Faculty Bibliography

Mean residual life (MRL) is the remaining life expectancy of a subject who has survived to a certain time point and can be used as an alternative to hazard function for characterizing the distribution of a time-to-event variable. Inference and application of MRL models have primarily focused on full-cohort studies. In practice, case-cohort and nested case-control designs have been commonly used within large cohorts that have long follow-up and study rare diseases, particularly when studying costly molecular biomarkers. They enable prospective inference as the full-cohort design with significant cost-saving benefits. In this paper, we study the modeling and inference of a family of generalized MRL models under case-cohort and nested case-control designs. Built upon the idea of inverse selection probability, the weighted estimating equations are constructed to estimate regression parameters and baseline MRL function. Asymptotic properties of the proposed estimators are established and finite-sample performance is evaluated by extensive numerical simulations. An application to the New York University Women's Health Study is presented to illustrate the proposed models and demonstrate a model diagnostic method to guide practical implementation.

BACKGROUND:Epithelial ovarian, fallopian tube, and primary peritoneal cancer have shared developmental pathways. Few studies have prospectively examined heterogeneity in risk factor associations across these three anatomic sites. METHODS:We identified 3,738 ovarian, 337 peritoneal, and 176 fallopian tube incident cancer cases in 891,731 women from 15 prospective cohorts in the Ovarian Cancer Cohort Consortium. Associations between 18 putative risk factors and risk of ovarian, peritoneal, and fallopian tube cancer, overall and for serous and high-grade serous tumors, were evaluated using competing risks Cox proportional hazards regression. Heterogeneity was assessed by likelihood ratio tests. RESULTS:Most associations did not vary by tumor site (phet≥0.05). Associations between first pregnancy (phet=0.04), tubal ligation (phet=0.01) and early-adult (age 18-21 years) body mass index (BMI) (phet=0.02) and risk differed between ovarian and peritoneal cancers. The association between early adult BMI and risk further differed between peritoneal and fallopian tube cancer (phet=0.03). First pregnancy and tubal ligation were inversely associated with ovarian, but not peritoneal, cancer. Higher early-adult BMI was associated with higher risk of peritoneal, but not ovarian or fallopian tube, cancer. Patterns were generally similar when restricted to serous and high-grade serous cases. CONCLUSIONS:Ovarian, fallopian tube, and primary peritoneal cancers appear to have both shared and distinct etiologic pathways, although most risk factors appear to have similar associations by anatomic site. IMPACT/CONCLUSIONS:Further studies on the mechanisms underlying the differences in risk profiles may provide insights regarding the developmental origins of tumors arising in the peritoneal cavity and inform prevention efforts.

BACKGROUND:Prevalence of Helicobacter pylori infection, the main risk factor for gastric cancer, has been decreasing in the US; however, there remains a substantial racial disparity. Moreover, the time-trends for prevalence of CagA-positive H. pylori infection, the most virulent form, are unknown in the US population. We sought to assess prevalence of CagA-positive Helicobacter pylori infection over time by race, in the US. METHODS:We utilized multiplex serology to quantify antibody responses to H. pylori antigens in 4,476 participants across 5 cohorts that sampled adults from 1985 to 2009. Using log-binomial regression models, we calculated prevalence ratios (PRs) and 95% confidence intervals (CIs) for the association between H. pylori-CagA sero-prevalence and birth year by race. RESULTS:African Americans were 3-times more likely to be H. pylori-CagA sero-positive than whites. After adjustment, H. pylori-CagA sero-prevalence was lower with increasing birth year among whites (Ptrend=0.001), but remained stable for African Americans. When stratified by sex and education separately, the decline in H. pylori-CagA sero-positivity among whites remained only for females (Ptrend<0.001) and was independent of educational attainment. Among African Americans, there was no difference by sex; further, sero-prevalence increased with increasing birth year among those with a high-school education or less (P=0.006). CONCLUSIONS:Among individuals in the US born from the 1920s to 1960s, H. pylori-CagA sero-prevalence has declined among whites, but not among African Americans. IMPACT/CONCLUSIONS:Our findings suggest a widening racial disparity in the prevalence of the most virulent form of Helicobacter pylori, the main cause of gastric cancer.

Registry-based epidemiologic studies suggest associations between chronic inflammatory intestinal diseases and pancreatic ductal adenocarcinoma (PDAC). As genetic susceptibility contributes to a large proportion of chronic inflammatory intestinal diseases, we hypothesize that the genomic regions surrounding established genome-wide associated variants for these chronic inflammatory diseases are associated with PDAC. We examined the association between PDAC and genomic regions (+/- 500 kb) surrounding established common susceptibility variants for ulcerative colitis, Crohn's disease, inflammatory bowel disease, celiac disease, chronic pancreatitis, and primary sclerosing cholangitis. We analyzed summary statistics from genome-wide association studies data for 8,384 cases and 11,955 controls of European descent from two large consortium studies using the summary data-based adaptive rank truncated product method to examine the overall association of combined genomic regions for each inflammatory disease group. Combined genomic susceptibility regions for ulcerative colitis, Crohn's disease, inflammatory bowel disease, and chronic pancreatitis were associated with PDAC at P-values < 0.05 (0.0040, 0.0057, 0.011, and 3.4 × 10-6, respectively). After excluding the 20 PDAC susceptibility regions (+/- 500 kb) previously identified by GWAS, the genomic regions for ulcerative colitis, Crohn's disease, and inflammatory bowel disease remained associated with PDAC (P-values = 0.0029, 0.0057, and 0.0098, respectively). Genomic regions for celiac disease (P-value = 0.22) and primary sclerosing cholangitis (P-value = 0.078) were not associated with PDAC. Our results support the hypothesis that genomic regions surrounding variants associated with inflammatory intestinal diseases, particularly, ulcerative colitis, Crohn's disease, inflammatory bowel disease, and chronic pancreatitis are associated with PDAC.

BACKGROUND:Excess body weight is an established cause of postmenopausal breast cancer, but it is unknown if weight loss reduces risk. METHODS:Associations between weight change and risk of breast cancer were examined among women aged ≥50 years in the Pooling Project of Prospective Studies of Diet and Cancer. In 10 cohorts, weight assessed on three surveys was used to examine weight change patterns over approximately 10 years (Interval 1 median= 5.2 years; Interval 2 median = 4.0 years). Sustained weight loss was defined as ≥ 2kg lost in Interval 1 that was not regained in Interval 2. Among 180,885 women, 6,930 invasive breast cancers were identified during follow-up. RESULTS:Compared with women with stable weight (± 2kg), women with sustained weight loss had a lower risk of breast cancer. This risk reduction was linear and specific to women not using postmenopausal hormones (>2-4.5kg lost: Hazard Ratio (HR)= 0.82, 95% confidence interval (CI): 0.70-0.96; >4.5-<9kg lost: HR = 0.75, 95% CI: 0.63-0.90; ≥9kg lost: HR = 0.68, 95% CI: 0.50-0.93). Women who lost ≥9kg and gained some (but not all) of it back were also at a lower risk of breast cancer. Other patterns of weight loss and gain over the two intervals had a similar risk of breast cancer to women with stable weight. CONCLUSIONS:These results suggest that sustained weight loss, even modest amounts, is associated with lower breast cancer risk for women aged ≥50 years. Breast cancer prevention may be a strong weight loss motivator for the two-thirds of American women who are overweight or obese.

Early-adulthood body size is strongly inversely associated with risk of premenopausal breast cancer. It is unclear whether subsequent changes in weight affect risk. We pooled individual-level data from 17 prospective studies to investigate the association of weight change with premenopausal breast cancer risk, considering strata of initial weight, timing of weight change, other breast cancer risk factors, and breast cancer subtype. Hazard ratios (HR) and 95% confidence intervals (CI) were obtained using Cox regression. Among 628,463 women, 10,886 were diagnosed with breast cancer before menopause. Models adjusted for initial weight at ages 18-24 years and other breast cancer risk factors showed that weight gain from ages 18-24 to 35-44 or to 45-54 years was inversely associated with breast cancer overall (e.g. HR per 5kg to ages 45-54: 0.96, 95% CI: 0.95-0.98) and with oestrogen-receptor(ER)-positive breast cancer (HR per 5kg to ages 45-54: 0.96, 95% CI: 0.94-0.98). Weight gain from ages 25-34 was inversely associated with ER-positive breast cancer only and weight gain from ages 35-44 was not associated with risk. None of these weight gains were associated with ER-negative breast cancer. Weight loss was not consistently associated with overall or ER-specific risk after adjusting for initial weight. Weight increase from early-adulthood to ages 45-54 years is associated with a reduced premenopausal breast cancer risk independently of early-adulthood weight. Biological explanations are needed to account for these two separate factors. This article is protected by copyright. All rights reserved.

BACKGROUND/OBJECTIVES/OBJECTIVE:Folates found in natural foods are thought to protect against cancer. However, folic acid (FA), a synthetic form of folate used in supplements and fortified foods, may increase breast cancer risk if present in unmetabolized form (UMFA) in the circulation. This study examined the associations of serum UMFA and 5-methyltetrahydrofolate (5-mTHF), the predominant form of circulating folate, with breast cancer risk. SUBJECTS/METHODS/METHODS:We conducted a nested case-control study in a prospective cohort. In total, 553 cases of invasive breast cancer, diagnosed before mandatory FA fortification of grain in the US in 1998, were individually-matched to 1059 controls. Serum UMFA and 5-mTHF were measured using liquid chromatography-tandem mass spectrometry in stored serum samples, and 5-mTHF was corrected for storage degradation. RESULTS: = 0.08). CONCLUSIONS:Circulating UMFA was not associated with breast cancer risk. These results apply to countries without mandatory FA food fortification. Studies are needed in countries with mandatory fortification, where levels of UMFA are much higher than in our study.

BACKGROUND:Obesity and diabetes are major modifiable risk factors for pancreatic cancer. Interactions between genetic variants and diabetes/obesity have not previously been comprehensively investigated in pancreatic cancer at the genome-wide level. METHODS:We conducted a gene-environment interaction (GxE) analysis including 8,255 cases and 11,900 controls from four pancreatic cancer GWAS datasets (PanScan I-III and PanC4). Obesity (BMI=30 kg/m2) and diabetes (duration = 3 years) were the environmental variables of interest. Approximately 870,000 SNPs were analyzed. Case-control (CC), case-only (CO), and joint-effect test methods were used for SNP-level GxE analysis. As a complementary approach, gene-based GxE analysis was also performed. Age, sex, study site and principal components accounting for population substructure were included as covariates. Meta-analysis was applied to combine individual-GWAS summary statistics. RESULTS:No genome-wide significant interactions with diabetes or obesity were detected at the SNP level by the CC or CO approaches. The joint-effect test detected numerous genome-wide significant GxE signals in the GWAS main effects top hit regions but the significance diminished after adjusting for the GWAS top hits. In the gene-based analysis, a significant interaction of diabetes with variants in the FAM63A (family with sequence similarity 63 member A) gene (significance threshold P<1.25E-6) was observed in the meta-analysis (PGxE= 1.2E-6, PJoint= 4.2E-7). CONCLUSIONS:Our current analyses did not find significant GxE interactions at the SNP level but found one significant interaction with diabetes at the gene level. A larger sample size might unveil additional genetic factors via GxE scans. IMPACT/CONCLUSIONS:This study may contribute to discovering the mechanism of diabetes-associated pancreatic cancer.

Obesity is known to be associated with primary liver cancer (PLC), but the separate effects of excess abdominal and gluteofemoral size are unclear. Thus, we examined the association between waist and hip circumference with risk of PLC overall and by histologic type - hepatocellular carcinoma (HCC) and intrahepatic cholangiocarcinoma (ICC). The Liver Cancer Pooling Project is a consortium of prospective cohort studies that includes data from 1,167,244 individuals (PLC n=2,208, HCC n=1,154, ICC n=335). Multivariable-adjusted hazards ratios (HRs) and 95% confidence intervals (CI) were estimated using proportional hazards regression. Waist circumference, per 5 cm increase, was associated with an 11% increased PLC risk (HR=1.11, 95%CI: 1.09-1.14), including when adjusted for hip circumference (HR=1.12, 95%CI: 1.08-1.17) and also when restricted to individuals in a normal body mass index (BMI) range (18.5-<25 kg/m² ; HR=1.14, 95%CI: 1.07-1.21). Hip circumference, per 5 cm increase, was associated with a 9% increased PLC risk (HR=1.09, 95%CI: 1.06-1.12), but no association remained after adjustment for waist circumference (HR=0.99, 95%CI: 0.94-1.03). HCC and ICC results were similar. These findings suggest that excess abdominal size is associated with an increased risk of liver cancer, even among individuals considered to have a normal BMI. However, excess gluteofemoral size alone confers no increased risk. Our findings extend prior analyses, which found an association between excess adiposity and risk of liver cancer, by disentangling the separate effects of excess abdominal and gluteofemoral size through utilization of both waist and hip circumference measurements. This article is protected by copyright. All rights reserved.

BACKGROUND:In almost all countries, incidence rates of liver cancer are 100-200% higher in males than in females. However, this difference is predominantly driven by hepatocellular carcinoma (HCC), which accounts for 75% of liver cancer cases. Intrahepatic cholangiocarcinoma (ICC) accounts for 12% of cases and has rates only 30% higher in males. Hormones are hypothesized to underlie observed sex differences. We investigated whether prediagnostic circulating hormone and sex hormone binding globulin (SHBG) levels were associated with liver cancer risk, overall and by histology, by leveraging resources from five prospective cohorts. METHODS:hormone value (approximate doubling of circulating concentration) and liver cancer were calculated using multivariable-adjusted conditional logistic regression. RESULTS:A doubling in the concentration of 4-androstenedione was associated with a 50% decreased liver cancer risk (OR=0.50,95%CI=0.30-0.82), while SHBG was associated with a 31% increased risk (OR=1.31,95%CI=1.05-1.63). Examining histology, a doubling of estradiol was associated with a 40% increased risk of ICC (OR=1.40,95%CI=1.05-1.89), but not HCC (OR=1.12,95%CI=0.81-1.54). CONCLUSIONS:This study provides the first evidence that higher levels of 4-androstenedione may be associated with lower, and SHBG with higher, liver cancer risk in women. However, this study does not support the hypothesis that higher estrogen levels decrease liver cancer risk. Indeed, estradiol may be associated with an increased ICC risk.