Background The UK Biobank is a large prospective cohort, based in the UK, that has deep phenotypic and genomic data on roughly a half a million individuals. Included in this resource are data on approximately 78,000 individuals with “non-white British ancestry.” While most epidemiology studies have focused predominantly on populations of European ancestry, there is an opportunity to contribute to the study of health and disease for a broader segment of the population by making use of the UK Biobank’s “non-white British ancestry” samples. Here, we present an empirical description of the continental ancestry and population structure among the individuals in this UK Biobank subset. Results Reference populations from the 1000 Genomes Project for Africa, Europe, East Asia, and South Asia were used to estimate ancestry for each individual. Those with at least 80% ancestry in one of these four continental ancestry groups were taken forward (N = 62,484). Principal component and K-means clustering analyses were used to identify and characterize population structure within each ancestry group. Of the approximately 78,000 individuals in the UK Biobank that are of “non-white British” ancestry, 50,685, 6653, 2782, and 2364 individuals were associated to the European, African, South Asian, and East Asian continental ancestry groups, respectively. Each continental ancestry group exhibits prominent population structure that is consistent with self-reported country of birth data and geography. Conclusions Methods outlined here provide an avenue to leverage UK Biobank’s deeply phenotyped data allowing researchers to maximize its potential in the study of health and disease in individuals of non-white British ancestry.

Introduction: Severe malaria remains a deadly disease for many young children in low – and middle–income countries. Levels of Interleukin–6 (IL–6) have been shown to identify cases of severe malaria and associate with severity, but it is unknown if this association is causal, or whether manipulation of IL–6 might alter outcomes in severe malaria. Methods: A single nucleotide polymorphism (SNP, rs2228145) in the IL–6 receptor (IL6R) was chosen as a genetic variant that is known to alter IL–6 signalling. We measured the association between the minor allele of this SNP (C) and C–reactive protein (CRP) levels, a marker of IL–6 signalling in the non–European ancestry population recruited to UK Biobank. We then took this forward as an instrument to perform Mendelian randomisation (MR) in MalariaGEN, a large cohort study of patients with severe malaria at eleven worldwide sites. As a secondary approach, we identified cis protein quantitative trait loci (cis–pQTL) for IL6R itself and other markers of IL–6 signalling in a recently published GWAS of the plasma proteome performed in African Americans. We then performed MR using these instruments in the African MalariaGEN sites (9/11). Analyses were performed at each site, and meta–analysed using inverse variance weighting. Additional analyses were performed for specific sub–phenotypes of severe malaria: cerebral malaria and severe malarial anaemia. Results: The minor allele (C) of rs2228145 was associated with decreased CRP across all tested continental ancestries in UK Biobank. There was no evidence of heterogeneity of effect and a large overall effect (beta -0.11 per standard deviation of normalised CRP per C allele, p = 7.55 x 10-255) In Mendelian randomisation studies using this SNP, we did not identify an effect of decreased IL–6 signalling on severe malaria case status (Odds ratio 1.14, 95% CI 0.56 – 2.34, p = 0.713). Estimates of the association with any severe malaria sub–phenotype were similarly null although there was significant imprecision in all estimates. Using an alternative instrument (cis–pQTLs for IL6R), which included 3 SNPS (including rs2228145), we identified the same null effect, but with greater precision (Odds ratio 1.02, 95% CI 0.95 1.10), and no effect on any severe malaria subtypes. Conclusions: Mendelian randomisation analyses using a SNP in the IL–6 receptor known to alter IL–6 signalling do not support a causal role for IL–6 signalling in the development of severe malaria, or any severe malaria sub–phenotype. This result suggests IL–6 may not be causal for severe outcomes in malaria, and that therapeutic manipulation of IL–6 may not be a suitable treatment for severe malaria.

Aims/hypothesis: Epidemiological studies have generated conflicting findings on the relationship between anti-diabetic medication use and cancer risk. Naturally occurring variation in genes encoding anti-diabetic drug targets can be used to investigate the effect of their pharmacological perturbation on cancer risk. Methods: We developed genetic instruments for three anti-diabetic drug targets (peroxisome proliferator activated receptor gamma, PPARG; sulfonylurea receptor 1, ABCC8; glucagon-like peptide 1 receptor, GLP1R) using summary genetic association data from a genome-wide association study (GWAS) of type 2 diabetes in 69,869 cases and 127,197 controls in the Million Veteran Program. Genetic instruments were constructed using cis-acting genome-wide significant (P<5x10-8) single-nucleotide polymorphisms (SNPs) permitted to be in weak linkage disequilibrium (r2<0.20). Summary genetic association estimates for these SNPs were obtained from GWAS consortia for the following cancers: breast (122,977 cases, 105,974 controls), colorectal (58,221 cases, 67,694 controls), prostate (79,148 cases, 61,106 controls), and overall (i.e. site-combined) cancer (27,483 cases, 372,016 controls). Inverse-variance weighted random-effects models adjusting for linkage disequilibrium were employed to estimate causal associations between genetically-proxied drug target perturbation and cancer risk. Colocalisation analysis was employed to examine robustness of findings to violations of Mendelian randomization (MR) assumptions. A Bonferroni correction was employed as a heuristic to define associations from MR analyses as "strong" and "weak" evidence. Results: In Mendelian randomization analysis, genetically-proxied PPARG perturbation was weakly associated with higher risk of prostate cancer (OR for PPARG perturbation equivalent to a 1 unit decrease in inverse-rank normal transformed HbA1c: 1.75, 95% CI 1.07-2.85, P=0.02). In histological subtype-stratified analyses, genetically-proxied PPARG perturbation was weakly associated with lower risk of ER+ breast cancer (OR 0.57, 95% CI 0.38-0.85; P=6.45 x 10-3). In colocalisation analysis however, there was little evidence of shared causal variants for type 2 diabetes liability and cancer endpoints in the PPARG locus, though these analyses were likely underpowered. There was little evidence to support associations of genetically-proxied PPARG perturbation with colorectal or overall cancer risk or genetically-proxied ABCC8 or GLP1R perturbation with risk across cancer endpoints. Conclusions/interpretation: Our drug-target MR analyses did not find consistent evidence to support an association of genetically-proxied PPARG, ABCC8 or GLP1R perturbation with breast, colorectal, prostate or overall cancer risk. Further evaluation of these drug targets using alternative molecular epidemiological approaches may help to further corroborate the findings presented in this analysis.

Observational studies have suggested a protective role for eosinophils in colorectal cancer (CRC) development and implicated neutrophils, but the causal relationships remain unclear. Here, we aimed to estimate the causal effect of circulating white blood cell (WBC) counts (N = ∼550,000) for basophils, eosinophils, monocytes, lymphocytes and neutrophils on CRC risk (N = 52,775 cases and 45,940 controls) using Mendelian randomization (MR). For comparison, we also examined this relationship using individual-level data from UK Biobank (4,043 incident CRC cases and 332,773 controls) in a longitudinal cohort analysis. The inverse-variance weighted (IVW) MR analysis suggested a protective effect of increased basophil count and eosinophil count on CRC risk [OR per 1-SD increase: 0.88, CI(95%): 0.78-0.99,P=0.04; OR: 0.93, CI(95%): 0.88-0.98,P=0.01]. The protective effect of eosinophils remained [OR per 1-SD increase: 0.88, CI(95%): 0.80-0.97,P=0.01] following adjustments for all other WBC subtypes, to account for genetic correlation between the traits, using multivariable MR. A protective effect of increased lymphocyte count on CRC risk was also found [OR: 0.84, CI(95%): 0.76-0.93,P=6.70e-4] following adjustment. Consistent with MR results, a protective effect for eosinophils in the cohort analysis in the fully adjusted model [RR per 1-SD increase: 0.96, CI(95%): 0.93-0.99,P=0.02] and following adjustment for the other WBC subtypes [RR: 0.96, CI(95%): 0.93-0.99,P=0.001] was observed. Our study implicates peripheral blood immune cells, in particular eosinophils and lymphocytes, in CRC development, highlighting a need for mechanistic studies to interrogate these relationships.What is already known of this topicWhile previous observational studies have suggested a protective role for eosinophils in colorectal cancer development and implicated neutrophils, whether changes in the levels of circulating white blood cells causes colorectal cancer has not been explored.What this study addsOur study is the first to use Mendelian randomization (MR) to investigate this relationship. In parallel, for comparison, we also conduct the largest cohort study to date on the topic. We found evidence to suggest that elevated eosinophil and lymphocyte count may have a protective effect on CRC risk, adding new insights into the pathogenesis of CRC.How this study might affect research, practice or policyOur findings will encourage further mechanistic exploration to understand the biological mechanisms underpinning our findings, which may lead to new therapeutic approaches or risk reduction strategies.

Observational studies have suggested a protective role for eosinophils in colorectal cancer (CRC) development and implicated neutrophils, but the causal relationships remain unclear. Here, we aimed to estimate the causal effect of circulating white blood cell (WBC) counts (N = ~550 000) for basophils, eosinophils, monocytes, lymphocytes and neutrophils on CRC risk (N = 52 775 cases and 45 940 controls) using Mendelian randomisation (MR). For comparison, we also examined this relationship using individual‐level data from UK Biobank (4043 incident CRC cases and 332 773 controls) in a longitudinal cohort analysis. The inverse‐variance weighted (IVW) MR analysis suggested a protective effect of increased basophil count and eosinophil count on CRC risk [OR per 1‐SD increase: 0.88, 95% CI: 0.78‐0.99, P = .04; OR: 0.93, 95% CI: 0.88‐0.98, P = .01]. The protective effect of eosinophils remained [OR per 1‐SD increase: 0.88, 95% CI: 0.80‐0.97, P = .01] following adjustments for all other WBC subtypes, to account for genetic correlation between the traits, using multivariable MR. A protective effect of increased lymphocyte count on CRC risk was also found [OR: 0.84, 95% CI: 0.76‐0.93, P = 6.70e‐4] following adjustment. Consistent with MR results, a protective effect for eosinophils in the cohort analysis in the fully adjusted model [RR per 1‐SD increase: 0.96, 95% CI: 0.93‐0.99, P = .02] and following adjustment for the other WBC subtypes [RR: 0.96, 95% CI: 0.93‐0.99, P = .001] was observed. Our study implicates peripheral blood immune cells, in particular eosinophils and lymphocytes, in CRC development, highlighting a need for mechanistic studies to interrogate these relationships.

BackgroundThe UK Biobank is a large prospective cohort, based in the United Kingdom, that has deep phenotypic and genomic data on roughly a half a million individuals. Included in this resource are data on approximately 78,000 individuals with “non-white British ancestry”. Whilst most epidemiology studies have focused predominantly on populations of European ancestry, there is an opportunity to contribute to the study of health and disease for a broader segment of the population by making use of the UK Biobank’s “non-white British ancestry” samples. Here we present an empirical description of the continental ancestry and population structure among the individuals in this UK Biobank subset.ResultsReference populations from the 1000 Genomes Project for Africa, Europe, East Asia, and South Asia were used to estimate ancestry for each individual. Those with at least 80% ancestry in one of these four continental ancestry groups were taken forward (N=62,484). Principal component and K-means clustering analyses were used to identify and characterize population structure within each ancestry group. Of the approximately 78,000 individuals in the UK Biobank that are of “non-white British” ancestry, 50,685, 6,653, 2,782, and 2,364 individuals were associated to the European, African, South Asian, and East Asian continental ancestry groups, respectively. Each continental ancestry group exhibits prominent population structure that is consistent with self-reported country of birth data and geography.ConclusionsMethods outlined here provide an avenue to leverage UK Biobank’s deeply phenotyped data allowing researchers to maximise its potential in the study of health and disease in individuals of non-white British ancestry.

Recognizing the early signs of cancer risk is vital for informing prevention, early detection, and survival. To investigate whether changes in circulating metabolites characterise the early stages of colorectal cancer (CRC) development, we examined associations between a genetic risk score (GRS) associated with CRC liability (72 single nucleotide polymorphisms) and 231 circulating metabolites measured by nuclear magnetic resonance spectroscopy in the Avon Longitudinal Study of Parents and Children (N=6,221). Linear regression models were applied to examine associations between genetic liability to colorectal cancer and circulating metabolites measured in the same individuals at age 8, 16, 18 and 25 years. The GRS for CRC was associated with up to 28% of the circulating metabolites at FDR-P<0.05 across all time points, particularly with higher fatty acids and very-low- and low-density lipoprotein subclass lipids. Two-sample reverse Mendelian randomization (MR) analyses investigating CRC liability (52,775 cases, 45,940 controls) and metabolites measured in a random subset of UK Biobank participants (N=118,466, median age 58y) revealed broadly consistent effect estimates with the GRS analysis. In conventional (forward) MR analyses, genetically predicted polyunsaturated fatty acid concentrations were most strongly associated with higher CRC risk. These analyses suggest that higher genetic liability to CRC can cause early alterations in systemic metabolism, and suggest that fatty acids may play an important role in CRC development.

ResultsForty-seven exposures passed an exposure-exposure correlation-adjusted Bonferroni P-value threshold (5.43E-05). These included previously hypothesised risk factors for DVT (e.g. body mass index, varicose veins, height, hyperthyroidism) and novel associations (e.g. prospective memory, basal metabolic rate). ConclusionOur analyses confirmed causal associations of risk factors previously associated with DVT and highlighted several novel risk factors for the disease. Our study demonstrates the utility of using a hypothesis free Mendelian randomization approach for the identification of novel disease risk factors.