Evaluating approaches for constructing polygenic risk scores for prostate cancer in men of African and European ancestry.

Genome-wide polygenic risk scores (GW-PRSs) have been reported to have better predictive ability than PRSs based on genome-wide significance thresholds across numerous traits. We compared the predictive ability of several GW-PRS approaches to a recently developed PRS of 269 established prostate cancer-risk variants from multi-ancestry GWASs and fine-mapping studies (PRS269). GW-PRS models were trained with a large and diverse prostate cancer GWAS of 107,247 cases and 127,006 controls that we previously used to develop the multi-ancestry PRS269. Resulting models were independently tested in 1,586 cases and 1,047 controls of African ancestry from the California Uganda Study and 8,046 cases and 191,825 controls of European ancestry from the UK Biobank and further validated in 13,643 cases and 210,214 controls of European ancestry and 6,353 cases and 53,362 controls of African ancestry from the Million Veteran Program. In the testing data, the best performing GW-PRS approach had AUCs of 0.656 (95% CI = 0.635-0.677) in African and 0.844 (95% CI = 0.840-0.848) in European ancestry men and corresponding prostate cancer ORs of 1.83 (95% CI = 1.67-2.00) and 2.19 (95% CI = 2.14-2.25), respectively, for each SD unit increase in the GW-PRS. Compared to the GW-PRS, in African and European ancestry men, the PRS269 had larger or similar AUCs (AUC = 0.679, 95% CI = 0.659-0.700 and AUC = 0.845, 95% CI = 0.841-0.849, respectively) and comparable prostate cancer ORs (OR = 2.05, 95% CI = 1.87-2.26 and OR = 2.21, 95% CI = 2.16-2.26, respectively). Findings were similar in the validation studies. This investigation suggests that current GW-PRS approaches may not improve the ability to predict prostate cancer risk compared to the PRS269 developed from multi-ancestry GWASs and fine-mapping.

Occupational exposure to radiofrequency electromagnetic fields and brain tumor risk: Application of the INTEROCC job-exposure matrix.

Radiofrequency electromagnetic fields (RF-EMF, 100 kHz to 300 GHz) are classified by IARC as possibly carcinogenic to humans (Group 2B). This study evaluates the potential association between occupational RF-EMF exposure and brain tumor risk, utilizing for the first time, a RF-EMF job-exposure matrix (RF-JEM) developed in the multi-country INTEROCC case-control study. Cumulative and time-weighted average (TWA) occupational RF-EMF exposures were estimated for study participants based on lifetime job histories linked to the RF-JEM using three different methods: (1) by considering RF-EMF intensity among all exposed jobs, (2) by considering RF-EMF intensity among jobs with an exposure prevalence ≥ the median exposure prevalence of all exposed jobs, and (3) by considering RF-EMF intensity of jobs of participants who reported RF-EMF source use. Stratified conditional logistic regression models were used, considering various lag periods and exposure time windows defined a priori. Generally, no clear associations were found for glioma or meningioma risk. However, some statistically significant positive associations were observed including in the highest exposure categories for glioma for cumulative and TWA exposure in the 1- to 4-year time window for electric fields (E) in the first JEM application method (odds ratios [ORs] = 1.36, 95% confidence interval [95% CI] 1.08, 1.72 and 1.27, 95% CI 1.01, 1.59, respectively), as well as for meningioma for cumulative exposure in the 5- to 9-year time window for electric fields (E) in the third JEM application method (OR = 2.30, 95% CI 1.11, 4.78). We did not identify convincing associations between occupational RF-EMF exposure and risk of glioma or meningioma.

Prevalent occupational exposures and risk of lung cancer among women: Results from the application of the Canadian Job-Exposure Matrix (CANJEM) to a combined set of ten case-control studies.

BACKGROUND: Worldwide, lung cancer is the second leading cause of cancer death in women. The present study explored associations between occupational exposures that are prevalent among women, and lung cancer. METHODS: Data from 10 case-control studies of lung cancer from Europe, Canada, and New Zealand conducted between 1988 and 2008 were combined. Lifetime occupational history and information on nonoccupational factors including smoking were available for 3040 incident lung cancer cases and 4187 controls. We linked each reported job to the Canadian Job-Exposure Matrix (CANJEM), which provided estimates of probability, intensity, and frequency of exposure to each selected agent in each job. For this analysis, we selected 15 agents (cleaning agents, biocides, cotton dust, synthetic fibers, formaldehyde, cooking fumes, organic solvents, cellulose, polycyclic aromatic hydrocarbons from petroleum, ammonia, metallic dust, alkanes C18+, iron compounds, isopropanol, and calcium carbonate) that had lifetime exposure prevalence of at least 5% in the combined study population. For each agent, we estimated lung cancer risk in each study center for ever-exposure, by duration of exposure, and by cumulative exposure, using separate logistic regression models adjusted for smoking and other covariates. We then estimated the meta-odds ratios using random-effects meta-analysis. RESULTS AND CONCLUSIONS: None of the agents assessed showed consistent and compelling associations with lung cancer among women. The following agents showed elevated odds ratio in some analyses: metallic dust, iron compounds, isopropanol, and organic solvents. Future research into occupational lung cancer risk factors among women should prioritize these agents.

Occupational exposure to nickel and hexavalent chromium and the risk of lung cancer in a pooled analysis of case-control studies (SYNERGY).

There is limited evidence regarding the exposure-effect relationship between lung-cancer risk and hexavalent chromium (Cr(VI)) or nickel. We estimated lung-cancer risks in relation to quantitative indices of occupational exposure to Cr(VI) and nickel and their interaction with smoking habits. We pooled 14 case-control studies from Europe and Canada, including 16 901 lung-cancer cases and 20 965 control subjects. A measurement-based job-exposure-matrix estimated job-year-region specific exposure levels to Cr(VI) and nickel, which were linked to the subjects' occupational histories. Odds ratios (OR) and associated 95% confidence intervals (CI) were calculated by unconditional logistic regression, adjusting for study, age group, smoking habits and exposure to other occupational lung carcinogens. Due to their high correlation, we refrained from mutually adjusting for Cr(VI) and nickel independently. In men, ORs for the highest quartile of cumulative exposure to CR(VI) were 1.32 (95% CI 1.19-1.47) and 1.29 (95% CI 1.15-1.45) in relation to nickel. Analogous results among women were: 1.04 (95% CI 0.48-2.24) and 1.29 (95% CI 0.60-2.86), respectively. In men, excess lung-cancer risks due to occupational Cr(VI) and nickel exposure were also observed in each stratum of never, former and current smokers. Joint effects of Cr(VI) and nickel with smoking were in general greater than additive, but not different from multiplicative. In summary, relatively low cumulative levels of occupational exposure to Cr(VI) and nickel were associated with increased ORs for lung cancer, particularly in men. However, we cannot rule out a combined classical measurement and Berkson-type of error structure, which may cause differential bias of risk estimates.

Neighbourhood social deprivation and risk of prostate cancer.

BACKGROUND: Striking geographic variations in prostate cancer incidence suggest an aetiological role for spatially-distributed factors. We assessed whether neighbourhood social deprivation, which can reflect limited social contacts, unfavourable lifestyle and environmental exposures, is associated with prostate cancer risk. METHODS: In 2005-2012, we recruited 1931 incident prostate cancer cases and 1994 controls in a case-control study in Montreal, Canada. Lifetime residential addresses were linked to an area-based social deprivation index around recruitment (2006) and about 10 years earlier (1996). Logistic regression estimated adjusted odds ratios (ORs) and 95% confidence intervals (CIs). RESULTS: Men residing in areas characterised by greater social deprivation had elevated prostate cancer risks (ORs of 1.54 and 1.60 for recent and past exposures, respectively; highest vs lowest quintiles), independently from area- and individual-level confounders and screening patterns. The increase in risk with recent high social deprivation was particularly elevated for high-grade prostate cancer at diagnosis (OR 1.87, 95% CI 1.32-2.64). Associations were more pronounced for neighbourhoods with higher proportions of separated/divorced or widowed individuals in the past, and with higher percentages of residents living alone recently. CONCLUSIONS: These novel findings, suggesting that neighbourhood-level social deprivation increases the risk of prostate cancer, point out to potential targeted public health interventions.

Family structure and living arrangements as indicators of social isolation, and prostate cancer risk.

Social isolation has been linked to a poorer prostate cancer prognosis. Little is known about how it could also influence its incidence. We investigated the association between family structure and living arrangements as potential indicators of social isolation, and prostate cancer risk, globally and according to disease aggressiveness. Data from the Prostate Cancer & Environment Study (PROtEuS), a case-control population-based study conducted between 2005 and 2012 in Montreal, Canada, were used. The study population comprised 1931 incident cases of prostate cancer, aged ≤75 years, and 1994 age-matched (±5 years) population controls. In-person interviews collected information on family composition and living arrangements recently and at age 40. Logistic regression estimated odds ratios (ORs) and 95% confidence intervals (CIs), adjusting for potential confounders. Single men had an increased risk of high-grade prostate cancer at diagnosis (OR 1.80; 95% CI 1.29-2.51), compared to men currently married or with a partner. Having at least one daughter was associated with a lower risk of aggressive cancer (OR 0.76; 95% CI 0.61-0.96) while no association was found with having son(s). An inverse dose-response relationship was observed between the number of people living with the subject 2 years before diagnosis/interview and prostate cancer risk (p-trend<0.001). These results suggest a protective role of a rich personal environment on the risk of developing prostate cancer. As several of the associations studied here are novel, replication is required.

Ascertaining asthma status in epidemiologic studies: a comparison between administrative health data and self-report.

BACKGROUND: Studies have suggested that agreement between administrative health data and self-report for asthma status ranges from fair to good, but few studies benefited from administrative health data over a long period. We aimed to (1) evaluate agreement between asthma status ascertained in administrative health data covering a period of 30 years and from self-report, and (2) identify determinants of agreement between the two sources. METHODS: We used administrative health data (1983-2012) from the Quebec Birth Cohort on Immunity and Health, which included 81,496 individuals born in the province of Quebec, Canada, in 1974. Additional information, including self-reported asthma, was collected by telephone interview with 1643 participants in 2012. By design, half of them had childhood asthma based on health services utilization. Results were weighted according to the inverse of the sampling probabilities. Five algorithms were applied to administrative health data (having ≥ 2 physician claims over a 1-, 2-, 3-, 5-, or 30-year interval or ≥ 1 hospitalization), to enable comparisons with previous studies. We estimated the proportion of overall agreement and Kappa, between asthma status derived from algorithms and self-reports. We used logistic regression to identify factors associated with agreement. RESULTS: Applying the five algorithms, the prevalence of asthma ranged from 49 to 55% among the 1643 participants. At interview (mean age = 37 years), 49% and 47% of participants respectively reported ever having asthma and asthma diagnosed by a physician. Proportions of agreement between administrative health data and self-report ranged from 88 to 91%, with Kappas ranging from 0.57 (95% CI: 0.52-0.63) to 0.67 (95% CI: 0.62-0.72); the highest values were obtained with the [≥ 2 physician claims over a 30-year interval or ≥ 1 hospitalization] algorithm. Having sought health services for allergic diseases other than asthma was related to lower agreement (Odds ratio = 0.41; 95% CI: 0.25-0.65 comparing ≥ 1 health services to none). CONCLUSIONS: These findings indicate good agreement between asthma status defined from administrative health data and self-report. Agreement was higher than previously observed, which may be due to the 30-year lookback window in administrative data. Our findings support using both administrative health data and self-report in population-based epidemiological studies.

Occupational heat exposure and prostate cancer risk: A pooled analysis of case-control studies.

BACKGROUND: Heat exposures occur in many occupations. Heat has been linked to key carcinogenic processes, however, evidence for associations with cancer risk is sparse. We examined potential associations between occupational heat exposure and prostate cancer risk in a multi-country study. METHODS: We analysed a large, pooled dataset of 3142 histologically confirmed prostate cancer cases and 3512 frequency-matched controls from three countries: Canada, France, and Spain. Three exposure indices: ever exposure, lifetime cumulative exposure and duration of exposure, were developed using the Finnish Job-Exposure Matrix, FINJEM, applied to the lifetime occupational history of participants. We estimated odds ratios (ORs) and 95% confidence intervals (CIs), using conditional logistic regression models stratified by 5-year age groups and study, adjusting for potential confounders. Potential interactions with exposure to other occupational agents were also explored. RESULTS: Overall, we found no association for ever occupational heat exposure (OR 0.97; 95% CI 0.87, 1.09), nor in the highest categories of lifetime cumulative exposure (OR 1.04; 95% CI 0.89, 1.23) or duration (OR 1.03; 95% CI 0.88, 1.22). When using only the Spanish case-control study and a Spanish Job Exposure Matrix (JEM), some weakly elevated ORs were observed. CONCLUSIONS: Findings from this study provide no clear evidence for an association between occupational heat exposure and prostate cancer risk.

Comparison of mixed model based approaches for correcting for population substructure with application to extreme phenotype sampling.

BACKGROUND: Mixed models are used to correct for confounding due to population stratification and hidden relatedness in genome-wide association studies. This class of models includes linear mixed models and generalized linear mixed models. Existing mixed model approaches to correct for population substructure have been previously investigated with both continuous and case-control response variables. However, they have not been investigated in the context of extreme phenotype sampling (EPS), where genetic covariates are only collected on samples having extreme response variable values. In this work, we compare the performance of existing binary trait mixed model approaches (GMMAT, LEAP and CARAT) on EPS data. Since linear mixed models are commonly used even with binary traits, we also evaluate the performance of a popular linear mixed model implementation (GEMMA). RESULTS: We used simulation studies to estimate the type I error rate and power of all approaches assuming a population with substructure. Our simulation results show that for a common candidate variant, both LEAP and GMMAT control the type I error rate while CARAT's rate remains inflated. We applied all methods to a real dataset from a Québec, Canada, case-control study that is known to have population substructure. We observe similar type I error control with the analysis on the Québec dataset. For rare variants, the false positive rate remains inflated even after correction with mixed model approaches. For methods that control the type I error rate, the estimated power is comparable. CONCLUSIONS: The methods compared in this study differ in their type I error control. Therefore, when data are from an EPS study, care should be taken to ensure that the models underlying the methodology are suitable to the sampling strategy and to the minor allele frequency of the candidate SNPs.

Bacillus Calmette-Guerin vaccination and multiple sclerosis: A population-based birth cohort study in Quebec, Canada.

BACKGROUND AND PURPOSE: The bacillus Calmette-Guerin (BCG) vaccine could reduce the incidence of multiple sclerosis (MS) through immunomodulation. Previous studies, presenting some limitations, reported no association. We re-examined this association in a large cohort focusing on relapsing-remitting MS (RRMS). METHODS: The cohort included 400,563 individuals, and was linked with the Quebec provincial BCG vaccination registry and administrative health data. Individuals were followed up from 1983 to 2014 and then within Period 1 (1983-1996) and Period 2 (1997-2014), for the occurrence of MS. Incident MS cases were defined as those with ≥3 hospital or physician claims for MS. Subjects with ≥1 drug reimbursement for MS disease-modifying therapies were classified as RRMS. Cox proportional hazards regression was used to estimate hazard ratios (HRs) over the follow-ups, adjusting for potential confounders. Possible effect modification due to sex was assessed. RESULTS: A total of 178,335 (46%) individuals were BCG vaccinated. There were 274 (0.06%) incident MS cases identified in 1983-1996, and 1433 (0.4%) in 1997-2014. No association was found with RRMS, either in Period 1 (adjusted HR [HRadj ] = 0.96, 95% confidence interval [CI] = 0.63-1.45; 96 cases) or in Period 2 (HRadj  = 1.02, 95% CI = 0.85-1.23; 480 cases). The remaining MS cases, for whom the phenotype was unknown, were positively associated with BCG over the entire follow-up (HRadj  = 1.25, 95% CI = 1.10-1.41; 1131 cases) and in Period 2 (HRadj  = 1.33, 95% CI = 1.17-1.52; 953 cases). No interaction with sex was found. CONCLUSIONS: Findings suggest that BCG vaccination does not decrease the risk of RRMS, and that future studies should consider phenotypes of MS.