Polygenic Prediction of Weight and Obesity Trajectories from Birth to Adulthood.

Severe obesity is a rapidly growing global health threat. Although often attributed to unhealthy lifestyle choices or environmental factors, obesity is known to be heritable and highly polygenic; the majority of inherited susceptibility is related to the cumulative effect of many common DNA variants. Here we derive and validate a new polygenic predictor comprised of 2.1 million common variants to quantify this susceptibility and test this predictor in more than 300,000 individuals ranging from middle age to birth. Among middle-aged adults, we observe a 13-kg gradient in weight and a 25-fold gradient in risk of severe obesity across polygenic score deciles. In a longitudinal birth cohort, we note minimal differences in birthweight across score deciles, but a significant gradient emerged in early childhood and reached 12 kg by 18 years of age. This new approach to quantify inherited susceptibility to obesity affords new opportunities for clinical prevention and mechanistic assessment.

Trial arm outcome variance difference after dropout as an indicator of missing-not-at-random bias in randomized controlled trials.

Randomized controlled trials (RCTs) are vulnerable to bias from missing data. When outcomes are missing not at random (MNAR), estimates from complete case analysis (CCA) and multiple imputation (MI) may be biased. There is no statistical test for distinguishing between outcomes missing at random (MAR) and MNAR. Current strategies rely on comparing dropout proportions and covariate distributions, and using auxiliary information to assess the likelihood of dropout being associated with the outcome. We propose using the observed variance difference across trial arms as a tool for assessing the risk of dropout being MNAR in RCTs with continuous outcomes. In an RCT, at randomization, the distributions of all covariates should be equal in the populations randomized to the intervention and control arms. Under the assumption of homogeneous treatment effects and homoskedastic outcome errors, the variance of the outcome will also be equal in the two populations over the course of follow-up. We show that under MAR dropout, the observed outcome variances, conditional on the variables included in the model, are equal across trial arms, whereas MNAR dropout may result in unequal variances. Consequently, unequal observed conditional trial arm variances are an indicator of MNAR dropout and possible bias of the estimated treatment effect. Heterogeneous treatment effects or heteroskedastic outcome errors are another potential cause of observing different outcome variances. We show that for longitudinal data, we can isolate the effect of MNAR outcome-dependent dropout by considering the variance difference at baseline in the same set of patients who are observed at final follow-up. We illustrate our method in simulation for CCA and MI, and in applications using individual-level data and summary data.

Common variation at 16p11.2 is associated with glycosuria in pregnancy: findings from a genome-wide association study in European women.

Glycosuria is a condition where glucose is detected in urine at higher concentrations than normal (i.e. not detectable). Glycosuria at some point during pregnancy has an estimated prevalence of 50% and is associated with adverse outcomes in both mothers and offspring. Little is currently known about the genetic contribution to this trait or the extent to which it overlaps with other seemingly related traits, e.g. diabetes. We performed a genome-wide association study (GWAS) for self-reported glycosuria in pregnant mothers from the Avon Longitudinal Study of Parents and Children (cases/controls = 1249/5140). We identified two loci, one of which (lead SNP = rs13337037; chromosome 16; odds ratio of glycosuria per effect allele: 1.42; 95% CI: 1.30, 1.56; P = 1.97 × 10-13) was then validated using an obstetric measure of glycosuria measured in the same cohort (227/6639). We performed a secondary GWAS in the 1986 Northern Finland Birth Cohort (NFBC1986; 747/2991) using midwife-reported glycosuria and offspring genotype as a proxy for maternal genotype. The combined results revealed evidence for a consistent effect on glycosuria at the chromosome 16 locus. In follow-up analyses, we saw little evidence of shared genetic underpinnings with the exception of urinary albumin-to-creatinine ratio (Rg = 0.64; SE = 0.22; P = 0.0042), a biomarker of kidney disease. In conclusion, we identified a genetic association with self-reported glycosuria during pregnancy, with the lead SNP located 15kB upstream of SLC5A2, a target of antidiabetic drugs. The lack of strong genetic correlation with seemingly related traits such as type 2 diabetes suggests different genetic risk factors exist for glycosuria during pregnancy.

Applying Mendelian randomization to appraise causality in relationships between nutrition and cancer.

Dietary factors are assumed to play an important role in cancer risk, apparent in consensus recommendations for cancer prevention that promote nutritional changes. However, the evidence in this field has been generated predominantly through observational studies, which may result in biased effect estimates because of confounding, exposure misclassification, and reverse causality. With major geographical differences and rapid changes in cancer incidence over time, it is crucial to establish which of the observational associations reflect causality and to identify novel risk factors as these may be modified to prevent the onset of cancer and reduce its progression. Mendelian randomization (MR) uses the special properties of germline genetic variation to strengthen causal inference regarding potentially modifiable exposures and disease risk. MR can be implemented through instrumental variable (IV) analysis and, when robustly performed, is generally less prone to confounding, reverse causation and measurement error than conventional observational methods and has different sources of bias (discussed in detail below). It is increasingly used to facilitate causal inference in epidemiology and provides an opportunity to explore the effects of nutritional exposures on cancer incidence and progression in a cost-effective and timely manner. Here, we introduce the concept of MR and discuss its current application in understanding the impact of nutritional factors (e.g., any measure of diet and nutritional intake, circulating biomarkers, patterns, preference or behaviour) on cancer aetiology and, thus, opportunities for MR to contribute to the development of nutritional recommendations and policies for cancer prevention. We provide applied examples of MR studies examining the role of nutritional factors in cancer to illustrate how this method can be used to help prioritise or deprioritise the evaluation of specific nutritional factors as intervention targets in randomised controlled trials. We describe possible biases when using MR, and methodological developments aimed at investigating and potentially overcoming these biases when present. Lastly, we consider the use of MR in identifying causally relevant nutritional risk factors for various cancers in different regions across the world, given notable geographical differences in some cancers. We also discuss how MR results could be translated into further research and policy. We conclude that findings from MR studies, which corroborate those from other well-conducted studies with different and orthogonal biases, are poised to substantially improve our understanding of nutritional influences on cancer. For such corroboration, there is a requirement for an interdisciplinary and collaborative approach to investigate risk factors for cancer incidence and progression.

Genomic analysis of diet composition finds novel loci and associations with health and lifestyle.

We conducted genome-wide association studies (GWAS) of relative intake from the macronutrients fat, protein, carbohydrates, and sugar in over 235,000 individuals of European ancestries. We identified 21 unique, approximately independent lead SNPs. Fourteen lead SNPs are uniquely associated with one macronutrient at genome-wide significance (P < 5 × 10-8), while five of the 21 lead SNPs reach suggestive significance (P < 1 × 10-5) for at least one other macronutrient. While the phenotypes are genetically correlated, each phenotype carries a partially unique genetic architecture. Relative protein intake exhibits the strongest relationships with poor health, including positive genetic associations with obesity, type 2 diabetes, and heart disease (rg ≈ 0.15-0.5). In contrast, relative carbohydrate and sugar intake have negative genetic correlations with waist circumference, waist-hip ratio, and neighborhood deprivation (|rg| ≈ 0.1-0.3) and positive genetic correlations with physical activity (rg ≈ 0.1 and 0.2). Relative fat intake has no consistent pattern of genetic correlations with poor health but has a negative genetic correlation with educational attainment (rg ≈-0.1). Although our analyses do not allow us to draw causal conclusions, we find no evidence of negative health consequences associated with relative carbohydrate, sugar, or fat intake. However, our results are consistent with the hypothesis that relative protein intake plays a role in the etiology of metabolic dysfunction.

Sensitivity to missing not at random dropout in clinical trials: Use and interpretation of the trimmed means estimator.

Outcome values in randomized controlled trials (RCTs) may be missing not at random (MNAR), if patients with extreme outcome values are more likely to drop out (eg, due to perceived ineffectiveness of treatment, or adverse effects). In such scenarios, estimates from complete case analysis (CCA) and multiple imputation (MI) will be biased. We investigate the use of the trimmed means (TM) estimator for the case of univariable missingness in one continuous outcome. The TM estimator operates by setting missing values to the most extreme value, and then "trimming" away equal fractions of both groups, estimating the treatment effect using the remaining data. The TM estimator relies on two assumptions, which we term the "strong MNAR" and "location shift" assumptions. We derive formulae for the TM estimator bias resulting from the violation of these assumptions for normally distributed outcomes. We propose an adjusted TM estimator, which relaxes the location shift assumption and detail how our bias formulae can be used to establish the direction of bias of CCA and TM estimates, to inform sensitivity analyses. The TM approach is illustrated in a sensitivity analysis of the CoBalT RCT of cognitive behavioral therapy (CBT) in 469 individuals with 46 months follow-up. Results were consistent with a beneficial CBT treatment effect, with MI estimates closer to the null and TM estimates further from the null than the CCA estimate. We propose using the TM estimator as a sensitivity analysis for data where extreme outcome value dropout is plausible.

Body muscle gain and markers of cardiovascular disease susceptibility in young adulthood: A cohort study.

BACKGROUND: The potential benefits of gaining body muscle for cardiovascular disease (CVD) susceptibility, and how these compare with the potential harms of gaining body fat, are unknown. We compared associations of early life changes in body lean mass and handgrip strength versus body fat mass with atherogenic traits measured in young adulthood. METHODS AND FINDINGS: Data were from 3,227 offspring of the Avon Longitudinal Study of Parents and Children (39% male; recruited in 1991-1992). Limb lean and total fat mass indices (kg/m2) were measured using dual-energy X-ray absorptiometry scans performed at age 10, 13, 18, and 25 y (across clinics occurring from 2001-2003 to 2015-2017). Handgrip strength was measured at 12 and 25 y, expressed as maximum grip (kg or lb/in2) and relative grip (maximum grip/weight in kilograms). Linear regression models were used to examine associations of change in standardised measures of these exposures across different stages of body development with 228 cardiometabolic traits measured at age 25 y including blood pressure, fasting insulin, and metabolomics-derived apolipoprotein B lipids. SD-unit gain in limb lean mass index from 10 to 25 y was positively associated with atherogenic traits including very-low-density lipoprotein (VLDL) triglycerides. This pattern was limited to lean gain in legs, whereas lean gain in arms was inversely associated with traits including VLDL triglycerides, insulin, and glycoprotein acetyls, and was also positively associated with creatinine (a muscle product and positive control). Furthermore, this pattern for arm lean mass index was specific to SD-unit gains occurring between 13 and 18 y, e.g., -0.13 SD (95% CI -0.22, -0.04) for VLDL triglycerides. Changes in maximum and relative grip from 12 to 25 y were both positively associated with creatinine, but only change in relative grip was also inversely associated with atherogenic traits, e.g., -0.12 SD (95% CI -0.18, -0.06) for VLDL triglycerides per SD-unit gain. Change in fat mass index from 10 to 25 y was more strongly associated with atherogenic traits including VLDL triglycerides, at 0.45 SD (95% CI 0.39, 0.52); these estimates were directionally consistent across sub-periods, with larger effect sizes with more recent gains. Associations of lean, grip, and fat measures with traits were more pronounced among males. Study limitations include potential residual confounding of observational estimates, including by ectopic fat within muscle, and the absence of grip measures in adolescence for estimates of grip change over sub-periods. CONCLUSIONS: In this study, we found that muscle strengthening, as indicated by grip strength gain, was weakly associated with lower atherogenic trait levels in young adulthood, at a smaller magnitude than unfavourable associations of fat mass gain. Associations of muscle mass gain with such traits appear to be smaller and limited to gains occurring in adolescence. These results suggest that body muscle is less robustly associated with markers of CVD susceptibility than body fat and may therefore be a lower-priority intervention target.

Investigating the relationships between unfavourable habitual sleep and metabolomic traits: evidence from multi-cohort multivariable regression and Mendelian randomization analyses.

BACKGROUND: Sleep traits are associated with cardiometabolic disease risk, with evidence from Mendelian randomization (MR) suggesting that insomnia symptoms and shorter sleep duration increase coronary artery disease risk. We combined adjusted multivariable regression (AMV) and MR analyses of phenotypes of unfavourable sleep on 113 metabolomic traits to investigate possible biochemical mechanisms linking sleep to cardiovascular disease. METHODS: We used AMV (N = 17,368) combined with two-sample MR (N = 38,618) to examine effects of self-reported insomnia symptoms, total habitual sleep duration, and chronotype on 113 metabolomic traits. The AMV analyses were conducted on data from 10 cohorts of mostly Europeans, adjusted for age, sex, and body mass index. For the MR analyses, we used summary results from published European-ancestry genome-wide association studies of self-reported sleep traits and of nuclear magnetic resonance (NMR) serum metabolites. We used the inverse-variance weighted (IVW) method and complemented this with sensitivity analyses to assess MR assumptions. RESULTS: We found consistent evidence from AMV and MR analyses for associations of usual vs. sometimes/rare/never insomnia symptoms with lower citrate (- 0.08 standard deviation (SD)[95% confidence interval (CI) - 0.12, - 0.03] in AMV and - 0.03SD [- 0.07, - 0.003] in MR), higher glycoprotein acetyls (0.08SD [95% CI 0.03, 0.12] in AMV and 0.06SD [0.03, 0.10) in MR]), lower total very large HDL particles (- 0.04SD [- 0.08, 0.00] in AMV and - 0.05SD [- 0.09, - 0.02] in MR), and lower phospholipids in very large HDL particles (- 0.04SD [- 0.08, 0.002] in AMV and - 0.05SD [- 0.08, - 0.02] in MR). Longer total sleep duration associated with higher creatinine concentrations using both methods (0.02SD per 1 h [0.01, 0.03] in AMV and 0.15SD [0.02, 0.29] in MR) and with isoleucine in MR analyses (0.22SD [0.08, 0.35]). No consistent evidence was observed for effects of chronotype on metabolomic measures. CONCLUSIONS: Whilst our results suggested that unfavourable sleep traits may not cause widespread metabolic disruption, some notable effects were observed. The evidence for possible effects of insomnia symptoms on glycoprotein acetyls and citrate and longer total sleep duration on creatinine and isoleucine might explain some of the effects, found in MR analyses of these sleep traits on coronary heart disease, which warrant further investigation.

Assessing the causal role of body mass index on cardiovascular health in young adults: Mendelian randomization and recall-by-genotype analyses.

Background: Body mass index (BMI) has been suggested to be causally related to cardiovascular health in mid-to-late life, but this has not been explored systematically at younger ages - nor with detailed cardiovascular phenotyping. Recall-by-Genotype (RbG) is an approach that enables the collection of precise phenotypic measures in smaller studies, whilst maintaining statistical power and ability for causal inference. Methods: In this study, we used a combination of conventional multivariable regression analysis, Mendelian randomization (MR) and sub-sample RbG methodologies to estimate the causal effect of BMI on gross-level and detailed cardiovascular health in healthy participants from the Avon Longitudinal Study of Parents and Children at age 17 (N=1420-3108 for different outcomes) and an independent sample from the same cohort (for RbG) study at age 21 (N=386-418). Results: In both MR and RbG analyses, results suggested that higher BMI causes higher blood pressure (BP) and left ventricular mass index (LVMI) in young adults (e.g., difference in LVMI per kg/m2 using MR: 1.07g/m2.7; 95% CI: 0.62, 1.52; P=3.87x10-06 and per 3.58kg/m2 using RbG: 1.65g/m2.7 95% CI: 0.83, 2.47; P=0.0001). Additionally, RbG results suggested a causal role of higher BMI on higher stroke volume (SV: difference per 3.58kg/m2: 1.49ml/m2.04; 95% CI: 0.62, 2.35; P=0.001) and cardiac output (CO: difference per 3.58kg/m2: 0.11l/min/m1.83; 95% CI: 0.03, 0.19; P=0.01) but no strong evidence for a causal role on systemic vascular resistance or total arterial compliance. Neither analysis supported a causal role of higher BMI on heart rate. Conclusions: Complementary MR and RbG causal methodologies, together with a range of sensitivity analyses, suggest that higher BMI is likely to cause worse cardiovascular health, specifically higher BP and LVMI, even in youth. Higher BMI also resulted in increased CO in the RbG study, which appeared to be solely driven by SV, as neither MR nor RbG analyses suggested a causal effect of BMI on heart rate. These consistent results support efforts to reduce BMI from a young age to prevent later adverse cardiovascular health and illustrate the potential for phenotypic resolution with maintained analytical power using RbG.

Variation of all-cause and cause-specific mortality with body mass index in one million Swedish parent-son pairs: An instrumental variable analysis.

BACKGROUND: High body mass index (BMI) is associated with mortality, but the pervasive problem of confounding and reverse causality in observational studies limits inference about the direction and magnitude of causal effects. We aimed to obtain estimates of the causal association of BMI with all-cause and cause-specific mortality. METHODS AND FINDINGS: In a record-linked, intergenerational prospective study from the general population of Sweden, we used two-sample instrumental variable (IV) analysis with data from 996,898 fathers (282,407 deaths) and 1,013,083 mothers (153,043 deaths) and their sons followed up from January 1, 1961, until December 31, 2004. Sons' BMI was used as the instrument for parents' BMI to compute hazard ratios (HRs) for risk of mortality per standard deviation (SD) higher parents' BMI. Using offspring exposure as an instrument for parents' exposure is unlikely to be affected by reverse causality (an important source of bias in this context) and reduces confounding. IV analyses supported causal associations between higher BMI and greater risk of all-cause mortality (HR [95% confidence interval (CI)] per SD higher fathers' BMI: 1.29 [1.26-1.31] and mothers' BMI: 1.39 [1.35-1.42]) and overall cancer mortality (HR per SD higher fathers' BMI: 1.20 [1.16-1.24] and mothers' BMI: 1.29 [1.24-1.34]), including 9 site-specific cancers in men (bladder, colorectum, gallbladder, kidney, liver, lung, lymphatic system, pancreas, and stomach) and 11 site-specific cancers in women (gallbladder, kidney, liver, lung, lymphatic system, ovaries, pancreas, stomach, uterus, cervix, and endometrium). There was evidence supporting causal associations between higher BMI in mothers and greater risk of mortality from kidney disease (HR: 2.17 [1.68-2.81]) and lower risk of mortality from suicide (HR: 0.77 [0.65-0.90]). In both sexes, there was evidence supporting causal associations between higher BMI and mortality from cardiovascular diseases (CVDs), stroke, diabetes, and respiratory diseases. We were unable to test the association between sons' and mothers' BMIs (as mothers' data were unavailable) or whether the instrument was independent of unmeasured or residual confounding; however, the associations between parents' mortality and sons' BMI were negligibly influenced by adjustment for available confounders. CONCLUSIONS: Consistent with previous large-scale meta-analyses and reviews, results supported the causal role of higher BMI in increasing the risk of several common causes of death, including cancers with increasing global incidence. We also found positive effects of BMI on mortality from respiratory disease, prostate cancer, and lung cancer, which has been inconsistently reported in the literature, suggesting that the causal role of higher BMI in mortality from these diseases may be underestimated. Furthermore, we expect different patterns of bias in the current observational and IV analyses; therefore, the similarities between our findings from both methods increases confidence in the results. These findings support efforts to understand the mechanisms underpinning these effects to inform targeted interventions and develop population-based strategies to reduce rising obesity levels for disease prevention.

Influence of puberty timing on adiposity and cardiometabolic traits: A Mendelian randomisation study.

BACKGROUND: Earlier puberty is widely linked with future obesity and cardiometabolic disease. We examined whether age at puberty onset likely influences adiposity and cardiometabolic traits independent of childhood adiposity. METHODS AND FINDINGS: One-sample Mendelian randomisation (MR) analyses were conducted on up to 3,611 white-European female and male offspring from the Avon Longitudinal Study of Parents and Children (ALSPAC) cohort recruited at birth via mothers between 1 April 1991 and 31 December 1992. Time-sensitive exposures were age at menarche and age at voice breaking. Outcomes measured at age 18 y were body mass index (BMI), dual-energy X-ray absorptiometry-based fat and lean mass indices, blood pressure, and 230 cardiometabolic traits derived from targeted metabolomics (150 concentrations plus 80 ratios from nuclear magnetic resonance [NMR] spectroscopy covering lipoprotein subclasses of cholesterol and triglycerides, amino acids, inflammatory glycoproteins, and others). Adjustment was made for pre-pubertal BMI measured at age 8 y. For negative control MR analyses, BMI and cardiometabolic trait measures taken at age 8 y (before puberty, and which therefore cannot be an outcome of puberty itself) were used. For replication analyses, 2-sample MR was conducted using summary genome-wide association study data on up to 322,154 adults for post-pubertal BMI, 24,925 adults for post-pubertal NMR cardiometabolic traits, and 13,848 children for pre-pubertal obesity (negative control). Like observational estimates, 1-sample MR estimates in ALSPAC using 351 polymorphisms for age at menarche (explaining 10.6% of variance) among 2,053 females suggested that later age at menarche (per year) was associated with -1.38 kg/m2 of BMI at age 18 y (or -0.34 SD units, 95% CI -0.46, -0.23; P = 9.77 × 10-09). This coefficient attenuated 10-fold upon adjustment for BMI at age 8 y, to -0.12 kg/m2 (or -0.03 SDs, 95% CI -0.13, 0.07; P = 0.55). Associations with blood pressure were similar, but associations across other traits were small and inconsistent. In negative control MR analyses, later age at menarche was associated with -0.77 kg/m2 of pre-pubertal BMI measured at age 8 y (or -0.39 SDs, 95% CI -0.50, -0.29; P = 6.28 × 10-13), indicating that variants influencing menarche also influence BMI before menarche. Cardiometabolic trait associations were weaker and less consistent among males and both sexes combined. Higher BMI at age 8 y (per 1 kg/m2 using 95 polymorphisms for BMI explaining 3.4% of variance) was associated with earlier menarche among 2,648 females (by -0.26 y, 95% CI -0.37, -0.16; P = 1.16 × 10-06), likewise among males and both sexes combined. In 2-sample MR analyses using 234 polymorphisms and inverse variance weighted (IVW) regression, each year later age at menarche was associated with -0.81 kg/m2 of adult BMI (or -0.17 SD units, 95% CI -0.21, -0.12; P = 4.00 × 10-15). Associations were weaker with cardiometabolic traits. Using 202 polymorphisms, later menarche was associated with lower odds of childhood obesity (IVW-based odds ratio = 0.52 per year later, 95% CI 0.48, 0.57; P = 6.64 × 10-15). Study limitations include modest sample sizes for 1-sample MR, lack of inference to non-white-European populations, potential selection bias through modest completion rates of puberty questionnaires, and likely disproportionate measurement error of exposures by sex. The cardiometabolic traits examined were heavily lipid-focused and did not include hormone-related traits such as insulin and insulin-like growth factors. CONCLUSIONS: Our results suggest that puberty timing has a small influence on adiposity and cardiometabolic traits and that preventive interventions should instead focus on reducing childhood adiposity.

Polygenic risk scores associate with blood pressure traits across the lifespan.

AIMS: Hypertension is a major modifiable cause of morbidity and mortality that affects over 1 billion people worldwide. Blood pressure (BP) traits have a strong genetic component that can be quantified with polygenic risk scores (PRSs). To date, the performance of BP PRSs has mainly been assessed in adults, and less is known about polygenic hypertension risk in childhood. METHODS AND RESULTS: Multiple PRSs for systolic BP (SBP), diastolic BP (DBP), and pulse pressure were developed using either genome-wide significant weights, pruning and thresholding, or Bayesian regression. Among 87 total PRSs, the top performer for each trait was applied in independent cohorts of children and adult to assess genotype-phenotype associations and disease risk across the lifespan. Differences between those with low (1st decile), average (2nd-9th decile), and high (10th decile) PRS emerge in the first years of life and are maintained throughout adulthood. These diverging BP trajectories also seem to affect cardiovascular and renal disease risk, with increased risk observed among those in the top decile and reduced risk among those in the bottom decile of the polygenic risk distribution compared with the rest of the population. CONCLUSION: Genetic risk factors are associated with BP traits across the lifespan, beginning in the first years of life. Given the importance of exposure time in disease pathogenesis and the early rise in BP levels among those genetically susceptible, PRSs may help identify high-risk individuals prior to hypertension onset, facilitate primordial prevention, and reduce the burden of this public health challenge.

A high genetic risk of elevated blood pressure (BP) is associated with increased BP from early childhood and throughout the lifespan. Inherited predispositions also affect the risk of cardiovascular morbidity and mortality, yet this appears to be modified by the absence or presence of hypertension, indicating that genetic hypertension risk is not deterministic, and that controlling BP can and should be done across the polygenic risk distribution. Given that differences in BP emerge early in life as a function of genetic risk, polygenic risk scores have the potential to reduce the duration of exposure to high BP by identifying high-risk individuals from birth, and thereby attenuate lifelong disease risk.

Trans-ancestral genome-wide association study of longitudinal pubertal height growth and shared heritability with adult health outcomes.

BACKGROUND: Pubertal growth patterns correlate with future health outcomes. However, the genetic mechanisms mediating growth trajectories remain largely unknown. Here, we modeled longitudinal height growth with Super-Imposition by Translation And Rotation (SITAR) growth curve analysis on ~ 56,000 trans-ancestry samples with repeated height measurements from age 5 years to adulthood. We performed genetic analysis on six phenotypes representing the magnitude, timing, and intensity of the pubertal growth spurt. To investigate the lifelong impact of genetic variants associated with pubertal growth trajectories, we performed genetic correlation analyses and phenome-wide association studies in the Penn Medicine BioBank and the UK Biobank. RESULTS: Large-scale growth modeling enables an unprecedented view of adolescent growth across contemporary and 20th-century pediatric cohorts. We identify 26 genome-wide significant loci and leverage trans-ancestry data to perform fine-mapping. Our data reveals genetic relationships between pediatric height growth and health across the life course, with different growth trajectories correlated with different outcomes. For instance, a faster tempo of pubertal growth correlates with higher bone mineral density, HOMA-IR, fasting insulin, type 2 diabetes, and lung cancer, whereas being taller at early puberty, taller across puberty, and having quicker pubertal growth were associated with higher risk for atrial fibrillation. CONCLUSION: We report novel genetic associations with the tempo of pubertal growth and find that genetic determinants of growth are correlated with reproductive, glycemic, respiratory, and cardiac traits in adulthood. These results aid in identifying specific growth trajectories impacting lifelong health and show that there may not be a single "optimal" pubertal growth pattern.

Comparison of intergenerational instrumental variable analyses of body mass index and mortality in UK Biobank.

BACKGROUND: An increasing proportion of people have a body mass index (BMI) classified as overweight or obese and published studies disagree whether this will be beneficial or detrimental to health. We applied and evaluated two intergenerational instrumental variable methods to estimate the average causal effect of BMI on mortality in a cohort with many deaths: the parents of UK Biobank participants. METHODS: In Cox regression models, parental BMI was instrumented by offspring BMI using an 'offspring as instrument' (OAI) estimation and by offspring BMI-related genetic variants in a 'proxy-genotype Mendelian randomization' (PGMR) estimation. RESULTS: Complete-case analyses were performed in parents of 233 361 UK Biobank participants with full phenotypic, genotypic and covariate data. The PGMR method suggested that higher BMI increased mortality with hazard ratios per kg/m2 of 1.02 (95% CI: 1.01, 1.04) for mothers and 1.04 (95% CI: 1.02, 1.05) for fathers. The OAI method gave considerably higher estimates, which varied according to the parent-offspring pairing between 1.08 (95% CI: 1.06, 1.10; mother-son) and 1.23 (95% CI: 1.16, 1.29; father-daughter). CONCLUSION: Both methods supported a causal role of higher BMI increasing mortality, although caution is required regarding the immediate causal interpretation of these exact values. Evidence of instrument invalidity from measured covariates was limited for the OAI method and minimal for the PGMR method. The methods are complementary for interrogating the average putative causal effects because the biases are expected to differ between them.

Challenges in undertaking nonlinear Mendelian randomization.

Mendelian randomization (MR) is a widely used method that exploits the unique properties of germline genetic variation to strengthen causal inference in relationships between exposures and outcomes. Nonlinear MR allows estimation of the shape of these relationships. In a previous paper, the authors applied linear and nonlinear MR to estimate the effect of BMI on mortality in UK Biobank, providing evidence for a J-shaped association. However, it is now clear that there are problems with widely used nonlinear MR methods, which draws attention to the likely erroneous nature of the conclusions regarding the shapes of several explored relationships. Here, the authors explore the utility and likely biases of these nonlinear MR methods with the use of a negative control design. Although there remains good evidence for a causal effect of higher BMI increasing the risk of mortality, the pattern of this association across different levels of BMI requires further characterization.

Genetic Predictors for Fecal Propionate and Butyrate-Producing Microbiome Pathway Are Not Associated with Colorectal Cancer Risk: A Mendelian Randomization Analysis.

BACKGROUND: Mechanistic data indicate the benefit of short-chain fatty acids (SCFA) produced by gut microbial fermentation of fiber on colorectal cancer, but direct epidemiologic evidence is limited. A recent study identified SNPs for two SCFA traits (fecal propionate and butyrate-producing microbiome pathway PWY-5022) in Europeans and showed metabolic benefits. METHODS: We conducted a two-sample Mendelian randomization analysis of the genetic instruments for the two SCFA traits (three SNPs for fecal propionate and nine for PWY-5022) in relation to colorectal cancer risk in three large European genetic consortia of 58,131 colorectal cancer cases and 67,347 controls. We estimated the risk of overall colorectal cancer and conducted subgroup analyses by sex, age, and anatomic subsites of colorectal cancer. RESULTS: We did not observe strong evidence for an association of the genetic predictors for fecal propionate levels and the abundance of PWY-5022 with the risk of overall colorectal cancer, colorectal cancer by sex, or early-onset colorectal cancer (diagnosed at <50 years), with no evidence of heterogeneity or pleiotropy. When assessed by tumor subsites, we found weak evidence for an association between PWY-5022 and risk of rectal cancer (OR per 1-SD, 0.95; 95% confidence intervals, 0.91-0.99; P = 0.03) but it did not surpass multiple testing of subgroup analysis. CONCLUSIONS: Genetic instruments for fecal propionate levels and the abundance of PWY-5022 were not associated with colorectal cancer risk. IMPACT: Fecal propionate and PWY-5022 may not have a substantial influence on colorectal cancer risk. Future research is warranted to comprehensively investigate the effects of SCFA-producing bacteria and SCFAs on colorectal cancer risk.

Application of Mendelian randomization to explore the causal role of the human gut microbiome in colorectal cancer.

The role of the human gut microbiome in colorectal cancer (CRC) is unclear as most studies on the topic are unable to discern correlation from causation. We apply two-sample Mendelian randomization (MR) to estimate the causal relationship between the gut microbiome and CRC. We used summary-level data from independent genome-wide association studies to estimate the causal effect of 14 microbial traits (n = 3890 individuals) on overall CRC (55,168 cases, 65,160 controls) and site-specific CRC risk, conducting several sensitivity analyses to understand the nature of results. Initial MR analysis suggested that a higher abundance of Bifidobacterium and presence of an unclassified group of bacteria within the Bacteroidales order in the gut increased overall and site-specific CRC risk. However, sensitivity analyses suggested that instruments used to estimate relationships were likely complex and involved in many potential horizontal pleiotropic pathways, demonstrating that caution is needed when interpreting MR analyses with gut microbiome exposures. In assessing reverse causality, we did not find strong evidence that CRC causally affected these microbial traits. Whilst our study initially identified potential causal roles for two microbial traits in CRC, importantly, further exploration of these relationships highlighted that these were unlikely to reflect causality.

Mendelian randomization: application to cardiovascular disease.

In the absence of an ethical, practical, and economical randomized trial, the epidemiologist is left to explore other methods in efforts to assert causality. An approach based on genotypic variation has the potential to mitigate against some of the problems found within conventional observational studies. Genetic variations associated with risk factors of interest at the population level can be used as proxy measures for these risk factors and to generate estimates of causal effect. The potential and the possible limitations of this approach within the cardiovascular field are presented in this review.

Mendelian randomization analysis of the causal impact of body mass index and waist-hip ratio on rates of hospital admission.

We analyze how measures of adiposity - body mass index (BMI) and waist hip ratio (WHR) - causally influence rates of hospital admission. Conventional analyses of this relationship are susceptible to omitted variable bias from variables that jointly influence both hospital admission and adipose status. We implement a novel quasi-Poisson instrumental variable model in a Mendelian randomization framework, identifying causal effects from random perturbations to germline genetic variation. We estimate the individual and joint effects of BMI, WHR, and WHR adjusted for BMI. We also implement multivariable instrumental variable methods in which the causal effect of one exposure is estimated conditionally on the causal effect of another exposure. Data on 310,471 participants and over 550,000 inpatient admissions in the UK Biobank were used to perform one-sample and two-sample Mendelian randomization analyses. The results supported a causal role of adiposity on hospital admissions, with consistency across all estimates and sensitivity analyses. Point estimates were generally larger than estimates from comparable observational specifications. We observed an attenuation of the BMI effect when adjusting for WHR in the multivariable Mendelian randomization analyses, suggesting that an adverse fat distribution, rather than a higher BMI itself, may drive the relationship between adiposity and risk of hospital admission.