality: a comparison of transdiagnostic vs. diagnosis‐based risk prediction

Previous studies have documented the utility of a transdiagnostic internalizing factor in predicting important future outcomes (e.g., subsequent mental disorder diagnoses). To date, however, no study has investigated whether an internalizing factor predicts mortality risk. Also, while pre­vious studies of mortality risk have emphasized its associations with particular internalizing disorders, no study has assessed how the transdiagnostic internalizing factor vs. disorder‐specific variance differently predict that risk. The primary aims of this study were to explore: a) whether the internalizing factor predicts mortality risk, b) whether particular internalizing psychopathologies uniquely predict mortality risk over and beyond the transdiagnostic internalizing factor, and c) whether there is a significant interaction of internalizing with self‐reported health in the prediction of mortality risk. We utilized a large national sample of American adults from the Midlife in the United States (MIDUS), a longitudinal study that examined midlife development of individuals across multiple waves between 1995 and 2015. Data were analyzed for the 6,329 participants who completed the phone interview and self‐administered questionnaire in MIDUS 1 (1995‐1996) and were then followed up until October 31, 2015 or until death. To investigate the association between internalizing and mortality risk, we used the semi‐parametric proportional hazards Cox model, where survival time was regressed on a latent internalizing factor. Overall findings indicate that a transdiagnostic internalizing factor significantly predicts mortality risk over a 20‐year period (hazard ratio, HR=1.12, 95% CI: 1.05‐1.16, p<0.01) and that internalizing outperforms disorder‐specific variance (e.g., depression‐specific variance) in the prediction of that risk. Further, there was a significant interaction between transdiagnostic internalizing and self‐reported health, whereby internalizing psychopathology had a specific association with early death for individuals with excellent self‐reported health condition (HR=1.50, 95% CI: 1.17‐1.84, p<0.05). This highlights the clinical utility of using the transdiagnostic internalizing factor for prediction of an important future outcome, and supports the argument that internalizing psychopathology can be a meaningful liability to explore in public health practice.     

IL-6 gene variation is associated with IL-6 and C-reactive protein levels but not cardiovascular outcomes in the Cardiovascular Health Study

Interleukin-6 (IL-6) and C-reactive protein (CRP) levels increase with age and likely play a role in adverse health outcomes in older adults. The relationship between IL-6 gene tag single nucleotide polymorphisms (SNPs) and circulating IL-6 and CRP levels, cardiovascular disease (CVD) outcomes, and mortality in Caucasian (CA) and African American (AA) participants of the Cardiovascular Health Study (CHS) was evaluated using ANCOVA and Cox proportional hazards models. The minor allele of the promoter SNP 1510 and intronic SNP 3572 associates with significantly higher serum IL-6 and CRP levels in CA but not AA. The CRP association persisted after CA and AA populations were combined and after accounting for multiple comparisons. These associations did not carry through to cardiovascular disease outcomes. Decreased risk of stroke was identified in CA, with the minor allele of SNP 1111 (HRR 0.71, 95% CI 0.52, 0.95), P = 0.02, and increased risk of CVD and all-cause mortality (HRR 1.31, 95% CI 1.05–1.64) in AAs heterozygote for SNP 2989. While genetic variation in the IL-6 gene was associated with circulating IL-6 and especially with CRP concentrations in this study, there is little evidence for association between common IL-6 gene variation and adverse health outcomes in this population of older adults.     

PLINK: a tool set for whole-genome association and population-based linkage analyses

Whole-genome association studies (WGAS) bring new computational, as well as analytic, challenges to researchers. Many existing genetic-analysis tools are not designed to handle such large data sets in a convenient manner and do not necessarily exploit the new opportunities that whole-genome data bring. To address these issues, we developed PLINK, an open-source C/C++ WGAS tool set. With PLINK, large data sets comprising hundreds of thousands of markers genotyped for thousands of individuals can be rapidly manipulated and analyzed in their entirety. As well as providing tools to make the basic analytic steps computationally efficient, PLINK also supports some novel approaches to whole-genome data that take advantage of whole-genome coverage. We introduce PLINK and describe the five main domains of function: data management, summary statistics, population stratification, association analysis, and identity-by-descent estimation. In particular, we focus on the estimation and use of identity-by-state and identity-by-descent information in the context of population-based whole-genome studies. This information can be used to detect and correct for population stratification and to identify extended chromosomal segments that are shared identical by descent between very distantly related individuals. Analysis of the patterns of segmental sharing has the potential to map disease loci that contain multiple rare variants in a population-based linkage analysis.     

C677T polymorphism of the methylenetetrahydrofolate reductase gene and the risk of ischemic stroke in Polish subjects

Hyperhomocysteinemia is reported to be an independent risk factor for the development of ischemic stroke. Several studies on genetic variants of methylenetetrahydrofolate reductase (MTHFR, which plays a crucial role in regulation of plasma homocysteine concentration) reported an association between C677T gene polymorphism and stroke in some Asian populations. No study but one detected this association in Caucasians. The purpose of the present case-control study was to find a relationship betweenMTHFR genotypes and stroke in a Polish population.MTHFR genotypes were determined by PCR in 152 patients with ischemic stroke from northwestern Poland and in 135 consecutive newborns from the same population. The TT genotype and the T allele were significantly more frequent in patients than in the control group (11.8% vs. 4.4%, and 34.5% vs. 21.5%,P < 0.01). When males and females were analyzed separately, the differences were statistically significant in both genders. It is concluded that presence of the T allele is a risk factor for ischemic stroke in Polish subjects.     

Lack of association between pro-inflammatory genotypes of the interleukin-1 (IL-1B -31 C/+ and IL-1RN *2/*2) and gastric cancer/duodenal ulcer in Korean population

IL-1beta is a pro-inflammatory cytokine with multiple biological effects and is a potent inhibitor of gastric acid secretion, and IL-1RN has been shown to be associated with enhanced IL-1beta production in vitro. Recently, it was reported that the pro-inflammatory genotypes, IL-1B -31 C/+ and IL-1RN *2/*2, were associated with an increased risk of gastric cancer in a Caucasian population. We tested the association between the polymorphisms and 190 gastric cancer, 117 duodenal ulcer, and 172 healthy subjects as controls in the Korean population. The allele frequency of IL-1B -31 C was more prevalent in Korean (51%) than in Caucasian (30%), while the frequency of IL-1RN *2 allele was less in Korean (6%) than in Caucasian (27%). Using the IL-1B TT genotype as a reference group, the CC genotype was not associated with an increased risk of gastric cancer or duodenal ulcer in the Korean population (odds ratios (OR)=0.90, 95% confidence interval (CI)=0.50-1.64; OR=0.72, 95% CI=0.36-1.46, respectively). Similarly, IL-1RN*2 was not a risk genotype for either gastric cancer or duodenal ulcer. No association was recognized on the haplotype analysis of the two genes, either. Our results did not support the previous report that IL-1B -31 C/IL-1RN*2 polymorphisms were associated with an increased risk of gastric cancer. The lack of association with duodenal ulcer also suggested that the polymorphisms were not directly related to the acid-secreting capability.