Proteomic Analyses in Diverse Populations Improved Risk Prediction and Identified New Drug Targets for Type 2 Diabetes

OBJECTIVE

Integrated analyses of plasma proteomics and genetic data in prospective studies can help assess the causal relevance of proteins, improve risk prediction, and discover novel protein drug targets for type 2 diabetes (T2D).

RESEARCH DESIGN AND METHODS

We measured plasma levels of 2,923 proteins using Olink Explore among ∼2,000 randomly selected participants from China Kadoorie Biobank (CKB) without prior diabetes at baseline. Cox regression assessed associations of individual protein with incident T2D (n = 92 cases). Proteomic-based risk models were developed with discrimination, calibration, reclassification assessed using area under the curve (AUC), calibration plots, and net reclassification index (NRI), respectively. Two-sample Mendelian randomization (MR) analyses using cis-protein quantitative trait loci identified in a genome-wide association study of CKB and UK Biobank for specific proteins were conducted to assess their causal relevance for T2D, along with colocalization analyses to examine shared causal variants between proteins and T2D.

RESULTS

Overall, 33 proteins were significantly associated (false discovery rate <0.05) with risk of incident T2D, including IGFBP1, GHR, and amylase. The addition of these 33 proteins to a conventional risk prediction model improved AUC from 0.77 (0.73-0.82) to 0.88 (0.85-0.91) and NRI by 38%, with predicted risks well calibrated with observed risks. MR analyses provided support for the causal relevance for T2D of ENTR1, LPL, and PON3, with replication of ENTR1 and LPL in Europeans using different genetic instruments. Moreover, colocalization analyses showed strong evidence (pH4 > 0.6) of shared genetic variants of LPL and PON3 with T2D.

CONCLUSIONS

Proteomic analyses in Chinese adults identified novel associations of multiple proteins with T2D with strong genetic evidence supporting their causal relevance and potential as novel drug targets for prevention and treatment of T2D.

Ethnic differences of genetic risk and smoking in lung cancer: two prospective cohort studies

BACKGROUND

The role of genetic background underlying the disparity of relative risk of smoking and lung cancer between European populations and East Asians remains unclear.

METHODS

To assess the role of ethnic differences in genetic factors associated with smoking-related risk of lung cancer, we first constructed ethnic-specific polygenic risk scores (PRSs) to quantify individual genetic risk of lung cancer in Chinese and European populations. Then, we compared genetic risk and smoking as well as their interactions on lung cancer between two cohorts, including the China Kadoorie Biobank (CKB) and the UK Biobank (UKB). We also evaluated the absolute risk reduction over a 5-year period.

RESULTS

Differences in compositions and association effects were observed between the Chinese-specific PRSs and European-specific PRSs, especially for smoking-related loci. The PRSs were consistently associated with lung cancer risk, but stronger associations were observed in smokers of the UKB [hazard ratio (HR) 1.26 vs 1.15, P  =  0.028]. A significant interaction between genetic risk and smoking on lung cancer was observed in the UKB (RERI, 11.39 (95% CI, 7.01-17.94)], but not in the CKB. Obvious higher absolute risk was observed in nonsmokers of the CKB, and a greater absolute risk reduction was found in the UKB (10.95 vs 7.12 per 1000 person-years, P <0.001) by comparing heavy smokers with nonsmokers, especially for those at high genetic risk.

CONCLUSIONS

Ethnic differences in genetic factors and the high incidence of lung cancer in nonsmokers of East Asian ethnicity were involved in the disparity of smoking-related risk of lung cancer.

Life-course adiposity and severe liver disease: a Mendelian randomization analysis

OBJECTIVE

There is little evidence on the genetic associations between life-course adiposity (including birth weight, childhood BMI, and adulthood BMI) and severe liver disease (SLD; including cirrhosis and liver cancer). The current study aimed to examine and contrast these associations.

METHODS

Genetic variants were obtained from genome-wide association studies. Two-sample Mendelian randomization (MR) analyses were performed to assess the genetic associations of life-course adiposity with SLD and liver biomarkers. Cox regression was used to estimate adjusted hazard ratios for SLD associated with genetic risk scores of life-course adiposity and adulthood weight change in the China Kadoorie Biobank.

RESULTS

In observational analyses, genetic predispositions to childhood adiposity and adulthood adiposity were each associated with SLD. There was a U-shaped association between adulthood weight change and risk of SLD. In meta-analyses of MR results, genetically predicted 1-standard deviation increase in birth weight was inversely associated with SLD at a marginal significance (odds ratio: 0.81 [95% CI: 0.65-1.00]), whereas genetically predicted 1-standard deviation higher childhood BMI and adulthood BMI were positively associated with SLD (odds ratio: 1.27 [95% CI: 1.05-1.55] and 1.79 [95% CI: 1.59-2.01], respectively). The results of liver biomarkers mirrored those of SLD.

CONCLUSIONS

The current study provided genetic evidence on the associations between life-course adiposity and SLD.

Conventional and genetic associations of adiposity with 1463 proteins in relatively lean Chinese adults

Adiposity is associated with multiple diseases and traits, but little is known about the causal relevance and mechanisms underlying these associations. Large-scale proteomic profiling, especially when integrated with genetic data, can clarify mechanisms linking adiposity with disease outcomes. We examined the associations of adiposity with plasma levels of 1463 proteins in 3977 Chinese adults, using measured and genetically-instrumented BMI. We further used two-sample bi-directional MR analyses to assess if certain proteins influenced adiposity, along with other (e.g. enrichment) analyses to clarify possible mechanisms underlying the observed associations. Overall, the mean (SD) baseline BMI was 23.9 (3.3) kg/m2, with only 6% being obese (i.e. BMI ≥ 30 kg/m2). Measured and genetically-instrumented BMI was significantly associated at FDR < 0.05 with levels of 1096 (positive/inverse: 826/270) and 307 (positive/inverse: 270/37) proteins, respectively, with FABP4, LEP, IL1RN, LSP1, GOLM2, TNFRSF6B, and ADAMTS15 showing the strongest positive and PON3, NCAN, LEPR, IGFBP2 and MOG showing the strongest inverse genetic associations. These associations were largely linear, in adiposity-to-protein direction, and replicated (> 90%) in Europeans of UKB (mean BMI 27.4 kg/m2). Enrichment analyses of the top > 50 BMI-associated proteins demonstrated their involvement in atherosclerosis, lipid metabolism, tumour progression and inflammation. Two-sample bi-directional MR analyses using cis-pQTLs identified in CKB GWAS found eight proteins (ITIH3, LRP11, SCAMP3, NUDT5, OGN, EFEMP1, TXNDC15, PRDX6) significantly affect levels of BMI, with NUDT5 also showing bi-directional association. The findings among relatively lean Chinese adults identified novel pathways by which adiposity may increase disease risks and novel potential targets for treatment of obesity and obesity-related diseases.

Improving the generalizability of protein-ligand binding predictions with AI-Bind

Identifying novel drug-target interactions is a critical and rate-limiting step in drug discovery. While deep learning models have been proposed to accelerate the identification process, here we show that state-of-the-art models fail to generalize to novel (i.e., never-before-seen) structures. We unveil the mechanisms responsible for this shortcoming, demonstrating how models rely on shortcuts that leverage the topology of the protein-ligand bipartite network, rather than learning the node features. Here we introduce AI-Bind, a pipeline that combines network-based sampling strategies with unsupervised pre-training to improve binding predictions for novel proteins and ligands. We validate AI-Bind predictions via docking simulations and comparison with recent experimental evidence, and step up the process of interpreting machine learning prediction of protein-ligand binding by identifying potential active binding sites on the amino acid sequence. AI-Bind is a high-throughput approach to identify drug-target combinations with the potential of becoming a powerful tool in drug discovery.

Genetically Predicted Differences in Systolic Blood Pressure and Risk of Cardiovascular and Noncardiovascular Diseases: A Mendelian Randomization Study in Chinese Adults

BACKGROUND

Mendelian randomization studies of systolic blood pressure (SBP) can assess the shape and strength of the associations of genetically predicted differences in SBP with major disease outcomes and are less constrained by biases in observational analyses. This study aimed to compare the associations of usual and genetically predicted SBP with major cardiovascular disease (CVD) outcomes, overall and by levels of SBP, age, and sex.

METHODS

The China Kadoorie Biobank involved a 12-year follow-up of a prospective study of 489 495 adults aged 40 to 79 years with no prior CVD and 86 060 with genetic data. Outcomes included major vascular events (59 490/23 151 in observational/genetic analyses), and its components (ischemic stroke [n=39 513/12 043], intracerebral hemorrhage [7336/5243], and major coronary events [7871/4187]). Genetically predicted SBP used 460 variants obtained from European ancestry genome-wide studies. Cox regression estimated adjusted hazard ratios for incident CVD outcomes down to usual SBP levels of 120 mm Hg.

RESULTS

Both observational and genetic analyses demonstrated log-linear positive associations of SBP with major vascular event and other major CVD types in the range of 120 to 170 mm Hg. Consistent with the observational analyses, the hazard ratios per 10 mm Hg higher genetically predicted SBP were 2-fold greater for intracerebral hemorrhage (1.71 [95% CI, 1.58-1.87]) than for ischemic stroke (1.37 [1.30-1.45]) or major coronary event (1.29 [1.18-1.42]). Genetic analyses also demonstrated 2-fold greater hazard ratios for major vascular event in younger (1.69 [95% CI, 1.54-1.86]) than in older people (1.28 [1.18-1.38]).

CONCLUSIONS

The findings provide support for initiation of blood pressure-lowering treatment at younger ages and below the conventional cut-offs for hypertension to maximize CVD prevention, albeit the absolute risks of CVD are far greater in older people.

Conventional and genetic risk factors for chronic Hepatitis B virus infection in a community-based study of 0.5 million Chinese adults

Despite universal vaccination of newborns, the prevalence of chronic hepatitis virus B (HBV) infection and the associated disease burden remain high among adults in China. We investigated risk factors for chronic HBV infection in a community-based study of 512,726 individuals aged 30-79 years recruited from ten diverse areas during 2004-2008. Multivariable logistic regression was used to estimate odds ratios (ORs) of hepatitis B surface antigen (HBsAg) positivity recorded at baseline by sociodemographic and lifestyle factors, and medical history. In a random subset (n = 69,898) we further assessed the association of 18 single nucleotide polymorphisms (SNPs) previously shown to be associated with HBsAg positivity and development of chronic liver disease (CLD) (1600 cases). Several factors showed strong associations with HBsAg positivity, particularly younger age (< 40 vs. ≥ 60 years: OR 1.48, 95% CI 1.32-1.66), male sex (1.40, 1.34-1.46) and urban residency (1.55, 1.47-1.62). Of the 18 SNPs selected, 17 were associated with HBsAg positivity, and 14 with CLD, with SNPs near HLA-DPB1 were most strongly associated with both outcomes. In Chinese adults a range of genetic and non-genetic factors were associated with chronic HBV infection and CLD, which can inform targeted screening to help prevent disease progression.

Circulating proteins and risk of pancreatic cancer: a case-subcohort study among Chinese adults

BACKGROUND

Pancreatic cancer has a very poor prognosis. Biomarkers that may help predict or diagnose pancreatic cancer may lead to earlier diagnosis and improved survival.

METHODS

The prospective China Kadoorie Biobank (CKB) recruited 512 891 adults aged 30-79 years during 2004-08, recording 702 incident cases of pancreatic cancer during 9 years of follow-up. We conducted a case-subcohort study measuring 92 proteins in 610 cases and a subcohort of 623 individuals, using the OLINK immuno-oncology panel in stored baseline plasma samples. Cox regression with the Prentice pseudo-partial likelihood was used to estimate adjusted hazard ratios (HRs) for risk of pancreatic cancer by protein levels.

RESULTS

Among 1233 individuals (including 610 cases), several chemokines, interleukins, growth factors and membrane proteins were associated with risk of pancreatic cancer, with adjusted HRs per 1 standard deviation (SD) of 0.86 to 1.86, including monocyte chemotactic protein 3 (MCP3/CCL7) {1.29 [95% CI (confidence interval) (1.10, 1.51)]}, angiopoietin-2 (ANGPT2) [1.27 (1.10, 1.48)], interleukin-18 (IL18) [1.24 (1.07, 1.43)] and interleukin-6 (IL6) [1.21 (1.06, 1.38)]. Associations between some proteins [e.g. matrix metalloproteinase-7 (MMP7), hepatocyte growth factor (HGF) and tumour necrosis factor receptor superfamily member 9 [TNFRSF9)] and risk of pancreatic cancer were time-varying, with higher levels associated with higher short-term risk. Within the first year, the discriminatory ability of a model with known risk factors (age, age squared, sex, region, smoking, alcohol, education, diabetes and family history of cancer) was increased when several proteins were incorporated (weighted C-statistic changed from 0.85 to 0.99; P for difference = 4.5 × 10-5), although only a small increase in discrimination (0.77 to 0.79, P = 0.04) was achieved for long-term risk.

CONCLUSIONS

Several plasma proteins were associated with subsequent diagnosis of pancreatic cancer. The potential clinical utility of these biomarkers warrants further investigation.

Sero-prevalence of 19 infectious pathogens and associated factors among middle-aged and elderly Chinese adults: a cross-sectional study

OBJECTIVES

To systematically assess the sero-prevalence and associated factors of major infectious pathogens in China, where there are high incidence rates of certain infection-related cancers.

DESIGN

Cross-sectional study.

SETTING

10 (5 urban, 5 rural) geographically diverse areas in China.

PARTICIPANTS

A subcohort of 2000 participants from the China Kadoorie Biobank.

PRIMARY MEASURES

Sero-prevalence of 19 pathogens using a custom-designed multiplex serology panel and associated factors.

RESULTS

Of the 19 pathogens investigated, the mean number of sero-positive pathogens was 9.4 (SD 1.7), with 24.4% of participants being sero-positive for >10 pathogens. For individual pathogens, the sero-prevalence varied, being for example, 0.05% for HIV, 6.4% for human papillomavirus (HPV)-16, 53.5% for Helicobacter pylori (H. pylori) and 99.8% for Epstein-Barr virus . The sero-prevalence of human herpesviruses (HHV)-6, HHV-7 and HPV-16 was higher in women than men. Several pathogens showed a decreasing trend in sero-prevalence by birth cohort, including hepatitis B virus (HBV) (51.6% vs 38.7% in those born <1940 vs >1970), HPV-16 (11.4% vs 5.4%), HHV-2 (15.1% vs 8.1%), Chlamydia trachomatis (65.6% vs 28.8%) and Toxoplasma gondii (22.0% vs 9.0%). Across the 10 study areas, sero-prevalence varied twofold to fourfold for HBV (22.5% to 60.7%), HPV-16 (3.4% to 10.9%), H. pylori (16.2% to 71.1%) and C. trachomatis (32.5% to 66.5%). Participants with chronic liver diseases had >7-fold higher sero-positivity for HBV (OR=7.51; 95% CI 2.55 to 22.13).

CONCLUSIONS

Among Chinese adults, previous and current infections with certain pathogens were common and varied by area, sex and birth cohort. These infections may contribute to the burden of certain cancers and other non-communicable chronic diseases.

Abstract 154: Sex-specific Genome Wide Association Study Of Early-onset Ischemic Stroke

Introduction: Genetic studies of early-onset disease have been an effective strategy to identify novel pathways and drug targets generalizable also to later-onset disease. Few studies have investigated the sex-specific genetic associations with early-onset ischemic stroke even though several features of ischemic stroke differ between males and females. We hypothesized that stratifying the GWAS by sex would reveal novel stroke loci.

Methods: We performed a transethnic ischemic stroke GWAS of 3,056 female cases and 4,462 male cases < 60 years-old and 16,192 and 16,048 sex-matched controls, respectively, from the Early Onset Stroke Genetics Consortium.

Results: We identified a significant association in women with a locus in close proximity to TMX1 , a transmembrane platelet protein that inhibits platelet function. Additionally, we identified 2 other suggestive (P < 5 x 10 -6 ) loci in females (see Table), i.e., at APOH , which encodes beta2-glycoprotein I, an established GWAS locus for lipoprotein(a), and LRFN2 which has been previously reported to associate with obesity-related measures and type II diabetes. We observed suggestive evidence for association in males with MMP3/MMP12 , a known stroke susceptibility locus.

Conclusions: Despite a very modest sample size, sex-specific analyses identified suggestive associations at biologically important novel loci in females and a known stroke locus in males. Further studies of sex-specific associations in both early- and later-onset ischemic stroke are needed.

Trans-ethnic Mendelian-randomization study reveals causal relationships between cardiometabolic factors and chronic kidney disease

BACKGROUND

This study was to systematically test whether previously reported risk factors for chronic kidney disease (CKD) are causally related to CKD in European and East Asian ancestries using Mendelian randomization.

METHODS

A total of 45 risk factors with genetic data in European ancestry and 17 risk factors in East Asian participants were identified as exposures from PubMed. We defined the CKD by clinical diagnosis or by estimated glomerular filtration rate of <60 ml/min/1.73 m2. Ultimately, 51 672 CKD cases and 958 102 controls of European ancestry from CKDGen, UK Biobank and HUNT, and 13 093 CKD cases and 238 118 controls of East Asian ancestry from Biobank Japan, China Kadoorie Biobank and Japan-Kidney-Biobank/ToMMo were included.

RESULTS

Eight risk factors showed reliable evidence of causal effects on CKD in Europeans, including genetically predicted body mass index (BMI), hypertension, systolic blood pressure, high-density lipoprotein cholesterol, apolipoprotein A-I, lipoprotein(a), type 2 diabetes (T2D) and nephrolithiasis. In East Asians, BMI, T2D and nephrolithiasis showed evidence of causality on CKD. In two independent replication analyses, we observed that increased hypertension risk showed reliable evidence of a causal effect on increasing CKD risk in Europeans but in contrast showed a null effect in East Asians. Although liability to T2D showed consistent effects on CKD, the effects of glycaemic phenotypes on CKD were weak. Non-linear Mendelian randomization indicated a threshold relationship between genetically predicted BMI and CKD, with increased risk at BMI of >25 kg/m2.

CONCLUSIONS

Eight cardiometabolic risk factors showed causal effects on CKD in Europeans and three of them showed causality in East Asians, providing insights into the design of future interventions to reduce the burden of CKD.

Liver biomarkers, genetic and lifestyle risk factors in relation to risk of cardiovascular disease in Chinese

Background and aims

Liver biomarkers and metabolic associated fatty liver disease (MAFLD) have been shown to be associated with cardiovascular disease (CVD). However, there is limited evidence on CVD subtypes [myocardial infarction (MI), ischemic stroke (IS), and intracerebral hemorrhage (ICH)], especially in the Chinese population. We examined these associations overall, by genetic predisposition to non-alcoholic fatty liver disease (NAFLD), and by lifestyle risk factors.

Approach and results

This is a nested case-control study of CVD (10,298 cases and 5,388 controls) within the China Kadoorie Biobank. Cox regression was used to estimate adjusted hazard ratios (HRs) for CVD associated with liver biomarkers and MAFLD and by stratum of genetic risk and a combined high-risk lifestyle score. For liver enzymes, there were positive associations with MI and IS, but no associations with ICH or carotid plaque. There were positive associations of NAFLD with risks of MI, IS, and ICH (HR 1.43 [95% CI 1.30-1.57], 1.25 [1.16-1.35], and 1.12 [1.02-1.23]) as well as carotid plaque (odds ratio 2.36 [1.12-4.96]). The associations of NAFLD with CVD and carotid plaque were stronger among individuals with a high genetic risk (ICH: p-interaction < 0.05), while the associations with stroke were stronger among those with a favorable lifestyle (p-interaction < 0.05). The results for MAFLD mirrored those for NAFLD.

Conclusion

In Chinese adults, liver biomarkers and MAFLD were associated with risk of CVD, with different magnitudes of associations by CVD subtypes. Genetic predisposition to NAFLD and lifestyle factors modified the associations of fatty liver with stroke.

Laboratory safety evaluation of bedinvetmab, a canine anti-nerve growth factor monoclonal antibody, in dogs

Nerve growth factor (NGF), a critical mediator of nociception, is a novel analgesic therapeutic target. Bedinvetmab, a canine monoclonal antibody (mAb), binds NGF and inhibits its interaction with tropomyosin receptor kinase A (trkA) and p75 neurotrophin receptor (p75^NTR) receptors. The objective of three integrated laboratory studies was to demonstrate the safety of bedinvetmab in adult laboratory Beagle dogs. Daily health, veterinary, clinical pathology, systemic exposure, and anti-drug antibody evaluations were performed. Study 1 additionally included electrocardiography, neurologic, and ophthalmic assessments, and radiographic monitoring of joints of the appendicular skeleton. Study 2 evaluated T-lymphocyte-dependent immune function. Study 3 evaluated the safety of short-term concurrent administration of carprofen, a nonsteroidal anti-inflammatory drug (NSAID), with bedinvetmab. Studies 1 and 3 included terminal pathology and histopathology evaluations. Study designs and procedures included directed complementary morphologic and functional evaluations of a literature- and in vitro-based list of potential safety issues related to the NGF signaling pathway and characteristics engineered into this mAb. Screening-level general procedures evaluated effects associated with mAbs that target and inhibit soluble agonist cytokines. There were no treatment-related adverse changes in clinical evaluations, clinical neurological and ophthalmic examinations, joints, immune morphology or function, and no effects of short-term concurrent NSAID usage. Treatment-emergent immunogenicity was not observed. Bedinvetmab (1 mg/kg SC monthly; 3× and 10× dose multiples) was well tolerated in normal laboratory Beagle dogs for 6 months and with 2 weeks' concurrent NSAID administration.

Physics-informed machine learning: case studies for weather and climate modelling

Machine learning (ML) provides novel and powerful ways of accurately and efficiently recognizing complex patterns, emulating nonlinear dynamics, and predicting the spatio-temporal evolution of weather and climate processes. Off-the-shelf ML models, however, do not necessarily obey the fundamental governing laws of physical systems, nor do they generalize well to scenarios on which they have not been trained. We survey systematic approaches to incorporating physics and domain knowledge into ML models and distill these approaches into broad categories. Through 10 case studies, we show how these approaches have been used successfully for emulating, downscaling, and forecasting weather and climate processes. The accomplishments of these studies include greater physical consistency, reduced training time, improved data efficiency, and better generalization. Finally, we synthesize the lessons learned and identify scientific, diagnostic, computational, and resource challenges for developing truly robust and reliable physics-informed ML models for weather and climate processes. This article is part of the theme issue 'Machine learning for weather and climate modelling'.

Improved lipidomic profile mediates the effects of adherence to healthy lifestyles on coronary heart disease

Adherence to healthy lifestyles is associated with reduced risk of coronary heart disease (CHD), but uncertainty persists about the underlying lipid pathway. In a case-control study of 4681 participants nested in the prospective China Kadoorie Biobank, 61 lipidomic markers in baseline plasma were measured by targeted nuclear magnetic resonance spectroscopy. Baseline lifestyles included smoking, alcohol consumption, dietary habit, physical activity, and adiposity levels. Genetic instrument was used to mimic the lipid-lowering effect of statins. We found that 35 lipid metabolites showed statistically significant mediation effects in the pathway from healthy lifestyles to CHD reduction, including very low-density lipoprotein (VLDL) particles and their cholesterol, large-sized high-density lipoprotein (HDL) particle and its cholesterol, and triglyceride in almost all lipoprotein subfractions. The statins genetic score was associated with reduced intermediate- and low-density lipoprotein, but weak or no association with VLDL and HDL. Lifestyle interventions and statins may improve different components of the lipid profile.

Observational and Genetic Associations of Body Mass Index and Hepatobiliary Diseases in a Relatively Lean Chinese Population

IMPORTANCE

There is some support for the existence of genetic associations between adiposity and certain hepatobiliary diseases in Western populations. However, there is little evidence of such genetic associations in China, where the causes of these diseases may differ from those in Western populations and the mean body mass index (BMI) is much lower.

OBJECTIVES

To compare the observational associations of BMI with hepatobiliary diseases and liver biomarkers with the genetic associations between BMI and these factors and to assess whether the genetic associations of BMI with liver diseases differed by hepatitis B virus infection status.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study used data from the prospective China Kadoorie Biobank, including 473 938 adults aged 30 to 79 years without hepatobiliary diseases at baseline from 10 diverse areas in China from June 25, 2004, to July 15, 2008. A random sample of 75 736 participants with genotyping data was included in the Mendelian randomization analysis. Follow-up was completed January 1, 2017 (median [interquartile range] length of follow-up, 10.2 [9.2-11.1] years). Data were analyzed from January to October 2019.

EXPOSURES

Measured BMI obtained during the baseline survey and genetically instrumented BMI derived using 92 single-nucleotide variations.

MAIN OUTCOMES AND MEASURES

Incident cases of hepatobiliary diseases, liver enzymes, fatty liver index, and fibrosis score.

RESULTS

Among 473 938 individuals (276 041 [58.2%] women), the mean (SD) age was 52 (10.9) years and mean (SD) BMI was 23.8 (3.4). Baseline BMI was associated with higher risks of chronic liver disease (adjusted risk ratio per 1-SD increase, 1.14; 95% CI, 1.11 to 1.17) and gallbladder disease (adjusted risk ratio per 1-SD increase, 1.29; 95% CI, 1.27 to 1.31), with heterogeneity by disease subtype (P < .001). Genetically instrumented BMI was associated with higher risks of chronic liver disease (risk ratio per 1-SD increase, 1.55; 95% CI, 1.08 to 2.24) and gallbladder disease (risk ratio per 1-SD increase, 1.40; 95% CI, 1.11 to 1.76), with no heterogeneity between subtypes. A meta-analysis of the genetic associations in China Kadoorie Biobank and those calculated in UK Biobank gave a risk ratio of 1.55 (95% CI, 1.30 to 1.84) for chronic liver disease and 1.42 (95% CI, 1.22 to 1.64) for gallbladder disease. In the China Kadoorie Biobank study, there were positive genetic associations of BMI with liver enzymes, steatosis, and fibrosis scores, consistent with observational associations. The genetic associations of BMI with liver diseases and biomarkers did not differ by hepatitis B virus infection status.

CONCLUSIONS AND RELEVANCE

In this cohort study of a relatively lean Chinese population, there were positive genetic associations of BMI with hepatobiliary diseases. These results suggest that maintaining a healthy weight through diet and physical activity may help prevent hepatobiliary diseases.

Genetic risk, incident gastric cancer, and healthy lifestyle: a meta-analysis of genome-wide association studies and prospective cohort study

BACKGROUND

Genetic variants and lifestyle factors have been associated with gastric cancer risk, but the extent to which an increased genetic risk can be offset by a healthy lifestyle remains unknown. We aimed to establish a genetic risk model for gastric cancer and assess the benefits of adhering to a healthy lifestyle in individuals with a high genetic risk.

METHODS

In this meta-analysis and prospective cohort study, we first did a fixed-effects meta-analysis of the association between genetic variants and gastric cancer in six independent genome-wide association studies (GWAS) with a case-control study design. These GWAS comprised 21 168 Han Chinese individuals, of whom 10 254 had gastric cancer and 10 914 geographically matched controls did not. Using summary statistics from the meta-analysis, we constructed five polygenic risk scores in a range of thresholds (p=5 × 10^-4 p=5 × 10^-5 p=5 × 10^-6 p=5 × 10^-7, and p=5 × 10^-8) for gastric cancer. We then applied these scores to an independent, prospective, nationwide cohort of 100 220 individuals from the China Kadoorie Biobank (CKB), with more than 10 years of follow-up. The relative and absolute risk of incident gastric cancer associated with healthy lifestyle factors (defined as not smoking, never consuming alcohol, the low consumption of preserved foods, and the frequent intake of fresh fruits and vegetables), was assessed and stratified by genetic risk (low [quintile 1 of the polygenic risk score], intermediate [quintile 2-4 of the polygenic risk score], and high [quintile 5 of the polygenic risk score]). Individuals with a favourable lifestyle were considered as those who adopted all four healthy lifestyle factors, those with an intermediate lifestyle adopted two or three factors, and those with an unfavourable lifestyle adopted none or one factor.

FINDINGS

The polygenic risk score derived from 112 single-nucleotide polymorphisms (p<5 × 10^-5) showed the strongest association with gastric cancer risk (p=7·56 × 10^-10). When this polygenic risk score was applied to the CKB cohort, we found that there was a significant increase in the relative risk of incident gastric cancer across the quintiles of the polygenic risk score (p_trend<0·0001). Compared with individuals who had a low genetic risk, those with an intermediate genetic risk (hazard ratio [HR] 1·54 [95% CI 1·22-1·94], p=2·67 × 10^-4) and a high genetic risk (2·08 [1·61-2·69], p<0·0001) had a greater risk of gastric cancer. A similar increase in the relative risk of incident gastric cancer was observed across the lifestyle categories (p_trend<0·0001), with a higher risk of gastric cancer in those with an unfavourable lifestyle than those with a favourable lifestyle (2·03 [1·46-2·83], p<0·0001). Participants with a high genetic risk and a favourable lifestyle had a lower risk of gastric cancer than those with a high genetic risk and an unfavourable lifestyle (0·53 [0·29-0·99], p=0·048), with an absolute risk reduction of 1·12% (95% CI 0·62-1·56).

INTERPRETATION

Chinese individuals at an increased risk of incident gastric cancer could be identified by use of our newly developed polygenic risk score. Compared with individuals at a high genetic risk who adopt an unhealthy lifestyle, those who adopt a healthy lifestyle could substantially reduce their risk of incident gastric cancer.

FUNDING

National Key R&D Program of China, National Natural Science Foundation of China, 333 High-Level Talents Cultivation Project of Jiangsu Province, and China Postdoctoral Science Foundation.

Cardiorespiratory Events Associated With Ophthalmic Surgery: A Single-Center, Retrospective Records Review of 130 775 Patients, 1999–2015

Purpose:

The most recent study of ophthalmic surgery morbidity and mortality was published in 1995, with a patient study population from 1977 to 1988. The present study reports surgical outcomes from a single-center, retrospective analysis of patient records from 1999 to 2015.

Methods:

Three International Classification of Diseases–9-CM codes for cardiorespiratory events were searched in the discharge diagnoses in an eye hospital over a 16-year period. The overall mortality and preoperative risk factors were analyzed, including the type of anesthetic, type of surgery, medical comorbidities, and bradycardia preceding the cardiac events.

Results:

Between February 1, 1999 and October 1, 2015, a total of 130 775 patients presented for ophthalmic surgery. Fifty-nine patients (0.45 per 1000) experienced a cardiorespiratory event. Of the 59 patients, 14 patients had a cardiorespiratory arrest, 9 of whom died during the perioperative period. Of the remaining 45 patients, 29 had significant adverse events needing some form of advanced monitoring, evaluation, and/or intervention. There was a significantly greater prevalence of diabetes among patients who had a cardiorespiratory event ( P < .001).

Conclusions:

The major risk factor associated with ophthalmic surgery morbidity and mortality was diabetes with its associated complications of autonomic neuropathy, nephropathy, and retinopathy. Of the 9 patients who died, 8 were diabetic with proliferative diabetic retinopathy and renal insufficiency/failure. The ninth mortality was secondary to a venous air embolism during ocular air infusion. The adage that “the eye is the window to our overall health” seems to be correct.

0989 Examining The Role Of Toddler Sleep Quality On Wake EEG And Language Ability

Introduction

Sleep and the development of language are prominent concerns of many parents and until recently, many have examined these concerns tangentially. Children with developmental delays/disabilities have shown to have impaired sleep and poor sleep quality, and impairments or changes in sleep quality may play a prominent role in the acquisition of language and neuronal oscillatory patterns. This study examines the role of child sleep quality paired with a normed measure of language and wake electroencephalography (EEG). Examining the role of child sleep quality with language ability and wake EEG may provide nascent incremental utility to understanding the influences of sleep on healthy development.

Methods

Data from 109 toddlers (age range 24 to 30.5 m, M = 26.83 ± 1.58 m, 52% male) from the Brief Infant Sleep Questionnaire (BISQ), Mullen Scales of Early Learning (MSEL), and continuous EEG were collected and analyzed. EEG was recorded (32 electrode cap BioSemi) while toddlers sat in a booster seat and watched a silent video. Data were analyzed in RStudio and Matlab to examine toddler’s sleep quality (infant sleep and parent behaviors) and relations with the MSEL and EEG (controlling for child age and gender).

Results

Means and standard deviations appeared within expected limits based on the range of each variable. Toddlers with slow-developing language were associated with relatively poor sleep quality, explaining 9.75% of the variance. We find preliminary evidence to suggest a potential sleep disruption around the time when a child is undergoing a rapid expansion in their vocabulary (expressive language). Toddler’s sleep quality and language acquisition were also correlated with wake EEG (alpha and beta).

Conclusion

Sleep is regarded as an essential component supporting the myriad changes observed in early development. Sleep quality fundamentally influences healthy development across domains. Here, we showed child sleep quality is highly associated with toddler’s language ability, and wake EEG, providing new insights into the developing brain.

Support

National Institutes of Health R01DC016273, R01MH107652-03S1, and Johnson & Johnson Consumer Inc., Skillman, NJ, USA.

Identification of risk loci and a polygenic risk score for lung cancer: a large-scale prospective cohort study in Chinese populations

BACKGROUND

Genetic variation has an important role in the development of non-small-cell lung cancer (NSCLC). However, genetic factors for lung cancer have not been fully identified, especially in Chinese populations, which limits the use of existing polygenic risk scores (PRS) to identify subpopulations at high risk of lung cancer for prevention. We therefore aimed to identify novel loci associated with NSCLC risk, and generate a PRS and evaluate its utility and effectiveness in the prediction of lung cancer risk in Chinese populations.

METHODS

To systematically identify genetic variants for NSCLC risk, we newly genotyped 19 546 samples from Chinese NSCLC cases and controls from the Nanjing Medical University Global Screening Array Project and did a meta-analysis of genome-wide association studies (GWASs) of 27 120 individuals with NSCLC and 27 355 without NSCLC (13 327 cases and 13 328 controls of Chinese descent as well as 13 793 cases and 14 027 controls of European descent). We then built a PRS for Chinese populations from all reported single-nucleotide polymorphisms that have been reported to be associated with lung cancer risk at genome-wide significance level. We evaluated the utility and effectiveness of the generated PRS in predicting subpopulations at high-risk of lung cancer in an independent prospective cohort of 95 408 individuals from the China Kadoorie Biobank (CKB) with more than 10 years' follow-up.

FINDINGS

We identified 19 susceptibility loci to be significantly associated with NSCLC risk at p≤5·0 × 10^-8, including six novel loci. When applied to the CKB cohort, the PRS of the risk loci successfully predicted lung cancer incident cases in a dose-response manner in participants at a high genetic risk (top 10%) than those at a low genetic risk (bottom 10%; adjusted hazard ratio 1·96, 95% CI 1·53-2·51; p_trend=2·02 × 10^-9). Specially, we observed consistently separated curves of lung cancer events in individuals at low, intermediate, and high genetic risk, respectively, and PRS was an independent effective risk stratification indicator beyond age and smoking pack-years.

INTERPRETATION

We have shown for the first time that GWAS-derived PRS can be effectively used in discriminating subpopulations at high risk of lung cancer, who might benefit from a practically feasible PRS-based lung cancer screening programme for precision prevention in Chinese populations.

FUNDING

National Natural Science Foundation of China, the Priority Academic Program for the Development of Jiangsu Higher Education Institutions, National Key R&D Program of China, Science Foundation for Distinguished Young Scholars of Jiangsu, and China's Thousand Talents Program.

Physical Activity, Sedentary Leisure Time, Circulating Metabolic Markers, and Risk of Major Vascular Diseases

BACKGROUND

Physical inactivity and sedentary behavior are associated with higher risk of cardiovascular disease (CVD). Little is known about the relevance of circulating metabolites for these associations.

METHODS

A nested case-control study within the prospective China Kadoorie Biobank included 3195 incident CVD cases (2057 occlusive CVD and 1138 intracerebral hemorrhage) and 1465 controls aged 30 to 79 years without prior CVD or statin use at baseline. Nuclear magnetic resonance spectroscopy was used to measure 225 metabolic markers and derived traits in baseline plasma samples. Linear regression was used to relate self-reported physical activity and sedentary leisure time to biomarkers, adjusting for potential confounders. These were contrasted with associations of biomarkers with occlusive CVD risk.

RESULTS

Physical activity and sedentary leisure time were associated with >100 metabolic markers, with patterns of associations generally mirroring each other. Physical activity was inversely associated with very low and low-density and positively with large and very large HDL (high-density lipoprotein) particle concentrations. Physical activity was also inversely associated with alanine, glucose, lactate, acetoacetate, and the inflammatory marker glycoprotein acetyls. In general, associations of physical activity and sedentary leisure time with specific metabolic markers were directionally consistent with the associations of these metabolic markers with occlusive CVD risk. Overall, metabolic markers potentially explained ≈70% of the protective associations of physical activity and ≈50% of the positive associations of sedentary leisure time with occlusive CVD.

CONCLUSIONS

Among Chinese adults, physical activity and sedentary behavior have opposing associations with a diverse range of circulating metabolites, which may partially explain their associations with CVD risk.

Genetic influences on eight psychiatric disorders based on family data of 4 408 646 full and half-siblings, and genetic data of 333 748 cases and controls

BACKGROUND

Most studies underline the contribution of heritable factors for psychiatric disorders. However, heritability estimates depend on the population under study, diagnostic instruments, and study designs that each has its inherent assumptions, strengths, and biases. We aim to test the homogeneity in heritability estimates between two powerful, and state of the art study designs for eight psychiatric disorders.

METHODS

We assessed heritability based on data of Swedish siblings (N = 4 408 646 full and maternal half-siblings), and based on summary data of eight samples with measured genotypes (N = 125 533 cases and 208 215 controls). All data were based on standard diagnostic criteria. Eight psychiatric disorders were studied: (1) alcohol dependence (AD), (2) anorexia nervosa, (3) attention deficit/hyperactivity disorder (ADHD), (4) autism spectrum disorder, (5) bipolar disorder, (6) major depressive disorder, (7) obsessive-compulsive disorder (OCD), and (8) schizophrenia.

RESULTS

Heritability estimates from sibling data varied from 0.30 for Major Depression to 0.80 for ADHD. The estimates based on the measured genotypes were lower, ranging from 0.10 for AD to 0.28 for OCD, but were significant, and correlated positively (0.19) with national sibling-based estimates. When removing OCD from the data the correlation increased to 0.50.

CONCLUSIONS

Given the unique character of each study design, the convergent findings for these eight psychiatric conditions suggest that heritability estimates are robust across different methods. The findings also highlight large differences in genetic and environmental influences between psychiatric disorders, providing future directions for etiological psychiatric research.

Metabolomic consequences of genetic inhibition of PCSK9 compared with statin treatment

Background

Both statins and PCSK9 inhibitors lower blood low-density lipoprotein cholesterol (LDL-C) levels to reduce risk of cardiovascular events. To assess potential differences between metabolic effects of these two lipid-lowering therapies, we performed detailed lipid and metabolite profiling of a large randomized statin trial and compared the results with the effects of genetic inhibition of PCSK9, acting as a naturally occurring trial.

Methods

228 circulating metabolic measures were quantified by nuclear magnetic resonance spectroscopy, including lipoprotein subclass concentrations and their lipid composition, fatty acids, and amino acids, for 5,359 individuals (2,659 on treatment) in the PROspective Study of Pravastatin in the Elderly at Risk (PROSPER) trial at 6-months post-randomization. The corresponding metabolic measures were analyzed in eight population cohorts (N=72,185) using PCSK9 rs11591147 as an unconfounded proxy to mimic the therapeutic effects of PCSK9 inhibitors.

Results

Scaled to an equivalent lowering of LDL-C, the effects of genetic inhibition of PCSK9 on 228 metabolic markers were generally consistent with those of statin therapy (R2=0.88). Alterations in lipoprotein lipid composition and fatty acid distribution were similar. However, discrepancies were observed for very-low-density lipoprotein (VLDL) lipid measures. For instance, genetic inhibition of PCSK9 had weaker effects on lowering of VLDL-cholesterol compared with statin therapy (54% vs. 77% reduction, relative to the lowering effect on LDL-C; P=2x10-7 for heterogeneity). Genetic inhibition of PCSK9 showed no significant effects on amino acids, ketones, or a marker of inflammation (GlycA) whereas statin treatment weakly lowered GlycA levels.

Conclusions

Genetic inhibition of PCSK9 had similar metabolic effects to statin therapy on detailed lipid and metabolite profiles. However, PCSK9 inhibitors are predicted to have weaker effects on VLDL lipids compared with statins for an equivalent lowering of LDL-C, which potentially translate into smaller reductions in cardiovascular disease risk.

Abstract P1-07-26: Clinicopathological predictors of axillary lymph node metastasis in early breast cancer patients

Axillary lymph node (ALN) status is key in the prognosis of early breast cancer. Sentinel lymph node biopsy (SLNB) is the standard treatment in determining ALN status in clinically node negative patients with early breast cancer, but nearly 70% exhibit no ALN metastasis (a regional lymph node (RLN) metastasis). Likewise, selective use of SLNB has engendered modeling ALN metastasis. Using the National Cancer Database (NCDB), this study aimed to evaluate clinicopathological factors to help predict first and subsequent RLN metastasis in patients with early breast cancer.

Methods: We identified 660,258 women from 2004-2013 with early breast cancer stage (cT1-T3, cN0-N1, and M0). A two-part model was estimated using multivariable logistic regression to evaluate 1) predictors, by odds (OR), of having at least one RLN metastasis, and 2) predictors, by rate difference (RD), of having additional RLN metastasis relative to women with only one RLN metastasis. The same set of predictor variables were included in both parts of the model. All analyses were conducted using SAS v. 9.4 with p &lt;.05 indicating statistical significance.

Results: Adjusted ORs and RDs of RLN metastasis for selected variables from the model are shown in Table 1. Increased likelihood of at least one RLN metastasis was significantly associated with the presence of larger tumor size (p &lt;0.0001), a primary tumor in the nipple region (p &lt;0.0001) of the left breast (p &lt;0.0001), lobular or ductal type histology (p &lt;0.001, and p &lt;0.0001, respectively), positive estrogen (ER) and progesterone (PR) receptor statuses (p &lt;0.0001 for both), younger age (p &lt;0.0001), being white (p &lt;0.0001), and greater comorbidity (Charlson/Deyo - CD) (p &lt; 0.0001). Predictors of at least one RLN metastasis were also significantly associated with higher adjusted rates of further metastasis, except age, tumor size, ductal type histology, ER, and PR. However, a primary tumor in the central region (p =0.037), not the nipple region, was most associated with additional metastasis.

Table 1. Adjusted OR/RD (95% CI) for RLN Metastasis Events≥1 event &gt;1 event0.99 (0.99-0.99)Age (continuous)1.00 (-)1.10 (1.05-1.14)White vs. Other1.02 (1.00-1.04)0.98 (0.95-1.00)CD = 0 vs. 10.98 (0.97-0.99)0.89 (0.85-0.93)CD = 0 vs. ≥20.97 (0.95-0.99)1.02 (1.02-1.02)Tumor Size (continuous)1.00 (-) Primary Tumor Site - Nipple vs. (left), Central vs. (right) 1.80 (1.61-2.02)Overlap1.02 (1.01-1.04)1.87 (1.59-2.20)Tail0.98 (-)1.64 (1.46-1.84)LOQ1.00 (-)1.71 (1.53-1.92)UOQ1.02 (1.00-1.03)2.07 (1.84-2.33)LIQ1.07 (1.04-1.09)2.73 (2.43 -3.07)UIQ1.11 (1.09-1.13)1.44 (1.28 -1.63)Central (left), Nipple (right)1.06 (1.02-1.11) Histological Type - Lobular vs. 4.00 (3.72-4.30)Other1.52 (1.46-1.57)0.99 (-)Ductal1.18 (1.17-1.19)1.04 (1.04-1.04)ER (Positive vs. Negative)1.00 (-)1.02 (1.02-1.02)PR (Positive vs. Negative)1.00 (-)1.06 (1.05-1.08)Laterality (Left vs. Right)1.02 (1.01-1.02)≥1 event (vs. 0): OR; &gt;1 event (vs. 1): RD; (-) = insignificant

Conclusion: Utilizing a large dataset, several clinicopathological factors emerged from the NCDB as independent predictors of at least one, or additional RLN metastasis, supporting their weighted inclusion in prediction tools for ALN metastasis. Notably, different primary tumor sites of the breast predicted the two events modeled.

Citation Format: Satish M, Walters R, Silberstein PT. Clinicopathological predictors of axillary lymph node metastasis in early breast cancer patients [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-07-26.

iPTF16geu: A multiply imaged, gravitationally lensed type Ia supernova

We report the discovery of a multiply imaged, gravitationally lensed type Ia supernova, iPTF16geu (SN 2016geu), at redshift z = 0.409. This phenomenon was identified because the light from the stellar explosion was magnified more than 50 times by the curvature of space around matter in an intervening galaxy. We used high-spatial-resolution observations to resolve four images of the lensed supernova, approximately 0.3 arc seconds from the center of the foreground galaxy. The observations probe a physical scale of ~1 kiloparsec, smaller than is typical in other studies of extragalactic gravitational lensing. The large magnification and symmetric image configuration imply close alignment between the lines of sight to the supernova and to the lens. The relative magnifications of the four images provide evidence for substructures in the lensing galaxy.

Genome-wide association study of lifetime cannabis use based on a large meta-analytic sample of 32 330 subjects from the International Cannabis Consortium

Cannabis is the most widely produced and consumed illicit psychoactive substance worldwide. Occasional cannabis use can progress to frequent use, abuse and dependence with all known adverse physical, psychological and social consequences. Individual differences in cannabis initiation are heritable (40-48%). The International Cannabis Consortium was established with the aim to identify genetic risk variants of cannabis use. We conducted a meta-analysis of genome-wide association data of 13 cohorts (N=32 330) and four replication samples (N=5627). In addition, we performed a gene-based test of association, estimated single-nucleotide polymorphism (SNP)-based heritability and explored the genetic correlation between lifetime cannabis use and cigarette use using LD score regression. No individual SNPs reached genome-wide significance. Nonetheless, gene-based tests identified four genes significantly associated with lifetime cannabis use: NCAM1, CADM2, SCOC and KCNT2. Previous studies reported associations of NCAM1 with cigarette smoking and other substance use, and those of CADM2 with body mass index, processing speed and autism disorders, which are phenotypes previously reported to be associated with cannabis use. Furthermore, we showed that, combined across the genome, all common SNPs explained 13-20% (P<0.001) of the liability of lifetime cannabis use. Finally, there was a strong genetic correlation (rg=0.83; P=1.85 × 10(-8)) between lifetime cannabis use and lifetime cigarette smoking implying that the SNP effect sizes of the two traits are highly correlated. This is the largest meta-analysis of cannabis GWA studies to date, revealing important new insights into the genetic pathways of lifetime cannabis use. Future functional studies should explore the impact of the identified genes on the biological mechanisms of cannabis use.

Reliability assessment of a peer evaluation instrument in a team-based learning course

OBJECTIVE

To evaluate the reliability of a peer evaluation instrument in a longitudinal team-based learning setting.

METHODS

Student pharmacists were instructed to evaluate the contributions of their peers. Evaluations were analyzed for the variance of the scores by identifying low, medium, and high scores. Agreement between performance ratings within each group of students was assessed via intra-class correlation coefficient (ICC).

RESULTS

We found little variation in the standard deviation (SD) based on the score means among the high, medium, and low scores within each group. The lack of variation in SD of results between groups suggests that the peer evaluation instrument produces precise results. The ICC showed strong concordance among raters.

CONCLUSIONS

Findings suggest that our student peer evaluation instrument provides a reliable method for peer assessment in team-based learning settings.

Regional variations in the prevalence and misdiagnosis of air flow obstruction in China: baseline results from a prospective cohort of the China Kadoorie Biobank (CKB)

Background

Despite the great burden of chronic respiratory diseases in China, few large multicentre, spirometry-based studies have examined its prevalence, rate of underdiagnosis regionally or the relevance of socioeconomic and lifestyle factors.

Methods

We analysed data from 512 891 adults in the China Kadoorie Biobank, recruited from 10 diverse regions of China during 2004–2008. Air flow obstruction (AFO) was defined by the lower limit of normal criteria based on spirometry-measured lung function. The prevalence of AFO was analysed by region, age, socioeconomic status, body mass index (BMI) and smoking history and compared with the prevalence of self-reported physician-diagnosed chronic bronchitis or emphysema (CB/E) and its symptoms.

Findings

The prevalence of AFO was 7.3% in men (range 2.5–18.2%) and 6.4% in women (1.5–18.5%). Higher prevalence of AFO was associated with older age (p<0.0001), lower income (p<0.0001), poor education (p<0.001), living in rural regions (p<0.001), those who started smoking before the age of 20 years (p<0.001) and low BMI (p<0.001). Compared with self-reported diagnosis of CB/E, 88.8% of AFO was underdiagnosed; underdiagnosis proportion was highest in 30–39-year olds (96.7%) compared with the 70+ age group (81.1%), in women (90.7%), in urban areas (89.4%), in people earning 5K–10 K ¥ monthly (90.3%) and in those with middle or high school education (92.6%).

Interpretation

In China, the burden of AFO based on spirometry was high and significantly greater than that estimated based on self-reported physician-diagnosed CB/E, especially in rural areas, reflecting major issues with diagnosis of AFO that will impact disease treatment and management.

Associations of blood glucose and prevalent diabetes with risk of cardiovascular disease in 500 000 adult Chinese: the China Kadoorie Biobank

AIMS

To examine the relationship of self-reported diabetes, and of random blood glucose levels among individuals without known diabetes, with the prevalence of cardiovascular disease in Chinese adults.

METHODS

We examined cross-sectional data from the China Kadoorie Biobank of 0.5 million people aged 30-79 years recruited from 10 diverse regions of China in the period 2004-2008. Logistic regression was used to estimate the odds ratios of prevalent cardiovascular disease associated with self-reported diabetes, and with measured random blood glucose levels among participants with no history of diabetes, adjusting simultaneously for age, sex, area, education, smoking, alcohol, blood pressure and physical activity.

RESULTS

A total of 3.2% of participants had self-reported diabetes (men 2.9%; women 3.3%) and 2.8% had screen-detected diabetes (men 2.6%; women 2.8%), i.e. they had no self-reported history of diabetes but a blood glucose level suggestive of a diagnosis of diabetes. Compared with individuals without a history of diabetes, the odds ratios associated with self-reported diabetes were 2.18 (95% CI 2.06-2.30) and 1.88 (95% CI 1.75-2.01) for prevalent ischaemic heart disease and stroke/transient ischaemic attack, respectively. Among participants without self-reported diabetes there was a positive association between random blood glucose and ischaemic heart disease and stroke/transient ischaemic attack prevalence (P for trend <0.0001). Below the diabetic threshold (<11.1 mmol/l) each additional 1 mmol/l of random blood glucose was associated with 4% (95% CI 2-5%) and 5% (95% CI 3-7%) higher odds of prevalent ischaemic heart disease and stroke/transient ischaemic attack, respectively.

CONCLUSIONS

In this adult Chinese population, self-reported diabetes was associated with a doubling of the odds of prevalent cardiovascular disease. Below the threshold for diabetes there was still a modest, positive association between random blood glucose and prevalent cardiovascular disease.

Gradient Boosting as a SNP Filter: an Evaluation Using Simulated and Hair Morphology Data

Typically, genome-wide association studies consist of regressing the phenotype on each SNP separately using an additive genetic model. Although statistical models for recessive, dominant, SNP-SNP, or SNP-environment interactions exist, the testing burden makes an evaluation of all possible effects impractical for genome-wide data. We advocate a two-step approach where the first step consists of a filter that is sensitive to different types of SNP main and interactions effects. The aim is to substantially reduce the number of SNPs such that more specific modeling becomes feasible in a second step. We provide an evaluation of a statistical learning method called "gradient boosting machine" (GBM) that can be used as a filter. GBM does not require an a priori specification of a genetic model, and permits inclusion of large numbers of covariates. GBM can therefore be used to explore multiple GxE interactions, which would not be feasible within the parametric framework used in GWAS. We show in a simulation that GBM performs well even under conditions favorable to the standard additive regression model commonly used in GWAS, and is sensitive to the detection of interaction effects even if one of the interacting variables has a zero main effect. The latter would not be detected in GWAS. Our evaluation is accompanied by an analysis of empirical data concerning hair morphology. We estimate the phenotypic variance explained by increasing numbers of highest ranked SNPs, and show that it is sufficient to select 10K-20K SNPs in the first step of a two-step approach.

An overview of internet biosurveillance

Internet biosurveillance utilizes unstructured data from diverse web-based sources to provide early warning and situational awareness of public health threats. The scope of source coverage ranges from local media in the vernacular to international media in widely read languages. Internet biosurveillance is a timely modality that is available to government and public health officials, healthcare workers, and the public and private sector, serving as a real-time complementary approach to traditional indicator-based public health disease surveillance methods. Internet biosurveillance also supports the broader activity of epidemic intelligence. This overview covers the current state of the field of Internet biosurveillance, and provides a perspective on the future of the field.

A retrospective review of factors associated with vasovasostomies in United States military members

BACKGROUND

Men seeking a vasectomy should receive counseling prior to the procedure that includes discussion of later seeking a reversal. We sought to determine demographic factors that may predispose patients to possibly later seek a vasectomy reversal.

METHODS

All U.S. Military electronic health records were searched between 2000 and 2009 for either a vasectomy or vasovasostomy procedure code. Aggregate demographic information was collected and statistical analysis performed.

RESULT

A total of 82,945 patients had a vasectomy of which 4,485 had a vasovasostomy resulting in a vasovasostomy-to-vasectomy rate of 5.04%. The average age at vasovasostomy was 34.9±5.0, with an average interval of 4.1±2.2 years. Men undergoing a vasectomy at a younger age were more likely to have a vasovasostomy. Various religions did have statistically significant differences. Within ethnic groups, only Native Americans [OR=1.39 (95% CI 1.198-1.614)] and Asians [OR=0.501 (95% CI 0.364-0.690)] had statistically significant differences when compared to Caucasians. Men with more children at the time of vasectomy were more likely to have a vasovasostomy.

CONCLUSION

Younger men, Native Americans, and men with more children at vasectomy were more likely to undergo a vasovasostomy. The reason for these differences is unknown, but this information may assist during pre-vasectomy counseling.