Faculty Bibliography

Quality of exposure assessment has been shown to be related to the ability to detect risk of lymphohematopoietic disorders in epidemiological investigations of benzene, especially at low levels of exposure. We set out to build a statistical model for reconstructing exposure levels for 2898 subjects from 501 factories that were part of a nested case-cohort study within the NCI-CAPM cohort of more than 110,000 workers. We used a hierarchical model to allow for clustering of measurements by factory, workshop, job, and date. To calibrate the model we used historical routine monitoring data. Measurements below the limit of detection were accommodated by constructing a censored data likelihood. Potential non-linear and industry-specific time-trends and predictor effects were incorporated using regression splines and random effects. A partial validation of predicted exposures in 2004/2005 was performed through comparison with full-shift measurements from an exposure survey in facilities that were still open. Median cumulative exposure to benzene at age 50 for subjects that ever held an exposed job (n=1175) was 509 mg/m3 years. Direct comparison of model estimates with measured full-shift personal exposure in the 2004/2005 survey showed moderate correlation and a potential downward bias at low (<1 mg/m3) exposure estimates. The modeling framework enabled us to deal with the data complexities generally found in studies using historical exposure data in a comprehensive way and we therefore expect to be able to investigate effects at relatively low exposure levels.Journal of Exposure Science and Environmental Epidemiology advance online publication, 12 August 2015; doi:10.1038/jes.2015.44.

Most mouthwashes contain alcohol, a known cause of head and neck cancer (oral cavity, pharynx, larynx), likely through the carcinogenic activity of acetaldehyde, formed in the oral cavity from alcohol. We carried out a pooled analysis of 8981 cases of head and neck cancer and 10 090 controls from 12 case-control studies with comparable information on mouthwash use in the International Head and Neck Cancer Epidemiology Consortium. Logistic regression was used to assess the association of mouthwash use with cancers of the oral cavity, oropharynx, hypopharynx, and larynx, adjusting for study, age, sex, pack-years of tobacco smoking, number of alcoholic drinks/day, and education. Compared with never users of mouthwash, the odds ratio (OR) of all head and neck cancers was 1.01 [95% confidence interval (CI): 0.94-1.08] for ever users, based on 12 studies. The corresponding ORs of cancer of the oral cavity and oropharynx were 1.11 (95% CI: 1.00-1.23) and 1.28 (95% CI: 1.06-1.56), respectively. OR for all head and neck cancer was 1.15 (95% CI: 1.01-1.30) for use for more than 35 years, based on seven studies (P for linear trend=0.01), and OR 1.31 (95% CI: 1.09-1.58) for use more than one per day, based on five studies (P for linear trend <0.001). Although limited by the retrospective nature of the study and the limited ability to assess risks of mouthwash use in nonusers of tobacco and alcohol, this large investigation shows potential risks for head and neck cancer subsites and in long-term and frequent users of mouthwash. This pooled analysis provides the most precise estimate of the association between mouthwash use and head and neck cancer.

BACKGROUND: Cigarette smoking is a major risk factor for head and neck cancer (HNC). To our knowledge, low cigarette smoking (<10 cigarettes per day) has not been extensively investigated in fine categories or among never alcohol drinkers. METHODS: We conducted a pooled analysis of individual participant data from 23 independent case-control studies including 19 660 HNC cases and 25 566 controls. After exclusion of subjects using other tobacco products including cigars, pipes, snuffed or chewed tobacco and straw cigarettes (tobacco product used in Brazil), as well as subjects smoking more than 10 cigarettes per day, 4093 HNC cases and 13 416 controls were included in the analysis. The lifetime average frequency of cigarette consumption was categorized as follows: never cigarette users, >0-3, >3-5, >5-10 cigarettes per day. RESULTS: Smoking >0-3 cigarettes per day was associated with a 50% increased risk of HNC in the study population [odds ratio (OR) = 1.52, 95% confidence interval (CI): (1.21, 1.90). Smoking >3-5 cigarettes per day was associated in each subgroup from OR = 2.01 (95% CI: 1.22, 3.31) among never alcohol drinkers to OR = 2.74 (95% CI: 2.01, 3.74) among women and in each cancer site, particularly laryngeal cancer (OR = 3.48, 95% CI: 2.40, 5.05). However, the observed increased risk of HNC for low smoking frequency was not found among smokers with smoking duration shorter than 20 years. CONCLUSION: Our results suggest a public health message that low frequency of cigarette consumption contributes to the development of HNC. However, smoking duration seems to play at least an equal or a stronger role in the development of HNC.

BACKGROUND:Many SNPs influence prostate cancer risk. To what extent genetic risk can be reduced by environmental factors is unknown. METHODS:We evaluated effect modification by environmental factors of the association between susceptibility SNPs and prostate cancer in 1,230 incident prostate cancer cases and 1,361 controls, all white and similar ages, nested in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Trial. Genetic risk scores were calculated as number of risk alleles for 20 validated SNPs. We estimated the association between higher genetic risk (≥12 SNPs) and prostate cancer within environmental factor strata and tested for interaction. RESULTS:Men with ≥12 risk alleles had 1.98, 2.04, and 1.91 times the odds of total, advanced, and nonadvanced prostate cancer, respectively. These associations were attenuated with the use of selenium supplements, aspirin, ibuprofen, and higher vegetable intake. For selenium, the attenuation was most striking for advanced prostate cancer: compared with <12 alleles and no selenium, the OR for ≥12 alleles was 2.06 [95% confidence interval (CI), 1.67-2.55] in nonusers and 0.99 (0.38-2.58) in users (Pinteraction = 0.031). Aspirin had the most marked attenuation for nonadvanced prostate cancer: compared with <12 alleles and nonusers, the OR for ≥12 alleles was 2.25 (1.69-3.00) in nonusers and 1.70 (1.25-2.32) in users (Pinteraction = 0.009). This pattern was similar for ibuprofen (Pinteraction = 0.023) and vegetables (Pinteraction = 0.010). CONCLUSIONS:This study suggests that selenium supplements may reduce genetic risk of advanced prostate cancer, whereas aspirin, ibuprofen, and vegetables may reduce genetic risk of nonadvanced prostate cancer. IMPACT/CONCLUSIONS:The effect of genetic factors on prostate cancer risk may vary by lifestyle interventions.

BACKGROUND: Organochlorine (OC) insecticides and polychlorinated biphenyls (PCBs) have been shown to have estrogenic, antiestrogenic, or antiandrogenic properties; as a result, the impact of exposure to these compounds and risk of hormonal cancers, such as prostate cancer, is a concern. OBJECTIVES: We conducted a nested case-control study, using prospectively collected serum, to estimate associations between OC exposures and metastatic prostate cancer in a population-based cohort from Norway. METHODS: Sera from 150 cases and 314 controls matched on date of blood draw, age at blood draw, and region was used to determine concentrations of 11 OC pesticide metabolites and 34 PCB congeners. Odds ratios (ORs) and 95% confidence intervals (95% CIs) for quartiles of lipid-corrected metabolite levels were calculated using conditional logistic regression. RESULTS: Metastatic prostate cancer was two times as likely among men with serum concentrations of oxychlordane in the highest quartile compared with those in the lowest quartile (OR = 2.03; 95% CI: 1.03, 4.03; p-trend 0.05). Elevated but nonsignificant ORs were estimated for the highest versus lowest quartile of heptachlor epoxide, HCB, and mirex, although these exposures were correlated with oxychlordane. Findings for specific PCB congeners showed a significant inverse association between natural log-transformed lipid-adjusted PCB 44 and metastatic prostate cancer (OR = 0.74; 95% CI: 0.56, 0.97; p-trend = 0.02). CONCLUSIONS: Our study highlights the importance of estimating associations with specific OC chemicals and suggests a possible role of OC insecticides and PCBs in the etiology of metastatic prostate cancer. CITATION: Koutros S, Langseth H, Grimsrud TK, Barr DB, Vermeulen R, Portengen L, Wacholder S, Beane Freeman LE, Blair A, Hayes RB, Rothman N, Engel LS. 2015. Prediagnostic serum organochlorine concentrations and metastatic prostate cancer: a nested case-control study in the Norwegian Janus Serum Bank cohort. Environ Health Perspect 123:867-872; http://dx.doi.org/10.1289/ehp.1408245.

SCOPE: Only a few studies analyzed the role of allium vegetables with reference to head and neck cancers (HNC), with mixed results. We investigated the potential favorable role of garlic and onion within the International Head and Neck Cancer Epidemiology (INHANCE) Consortium. METHODS AND RESULTS: We analyzed pooled individual-level data from eight case-control studies, including 4590 cases and 7082 controls. We estimated odds ratios (ORs) and 95% confidence intervals (CIs) for associations between garlic and onion intakes and HNC risk. Compared with no or low garlic use, the ORs of HNC were 0.95 (95% CI 0.71-1.27) for intermediate and 0.74 (95% CI 0.55-0.99) for high garlic use (p for trend = 0.02). The ORs of HNC for increasing categories of onion intake were 0.91 (95% CI 0.68-1.21) for >1 to 3 portions per week (p for trend = 0.02), as compared to <1 portion per week. We found an inverse association between high onion intake and laryngeal cancer risk (OR = 0.69; 95% CI 0.54-0.88), but no significant association for other subsites. CONCLUSION: The results of this pooled-analysis support a possible moderate inverse association between garlic and onion intake and HNC risk.

Although genome-wide association studies (GWAS) have separately identified many genetic susceptibility loci for ulcerative colitis (UC), Crohn's disease (CD) and colorectal cancer (CRC), there has been no large-scale examination for pleiotropy, or shared genetic susceptibility, for these conditions. We used logistic regression modeling to examine the associations of 181 UC and CD susceptibility variants previously identified by GWAS with risk of CRC using data from the Genetics and Epidemiology of Colorectal Cancer Consortium and the Colon Cancer Family Registry. We also examined associations of significant variants with clinical and molecular characteristics in a subset of the studies. Among 11794 CRC cases and 14190 controls, rs11676348, the susceptibility single nucleotide polymorphism (SNP) for UC, was significantly associated with reduced risk of CRC (P = 7E-05). The multivariate-adjusted odds ratio of CRC with each copy of the T allele was 0.93 (95% CI 0.89-0.96). The association of the SNP with risk of CRC differed according to mucinous histological features (P heterogeneity = 0.008). In addition, the (T) allele was associated with lower risk of tumors with Crohn's-like reaction but not tumors without such immune infiltrate (P heterogeneity = 0.02) and microsatellite instability-high (MSI-high) but not microsatellite stable or MSI-low tumors (P heterogeneity = 0.03). The minor allele (T) in SNP rs11676348, located downstream from CXCR2 that has been implicated in CRC progression, is associated with a lower risk of CRC, particularly tumors with a mucinous component, Crohn's-like reaction and MSI-high. Our findings offer the promise of risk stratification of inflammatory bowel disease patients for complications such as CRC.

Benzene exposure has been causally linked with acute myeloid leukemia (AML), but inconsistently associated with other hematopoietic, lymphoproliferative and related disorders (HLD) or solid tumors in humans. Many neoplasms have been described in experimental animals exposed to benzene. We used Poisson regression to estimate adjusted relative risks (RR) and the likelihood ratio statistic to derive confidence intervals for cause-specific mortality and HLD incidence in 73,789 benzene-exposed compared with 34,504 unexposed workers in a retrospective cohort study in 12 cities in China. Follow-up and outcome assessment was based on factory, medical and other records. Benzene-exposed workers experienced increased risks for all-cause mortality (RR = 1.1, 95% CI = 1.1, 1.2) due to excesses of all neoplasms (RR = 1.3, 95% CI = 1.2, 1.4), respiratory diseases (RR = 1.7, 95% CI = 1.2, 2.3) and diseases of blood forming organs (RR = infinity, 95% CI = 3.4, infinity). Lung cancer mortality was significantly elevated (RR = 1.5, 95% CI = 1.2, 1.9) with similar RRs for males and females, based on three-fold more cases than in our previous follow-up. Significantly elevated incidence of all myeloid disorders reflected excesses of myelodysplastic syndrome/acute myeloid leukemia (RR = 2.7, 95% CI = 1.2, 6.6) and chronic myeloid leukemia (RR = 2.5, 95% CI = 0.8, 11), and increases of all lymphoid disorders included excesses of non-Hodgkin lymphoma (RR = 3.9, 95%CI = 1.5, 13) and all lymphoid leukemia (RR = 5.4, 95%CI = 1.0, 99). The 28-year follow-up of Chinese benzene-exposed workers demonstrated increased risks of a broad range of myeloid and lymphoid neoplasms, lung cancer, and respiratory diseases and suggested possible associations with other malignant and non-malignant disorders.

BACKGROUND: Concordance with the Dietary Approaches to Stop Hypertension (DASH) diet has been shown to reduce blood pressure (BP) in short-term intervention studies, but long-term effects are unclear. We evaluated the association of DASH diet concordance with BP trajectories and incidence of hypertension, in 2187 men and women (mean age 52.5 years at baseline) participating in the Framingham Offspring cohort. METHOD: Diet and BP were assessed from 1991 to 2008, with a median follow-up time of 13.4 years. DASH scores (ranging from 0 for worst to 10 for best concordance with DASH diet) were calculated by summing 10 food components that comprise the DASH diet pattern, including fruits and vegetables, low-fat dairy products, lean meat, and plant-based protein. Mixed-effect and Cox regression models were applied, to assess the association of DASH diet concordance with BP longitudinal change and with incidence of hypertension, respectively. All analyses were adjusted for age, sex, smoking status, history of diabetes, BMI, and physical activity. RESULT: Overall, SBP increased by 0.34 mmHg and DBP by 0.10 mmHg annually, in the Framingham Offspring cohort. Every unit increase in the DASH score resulted in a modest increase in SBP of 0.054 mmHg/year (P = 0.028). No associations were observed between DASH diet concordance and DBP or incidence of hypertension. CONCLUSION: Long-term concordance with the DASH diet was not associated with a decreasing BP trajectory over time, or with decreased incidence of hypertension, in this population of middle-aged adults.