Search

Background: Chronic inflammation plays a significant role in hepatocellular carcinoma (HCC). With rising incidence and poor prognosis of HCC, identifying relatively less invasive biomarkers, especially inflammatory ones, may improve the assessment and stratification of HCC risk. The objective of our studies was to assess if CD8+ T cell cytokines in pre-diagnostic serum are associated with risk of HCC development. Methods: We conducted two parallel case-control studies of HCC nested within the Shanghai Cohort Study (SCS) and the Singapore Chinese Health Study (SCHS), two prospective cohorts of 81,000 individuals with 25+ years of follow-up. The serum concentrations of five CD8+ T cell cytokines ─ soluble CD137 (sCD137), soluble Fas (sFas), perforin, macrophage inflammatory protein 1-beta (MIP-1β), and tumor necrosis factor alpha (TNF-α) ─ were determined using Luminex bead-based immunoassay on 315 HCC cases and 315 individually matched controls in the SCS, and on 197 HCC case-control pairs in the SCHS. Conditional logistic regression was used to estimate odds ratio (OR) and 95% confidence interval (CI) for HCC with elevated levels of cytokines with adjustment for body mass index, alcohol drinking, cigarette smoking, seropositivity of hepatitis surface antigen (HBsAg), and history of diabetes. Results: sCD137 levels were statistically significantly higher in HCC cases than controls in both cohorts. Compared with the 1st quartile (Q1) of sCD137, multivariable-adjusted ORs (95% CI) of HCC for the 4th quartile (Q4) were 3.8 (1.7-8.3) in the SCS and 3.5 (1.4 -8.5) in the SCHS (both Ptrend’s < 0.001). Among HBsAg-negative individuals, ORs (95% CIs) for Q2, Q3, & Q4 of sCD137 were 2.7 (1.5-4.9), 2.7 (1.4-5.0), and 4.5 (2.4-8.5), respectively, compared with Q1 (Ptrend < 0.001) in both SCS and SCHS combined. The corresponding ORs (95% CIs) among HBsAg- positive individuals were 21.0 (9.3-47.6), 34.1 (13.8-84.2), and 56.7 (23.3-137.9), respectively, with a P=0.095 for multiplicative interaction. The sCD137-HCC risk association remained constant over different time periods from blood draw to HCC diagnosis: ORs (95% CIs) for Q4 vs. Q1 were 4.7 (1.2-17.5) for Conclusion: These novel and validated findings demonstrated that serum sCD137 levels were significantly elevated many years prior to HCC diagnosis, and had a potential synergistic effect with chronic viral infection on the HCC risk. sCD137 may be developed as a immune monitoring biomarker for risk stratification and assessment, which can lead to early diagnosis and improve prognosis of HCC patients. Citation Format: Claire E. Thomas, Jennifer J. Adibi, Allison L. Kuipers, Brenda Diergaarde, Hung N. Luu, Aizhen Jin, Woon-Puay Koh, Yu-Tang Gao, Jennifer Adams-Haduch, Renwei Wang, Anna Lokshin, Jaideep Behari, Jian-Min Yuan. Soluble CD137 and risk of hepatocellular carcinoma: nested case-control studies in cohorts in Shanghai and Singapore [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3015.

Background: Exogenous sex hormone supplements had been associated with increased lung cancer mortality in a large clinical trial. No study to date has evaluated temporal trajectories of circulating sex hormones and aromatase activity and their associations with lung cancer survival. Objective: To characterize temporal changes in prediagnostic levels of circulating sex hormones and aromatase activity and to evaluate their associations with overall survival in lung cancer patients. Methods: Included in the analysis were 385 incident lung cancer patients identified in a large prospective cohort among postmenopausal women who had never used cigarette products nor exogenous sex hormone supplements. Concentrations of sex hormones were quantitated using LC-MS/MS assays in prediagnostic plasma samples. The product-substrate molar ratio of estrone to androstenedione was used as an index of aromatase activity (IAA). A multivariable Cox model with restricted cubic spline functions was used to calculate hazard ratio (HR) for overall survival. Results: Of 385 patients with lung cancer, 308 died during a median follow-up of 18.1 years. Initial analyses in all patients showed that higher levels of circulating estrone and IAA and lower levels of androstenedione and testosterone were associated with poorer overall survival after adjusting for nonclinical covariates. In analyses restricted to those with clinical data (n=281), stronger associations were found after further adjustment for tumor stage and treatment regimens. Compared with patients at the median level of IAA, adjusted HRs (95% CI) for total mortality in those at the 30th, 70th, 90th and 95th percentiles were 0.98 (0.84-1.13), 1.05 (0.94-1.16), 1.28 (1.06-1.54), and 1.64 (1.27-2.12), respectively, with P-overall < 0.001. A similar positive association was observed for estrone. In contrast, inverse associations were seen for androgens. For example, compared with the median level of testosterone, HRs (95% CI) for those at the 5th, 10th, 30th, 70th percentiles were 2.10 (1.43-3.09), 1.69 (1.29-2.23), 1.45 (1.19-1.76), and 0.86 (0.78-0.94), respectively, with P-overall = 0.001. Further, we found that circulating levels of sex hormones changed during disease progression. The closer the hormone measurement to cancer diagnosis the higher the IAA and estrone levels (P-overall < 0.001 for both), and the temporal change plateaued 10 years before cancer diagnosis (P-nonlinearity < 0.001 for both). An opposite temporal pattern was found for testosterone. Moreover, the significant association between sex hormones and lung cancer survival was only observed when sex hormones were measured within 10 years before diagnosis. Conclusions: Findings from our study, for the first time, demonstrate temporal changes in circulating sex hormones and their associations with lung cancer survival in postmenopausal women. Citation Format: Yingya Zhao, Xiao-Ou Shu, Yu-Tang Gao, Mark M. Kushnir, Qiuyin Cai, Hui Cai, Qing Lan, Nathaniel Rothman, Wei Zheng, Gong Yang. Temporal changes in circulating sex hormones and aromatase activity and associations with lung cancer survival in postmenopausal women. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6491.

Background: Previous studies have investigated the association between reproductive factors and lung cancer risk; however, findings have been inconsistent. This study aims to assess the association between reproductive factors with lung cancer incidence and mortality among Asian women. Methods: A total of 308,949 female participants with a mean age of 55.13 from 11 prospective cohorts and four Asian countries (Japan, Korea, China, and Singapore) in the Asia Cohort Consortium (ACC) were included. Cox proportional hazards regression models were used to estimate the hazard ratios (HR) and 95% confidence intervals (CIs). Results: A total of 3,119 primary lung cancer cases and 2,247 lung cancer deaths were identified with a mean follow-up of 16.4 years. Parous women had a lower risk of lung cancer incidence and mortality as compared with nulliparous women, with HRs of 0.82 (95% CI = 0.70 - 0.96) and 0.78 (95% CI = 0.65 - 0.94). Corresponding HRs were lowest among women with 1-2 children, with HRs of 0.78 (95% CI = 0.66 - 0.93) and 0.72 (95% CI = 0.59 - 0.87) for lung cancer incidence and mortality. The protective association of parity and lung cancer incidence was greater among ever-smokers (HR=0.66, 95% CI = 0.49 - 0.87) than in never-smokers (HR=0.90, 95% CI = 0.74 - 1.09) (P-interaction = 0.029). Compared with age at first delivery ≤20 years, older age at first delivery (≥26 years) was associated with a lower risk of lung cancer incidence and mortality. Compared with age at menopause Conclusions: Distinct from Western women, Asian parous women, especially those who have 1-2 children had a lower risk of lung cancer incidence and mortality compared with nulliparous women. Future studies are needed to assess the underlying mechanisms, the relationships within these female reproductive factors, and the potential changes in smoking habits over time. Citation Format: Xin Yin, Rie Kishida, Sarah Krull Abe, Md. Rashedul Islam, Md. Shafiur Rahman, Eiko Saito, Qing Lan, Batel Bletcher, Melissa Merritt, Ji-Yeob Choi, Aesun Shin, Ryoko Katagiri, Xiao-Ou Shu, Norie Sawada, Akiko Tamakoshi, Woon-Puay Koh, Ichiro Tsuji, Chisato Nagata, Sue K. Park, Sun-Seog Kweon, Yu-Tang Gao, Shoichiro Tsugane, Takashi Kimura, Jian-Min Yuan, Yukai Lu, Seiki Kanemura, Yumi Sugawara, Keiko Wada, Min-Ho Shin, Habibul Ahsan, Paolo Boffetta, Kee Seng Chia, Keitaro Matsuo, You-Lin Qiao, Nathaniel Rothman, Wei Zheng, Manami Inoue, Daehee Kang, Wei Jie Seow. Association between reproductive factors with lung cancer incidence and mortality: A pooled analysis of over 308,000 females in the Asia Cohort Consortium. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 4200.

Background: The etiology of lung cancer among never-smokers has not been adequately elucidated despite that globally15% of lung cancer cases in men and 53% in women are not smoking-related. Epigenetic modifications, including changes in DNA methylation (DNAm), have been suggested as possible underlying mechanisms. However, only a few prospective epigenome-wide association studies (EWAS) of lung cancer incidence have been conducted, all exclusively focused on DNAm in peripheral blood cells and included a minimal number of never-smokers. We aimed to investigate genome-wide DNAm associations and epigenetic age acceleration with future risk of lung cancer among never-smokers using pre-diagnostic oral rinse samples. Methods: We conducted a case-control study of 80 never-smoking incident lung cancer cases and 83 comparable never-smoking controls nested in two large prospective cohorts: the Shanghai Women’s Health Study and Shanghai Men’s Health Study. DNAm was measured using the Illumina EPIC array. The top 50 differentially methylated positions (DMPs) were identified from a discovery sample and tested for replication in a validation sample using robust linear regression models. We also conducted an EWAS in the pooled sample. We examined functional overlap enrichment across chromatin states and histone mark broadPeaks for the top 1000 DMPs using eFORGE and constructed enrichment biological pathways analyses. Results: Across discovery and pooled EWAS, we identified four DMPs associated with lung cancer at the epigenome-wide significance level of P Conclusions: To our knowledge, this is the first prospective EWAS of lung cancer among never-smokers using oral rinse samples. Our results show that DNAm in pre-diagnostic oral rinse samples can provide new insights into lung cancer etiology and risk factors. Citation Format: Mohammad L. Rahman, Charles E. Breeze, Xiao-Ou Shu, Jason YY Wong, Andres Cardenas, Xuting Wang, Bu-Tian Ji, Wei Hu, Batel Blechter, Qiuyin Cai, H Dean Hosgood, Gong Yang, Jianxin Shi, Jirong Long, Yu-Tang Gao, Douglas Bell, Wei Zheng, Qing Lan, Nathaniel Rothman. Epigenome-wide association study of lung cancer among never-smokers in two prospective cohorts in Shanghai [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3483.

Biliary tract cancers are rare but highly fatal with poorly understood etiology. Identifying potentially modifiable risk factors for these cancers is essential for prevention. Here we estimated the relationship between adiposity and cancer across the biliary tract, including cancers of the gallbladder (GBC), intrahepatic bile ducts (IHBDC), extrahepatic bile ducts (EHBDC), and the ampulla of Vater (AVC). We pooled data from 27 prospective cohorts with over 2.7 million adults. Adiposity was measured using baseline body mass index (BMI), waist circumference, hip circumference, waist-to-hip, and waist-to-height ratios. HRs and 95% confidence intervals (95% CI) were estimated using Cox proportional hazards models adjusted for sex, education, race, smoking, and alcohol consumption with age as the time metric and the baseline hazard stratified by study. During 37,883,648 person-years of follow-up, 1,343 GBC cases, 1,194 EHBDC cases, 784 IHBDC cases, and 623 AVC cases occurred. For each 5 kg/m2 increase in BMI, there were risk increases for GBC (HR = 1.27; 95% CI, 1.19–1.36), IHBDC (HR = 1.32; 95% CI, 1.21–1.45), and EHBDC (HR = 1.13; 95% CI, 1.03–1.23), but not AVC (HR = 0.99; 95% CI, 0.88–1.11). Increasing waist circumference, hip circumference, waist-to-hip ratio, and waist-to-height ratio were associated with GBC and IHBDC but not EHBDC or AVC. These results indicate that adult adiposity is associated with an increased risk of biliary tract cancer, particularly GBC and IHBDC. Moreover, they provide evidence for recommending weight maintenance programs to reduce the risk of developing these cancers. Significance: These findings identify a correlation between adiposity and biliary tract cancers, indicating that weight management programs may help minimize the risk of these diseases.

Pancreatic cancer is one of the most fatal human cancers, with an overall 5-year survival rate of 10.8%. Early detection is critical for improving pancreatic cancer prognosis, but biomarkers for early detection are lacking. We conducted a two-stage study to identify circulating miRNAs as biomarkers for pancreatic cancer early detection using pre-diagnostic plasma samples, collected within 5 years prior to cancer diagnosis, from case-control studies nested in five prospective studies. The discovery stage included 185 case-control pairs from the Prostate, Lung, Colorectal and Ovarian Cancer (PLCO) Screening Trial. Replication stage samples comprised 277 case-control pairs from diverse cohorts: Shanghai Women’s and Men’s Health Studies, Southern Community Cohort Study (SCCS), and Multiethnic Cohort. Controls were individually matched on age at enrollment, sex, recruitment site (SCCS), race/ethnicity, and date of blood draw in each cohort. Cell-free small RNAs were extracted from plasma samples and miRNAs were measured by the NanoString nCounter Analysis System using the Human v3 miRNA Expression panel (a total of 798 miRNAs). Normalized miRNAs were categorized by decile. For miRNAs that have ≥10% samples with an undetectable level (0), the non-zero level was categorized by approximately 10% increment. Associations of circulating miRNAs with pancreatic cancer risk, measured in odds ratios (ORs) and 95% confidence intervals (CIs) per decile change, were calculated using conditional logistic regression analyses in discovery and replication studies, separately within each cohort, and results meta-analyzed. We identified three miRNAs, hsa-miR-199a-3p+/hsa-miR-199b-3p, hsa-miR-191-5p, hsa-miR-767-5p, being consistently associated with pancreatic risk in both discovery and replication sets, with combined ORs (95% CIs) of 0.89 (0.84-0.95), 0.90 (0.85-0.95), and 1.08 (1.02-1.13), and P of 9.09E-05, 6.95E-05 and 4.03E-03, respectively. Adjustment for age, BMI, smoking, diabetes and family history of pancreatic cancer did not change the associations. Stratified analyses by age at diagnosis found five additional replicated miRNAs: hsa-miR-640, hsa-miR-1299, hsa-miR-22-3p, hsa-miR-874-5p, and hsa-miR-449b-5p among those 65 years or older, with combined ORs (95% CIs) of 1.33 (1.16-1.52), 1.28 (1.12-1.46), 0.76 (0.65-0.89), 1.25 (1.09-1.43), and 1.22 (1.07-1.39), and P-value ranging from 4.75E-05 to 0.003. These results suggest that circulating miRNA biomarkers may be useful in identifying individuals with high risk of developing pancreatic cancer for close surveillance and/or a screening test. Citation Format: Cong Wang, Hui Cai, Qiuyin Cai, Jie Wu, Rachael Stolzenberg-Solomon, Clair Zhu, Yu-Tang Gao, Jordan Berlin, Fei Ye, Wei Zheng, Veronica W. Setiawan, Xiao-Ou Shu. Circulating miRNA as biomarkers for pancreatic cancer early detection [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr LB109.

Background. Serine and glycine play an important role in the folate-dependent one-carbon metabolism. The metabolism of serine and glycine has been shown to be associated with cancer cell proliferation. No prior epidemiologic study has investigated the associations for serum levels of serine, glycine and their related-metabolites with pancreatic cancer risk. Methods. Two parallel nested case-control studies derived from the Shanghai Cohort Study (with 129 cases and 258 controls) and from the Singapore Chinese Health Study (with 58 cases and 104 controls), respectively, were conducted to examine these associations. Conditional logistic regression method was used to evaluate the associations between serine, glycine and related-matabolites with pancreatic cancer risk, adjusting for potential confounders (i.e., level of education, body mass index, smoking status, alcohol consumption, history of diabetes, serum cotinine concentration, serum pyridoxal 5'-phosphate concentration (PLP), and estimated glomerular filtration rate (eGFR)). Results. In the two studies combined data, odds ratios and 95% confidence intervals of pancreatic cancer for the highest quartile of serine and glycine were 0.42 (0.23-0.78) and 0.39 (0.21-0.72), respectively, compared with the lowest quartile. No significant association with risk of pancreatic cancer was observed for other serine- or glycine related metabolites including cystathionine, cysteine, and sarcosine. Conclusion. The novel observation for a statistically significant inverse association for serum serine and glycine with risk of pancreatic cancer supports a protective role of these metabolites in the folate pathway against the development of pancreatic cancer in humans. With the experimental data showing the impact of serine and glycine on gut microbiota composition, [YJM1] our findings suggest that the alteration of gut microbiota may play a contributing role in reducing the risk of pancreatic cancer. Funding: The Singapore Chinese Health Study was supported by the National Institutes of Health (NIH) of the United States (grants # R01 CA144034 and UM1 CA182876). HN Luu was partially supported by the University of Pittsburgh Medical Center Hillman Cancer Center start-up grant. P Paragomi was supported by the NIH T32CA186873 training grant in cancer epidemiology and prevention. W-P Koh was partially supported by the National Medical Research Council, Singapore (NMRC/CSA/0055/2013). Citation Format: Hung N. Luu, Renwei Wang, Pedram Paragomi, Joyce Y. Huang, Lesley M. Butler, Øivind Midttun, Arve Ulvik, Aizhen Jin, Yu-Tang Gao, Per M. Ueland, Woon-Puay Koh, Jian-Min Yuan. The association between serum serine and glycine and related-metabolites with pancreatic cancer in two prospective cohort studies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 756.

Background: Interleukin-27 (IL-27), which consists of p28 and Epstein-Barr virus induced gene 3 (EBI3) subunits, is a heterodimeric member of the IL-12/IL-23 cytokine family. Overexpression of IL-27 increases T cell expression of inhibitory receptors including PD-1, LAG3, TIGIT, and TIM3 that could limit intensity and duration of T cell responses to infection and cancer. IL-27 is overexpressed in HCC. The purpose of the study was to examine the role of IL-27 in the development of HCC in humans. Methods: Two parallel studies were conducted. The discovery study included 100 HCC cases and 100 individually matched controls from the Singapore Chinese Health Study, which enrolled 32,000 participants with blood samples free of cancer at entry into the study and subsequently followed up for >15 years. The validation study also included 100 HCC cases and 100 matched controls from the Shanghai Cohort Study, a prospective cohort of 18,000 participants who were free of cancer at entry and similarly followed up for >20 years. Human IL-27 protein was quantified in sera that were collected, on average, 4.5 year prior to the diagnosis of HCC. Conditional logistic regression method was used to estimate odds ratio (OR) and 95% confidence interval (CI) of HCC for higher tertiles of IL-27 with adjustment for hepatitis B surface antigen (HBsAg), alcohol intake, body mass index and cigarette smoking. The Cancer Genome Atlas (TCGA) data were analyzed for the IL-27 gene expression in liver and other cancers. Results: Serum IL-27 levels were significantly higher in persons who subsequently developed HCC than controls who remained free of HCC in both studies. In the discovery study, persons in the 2nd (1.39-1.95 ng/mL) and 3rd tertile of IL-27 (>1.95 ng/ml) had 9.3 (95% CI = 1.07-80.36) and 51.7 (95% CI = 5.1-527.8) times the risk of developing HCC, respectively, compared with the lowest IL-27 ( Conclusion: Higher serum levels of IL-27 are associated with significantly higher risk of HCC development. Implications: IL-27 can be further developed as a biomarker for identification of individuals at high risk for HCC and for diagnosis of HCC at an early stage, enabling more frequent surveillance. Anti-IL-27 agents may be developed for primary prevention of HCC. The blockade of IL-27 alone or in combination with other anti-immunosuppressive agents such as anti-PD1, anti-PD-L1, IL-2 and anti-LAG3 may represent novel immunotherapy strategies for HCC. Citation Format: Jian-Min Yuan, Wang Yue, Renwei Wang, Jennifer Admas-Haduch, Woon-Puay Koh, Yu-Tang Gao, Michael T. Lotze. Serum interleukin-27 serves as an early biomarker predicting development of hepatocellular carcinoma (HCC) in two large prospective cohorts [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1050.

We investigated the relationship between telomere length and lung cancer in a pooled analysis from three prospective cohort studies: the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, conducted among men and women in the United States, and previously published data from the Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) Trial conducted among male smokers in Finland, and the Shanghai Women's Health Study (SWHS), which is comprised primarily of never-smokers. The pooled population included 847 cases and 847 controls matched by study, age, and sex. Leukocyte telomere length was measured by a monochrome multiplex qPCR assay. We used conditional logistic regression models to calculate ORs and their 95% confidence intervals (CI) for the association between telomere length and lung cancer risk, adjusted for age and pack-years of smoking. Longer telomere length was associated with increased lung cancer risk in the pooled analysis [OR (95% CI) by quartile: 1.00; 1.24 (0.90–1.71); 1.27 (0.91–1.78); and 1.86 (1.33–2.62); P trend = 0.000022]. Findings were consistent across the three cohorts and strongest for subjects with very long telomere length, i.e., lung cancer risks for telomere length [OR (95% CI)] in the upper half of the fourth quartile were 2.41 (1.28–4.52), 2.16 (1.11–4.23), and 3.02(1.39–6.58) for the PLCO trial, the ATBC trial, and the SWHS, respectively. In addition, the association persisted among cases diagnosed more than 6 years after blood collection and was particularly evident for female adenocarcinoma cases. Telomere length in white blood cell DNA may be a biomarker of future increased risk of lung cancer in diverse populations. Cancer Res; 74(15); 4090–8. ©2014 AACR.

Background: Pancreatic cancer (PC) is one of the deadliest malignancies. Lab based research suggests that two types of polyunsaturated fatty acids (PUFA), n-3 and n-6 PUFA, may have opposing effects on PC carcinogenesis. However, human evidence on the role of dietary PUFA intake in PC risk is inconsistent. Previous epidemiological studies on this topic were conducted primarily among Whites and included few individuals from low socioeconomic status (SES). The exposure patterns of n-3 and n-6 PUFA differ across racial groups, with African Americans (AAs) in the high end and Asians in the low end of range. To better characterize the associations of PUFA with PC risk, we conducted a nested case-control study focusing on understudied Asians and low SES AAs and Whites. Methods: We performed a nested case-control study within three population-based prospective cohort studies: the Shanghai Women’s Health Study (SWHS), Shanghai Men’s Health Study (SMHS), and Southern Community Cohort Study (SCCS). Dietary PUFA intakes and other variables were collected from the baseline dietary and other questionnaires. Participants were followed up through annual linkage to cancer registries. Controls were matched to cases on age, sex, and blood sample availability in SWHS/SMHS, and age, sex, race, and enrollment location in SCCS, using incidence density sampling (case-to-control ratio:1:4). Included in the study are 244 cases and 973 controls of Asian females in SWHS, 189 cases and 755 controls of Asian males in SMHS, and 307 cases and 1,228 controls in SCCS (including 219 cases and 876 controls of AAs). The PUFA intake was adjusted for total energy intake using the residual method. Conditional logistic regression analyses were performed to estimate the association of PC risk with dietary PUFA consumption after adjusting for potential covariates. Results: The average dietary intake (mean ± SD; g/day) of n-3 and n-6 PUFA were 1.0 ± 0.5 and 6.3 ± 3.1, 1.1 ± 0.5 and 7.2 ± 3.0, and 2.0 ± 1.2 and 19.3 ± 11.7 in the controls of SWHS, SMHS and SCCS, respectively. We observed that in Chinese men, high intakes of certain n-3 PUFA were significantly associated with increased PC risk. The adjusted odds ratio (OR) (95% confidence interval (CI)) comparing the highest versus lowest quartile of docosapentaenoic acid (DPA) was 1.79 (1.10-2.93), with a p for trend across quartiles of 0.02. For docosahexaenoic acid (DHA), the OR 4th vs. 1st quartile (95% CI) = 1.71 (1.06-2.78) (p for trend = 0.04). For marine n-3 PUFA, the OR 4th vs. 1st quartile (95% CI) = 1.69 (1.03-2.76) (p for trend = 0.04). There was no significant association for n-6 PUFA. No significant association was observed for n-3 or n-6 PUFA with PC risk in SWHS or SCCS. Conclusions: In a nested case-control study, we observed that in Chinese men, the dietary intake of overall marine n-3 PUFA, DPA, and DHA were positively associated with PC risk. Such associations were not observed in Chinese women or low SES AAs/Whites. Citation Format: Lang Wu, Wei Zheng, William J. Blot, Mark D. Steinwandel, Wanqing Wen, Yong-Bing Xiang, Yu-Tang Gao, Hong-Lan Li, Rachael Stolzenberg-Solomon, Harvey J. Murff, Xiao-Ou Shu. Dietary polyunsaturated fatty acids intake and pancreatic cancer risk in understudied populations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 639.

Background: Organochlorines (OCs) are environmentally persistent compounds that have been extensively used as pesticides and for other industrial applications. Residues of OCs have been detected at hazardous waste sites and in various environmental media worldwide and serum levels of OCs continue to be detectable in the general population. Specific OCs have been associated with non-Hodgkin lymphoma (NHL) risk with varying degrees of evidence. These associations have not been evaluated in Asia, where the exposure patterns of substantially high levels of certain OC pesticides and lower levels of polychlorinated biphenyls (PCBs) are different from Western populations. China accounted for nearly 20% of the worldwide production of dichlorodiphenyltrichloroethane (DDT) through 1983 when it was restricted, and historical usage of hexachlorocyclohexane (HCH) in China has been among the highest in the world. Methods: We evaluated the risk of NHL in relation to pre-diagnostic blood levels of five OC pesticides/metabolites (hexachlorobenzene, β-HCH, oxychlordane, trans-nonachlor, and p,p’-DDE, the primary metabolite of DDT) and four PCB congeners (118, 138-158, 153, 180) in a case-control study of 167 NHL cases and 167 controls matched on age, sex, and blood collection date. The study was nested within three population-based cohorts of Chinese men and women in Shanghai and Singapore. Associations between lipid-adjusted OC concentrations and NHL risk were analyzed using conditional logistic regression. Results: Median levels of p,p’-DDE and β-HCH were up to 12 and 65 times higher, respectively, in Shanghai (blood collected between 1986-2000) compared to reported levels in population-based cohorts in the United States (CLUE; Nurse’s Health Study) and Norway (Janus) with blood collected between 1972-1990. Median levels of p,p’-DDE and β-HCH were more comparable in Singapore (blood collected between 2000-2004) relative to the Western cohorts (1-2 fold concentration differences). Levels of β-HCH were associated with increased risk of overall NHL (3rd vs. 1st tertile OR=1.78, 95%CI=0.98-3.23; ptrend =0.049) in the pooled analysis of three cohorts. No significant associations were observed for other OCs and NHL risk, including for p,p’-DDE. Results for β-HCH and p,p’-DDE were consistent across cohorts. Discussion: Associations between β-HCH and NHL risk have not been consistent in studies of Western populations. Our findings provide the first evidence suggesting associations between blood levels of β-HCH and NHL risk in a population in Asia that experienced far higher exposures. Although there is limited evidence that DDT (IARC Group 2A) is associated with NHL based on studies in Western populations, our findings among Asians, who had higher p,p’-DDE levels than reported in the general population in the West, do not support an association with environmental exposure. Citation Format: Bryan A. Bassig, Xiao-Ou Shu, Andreas Sjodin, Woon-Puay Koh, Yu-Tang Gao, Jennifer Adams-Haduch, Mark Davis, Renwei Wang, Yong-Bing Xiang, Mark Purdue, Bu-Tian Ji, Gong Yang, Richard Jones, H. Dean Hosgood, Wei Jie Seow, Wei Hu, Wei Zheng, Jian-Min Yuan, Qing Lan, Nathaniel Rothman. Pre-diagnostic blood levels of organochlorines and risk of non-Hodgkin lymphoma in three population-based cohorts in China and Singapore [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 5043.

Background Epidemiological studies investigating the associations of coffee and tea intake with lung cancer risk have yielded inconsistent results. These previous studies included mostly lung cancer patients diagnosed among smokers. Because coffee and tea consumption are closely related to smoking behavior, these previous studies could suffer from biases due to residual confounding of smoking. To better characterize the relationship, a large study with a large number of lung cancer cases diagnosed among never smokers and detailed information on tea and coffee consumption, is needed. Using data from a large-scale pooled analysis that consists of over 1.2 million participants in the U.S. and Asia, we carried out a comprehensive evaluation on the association of coffee and tea intake with lung cancer risk. Methods Individual-level data from seven prospective cohort studies conducted in the U.S., and ten studies conducted in Asia, were included. Demographic, lifestyle, coffee and tea intake data were collected at the baseline survey for each study. Multivariable-adjusted Cox proportional hazard models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). The first 2 years of follow-up time was excluded to minimize potential influence of reverse causality on study results. Subgroup analyses by smoking status, sex, race, histologic subtype and coffee type (caffeinated or decaffeinated) were also conducted to assess potential effect modification, as well as heterogeneity of the association. Results After a median follow-up of 8.6 years, 20,519 incident lung cancer cases were identified. Both coffee and tea consumption were associated with an increased risk of lung cancer. Comparing non-coffee and non-tea consumption, HRs for lung cancer associated with exclusive coffee drinkers (≥2 cups/day) among current, former and never smokers were 1.30 (95% CI, 1.15-1.47), 1.49 (1.27-1.74) and 1.41 (95% CI, 1.21-1.63), respectively. Similar positive associations were observed for caffeinated and decaffeinated coffee. The HRs associated with exclusive tea drinkers (>2 cups/day) were 1.16 (95% CI, 1.02-1.32), 1.10 (0.92, 1.32) and 1.37 (95% CI, 1.17-1.60) for current, former and never smokers, respectively. These associations did not differ significantly by sex, race or histologic subtypes. Conclusion A high consumption of coffee or tea was both associated with an increased risk of lung cancer regardless of race or smoking status. Our study included a large number of never-smoker lung cancer patients, which minimized potential confounding effects due to smoking. The positive association observed for both caffeinated and decaffeinated coffee suggests that compounds other than caffeine may play a role in the etiology of lung cancer. Citation Format: Jingjing Zhu, Wei Zheng, Rashimi Sinha, Stephanie A. Smith-Warner, Yong-Bing Xiang, Yikyung Park, Shoichiro Tsugane, Emily White, Woon-Puay Koh, Sue K. Park, Norie Sawada, Seiki Kanemura, Yumi Sugawara, Ichiro Tsuji, Kim Robien, Yasutake Tomata, Keun-Young Yoo, Jeongseon Kim, Jian-Min Yuan, Yu-Tang Gao, Yumie Takata, Eiko Saito, William Blot, Xiao-Ou Shu. Associations of coffee and tea consumption with lung cancer risk: A pooled analysis of 17 cohort studies involving over 1.2 million participants [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 632.

Background: Chronic inflammation causes persistent liver injury and consecutive regeneration, potentially leading to fibrosis and cirrhosis, and consequently, to the development of hepatocellular carcinoma (HCC). Prostaglandin E2 (PGE2) is one of the major end-products of the cyclooxygenase-2 (COX-2) pathway, an enzyme that is an important mediator of inflammation. Oxidative stress, which results from the generation of reactive oxygen species (ROS) by environmental risk factors or cellular mitochondrial dysfunction, may be involved in the progression of chronic liver disease to the development of HCC. 8-epi-prostaglandin F2α (8-epi-PGF2α) is a product of lipid peroxidation, which has been recognized as a specific, chemically stable, quantitative marker of systemic and integrated measure of oxidative stress. Method: We evaluated the associations for urinary levels of 8-epi-PGF2α and PGE-M, a metabolite of PGE2, with HCC risk in a prospective cohort of 18,244 men in Shanghai, China. After 25 years of follow-up, 347 participants developed HCC. For each case, two control subjects were chosen and matched on age (±2 years), date of sample collection (±1 months), and neighborhood of residence. Urinary 8-epi-PGF2α and PGE-M were quantified using validated liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) assays. The logistic regression method was used to calculate the odds ratio (OR) and its 95% confidence interval (CI) associated with higher levels of both urinary 8-epi-PGF2α and PGE-M. Results: 8-epi-PGF2α levels were significantly higher in HCC cases than control subjects (geometric means 0.92 vs 0.80 pmol/mg creatinine, P 1 B surface antigen. This association remained significant after excluding all HCC cases diagnosed within the first 10 years of follow-up. The present study did not find any significant association between urinary PEG-M and HCC risk. Discussion: 8-Isoprostanes have been found to be significantly elevated in patients with non-alcoholic steatohepatitis (NASH). Our findings support a significant role of oxidative stress in the development of HCC, especially in those without traditional risk factors. The null finding on PGE-M and HCC risk is consistent with results of epidemiological studies that did not show a statistically significant association for HCC risk with use of COX-2 inhibitors. Citation Format: Jian-Min Yuan, Menno Grouls, Steven G. Carmella, Renwei Wang, Alicia Heskin, Yang Jiang, Yuting Tan, Jennifer Adams-Haduch, Yu-Tang Gao, Stephen S. Hecht. Prediagnostic levels of urinary 8-epi-prostaglandin F2α and prostaglandin E2 metabolite, biomarkers of oxidative damage and inflammation, and risk of hepatocellular carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2681.

The lung cancer rate among never-smokers is the highest among Asian women, however its etiology and any relevant non-smoking related biomarkers are still unclear. Pre-diagnostic lung cancer-related metabolic biomarkers may provide novel insights into lung cancer mechanisms, and may contribute to the discovery of etiologic factors for the high lung cancer prevalence among Asian women. We evaluated the role of the urinary metabolome in lung cancer development among female never-smokers in China by conducting a nested case-control study of 275 lung cancer cases and 289 healthy controls from the Shanghai Women's Health Study, a prospective cohort comprised of 73,363 Chinese female never-smokers. Metabolic profiling of urinary chemical features was conducted using ultrahigh-performance liquid chromatography - tandem mass spectrometry (UPLC-MS) (39,409 spectral features) and 600 MHz 1H nuclear magnetic resonance (NMR) spectroscopy (386 features). Unconditional logistic regression models were used to estimate the odds ratios (ORs) and 95% confidence intervals (CIs) for the association between each log-transformed metabolite level and lung cancer risk, adjusting for potential confounders such as age, body mass index, history of respiratory disease and passive smoking. Spearman correlation and linear regression were used to estimate associations between the most significant metabolites and pre-diagnosis dietary factors. Three detected UPLC-MS urinary metabolites were negatively associated with lung cancer risk with a false discovery rate of less than 10%: pos_2.61_127.0382m/z (OR = 0.57, 95% CI = 0.46-0.72, P = 1.98 x 10-6), neg_2.60_369.0408m/z (OR = 0.97, 95% CI = 0.96-0.98, P = 1.36 x 10-6), and pos_2.61_184.0325n (OR = 0.55, 95% CI = 0.43-0.71, P = 4.91 x 10-6). These were strongly correlated with each other (rho > 0.65, p < 0.0001). The significant metabolite (pos_2.61_127.0382m/z) was identified as 5-methyl-2-furoic acid and was moderately correlated with self-reported dietary intake of soy (rho = 0.21, p < 0.001). In conclusion, we identified a metabolite in urine (5-methyl-2-furoic acid) that provides support for the protective association of soy-based foods on lung cancer risk that was previously observed in this population of never-smoking women. Further studies are warranted to replicate these findings. Citation Format: Wei Jie Seow, Xiao-Ou Shu, Jeremy Nicholson, Elaine Holmes, Wei Hu, Qiuyin Cai, Yu-Tang Gao, Yong-Bing Xiang, Steve Moore, Bryan A. Bassig, Jason Yy Wong, Jinming Zhang, Bu-Tian Ji, Claire Boulange, Manuja Kaluarachchi, Kyrillos F. Adesina-Georgiadis, Anisha Wijeyesekera, Wei Zheng, Paul Elliot, Nathaniel Rothman, Qing Lan. Prospective study of untargeted urinary metabolomics and risk of lung cancer among female never-smokers in Shanghai, China [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4974.

Background: Findings from in vitro and animal studies suggest a protective role for green tea consumption against gastric cancer risk; however, data from observational studies are inconsistent and mostly based on self-reported dietary intake. To date, very few studies have investigated the association between tea catechins and gastric cancer risk using a biomarker-based approach. Methods: We analyzed data from 300 incident gastric cancer cases and 905 controls nested within a prospective population-based cohort of 18,244 middle-aged and older men in Shanghai, China. Control subjects were matched to case patients by age (± 2 years), date of sample collection (± 1 month), and neighborhood of residence at recruitment. Tea catechins including epicatechin (EC); epigallocatechin (EGC); methylated-EGC (Me-EGC); 5-(3′,4′,5′-trihydroxy-phenyl)-γ-valerolactone (M4); and 5-(3′,4′-dihydroxy-phenyl)-γ-valerolactone (M6) were measured in pre-diagnostic urine specimens collected between 1986 and 1989 using a validated high-performance liquid chromatography method. Odds ratios (ORs) and 95% confidence intervals (CIs) of gastric cancer by quintiles of tea catechins levels were estimated by unconditional logistic regression models, adjusting for potential confounders. Results: Compared to control subjects, cancer patients were less educated with higher prevalences of smoking, alcohol drinking and seropositivity of Helicobactor pylori. Although not statistically significant, cases showed higher levels of urinary tea catechins relative to controls. Among control subjects, levels of urinary catechins were significantly positively associated with consumption of tea. After adjusting for matching factors and selected known risk factors, men with the highest quintiles of urinary EC, Me-EGC, and M6 had significantly higher risk of gastric cancer compared to those in the lowest quintiles (OR = 1.59, 95% CI = 0.98 to 2.58, P trend = 0.04 for EC; OR = 1.47, 95% CI = 0.86 to 2.49, P trend = 0.02 for Me-EGC; OR = 1.64, 95% CI = 1.06 to 2.55, P trend = 0.01 for M6). Corresponding risks were slightly stronger among cases with Helicobactor pylori seropositivity status. When the analysis was limited to ever smokers or drinkers, the OR (95% CI) comparing top versus bottom quintiles of Me-EGC were 2.05 (1.06 to 3.97; P trend = 0.01) and 2.17 (0.98 to 4.79; P trend = 0.04), respectively; however, we did not observe significant interaction between Me-EGC and smoking or alcohol drinking status on risk of gastric cancer. Conclusion: In contrast to some previous findings, we found that tea catechins are associated with increased risk of gastric cancer. The impact of tea catechins on the risk of gastric cancer may differ by smoking or alcohol drinking status. Funding source: NIH/NCI (R01 CA144034 and UM1 CA182876). Citation Format: Hamed Samavat, Renwei Wang, Chung S. Yang, Yu-Tang Gao, Jian-Min Yuan. Association between urinary tea catechins and risk of gastric cancer: A nested case-control study within the Shanghai Cohort Study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5263.

Lung cancer (LC) is a global health burden that accounted for 1.69 million deaths worldwide in 2015. Asian women have high incidence rates of LC, with about half of all diagnoses occurring among never-smokers. Chromosomal copy-number (CN) aberrations can arise de novo in somatic cells with progressing age. These genomic alterations can change the dosage of critical genes and impact biological processes. Gains and losses of Chromosome X in a proportion of cells (ChrX mosaicism) are markers of genomic instability; however, their relationship with LC is unclear. We characterized leukocyte ChrX mosaicism among never-smoking Asian female LC patients and cancer-free controls. We investigated 5,137 LC cases (4,477 adenocarcinomas (AC) and 660 squamous cell carcinomas (SCC)) and 4,535 controls from 13 case-control studies and one cohort study in the Female Lung Cancer Consortium in Asia. Subjects were aged 19-91 years from China, Singapore, Taiwan, South Korea, and Japan. In case-control studies, blood was drawn after diagnosis and mostly before treatment. Leukocyte DNA was genotyped on Illumina 660W arrays, with a small subset on 610Q and 370K arrays. ChrX mosaicism was detected using normalized log R ratio (LRR) and B allele frequency (BAF) values from probes covering ChrX. ChrX was segmented for mosaic events (gains, losses, and copy-neutral) using circular binary segmentation on BAF values. Segments We found 18 detectable mosaic ChrX events (0.19%) in 12 women. LC cases had 5 gains, 3 losses, and 4 copy-neutral mosaic events; while controls had no gains, 1 loss, and 5 copy-neutral events (p=9.8E-2). AC patients had significantly more combined gains and losses (n=7) and fewer copy-neutral events (n=0) compared to controls who had 1 combined gain and loss and 5 copy-neutral events (p=4.7E-3). Notably, all 5 gains were among AC and none among controls (p=3.5E-3). Women with combined gain and loss events had 7 times the odds of AC compared to those without them (p=3.8E-2). Our preliminary findings suggest a possible role of leukocyte ChrX mosaicism, particularly CN gains, in the biological mechanism of LC development. However, given the rarity of these events and possible disease-effect, larger studies are needed to further evaluate ChrX mosaicism as a risk factor for future occurrence of LC. Citation Format: Jason Y. Wong, Mitchell Machiela, Weiyin Zhou, Wei Jie Seow, Bryan Bassig, Jinming Zhang, Meredith Yeager, Wei Zheng, Xiao-Ou Shu, Hongbing Shen, Keitaro Matsuo, Chao Agnes Hsiung, Maria P. Wong, Yun-Chul Hong, Jiu-cun Wang, Yi-Long Wu, Baosen Zhou, Robert Klein, Zhihua Yin, Tangchun Wu, Pan-Chyr Yang, Yong-Bing Xiang, Adeline Seow, Yu-Tang Gao, Chen Wu, Jianjun Liu, Zhibin Hu, Laurie Burdett, Qiuyin Cai, Juncheng Dai, Dongxin Lin, Kexin Chen, Stephen Chanock, Nathaniel Rothman, Qing Lan. Mosaic chromosome X copy-number aberrations in leukocytes of never-smoking lung cancer patients: The Female Lung Cancer Consortium in Asia (FLCCA) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3369.

Background: Higher educational levels have been linked to reduced risk of death in previous studies. Most of these studies, however, were conducted in the United States and Europe. Methods: We performed a pooled analysis to evaluate the association between educational levels and risk of death among 694,434 Asian individuals from 15 prospective cohort studies that participated in the Asia Cohort Consortium. The analysis included 103,023 deaths during a mean follow-up period of 12.5 years, among which 33,945 were due to cancer and 34,645 were due to cardiovascular diseases (CVD). We employed Cox proportional hazards regression models to estimate hazard ratios (HRs) and 95% confidence intervals (95% CIs) for the association between educational level and the risk of death after adjusting for potential confounders. Results: Higher educational levels were significantly associated with lower risk of death from all causes; compared to a low educational level (≤ primary school), HRs and 95% CIs for secondary education, trade/technical education, and ≥ university degree were 0.86 (0.83-0.90), 0.80 (0.71-0.89), and 0.68 (0.61-0.75), respectively (Ptrend Conclusions: Higher educational level was associated with substantially lower risk of death in Asian populations including both East and South Asians. Citation Format: Keming Yang, Ying Zhang, Eiko Saito, Yu Chen, Prakash C. Gupta, Yu-Tang Gao, Akiko Tamakoshi, Yong-Bing Xiang, Woon-Puay Koh, Norie Sawada, Kemmyo Sugiyama, Yasutake Tomata, Atsuko Sadakane, Chisato Nagata, San-Lin You, Renwei Wang, Myung-Hee Shin, Habibul Ahsan, Wen-Harn Pan, Mangesh S. Pednekar, Jian-Min Yuan, Xiao-Ou Shu, Hui Cai, Shoichiro Tsugane, Ichiro Tsuji, Seiki Kanemura, Keiko Wada, Yoon-Ok Ahn, Kotaro Ozasa, John D. Potter, Wei Zheng, Hongmei Nan. Association between educational level and mortality: A pooled analysis of over 694,000 individuals in the Asia Cohort Consortium [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4234.

To evaluate the role of chronic inflammation in the development of gallstones and biliary tract cancer, we examined the risk associated with 62 single nucleotide polymorphisms (SNPs), in 22 inflammation-related genes, in a population-based case-control study conducted in Shanghai, China, where the incidence of biliary tract cancer has been increasing in recent decades. The study included 411 cases with biliary tract cancer (237 gallbladder, 127 extrahepatic bile duct, and 47 ampulla of Vater), 895 with biliary stones, and 786 controls randomly selected from the population. Unconditional logistic regression was used to calculate odds ratios and 95% confidence intervals for the association of individual SNPs and haplotypes with biliary stones and biliary tract cancer. Of the 62 SNPs examined, 14 were related to the risk of biliary cancer and stones. Specifically, variants in the IL8, IL8RB, RNASEL, and NOS2 genes were associated with biliary stones, whereas VEGF variants were associated with gallbladder cancer. Of the 10 genes with multiple SNPs from which we inferred haplotypes, only one IL8RB haplotype, consisting of 3 SNPs (rs2230054, rs1126579, and rs1126580), was associated with the risk of bile duct cancer (P = 0.003) and biliary stones (P = 0.02), relative to the most frequent haplotype. In summary, common variants in genes that influence inflammatory responses may predispose to gallstones and biliary tract cancer, suggesting the need for future studies into the immunologic and inflammatory pathways that contribute to biliary diseases, including cancer. [Cancer Res 2008;68(15):6442–52]

Matrix metalloproteinase-7 (MMP-7) is a small secreted proteolytic enzyme with broad substrate specificity against ECM and non-ECM components. Known to be vital for tumor invasion and metastasis, accumulating evidence also implicates MMP-7 in cancer development. Using data from the Shanghai Breast Cancer Study, we conducted a two-stage study to evaluate the association of MMP-7 single nucleotide polymorphisms (SNPs) with breast cancer risk. Additionally, associated SNPs were characterized by laboratory assays. In stage 1, 11 SNPs were genotyped among 1,079 incident cases and 1,082 community controls using an Affymetrix Genotyping System. Promising SNPs were selected for stage 2 evaluation and genotyped by TaqMan allelic discrimination assays in an independent set of 1,911 cases and 1,811 controls. Three SNPs were selected for stage 2 validation (rs880197, rs10895304, and rs12184413); one had highly consistent results between the two stages of the study. In combined analysis, homozygosity for the variant T allele for rs12184413 was associated with an odds ratio (OR) of 0.7 [95% confidence interval (95% CI), 0.6–0.9] compared with the common C allele. This effect was slightly more pronounced in postmenopausal women (OR, 0.6; 95% CI, 0.4–0.8) than in premenopausal women (OR, 0.8; 95% CI, 0.6–1.1). This SNP is located 3′ of the MMP-7 gene, in an area enriched with CTCF binding sites. In silico analysis suggested a regulatory role for this region, and our in vitro assays showed an allelic difference in nuclear protein binding capacity. Results from our study suggest that common MMP-7 genetic polymorphisms may contribute to breast cancer susceptibility. [Cancer Res 2008;68(15):6453–9]

The CYP11A1 gene encodes the cholesterol side chain cleavage enzyme that catalyzes the initial and rate-limiting step of steroidogenesis. A large number of epidemiologic studies have implicated the duration and degree of endogenous estrogen exposure in the development of breast cancer in women. Here, we conduct a systematic investigation of the role of genetic variation of the CYP11A1 gene in breast cancer risk in a study of 1193 breast cancer cases and 1310 matched controls from the Shanghai Breast Cancer Study. We characterize the genetic architecture of the CYP11A1 gene in a Chinese study population. We then genotype tagging polymorphisms to capture common variation at the locus for tests of association. Variants designating a haplotype encompassing the gene promoter are significantly associated with both increased expression (P = 1.6e−6) and increased breast cancer risk: heterozygote age-adjusted odds ratio (OR), 1.51 [95% confidence interval (95% CI), 1.19–1.91]; homozygote age-adjusted OR, 2.94 (95% CI, 1.22–7.12), test for trend, P = 5.0e−5. Among genes controlling endogenous estrogen metabolism, CYP11A1 harbors common variants that may influence expression to significantly modify risk of breast cancer. [Cancer Res 2007;67(12):5673–82]

The effect of genetic polymorphisms in the TGF-β1 gene at codon 10 (T+29C), codon 25 (G+74C), and the promoter region [C → T at −509 from the transcription site, (C-509T)] on breast cancer survival was evaluated among a cohort of 1111 patients. The median follow-up time for the cohort was 5.17 years after cancer diagnosis. No DNA sequence variation at codon 25 of the TGF-β1 gene was found, whereas polymorphisms in C-509T and T+29C were in strong linkage disequilibrium. Patients who carried the C allele of T+29C polymorphism had a reduced 5-year disease-free survival rate (75.6% for T/C, and 78.2% for C/C) compared with the T/T genotype (85.1%; P, 0.04); the age-adjusted hazard ratio was 1.5 (95% confidence interval, 1.1–2.2). Adjustment for clinical prognostic factors slightly attenuated the association (hazard ratio, 1.4, 95% confidence interval, 1.0–1.9). Our study suggests that genetic polymorphisms in the TGF-β1 gene may play a role in breast cancer progression.

Background: Aflatoxin, which is a risk factor for developing hepatocellular carcinoma, may also cause gallbladder cancer (GBC). We hypothesized that GBC patients have higher aflatoxin exposure than gallstone patients. Methods: We used samples and data from the Shanghai Biliary Tract Cancer Case-control Study conducted from 1997-2001. Of the 267 GBC and 730 gallstone participants with plasma available, 250 GBC patients and 250 controls with gallstones were randomly selected for measurement of aflatoxin B1 (AFB1)-lysine plasma adducts using isotope dilution mass spectrometry. In 54 GBC patients, tumor tissue was examined for the aflatoxin-related R249S TP53 mutation through targeted sequencing. Results: We calculated age- and sex-adjusted odds ratios (ORs) and 95% confidence intervals (CIs), as well as the population attributable fraction (PAF). AFB1-lysine adduct detection was nearly 3-fold higher in GBC patients (OR, 2.96; 95% CI, 1.90 to 4.61). Among the 85 GBC patients (34%) with detectable AFB1-lysine adduct levels, the median was 4.6 pg/mg, compared to 1.2 pg/mg among the 37 gallstone-patient controls (15%) with detectable levels, and the OR was 9.97(95% CI, 2.73 to 36.44) for quartile 4 versus 1 in GBC compared to gallstone patients. The PAF for cancer related to aflatoxin was 22% (95% CI: 17%-27%). In 54 GBC patients, tumor tissue was examined for the aflatoxin-related R249S TP53 mutation through targeted sequencing, however, all patients were negative for the mutation. Conclusions: Aflatoxin is significantly associated with GBC, although it does not appear to be correlated with the R249S TP53 mutation in the gallbladder. If aflatoxin is shown to cause GBC, it may have accounted for up to 22% of the gallbladder cancers in Shanghai during the study period and may account for an even higher proportion in high-risk areas. If verified, reducing aflatoxin exposure may reduce GBC incidence. Citation Format: Jill E. Koshiol, Yu-Tang Gao, Michael Dean, Patricia Egner, Chirag Nepal, Kristine Jones, Bingsheng Wang , Asif Rashid, Wen Lou, Alison Van Dyke, Catterina Ferreccio, Michael Malasky, Ming-Chang Shen, Bin zhu, Jesper Andersen, Allan Hildesheim, Ann W. Hsing, John Groopman. Association of aflatoxin and gallbladder cancer in Shanghai [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-157. doi:10.1158/1538-7445.AM2017-LB-157

Background: Infections with some viruses, including HIV and Epstein-Barr Virus (EBV) in immunocompromised individuals, are known risk factors for specific subtypes of non-Hodgkin lymphoma (NHL). Epidemiological studies have also provided evidence that hepatitis viruses, and EBV in immunocompetent individuals, may also be associated with NHL. There are limited population-based prospective studies with pre-diagnostic blood samples that have comprehensively evaluated viral biomarkers and NHL risk in East Asians, in whom the descriptive characteristics of NHL and the prevalence of certain viral infections differ compared to Western populations. Methods: We conducted a nested case-control study of 214 NHL cases and 214 controls from three population-based prospective cohorts in Shanghai and Singapore. Cases and controls were individually matched by age, sex, date of blood draw, and cohort. Antibodies to 21 antigens associated with the evaluated viruses (herpesviruses, Hepatitis B (HBV) and C (HCV), and polyomaviruses) were measured in plasma/serum using fluorescent bead-based multiplex serology. Conditional logistic regression was used to evaluate associations between viral antibody levels measured as median fluorescence intensity and NHL. Results: For herpesviruses, an increased risk of NHL was observed for higher compared to lower early antigen diffuse (EA-D) (OR = 2.2, 95% CI = 1.2-4.1) and BZLF1-encoded replication activator (ZEBRA) (OR = 2.2, 95% CI = 1.0-4.9) antibodies (rsp = +0.5) associated with EBV. An increased risk of NHL was also observed among those seropositive for the intermediate-early 1A antigen (OR = 1.9, 95% CI = 1.0-3.3; rsp with EBV antibodies = +0.2) associated with human herpesvirus-6 (HHV-6). For hepatitis viruses, a significant NHL risk was observed for higher compared to lower antibodies to the HBV-associated core (HBc) antigen (OR = 1.8, 95% CI = 1.1-3.1), and this risk was particularly apparent in those with the highest HBc and EBV EA-D antibody levels (OR = 4.2, 95% CI = 1.4-12.7) compared to the lowest. Seropositivity to HCV was low (1.4% cases; 0.9% controls). No associations with NHL risk were observed for individual polyomaviruses (BK, JC, TSV, MCV). Discussion: Our study of serologic markers of infection and NHL risk in three prospective population-based studies of Chinese individuals suggests a role of specific viral agents in lymphomagenesis. The findings for EBV are consistent with some data from Western cohorts and indicate that EBV reactivation may be associated with NHL risk in the Chinese general population. HHV-6 is a lymphotropic virus that has been observed in some retrospective studies to be associated with lymphoma, but to our knowledge has not previously been associated with NHL prospectively in the general population. HBV is endemic to regions of East Asia, including China, and our data suggest that high levels of antibodies to the HBc antigen may be a marker for NHL risk. Citation Format: Bryan A. Bassig, Angelika Michel, Xiao-Ou Shu, Woon-Puay Koh, Yu-Tang Gao, Lesley M. Butler, Mark Purdue, Yong-Bing Xiang, Jennifer Adams-Haduch, Renwei Wang, Nicole Brenner, Tim Waterboer, Martina Willhauck-Fleckenstein, Bu-Tian Ji, H. Dean Hosgood, Charles S. Rabkin, Jason Y.Y. Wong, Jinming Zhang, Wei Hu, Gong Yang, Wong-Ho Chow, Michael Pawlita, Wei Zheng, Jian-Min Yuan, Qing Lan, Nathaniel Rothman. Serologic markers of infection and risk of non-Hodgkin lymphoma in a pooled prospective study of three Chinese cohorts [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4245. doi:10.1158/1538-7445.AM2017-4245

Background. Animal studies have shown that polyunsaturated fatty acids (PUFAs) have antineoplastic and anti-inflammatory properties. Results from epidemiologic studies, however, have been inconclusive. We prospectively evaluated the association of dietary PUFA intakes and lung cancer risk in two population-based cohort studies, the Shanghai Men’s Health Study (SMHS) and Shanghai Women’s Health Study (SWHS). Methods. A total of 130,823 study participants (i.e., 60,427 men and 70,396 women) were included in the current analysis. Dietary fatty acid intakes were derived from data collected at the baseline by validated food frequency questionnaires. Cox proportional hazards model was applied to assess the association between PUFAs intake and lung cancer risk with adjustment for age, smoking status, smoking packs-year (men only), drinking status, BMI, physical activity status, total energy, red meat intake, vegetable intake, vitamin supplemental use, menopausal status and hormone replacement therapy (women only). Results. We found an inverse association between total PUFA intakes and lung cancer risk [hazard ratios (HRs) and respective 95% confidence intervals (CIs) for quintiles 2, 3, 4, and 5 versus quintile 1 were 0.73 (0.56-0.95), 0.78 (0.59-1.03), 0.64 (0.47-0.87) and 0.55 (0.37-0.82) in SMHS and 0.69 (0.53-0.90), 0.76 (0.58-0.99), 0.60 (0.44-0.81), and 0.53 (0.36-0.79) in SWHS)]. A similar pattern was observed for consumption of linoleic acid, total n-6 PUFAs and the ratio n-6 PUFAs/n-3 PUFAs in adenocarcinoma patients. This inverse association was more likely in male ever-smokers for total PUFAs, linoleic acid and total n-6 PUFAs. On the other hand, EPA intake was positively associated with lung cancer risk in female never-smokers [HRs and 95% CIs: 1.10 (0.85-1.42), 1.36 (1.05-1.77), 1.28 (0.97-1.88) and 1.30 (0.97-1.74), for quintiles 2-5 versus quintile 1]. A similar positive association between DHA intake and lung cancer risk in female never-smokers was also observed. Conclusions. Total polyunsaturated fatty acid, linoleic and total n-6 PUFA intakes were associated with decreased risk of lung cancer. EPA and DHA intakes were associated with an increased risk of lung cancer among female never-smokers. Financial Support: This work was supported by grants from the US National Institutes of Health/National Cancer Institute (R37 CA070867 and UM1 CA182910 - to Wei Zheng; R01 CA082729, UM1 CA173640 and R25 CA160056 - to Xiao-Ou Shu) Citation Format: Hung N. Luu, Harvey J. Murff, Honglan Li, Qiuyin Cai, Yu-Tang Gao, Jing Gao, Hui Cai, Gong Yang, Qing Lan, Yong-Bing Xiang, Wei Zheng, Xiao-Ou Shu. A prospective study of dietary polyunsaturated fatty acid intakes and lung cancer risk [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5310. doi:10.1158/1538-7445.AM2017-5310

Background. Previous studies suggest that a stable end-product of prostaglandin E2, the urinary metabolite PGE-M, is excreted in the urine and can be used as an index of systemic prostaglandin E2 (PGE2) production. In the present study we investigate the PGE-M, Helicobacter pylori (H. pylori), and gastric cancer association. Methods. The present analysis included 359 prospectively ascertained gastric cancer cases and 700 individually matched controls from two population-based prospective cohort studies, the Shanghai Women’s Health Study and Shanghai Men’s Health Study. Urinary PGE-M was measured by a liquid chromatography/tandem mass spectrometric method. Sero-positivity to 15 H. pylori recombinantly expressed fusion proteins was detected by H. pylori multiplex serology. Results. Adjusting for H. pylori, increasing PGE-M was associated with higher risk of gastric cancer (Quartile 4 vs. 1, OR=1.76, 95% CI: 1.17-2.66, Ptrend =0.004). This association remained after excluding those diagnosed within two years from sample collection (OR=1.73, 95% CI: 1.12-2.65, Ptrend =0.007). However it was no longer present among individuals with 10 or more years of follow-up (2-4.9 years, OR=3.15, 95% CI: 1.11-8.91; 5-9.9 years, OR=2.23, 95% CI: 1.22-4.06; ≥10 years, OR=0.73, 95% CI: 0.31-1.70). The association of PGE-M with gastric cancer risk was not modified by H. pylori status, but added predictive ability beyond H. pylori; compared to H. pylori-negative individuals with below-median PGE-M levels, H. pylori-positive individuals with above-median PGE-M levels had a 5-fold increase in the odds ratio of gastric cancer (OR=5.08, 95% CI: 2.47-10.43). Conclusion. In China, higher PGE-M levels may indicate an increased risk of gastric cancer independent of the risk conferred by H. pylori infection status, particularly for cancers diagnosed within 10 years of sample collection. Citation Format: Tianyi Wang, Hui Cai, Wei Zheng, Angelika Michel, Michael Pawlita, Ginger Milne, Yong-Bing Xiang, Yu-Tang Gao, Hong-Lan Li, Nathaniel Rothman, Qing Lan, Xiao-Ou Shu, Meira Epplein. A prospective study of urinary prostaglandin E2 metabolite, Helicobacter pylori antibodies, and gastric cancer risk [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2272. doi:10.1158/1538-7445.AM2017-2272

Background: Vitamin B6 is an important enzymatic cofactor in the synthesis of one-carbon units, amino acids, and carbohydrates. Epidemiological studies provided inconsistent results on the associations between circulating pyridoxal 5’-phosphate (PLP) and pancreatic cancer risk. Method: Two nested case-control studies were conducted within the Shanghai Cohort Study (SCS) (129 cases and 258 matched controls) and the Singapore Chinese Health Study (SCHS) (58 cases and 104 matched controls). Concentrations of B6 vitamers [PLP, pyridoxal (PL), and pyridoxic acid (PA)] were measured in serum samples that were collected an average of 12.5 years in SCS and 6.8 years in SCHS before pancreatic cancer diagnosis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated using conditional logistic regression with the adjustment for potential confounders. Restricted cubic splines were used to explore potential non-linear relationships between serum B6 vitamer levels and pancreatic cancer risk. Results: The median (5th-95th percentiles) levels of serum PLP were 25.7 (10.0-91.7) nmol/L among SCS controls and 58.1 (20.8-563.0) nmol/L among SCHS controls. There was a statistically significant inverse association between serum PLP levels and pancreatic cancer risk in SCS [comparing highest (&gt35.6 nmol/L) with lowest quartile (&lt18.1 nmol/L): OR = 0.44; 95%CI: 0.21, 0.91; P for trend = 0.01]. A weaker, nonsignificant association was observed in SCHS [comparing highest (&gt88.2 nmol/L) with lowest quartile (&lt39.2 nmol/L): OR = 0.78; 95%CI: 0.27, 2.25; P for trend = 0.64]. We combined the data from the two cohorts for the purpose of estimating the serum PLP-pancreatic cancer association using the widest PLP range available. Compared with PLP < 20 nmol/L (defined as vitamin B6 deficiency), PLP ≥ 45 nmol/L was associated with 50% reduced risk of pancreatic cancer (OR = 0.50; 95%CI: 0.26, 0.97). Although the linear trend test was statistically significant (P = 0.04), the spline curve modeling revealed a non-linear relationship between serum PLP and pancreatic cancer, where the strongest inverse relationship was seen among individuals with serum PLP of below approximately 30 nmol/L, while a weak association was seen for those with higher PLP levels. The inverse association between PLP and pancreatic cancer risk remained statistically significant after excluding cases whose blood samples were collected within 2 years prior to cancer diagnosis. No statistically significant association was found between serum level of PL or PA and pancreatic cancer risk. Conclusion: This study is consistent with the notion that higher circulating levels of PLP may protect against the development of pancreatic cancer. The inverse association between circulating PLP and pancreatic cancer risk may be masked in studies without a wide range of PLP levels, or enough subjects in deficient range. Citation Format: Joyce Y. Huang, Lesley M. Butler, Øivind Midttun, Per M. Ueland, Woon-Puay Koh, Yu-Tang Gao, Jian-Min Yuan. Serum B6 vitamers (pyridoxal-5’-phosphate, pyridoxal, and 4’-pyridoxic acid) and pancreatic cancer risk: Two nested case-control studies in Asian populations. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4297.

Background: Diabetes is associated with an increased risk of several cancers and overall mortality and cancer mortality. Previous studies have suggested that metformin use may decrease cancer mortality, though findings have been inconsistent. We examined metformin and other diabetes medication use and survival from breast, colorectal, lung, and gastric cancers with respect to timing of diabetes medication initiation. Methods: Electronic medical record (EMR) data on diabetes medication use was extracted for 2,890 participants from the Shanghai Men's Health Study (SMHS) and Shanghai Women's Health Study (SWHS) with incident breast (n = 633), colorectal (n = 892), lung (n = 822), or gastric (n = 543) cancers diagnosed after 2004. Individuals with and without diabetes diagnosis were analyzed for the association between diabetes medication use (metformin, sulfonylureas, and insulin) and cancer survival using Cox proportional hazards models. Results: After adjustment for patient and clinical characteristics, ever use of any diabetes medication was associated with a decrease in all-cause mortality among all four cancer types (HR = 0.84, 95% CI: 0.74, 0.94). Compared to non-users of any diabetes drug, lower mortality was observed among all cancer patients who ever took metformin (HR = 0.78, 95% CI: 0.65, 0.93) or sulfonylureas (HR = 0.80, 95% CI: 0.69, 0.93). When stratified by initiation of diabetes medication use with respect to cancer diagnosis, the association was significant for both metformin and sulfonylureas use, but only among those who initiated use after cancer diagnosis. When cancers were analyzed individually, significant associations were observed for lung and colorectal cancer cases for metformin or sulfonylureas use among those who initiated use after cancer diagnosis. The inverse associations were predominantly observed among those whose diabetes diagnosis could be verified by EMR. Diabetes medication use was not significantly associated with survival from breast or gastric cancer. Conclusions: Use of metformin or sulfonylureas was associated with improved survival among lung and colorectal cancer patients. The association was primarily observed among those who initiated diabetes medication use after cancer diagnosis. While a possible survival time bias can't be excluded, additional investigation on the topic is needed given the potential translational impact if our finding were proved to be true. Citation Format: Michelle L. Baglia, Yong-Bing Xiang, Gong Yang, Tao Zheng, Honglan Li, Mingrong You, Yong Cui, Yu-Tang Gao, Wei Zheng, Xiao-Ou Shu. Diabetes medication use in association with survival among patients of breast, colorectal, lung, and gastric cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3419.

Introduction: Epidemiological findings on the association between fruit and vegetable consumption and gastric cancer risk remain inconsistent, resulting in the designation of fruits and vegetables as “probable” and “possible” respectively, but not “convincing”, protective factors. However, intervention studies provide support for the effect of micronutrient supplementation in the prevention of gastric cancer and its precursor lesions, though the effects remain less substantial than for Helicobacter pylori (H. pylori) eradication. Objective: To ascertain the association between fruit and vegetable intake and non-cardia gastric cancer incidence with adjustment for H. pylori within East Asian cohort studies. Methods: The present analysis includes 1970 participants (810 prospectively ascertained non-cardia gastric cancer cases with 1160 matched controls) from 5 cohort studies in the Helicobacter pylori Biomarker Cohort Consortium. These cohorts collected blood samples as well as demographic, lifestyle, and dietary data at baseline. Pre-diagnostic antibody levels to 15 H. pylori proteins were assessed using multiplex serology. Conditional logistic regression, adjusting for total energy intake, smoking, and H. pylori status, was applied to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for gastric cancer risk across cohort- and sex-specific quartiles of fruit and vegetable intake. Results: Increasing fruit intake was significantly associated with decreasing risk of non-cardia gastric cancer, so that individuals in the highest quartile of fruit consumption had a 29% reduced odds of gastric cancer, compared to individuals in the lowest quartile (OR = 0.71, 95% CI, 0.52-0.95, P for trend = 0.02). Compared to CagA-positive H. pylori low fruit consumers, the strongest inverse association of gastric cancer risk was amongst those high fruit consumers without evidence of H. pylori antibodies (OR = 0.12, 95% CI: 0.06-0.25), whereby the inverse association by increasing fruit consumption was attenuated among individuals infected with CagA-positive H. pylori (OR = 0.82, 95% CI: 0.66-1.03). We observed a weaker, non-dose-response suggestion of an inverse association of vegetable intake with non-cardia gastric cancer risk. Conclusions: To our knowledge, this is the largest prospective study in the high-risk region of East Asia to examine the association of fruit and vegetable consumption with non-cardia gastric cancer risk adjusted for H. pylori. We have found that high fruit intake may play a role in decreasing risk of non-cardia gastric cancer, even after adjustment for H. pylori subtype-specific infection. Funding: R01 CA174853 Citation Format: Tianyi Wang, Hui Cai, Shizuka Sasazuki, Shoichiro Tsugane, Wei Zheng, Eo Rin Cho, Sun Ha Jee, Angelika Michel, Michael Pawlita, Yong-Bing Xiang, Yu-Tang Gao, Xiao-Ou Shu, Weicheng You, Meira Epplein. Fruit and vegetable consumption and risk of gastric cancer: a prospective nested case-control study in China, Japan and Korea. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1735.

Only two genome-wide association studies (GWAS) have been conducted to date to identify potential markers for total mortality after diagnosis of breast cancer. Here, we report the identification of two single-nucleotide polymorphisms (SNP) associated with total mortality from a two-stage GWAS conducted among 6,110 Shanghai-resident Chinese women with tumor–node–metastasis (TNM) stage I to IV breast cancer. The discovery stage included 1,950 patients and evaluated 613,031 common SNPs. The top 49 associations were evaluated in an independent replication stage of 4,160 Shanghai patients with breast cancer. A consistent and highly significant association with total mortality was documented for SNPs rs3784099 and rs9934948. SNP rs3784099, located in the RAD51L1 gene, was associated with total morality in both the discovery stage (P = 1.44 × 10−8) and replication stage (P = 0.06; P-combined = 1.17 × 10−7). Adjusted HRs for total mortality were 1.41 [95% confidence interval (CI), 1.18–1.68] for the AG genotype and 2.64 (95% CI, 1.74–4.03) for the AA genotype, when compared with the GG genotype. The variant C allele of rs9934948, located on chromosome 16, was associated with a similarly elevated risk of total mortality (P-combined = 5.75 × 10−6). We also observed this association among 1,145 patients with breast cancer of European ancestry from the Nurses’ Health Study (NHS; P = 0.006); the association was highly significant in a combined analysis of NHS and Chinese data (P = 1.39 × 10−7). Similar associations were observed for these two SNPs with breast cancer–specific mortality. This study provides strong evidence suggesting that the RAD51L1 gene and a chromosome 16 locus influence breast cancer prognosis. Cancer Res; 72(5); 1182–9. ©2012 AACR.

Tobacco smoking is the most important and well-established bladder cancer risk factor and a rich source of chemical carcinogens and reactive oxygen species that can induce damage to DNA in urothelial cells. Therefore, common variation in DNA repair genes might modify bladder cancer risk. In this study, we present results from meta-analyses and pooled analyses conducted as part of the International Consortium of Bladder Cancer. We included data on 10 single nucleotide polymorphisms corresponding to seven DNA repair genes from 13 studies. Pooled analyses and meta-analyses included 5,282 cases and 5,954 controls of non-Latino white origin. We found evidence for weak but consistent associations with ERCC2 D312N [rs1799793; per-allele odds ratio (OR), 1.10; 95% confidence interval (95% CI), 1.01-1.19; P = 0.021], NBN E185Q (rs1805794; per-allele OR, 1.09; 95% CI, 1.01-1.18; P = 0.028), and XPC A499V (rs2228000; per-allele OR, 1.10; 95% CI, 1.00-1.21; P = 0.044). The association with NBN E185Q was limited to ever smokers (interaction P = 0.002) and was strongest for the highest levels of smoking dose and smoking duration. Overall, our study provides the strongest evidence to date for a role of common variants in DNA repair genes in bladder carcinogenesis. The TBCS was supported by a grant to ECNIS, a network of excellence operating within the European Union 6th Framework Program, Priority 5: ‘‘Food Quality and Safety’’ (contract no. 513943), by a grant of the Italian Association for Cancer Research, of the Piedmont Region Progetti di Ricerca Sanitaria Finalizzata, and of the Ministry of Health, Oncology Integrated Project ‘‘Cancer Primary Prevention in Italy: a research-based approach,’’ Italy.

4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and its glucuronides (sum of which is denoted as total NNAL) are metabolites of 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK). NNK and NNAL can induce lung cancer in laboratory animals but human data are limited. The association between prediagnostic levels of urinary total NNAL and risk of lung cancer development was evaluated in two prospective cohorts of Chinese cigarette smokers. We conducted a nested case-control study involving 246 cases of incident lung cancer and 245 cohort controls who were individually matched to the index cases by age, gender, neighborhood of residence at cohort enrollment, and date of urine collection. Urinary levels of total NNAL were significantly associated with risk of lung cancer in a dose-dependent manner. Relative to the lowest tertile, risks associated with the second and third tertiles of total NNAL were 1.43 [95% confidence interval (95% CI), 0.86–2.37] and 2.11 (95% CI, 1.25–3.54), respectively (P for trend = 0.005) after adjustment for self-reported smoking history and urinary total cotinine. Smokers in the highest tertiles of urinary total NNAL and total cotinine exhibited a 8.5-fold (95% CI, 3.7–19.5) increased risk for lung cancer relative to smokers with comparable smoking history but possessing the lowest tertiles of urinary total NNAL and total cotinine. Findings of the present study directly link NNK exposure to lung cancer development in humans. [Cancer Res 2009;69(7):2990–5]

Animal and cell studies indicate an inhibitory effect of matrix metalloproteinase-8 (MMP8) on tumorigenesis and metastasis. We investigated whether MMP8 gene variation was associated with breast cancer metastasis and prognosis in humans. We first studied nine tagging single nucleotide polymorphisms (SNP) in the MMP8 gene in 140 clinically and pathologically well-characterized breast cancer patients. Four of the SNPs were found to be associated with lymph node metastasis, the most pronounced being a promoter SNP (rs11225395) with its minor allele (T) associating with reduced susceptibility to lymph node metastasis (P = 0.02). This SNP was further evaluated for association with cancer relapse and survival among a cohort of ∼1,100 breast cancer patients who had been followed for cancer recurrence and mortality for a median of 7.1 years. The T allele was associated with reduced cancer relapse and greater survival, particularly among patients with earlier stage cancer. Among patients of tumor-node-metastasis stage 0 to II, the adjusted hazard ratio of disease-free survival was 0.7 [95% confidence interval (95% CI), 0.5–0.9] for patients carrying T allele compared with those homozygous for the C allele (P = 0.02). In vitro experiments showed that the T allele had higher promoter activity than the C allele in breast cancer cells. Electrophoretic mobility shift assays showed binding of nuclear proteins to the DNA sequence at the SNP site of the T allele but not that of the C allele. The data suggest that MMP8 gene variation may influence breast cancer prognosis and support the notion that MMP8 has an inhibitory effect on cancer metastasis. [Cancer Res 2007;67(21):10214–21]

Breast cancer is the most common invasive cancer in females worldwide. Over the past 20 years, pathogenic mutations in high-penetrance genes (BRCA1, BRCA2, PTEN and TP3) and moderate-penetrance genes (CHEK2, ATM, BRIP1, PALB2, RAD51C, STK11, CDH1, RAD50, and NBN) have been identified for breast cancer. Most of these germline mutations were discovered in women of European ancestry. In the present study, we aimed to investigate whether there were additional variants in these genes associated with breast cancer among women of Asian ancestry. Genotype data of up to 584 SNPs from the Breast Cancer Association Consortium (BCAC) including 6,269 cases and 6,624 controls and Shanghai Breast Cancer Genetics Study (SBCGS) including 5,768 cases and 5,703 controls were analyzed. Among 18 reported pathogenic mutations included in the dataset, 12 were monomorphic. Rare allele was observed for the remaining 6 mutations and one of them showed a significant association with P value of 0.0096. Among the remaining SNPs that were not previously reported to be pathogenic, 17 SNPs showed nominal association with P10% in the RAD50 gene were also associated with breast cancer with P value of 1.0 × 10−3 and 2.1 × 10−3 for rs17772583 and rs2244012, respectively. Our study identified several new risk variants in the BRCA1, BRCA2 and RAD50 genes in relation to breast cancer risk in East Asian women and showed that many of the previously reported pathogenic mutations in the high/moderate penetrance genes of breast cancer are rare in our study population comprising largely sporadic breast cancer patients of East Asian ancestry. Citation Format: Mi-Ryung Han, Xiao-Ou Shu, Qiuyin Cai, Yu-Tang Gao, Ying Zheng, Kyriaki Michailidou, Joe Dennis, Manjeet K. Bolla, Qin Wang, Ji-Yeob Choi, Mikael Hartman, Daehee Kang, Artitaya Lophatananon, Hui Miao, Keitaro Matsuo, Kenneth Muir, Sulee Sangrajrang, Chen-Yang Shen, Soo Hwang Teo, Anna H Wu, Alison M. Dunning, Paul D. P. Pharoah, Douglas F. Easton, Wei Zheng, Jirong Long. Evaluation of genetic variants in high and moderate-penetrance breast cancer susceptibility genes in East Asians. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2780. doi:10.1158/1538-7445.AM2015-2780

Recent prospective studies conducted in Western populations among mostly Caucasians have suggested that higher levels of soluble CD27 (sCD27) and soluble CD30 (sCD30), two markers indicative of B-cell activation, are associated with risk of non-Hodgkin lymphoma (NHL), with significant associations persisting in individuals who were diagnosed with NHL >10 years after blood collection. However, there are currently no molecular epidemiologic data evaluating whether these biomarkers are associated with NHL risk in East Asian populations, in whom the descriptive characteristics of NHL in terms of subtype distributions and incidence rates are quite different from those in Western countries. To explore potential mechanistic commonalities for NHL in these populations, we conducted a pooled nested case-control study from three prospective studies of Chinese men and women including 218 NHL cases and 218 individually matched controls. Levels of sCD27 and sCD30 were measured in all study subjects at the same time using ELISA. Conditional logistic regression was used to calculate odds ratios (ORs) and 95% confidence intervals (CIs) according to quartile levels of sCD27 or sCD30 in all study controls. An increased risk of NHL in the pooled population was observed for elevated levels of both sCD27 and sCD30. Compared to the lowest quartile, ORs (95% CIs) for the 2nd, 3rd, and 4th quartiles of sCD27 were 1.60 (0.83-3.09), 1.94 (0.98-3.83), and 4.45 (2.25-8.81), respectively (ptrend = 0.000005). The corresponding ORs (95% CIs) for sCD30 were 1.74 (0.85-3.58), 1.86 (0.94-3.67), and 5.15 (2.62-10.12) (ptrend = 0.0000002). These associations remained statistically significant in individuals diagnosed with NHL 10 or more years after blood draw. Notably, the magnitude of the associations with risk of NHL was very similar to those in Western populations in previous studies. These findings of the similar association between sCD27 or sCD30 and NHL risk across different populations support an important underlying mechanism of B-cell activation in lymphomagenesis. Citation Format: Bryan A. Bassig, Xiao-Ou Shu, Woon-Puay Koh, Yu-Tang Gao, Mark P. Purdue, Lesley M. Butler, Jennifer Adams-Haduch, Yong-Bing Xiang, Troy J. Kemp, Renwei Wang, Ligia A. Pinto, Tongzhang Zheng, Bu-Tian Ji, H. Dean Hosgood, Wei Hu, Gong Yang, Heping Zhang, Wong-Ho Chow, Christopher Kim, Wei Jie Seow, Wei Zheng, Jian-Min Yuan, Qing Lan, Nathaniel Rothman. Soluble levels of CD27 and CD30 are associated with risk of non-Hodgkin lymphoma in a pooled analysis of three prospective cohorts of Chinese men and women in Shanghai and Singapore. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 846. doi:10.1158/1538-7445.AM2015-846

Aberrant DNA methylation in peripheral blood cells (PBCs) may be a potential noninvasive biomarker for breast cancer risk. Whether methylations at CpG sites in PBCs can serve as proxies in normal breast tissues are unknown. We evaluated the correlation of DNA methylation between PBC DNA and breast cancer adjacent normal tissue using data and samples from the Shanghai Breast Cancer Study. Genome-wide DNA methylation profiles in PBCs and matched tumor adjacent normal tissues from 31 breast cancer patients were assayed using Illumina HumanMethylation450 BeadChip, which includes >485,000 CpG sites and covers 99% of RefSeq genes with multiple probes per gene. After quality control filtering, a total of 432,513 CpG sites were retained for analysis. We evaluated the association of methylation for each CpG site between DNA samples for PBCs and normal breast tissue using Spearman's rank correlation (r). The overall median correlation (r) is 0.35. A total of 99,928 (23.1%) and 18,290 (4.2%) of CpG sites had a high correlation between PBC DNA and normal breast tissue DNA with r > 0.50 and r > 0.75, respectively. Interestingly, we found that the correlation was negatively associated with both methylation levels (mean) and inter-person variability (standard deviation) in normal breast tissues (ρ = -0.14 and -0.19; P < 2.2 x10-16 for both). In 926 previously reported methylation quantitative trait loci, the correlations between PBCs and normal breast tissue were significantly higher (median r = 0.42) than the remaining loci (median r = 0.34) (Wilcoxon test, P < 2.2 x10-16), suggesting that methylation variations controlled by genetic variations exhibited a robust correlation between PBC and normal breast tissue DNA. In summary, our results suggest that the methylation status of CpG sites with large inter-person variations or genetically-controlled variations in PBC DNA correlate well with those in normal breast tissue. This supports the conduct of a genome-wide methylation study for breast cancer using PBC DNA. Citation Format: Xingyi Guo, Xiao-Ou Shu, Chun Li, Jirong Long, Bingshan Li, Yu-Tang Gao, Wei Zheng, Qiuyin Cai. Correlation of DNA methylation pattern between peripheral blood cell and normal breast tissue. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1068. doi:10.1158/1538-7445.AM2015-1068

Background. Cytochrome P450 2A6 (CYP2A6) is polymorphic and catalyzes the metabolism of nicotine and certain tobacco-specific carcinogens, including 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), thus may affect smoking behavior and contribute to substantial inter-individual variation in risk of smoking-related lung cancer. CYP2A6 genetic profiling coupled with biomarkers of tobacco constituents may enhance our understanding of the gene-environmental interaction on lung cancer risk and identify high-risk smokers for targeted preventive interventions. Methods. A nested case-control study including 513 incident lung cancers and an equal number of individually matched controls, all current smokers, was conducted within two population-based prospective cohorts with >80,000 Chinese men and women aged 45-74 years in Shanghai and Singapore. Quantified were 4 common allelic variants of CYP2A6 (*4, *7, *9 and rs1137115), urinary total nicotine, total cotinine (COT), total 3-hydroxy cotinine (3HC), and total 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL, a NNK metabolite). Calculated were total nicotine equivalent (TNE = the sum of total nicotine, COT and 3HC), and the 3HC:COT ratio as a measure of CYP2A6 activity. Logistic regression method was used to calculate odd ratios and 95% confidence intervals for lung cancer associated CYP2A6 genotypes and urinary biomarkers. Results. The smoking-adjusted ORs (95% CIs) of lung cancer comparing the highest to lowest quartile of total nicotine, total COT, total 3HC, TNE, and total NNAL were 2.49 (1.72-3.62), 3.03 (2.35-5.41), 3.56 (2.35-5.41), 3.35 (2.25-5.00), and 2.43 (1.65-3.57), respectively (all Ps for trend < 0.001). All 4 individual mutant variants of CYP2A6 were associated with reduced 3HC:COT ratio (all Ps for trend < 0.05). When all 4 genotypes were combined to create a CYP2A6 index with 5 groups: G1 (most efficient) to G5 (least efficient), the 3HC:COT ratio for G1 was 4 times that of G5 with a strong dose-dependent genotype-phenotype relation (P trend < 0.001). Smokers with CYP2A6 G5 consumed significantly fewer cigarettes and excreted lower urinary TNE (P trend < 0.001). Compared with CYP2A6 G1, the multivariate-adjusted ORs (95% CIs) of lung cancer for G2, G3, G4 and G5 were 0.95 (0.64-1.42), 1.04 (0.67-1.63), 0.54 (0.34-0.86) and 0.69 (0.43-1.13), respectively, (P trend = 0.005). Furthermore, there was a suggestive interaction between CYP2A6 index and total NNAL on lung cancer risk (P interaction = 0.087). Conclusion. CYP2A6 genetic polymorphisms have a strong influence on cigarette consumption, nicotine uptake and metabolism. In addition to reduced carcinogen exposure, altered metabolism of the tobacco-specific carcinogen NNK may contribute to the reduced risk of lung cancer in smokers with low functional CYP2A6 genotypes. Grants: R01 CA129534, R01 CA144034, UM1 CA182876. Citation Format: Jian-Min Yuan, Heather H. Nelson, Lesley M. Butler, Steven G. Carmella, Renwei Wang, Jennifer Adams-Haduch, Stephen S. Hecht, Woon-Puay Koh, Yu-Tang Gao, Sharon E. Murphy. The impact of CYP2A6 genetic polymorphisms on nicotine metabolism and lung cancer risk in two prospective cohorts of smokers. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 4595. doi:10.1158/1538-7445.AM2015-4595

Background: There is growing evidence that longer telomere length is associated with higher risk of lung cancer. We investigated this association in the prospective Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, which was conducted in the United States. We also combined these data with two previously published prospective cohorts: the Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) trial conducted among ever smoking males in Finland and the Shanghai Women's Health Study (SWHS) which comprised primarily of never-smoking women, resulting in a pooled analysis on a total of 847 cases and 847 controls matched by age, sex and study. Methods: Blood samples were collected prior to diagnosis of lung cancer and telomere length was measured using the same monochrome multiplex quantitative PCR method in all three studies. We used conditional logistic regression models to calculate odds ratios (OR) and their 95% confidence intervals (CI) for the association between telomere length and lung cancer risk adjusted for age, to account for residual confounding, and pack-years of smoking as continuous variables. Analyses by telomere length quartile retaining the initial categorization used in each study, and using categorization based on telomere length in pooled controls, produced similar findings and results are presented for the former. Results: In the PLCO Trial, increasing telomere length was significantly associated with lung cancer risk (adjusted OR [95% CI] by quartile: 1.00; 1.11 [0.65-1.92]; 1.20 [0.66-2.15]; and 1.83 [1.05-3.19]; P-trend = 0.011), consistent with results from the ATBC and SWHS studies. In the pooled analyses, the adjusted OR (95% CI) by quartile was 1.00; 1.24 (0.90-1.72); 1.27 (0.91-1.77); and 1.87 (1.33-2.63); P-trend = 0.000022. This positive association was particularly evident for adenocarcinoma cases, especially those diagnosed more than 6 years after blood collection (n=115; adjusted OR [95% CI] by quartile: 1.00; 2.48 [0.85-7.23]; 2.05 [0.81-5.15]; and 3.59 [1.38-9.34]; P-trend = 0.0027). Conclusion: Telomere length in white blood cell DNA may be an important biomarker of future increased risk of lung cancer in diverse populations. Citation Format: Wei Jie Seow, Richard Cawthon, Mark Purdue, Wei Hu, Yu-Tang Gao, Wen-Yi Huang, Stephanie J. Weinstein, Bu-Tian Ji, Jarmo Virtamo, Dean Hosgood, Bryan Bassig, Xiaoou Shu, Qiuyin Cai, Yongbin Xiang, Shen Min, Wong-Ho Chow, Sonja Berndt, Christopher Kim, Unhee Lim, Demetrius Albanes, Neil E. Caporaso, Stephen Chanock, Wei Zheng, Nathaniel Rothman, Qing Lan. Telomere length in white blood cell DNA and lung cancer: a pooled analysis of three prospective cohorts. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4160. doi:10.1158/1538-7445.AM2014-4160

Background: Catechins from tea and non-tea sources are potential anti-carcinogenic compounds with potent antioxidant properties. Data supporting an effect of specific catechins on the risk of hepatocellular carcinoma (HCC) in humans are sparse. We examined the association between urinary biomarkers of catechins and their metabolites and risk of HCC. Methods: A nested case-control study of HCC (221 cases and 1072 matched controls) was conducted within the Shanghai Cohort Study, a cohort of 18,244 men who were enrolled in 1986-1989. Specific catechins and their metabolites including epicatechin (EC), epigallocatechin (EGC), 4′-O-methyl-epigallocatechin (4′-MeEGC), (-)-5-(3′, 4′, 5′-trihydroxyphenyl)-gamma-valerolactone (M4), and (-)-5-(3′, 4′-dihydroxyphenyl)-gamma-valerolactone (M6) were quantified in urine samples collected from study subjects at enrollment. Logistic regression methods were used to calculate odds ratios (OR) and 95% confidence intervals (CI) with adjustment for established risk factors for HCC, including hepatitis B surface antigen (HBsAg) serology status, self-reported physician-diagnosed cirrhosis, cigarette smoking, and alcohol drinking. Results: Compared with men who had no detectable EC, men with detectable levels of urinary EC had a statistically significant lower risk of HCC (OR=0.64; 95% CI: 0.45, 0.92). The inverse association between EC and HCC risk was not modified by smoking, alcohol intake or HBsAg/cirrhosis status. This inverse EC-HCC association strengthened in analyses restricted to cases that were diagnosed 10 or more years after specimen collection (OR=0.40; 95% CI: 0.22, 0.71). There was no association between urinary levels of EGC or other catechin metabolites and HCC risk. Conclusion: In our study population green tea is the sole source of dietary EGC while the major sources of EC include green tea, as well as apples, black grapes, soybeans, and broad beans. Thus, the findings of the present study support the notion that consumption of fruits and legumes may have a beneficial effect on HCC risk. Experimental studies are warranted to elucidate the biologic mechanisms of EC in hepatocarcinogenesis. Citation Format: Joyce Y. Huang, Lesley M. Butler, Renwei Wang, Chung S. Yang, Jian-Min Yuan, Yu-Tang Gao. Urinary biomarkers of catechins in relation to risk of hepatocellular carcinoma in the Shanghai Cohort Study. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2212. doi:10.1158/1538-7445.AM2014-2212

Introduction: Indoor air pollution (IAP), mostly from coal burning, was responsible for approximately 3.5 million deaths from all-causes and over 100,000 disability-adjusted life years in 2010. The majority of studies conducted in the past have been retrospective, in rural and under developed populations with recent and high levels of exposure. We assessed the association between historic kitchen coal use and various causes of mortality in the prospective Shanghai Women's Health Study cohort. Methods: A cohort of 73,363 women was followed through December 2009 with a combination of in-person surveys every 2-3 years and annual linkage to a vital statistics registry database where all causes of mortality was recorded. A total of 3808 deaths were identified during the follow-up period. Cox proportional hazards models were used to estimate risk of mortality associated with in-home coal use. Models were adjusted for smoking status, family income, environmental tobacco smoke, occupational history, shift work, BMI, hormone therapy, and parity. Results: All-cause mortality was elevated 13% among ever coal users (Hazard Ratio (HR): 1.13; 95% CI: 1.04-1.23) compared to never-coal users. Coal use was most strongly associated with all-cause mortality among women who had more than 15 years of coal use (15-30 years: HR: 1.14; 95% CI: 1.02-1.26; >30 years: HR: 1.15; 95% CI: 1.04-1.26). Women with 15-30 years of coal use had a 21% elevation of cancer mortality (95% CI: 1.04-1.42), but no elevation was observed in women with >30 years of coal use (HR: 1.06; 95% CI: 0.91-1.24) compared to never-coal users. More than 30 years of coal use was associated with a 32% elevated risk of cardiovascular mortality (HR: 1.32; 95% CI: 1.11-1.57), and 62% elevated risk of myocardial infarction mortality (HR: 1.62; 95% CI: 1.01-2.63) compared to never-coal users. All-cause mortality was elevated in women who last lived in a coal burning home up to 20 years ago (>0-10 years: HR: 1.16; 95% CI: 1.04-1.28; >10-20 years: HR: 1.21; 95% CI: 1.09-1.35). Discussion: This is the first study of mortality and in-home coal use in a prospective study after much of Shanghai transitioned into a developed and urban city. Evidence from this study suggests that previous coal use could be related to excess cardiovascular and cancer deaths among women in Shanghai. Citation Format: Christopher Kim, Xiao-Ou Shu, H. Dean Hosgood, Bryan A. Bassig, Wei Jie Seow, Yongbin Xiang, Bu-Tian Ji, Wei Hu, Wong-Ho Chow, Yutang Gao, Nathaniel Rothman, Qing Lan. Past use of coal for cooking is associated with all-cause mortality in the prospective Shanghai Women's Health Study. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2184. doi:10.1158/1538-7445.AM2014-2184

Socioeconomic development and urbanization have resulted in dramatic changes in cancer incidence and mortality in China. As one of the largest and most developed cities in China, Shanghai's cancer rates reflect the changing burden of cancer in China. This study was undertaken to evaluate trends and quantify changes in the incidence and mortality of breast, ovarian, endometrial, and cervical cancers among female residents of urban Shanghai from 1973 to 2009. Data on cancer incidence and mortality were collected by the Shanghai Cancer Registry. International Classification of Disease version 10 codes for breast (C50), cervical (C53), endometrial (corpus uteri and uterus unspecified; C54 and C55), and ovarian cancer (C56) were included in the current analysis. Annual population data was provided by the Department of Vital Statistics, Shanghai Municipal Center for Disease Control and Prevention. A total of 73,755 female cancers and 28,274 related deaths were documented between 1973 and 2009. Age-standardized incidence and mortality rates were calculated using the Segi/Doll 1960 world standard. Joinpoint regression models were used to fit joined straight lines on a log-scale for annual age-standardized rates and to calculate annual percent changes (APC). Between 1973 and 2009, the incidence of breast, ovarian, and endometrial cancers increased, with overall APCs of 2.93, 2.20, and 1.66, respectively. Significant trends were found for breast cancer between 1979 and 1998 (APC 3.07) and for ovarian cancer between 1983 and 2009 (APC 2.83). For endometrial cancer a significant decrease occurred from 1973 to 1983 (APC -4.27), followed by a significant increase from 1983 to 1996 (APC 2.58). The incidence of cervical cancer decreased, with an overall APC of -3.91; this included three significant decreasing trends between 1973 and 1996, followed by a significant increase through 2009 (APC 8.49). The mortality of breast and ovarian cancers increased, with overall APCs of 0.70 and 0.58, respectively. Significant trends were found for breast cancer between 1982 and 2004 (APC 1.55) and for ovarian cancer during the entire time period. The mortality of cervical and endometrial cancers decreased, with overall APCs of -5.36 and -2.33, respectively. For cervical cancer, this included a significant decrease between 1973 and 1999 (APC -7.34), followed by a significant increase through 2009 (APC 3.24). For endometrial cancer, this included a significant decrease between 1973 and 1988 (APC -5.59). In summary, increasing trends were found for breast, ovarian and endometrial cancer incidence, and for breast and ovarian cancer mortality, while decreasing trends were found for cervical cancer incidence and cervical and endometrial cancer mortality among women in urban Shanghai. Additional analyses to identify age-period-cohort effects in this data are underway. This study lays the ground work for the establishment of cancer prevention and control priorities in China. Citation Format: Zhezhou Huang, Alicia C. Beeghly-Fadiel, Ying Zheng, Wanqing Wen, Yutang Gao, Chunxiao Wu, Pingping Bao, Weijian Zhong, Fan Jin, Yongbing Xiang, Wei Zheng, Xiao Ou Shu, Wei Lu. Secular trends in incidence and mortality of female cancers in Shanghai, China (1973-2009). [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4140. doi:10.1158/1538-7445.AM2014-4140

The effect of genetic polymorphisms in the TGF-beta1 gene at codon 10 (T+29C), codon 25 (G+74C), and the promoter region [C --T at -509 from the transcription site, (C-509T)] on breast cancer survival was evaluated among a cohort of 1111 patients. The median follow-up time for the cohort was 5.17 years after cancer diagnosis. No DNA sequence variation at codon 25 of the TGF-beta1 gene was found, whereas polymorphisms in C-509T and T+29C were in strong linkage disequilibrium. Patients who carried the C allele of T+29C polymorphism had a reduced 5-year disease-free survival rate (75.6% for T/C, and 78.2% for C/C) compared with the T/T genotype (85.1%; P, 0.04); the age-adjusted hazard ratio was 1.5 (95% confidence interval, 1.1-2.2). Adjustment for clinical prognostic factors slightly attenuated the association (hazard ratio, 1.4, 95% confidence interval, 1.0-1.9). Our study suggests that genetic polymorphisms in the TGF-beta1 gene may play a role in breast cancer progression.

Long-term estrogen exposure and family history of breast cancer are the two factors that are most consistently found to be associated with breast cancer risk. Sequence variants in genes involved in estrogen synthesis, metabolism, and signal transduction may account, in part, for this observation. Using data and DNA samples from the Shanghai Breast Cancer Study, we tested the hypothesis that sequence variants of the estrogen receptor beta gene (ESR2) may be associated with increased risk for breast cancer, particularly among women who have a high level and long-term endogenous estrogen exposure. Direct sequencing of the ESR2 gene among 30 Chinese women revealed eight sequence variants. Association analysis of six common sequence variants in 1134 cases and 1235 controls provided evidence for positive associations between breast cancer risk and two single nucleotide polymorphisms (SNPs), [C(14206)T and C(33390)G], among postmenopausal women. Evidence of a stronger association was found for SNP [C(33390)G] among women with a long duration (or =34 years) of menstruation (odds ratio, 2.37; 95% confidence interval, 1.18-4.77). A potential synergistic effect between SNP [C(33390)G] and several steroid sex hormones was observed, and a 3-4-fold elevated risk of breast cancer was found among women with a CG or GG genotype in SNP [C(33390)G] combined with a high level of steroid sex hormone or a low level of sex hormone binding globulin. Our results are consistent with the hypothesis of a joint effect of estrogen receptor beta sequence variants and endogenous estrogen exposure on breast cancer risk.

Recent studies suggest that genetic polymorphisms of the estrogen receptor-alpha (ER-alpha) gene may be associated with breast cancer risk. To evaluate the role of this gene in the risk of breast cancer, we genotyped a newly identified GT dinucleotide repeat [(GT)(n)] polymorphism located in the promoter region (6.6 kb upstream of the transcription start site) in 947 breast cancer cases and 993 age frequency-matched community controls from a population-based case-control study conducted among Chinese in urban Shanghai. Sixteen alleles were identified, the most common one having 16 GT repeats [(GT)(16)]. Compared with subjects homozygous for this allele, subjects carrying the (GT)(17) or (GT)(18) allele had a decreased risk of breast cancer. The odds ratios (ORs) were 0.81 [95% confidence interval (CI), 0.62-1.06] and 0.58 (95% CI, 0.36-0.94), respectively, for one and two copies of the (GT)(17) or (GT)(18) allele. The inverse association with carrying either of these alleles was stronger among women with30 years of menstrual cycles (OR 0.66; 95% CI 0.51-0.85) than those with a shorter duration of menstrual cycles (OR 0.97; 95% CI 0.73-1.27), and the test for an interaction was statistically significant (P = 0.04). Among breast cancer patients, the presence of either the (GT)(17) or (GT)(18) allele was associated with a reduced expression of progesterone receptor. Results of this study indicate that the GT dinucleotide repeat polymorphism in ER-alpha gene promoter region may be a new biomarker for genetic susceptibility to breast cancer.

Brassica vegetable consumption (e.g., Chinese cabbage) provides isothiocyanates (ITC) and other glucosinolate derivatives capable of inducing Phase II enzymes [e.g., glutathione S-transferases (GSTM1, GSTT1, and GSTP1) and NADPH quinine oxidoreductase] and apoptosis, altering steroid hormone metabolism, regulating estrogen receptor response, and stabilizing cellular proliferation. Asian populations consuming large amounts of Brassica have a lower breast cancer incidence compared with Western populations; however, the association between Brassica consumption and breast cancer risk is uncertain. It is difficult to estimate glucosinolate exposure and degradation in humans, possibly limiting epidemiological investigations of Brassica and cancer associations. We conducted a case control investigation of breast cancer in Shanghai, China, using urinary ITC levels as a biological measure of glucosinolate intake and degradation in populations with habitual Brassica intake. A representative subgroup of 337 cases providing presurgery, fasting, and first-morning urine specimens was one-to-one matched (age, menopausal status, date of urine collection, and day of laboratory assay) to population controls. Urinary ITC levels were inversely associated with breast cancer [odds ratio (OR) (Quartile 1) = 1 (ref); OR(Q2) = 0.9, 95% confidence interval (0.6, 1.4); OR(Q3) = 0.7, (0.5, 1.1); OR(Q4) = 0.5, (0.3, 0.8), adjusted for age, menopausal status, soy protein, fibroadenoma history, family breast cancer, physical activity, waist-to-hip ratio, body mass index, age at menarche, and parity in conditional logistic model]. This protective association persisted within post and premenopausal women. In contrast, total Brassica intake estimated from a food frequency questionnaire was not associated with breast cancer. Trends in the association between urinary ITC and breast cancer were more consistent with homozygous deletion of GSTM1 or GSTT1, the AAgenotype of GSTP1 (A313G), or with the C allele of NADPH quinine oxidoreductase (C609T), although interactions were not statistically significant. In conclusion, greater Brassica vegetable consumption, as measured by the urinary ITC biomarker, was associated with significantly reduced breast cancer risk among Chinese women.

Lung cancer incidence has been on the rise in less-developed countries. Indoor air pollution caused by poor ventilation, burning of solid fuels, and cooking oils has been associated with lung cancer risk in developing poor and rural countries in many retrospective case-control studies. However, few studies have been conducted in more urban and more economically developed populations, even less in prospective studies. To assess the relationship of indoor air quality and risk of lung cancer in a modern population, a prospective cohort study was conducted in the Shanghai Women's Health Study. 73,442 women were followed through December 2009. A total of 469 lung cancer patients were diagnosed during the follow-up period, but after excluding ever smokers, 71,320 women in the cohort and 429 women with lung cancer were included in this analysis. Standardized and structured questionnaires were used to collect information on all household living and cooking practices: past three residences lived, cooking fuel utilized, cooking oil utilized, ventilation conditions, smoking patterns of family at home, and coworker smoking at work. Effect estimates for household ventilation conditions, cooking fuels, and cooking oils use for the risk of lung cancer are presented as hazard ratios (HR) with 95% confidence intervals (95% CI) from Cox proportional hazards regression. Ever having poor household ventilation was associated with a 61% increase in lung cancer risk (HR: 1.61; 95% CI: 1.19-2.18), and more than 20 years was associated with a 100% increase in risk (HR: 2.00; 95% CI: 1.32-3.01) compared to always having good ventilation. Ever use of coal with poor ventilation was associated with a 71% increased risk of lung cancer (HR: 1.71; 95% CI: 1.19-2.46), and more than 20 years of using coal with poor ventilation was associated with an 83% increase (HR: 1.83; 95% CI: 1.16-2.89) compared to gas users with good ventilation. There was some suggestion of lung cancer risk with vegetable oil use compared to soybean oil use, although the number of lung cancer cases was few and not statistically stable. In summary, this prospective cohort study of non-smoking Chinese women suggests that exposure to poor ventilation and poorly vented cooking coal increases the risk of lung cancer. These findings were consistent with past retrospective case-control studies and suggesting that past exposure to coal use with poorly ventilated conditions pose a health risk. These results suggest that current modern day cooking conditions, particularly in poorly ventilated homes, is still a public health issue. Citation Format: Christopher Kim, Yutang Gao, Yongbing Xiang, Francesco Barone-Adesi, Yawei Zhang, H. Dean Hosgood, Shuangge Ma, Xiao-ou Shu, Bu-Tian Ji, Wong-Ho Chow, Qiuyin Cai, Wei Zheng, Nathaniel Rothman, Qing Lan. Household ventilation, cooking fuels and oils, and lung cancer risk in a prospective cohort of non-smoking Chinese women. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3629. doi:10.1158/1538-7445.AM2013-3629

microRNA (miRNA) are small noncoding molecules that function as regulators of the genome. Genetic variants in miRNA binding sites and miRNA genes have begun to be evaluated in relation to disease risk; however, no studies have yet comprehensively evaluated miRNA variants in relation to breast cancer risk. Using a two stage study design, and genome-wide association study (GWAS) data, we evaluated genetic variants within 1,425 miRNA genes (±5kb). In Stage 1, a total of 2,213 common genetic variants were evaluated among 2,876 breast cancer cases and 2,285 controls using data from the Shanghai Breast Cancer Genetics Study. Nominally significant associations (P value ≤0.05) were found for 108 variants in 96 genetic loci (r²≥0.6). In Stage 2, 94 variants in 84 loci were evaluated among 2,225 cases and 2,052 controls from the Seoul Breast Cancer Study; nine variants in eight loci had nominally significant associations. In analyses of the two study stages combined, 14 loci had highly significant associations (P value ≤0.01) and 19 additional loci had significant associations (P value ≤0.05) with breast cancer risk, all with consistent directions of association between the two study stages. These 36 variants in 33 loci have been mapped to 16 miRNA genes, many in the promoter regions. Based on our findings, genetic variants in miRNA genes are likely to contribute to breast cancer susceptibility. Citation Format: Alicia C. Beeghly-Fadiel, Hyuna Sung, Ben Zhang, Ji-Yeob Choi, Ryan J. Delahanty, Jirong Long, Yu-Tang Gao, Wei Lu, Qiuyin Cai, Xiao Ou Shu, Daehee King, Wei Zheng. Genetic variants in microRNA genes and breast cancer risk. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2557. doi:10.1158/1538-7445.AM2013-2557

Pre-diagnostic levels of serum one-carbon metabolism biomarkers and risk of hepatocellular carcinoma BACKGROUND. It is well established that rats fed diets deficient in choline and methionine spontaneously develop hepatocellular carcinoma (HCC). Tumor DNA from these animals show a characteristic hypomethylation pattern, suggesting that disruption of the one-carbon metabolism (1CM) pathway may be an underlying mechanism for hepatocarcinogenesis. Prospective epidemiological data on circulating levels of 1CM biomarkers in relation to HCC risk have been lacking. METHODS. We prospectively examined the association between prediagnostic serum concentrations of betain, choline, cystathionine, homocysteine (Hcy), methionine, 5-methyltetrahydrofolate (5-MTHF), pyridoxal-5-phosphate (PLP, the bioactive form of vitamin B6), S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH), and risk of developing HCC among 297 incident HCC cases and 631 matched controls from a cohort of 18,244 men in Shanghai, China. Conditional logistic regression method was used to calculate odds ratios (OR) and 95% confidence intervals (CI) adjusted for smoking, alcohol use, hepatitis B surface antigen (HBsAg) seropositivity and chronic liver disease. RESULTS. Higher serum levels of PLP were associated with statistically significantly reduced risk of HCC. The ORs (95% CIs) for the 2nd, 3rd, and 4th quartiles of serum PLP were 0.52 (0.32, 0.85), 0.57 (0.35, 0.93) and 0.35 (0.20, 0.60), respectively, compared with the 1st quartile (Ptrend CONCLUSION. Choline and PLP may be protective against the development of HCC whereas circulating methionine may be a risk factor for developing HCC, especially among high-risk individuals. Future investigation is warranted to test whether diets high in vitamin B6 are effective in preventing HCC development in high-risk populations. Citation Format: Lesley M. Butler, Renwei Wang, Teodoro Bottiglieri, Yu-Tang Gao, Jian-Min Yuan. Pre-diagnostic levels of serum one-carbon metabolism biomarkers and risk of hepatocellular carcinoma. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 107. doi:10.1158/1538-7445.AM2013-107

Background: In the United States, 10-15% of lung cancer victims are lifelong never smokers. Non-smoking related lung cancer alone ranks as the 6th most frequently diagnosed cancer. The etiological factors for non-smoking lung cancer are unclear. Although strong evidence supports a major role for polycyclic aromatic hydrocarbon (PAH) as causes of lung cancer in smokers, epidemiological studies prospectively investigating biomarkers of PAH and volatile organic carcinogens in relation to lung cancer risk among never smokers are lacking. Methods: A nested case-control study of lung cancer among never smokers was conducted within the Shanghai Cohort Study, a prospective cohort of 18,244 Chinese men aged 45-64 years at enrollment during 1986-1989. After more than 20 years of follow-up, we identified 96 incident lung cancer cases among never smokers. For each case, we randomly chose one control among cohort participants who were never smokers. The control was matched to the index case by age (±2 years), date of urine collection (±1 month) and neighborhood of residence. We quantified urinary free cotinine, 3-hydroxyphenanthrene (3-OH-Phe), total hydroxyphenanthrenes (the sum of 1-, 2-, 3-, and 4-OH-Phe) (total OH-Phe), r-1,t-2,3,c-4-tetrahdroxy-1,2,3,4-tetrahydrophenanthrene (PheT), mercapturic acid metabolites of acrolein [3-hydroxypropyl mercapturic acid (HPMA)], benzene [S-phenyl mercapturic acid (SPMA)], crotonaldehyde [4-hydroxybut-2-yl mercapturic acid (HBMA)], and ethylene oxide [2-hydroxyethyl mercapturic acid (HEMA)]. Self-reported never smokers whose urinary free cotinine was greater than 20 ng/ml were excluded from the present analysis for their potential use of tobacco. The odds ratio (OR) and 95% confidence interval (CI) for lung cancer for each urinary biomarker was estimated using an unconditional logistic regression method. Results: Among never smokers, urinary levels of all 3 measured PAH biomarkers were statistically significantly higher in cases than controls. The geometric means (pmol/mg creatinine) of total OH-Phe, 3-OH-Phe and PheT were 19.10 (in cases) vs 16.08 (in controls) (P = 0.017); 6.64 vs 5.44 (P = 0.011); and 19.84 vs 16.10 (P = 0.035), respectively. Compared with the lowest quartile, ORs (95% CIs) for lung cancer for the highest quartile were 2.59 (1.01-6.65) for total OH-Phe, 3.10 (1.12-7.75) for 3-OH-Phe, and 2.98 (1.13-7.87) for PheT (all Ptrend Conclusion: This is the first prospective study that demonstrates a positive association between urinary biomarkers of PAH and lung cancer risk among never smokers. Given its ubiquitous presence in the environment, PAH could account for large proportion of non-smoking related lung cancer. 1 Citation Format: Jian-Min Yuan, Yu-Tang Gao, Sharon Murphy, Steven G. Carmella, Renwei Wang, Heather H. Nelson, Stephen S. Hecht. Urinary biomarkers of polycyclic aromatic hydrocarbon and volatile organic carcinogens in relation to lung cancer development in lifelong never smokers - The Shanghai Cohort Study. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1502. doi:10.1158/1538-7445.AM2013-1502

Cumulative evidence from both in vitro and animal studies suggests that cyclooxygenase-2 (COX-2) may be involved in the development and progression of lung cancer. Overproduction of prostaglandin E synthase and prostaglandin E2 (PGE2), one of the end products of the COX-2 pathway, has been implicated in the pathogenesis of non-small cell lung cancer. On the other hand, evidence suggests that prostacyclin (PGI2), another end products of the COX-2 pathway, may differ from other bioactive prostaglandins and may exert cancer-inhibitory effects on the development of lung cancer. PGE2 is quickly converted to 11α-hydroxy-9,15-dioxo-2,3,4,5-tetranor-prostane-1,20-dioic acid (PGE-M), a major PGE2 metabolite and excreted in urine. PGI2 is rapidly converted into 6-keto-PGF1a, a chemically stable but biologically inactive hydration product. Circulatory 6-keto-PGF1a undergoes ß-oxidation and appears in the urine as 2,3-dinor-6-keto-PGF1a (PGI-M), the primary breakdown product of PGI2. Urinary PGE-M and PGI-M are stable compounds and can be measured accurately and reliably with newly developed and validated LC-MS and GC-MS assays. In this study, we evaluated the association of pre-diagnostic urinary PGE-M and PGI-M levels with subsequent risk of developing lung cancer. We conducted a nested case-control study using resources of the Shanghai Men's Health Study, a prospective cohort study, in which 61,491 Chinese men aged 40-74 years were recruited from 2002 to 2006 and followed for up to 7 years. Urinary PGE-M and PGI-M in 349 incident lung cancer patients were compared to 349 matched controls. Overall, the geometric mean of baseline level of urinary PGE-M did not differ significantly between cases and controls. Compared with the lowest quartile, the risks (odds ratios, ORs) of developing lung cancer were non-significantly elevated (1.48, 95% CI=0.91-2.40 for the highest quartile). No dose-response relationship was evident. The baseline level of urinary PGI-M was more than 20% higher in cases than in controls. The ORs for developing lung cancer increased from 1.0 to 1.14 (95% CI=0.66-1.95), 1.45 (95% CI=0.86-2.45), and 1.70 (95% CI=0.99-2.92) across the quartiles of urinary PGI-M levels (P for trend=0.0056). Excluding 63 case-control pairs who were diagnosed with lung cancer within one year of follow-up did not change the associations for PGE-M and PGI-M. No interaction was found between PGE-M and PGI-M in relation to lung cancer risk. In summary, using data from a large prospective cohort study, we found that high levels of urinary PGI-M, and possibly PGE-M, were associated with subsequent risk of developing lung cancer. Our data suggest that high activity of the COX-2 pathway, rather than the prostaglandin E2 alone, may play a major role in lung cancer development. Further studies are needed to evaluate these two urinary markers as possible biomarkers for lung cancer risk assessment. Citation Format: Qiuyin Cai, Yong-Bing Xiang, Wanqing Wen, Ginger L. Milne, Hong-Lan Li, Jie Wu, Gong Yang, Jing Gao, Tsogzolmaa Dorjgochoo, Yu-Tang Gao, Wei Zheng, Xiao Ou Shu. Prospective study of urinary prostaglandin E2 and prostacyclin metabolites and lung cancer risk . [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 122. doi:10.1158/1538-7445.AM2013-122

Introduction Dietary glycemic index (GI) and glycemic load (GL) have been observed to have a positive relationship with obesity and diabetes which are risk factors for liver cancer. However, studies assessing the association between GI, GL and liver cancer have yielded inconsistent results. We evaluated the association of GI and GL with liver cancer using data from two prospective cohort studies, the Shanghai Women's Health Study (SWHS) and the Shanghai Men's Health Study (SMHS). Methods The study population consisted of 72,906 women from SWHS and 60,203 men from the SMHS who had not been previously diagnosed with cancer, did not have extreme total energy intake as assessed by a food frequency questionnaire (FFQ) and had complete data on all covariates of interest. Subject recruitment for the SWHS began in 1996 and for the SMHS began in 2002. Outcome follow-up data through December 31, 2009 from both cohorts were used in the current analysis. At baseline, a validated FFQ was administered and daily dietary GI and GL were calculated for each cohort member. These values were energy adjusted using the residuals method and categorized into quintiles. Confounders were assessed using data from the baseline questionnaire. Hazard ratios and 95% confidence intervals were calculated for the association between GI/GL and liver cancer incidence with adjustment for potential confounders. Results After a median follow-up of 11.2 and 5.3 years for the SWHS and SMHS, respectively, 346 incident liver cancer (ICD-9: 155) were identified. Women in the upper quintiles of GL intake were more likely to be older, have less education, have lower annual family incomes, be smokers, be post-menopausal, and were less likely to exercise, consume alcohol or have a history of diabetes or hepatitis. Men in the upper quintiles of GL intake were more likely to have less education, have lower annual per capita family incomes, and were less likely to exercise, to consume alcohol or have a history of diabetes. After adjustment for all confounders, no statistically significant trends were observed for men or women for the quintile of intake of GI or GL with the incidence of liver cancer (SWHS: GL p trend = 0.67, GI p trend = 0.11; SMHS: GL p trend = 0.64, GI p trend = 0.29). Similarly, no statistically significant trends were observed when the models were stratified by chronic hepatitis status (yes or no). Conclusions There is little evidence that GI or GL affect the incidence of liver cancer in this population. Further research about other modifiable risk factors for liver cancer is warranted. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 626. doi:1538-7445.AM2012-626