Although the abundant stromal reaction associated with pancreatic cancer has long been appreciated, the functions of the CAF cells that establish this stromal reaction remain unclear. An improved understanding of the transcriptional and functional heterogeneity of pancreatic CAFs, as well as their tumor-supportive versus tumor-suppressive capacity, may facilitate the development of effective therapies for this disease.

We conducted a systematic literature search in PubMed and Scopus to January 2020. Prospective cohort studies reporting the association between objectively-measured LPA using activity monitors (e.g., accelerometers) and risk of cancer mortality in the general population were included. The summary hazard ratios (HR) per 30 min/day of LPA and 95% confidence intervals (CI) were obtained using a random-effects model. Dose–response analysis was used to plot their relationship.

Five prospective cohort studies were included, in which the definition of LPA based on accelerometer readings was mainly set within 100 to 2,100 counts/min. The summary HR for cancer mortality per 30 min/day of LPA was 0.86 (95% CI, 0.79–0.95; I² < 1%), and the association between LPA and risk reduction in cancer mortality was linearly shaped (P_nonlinearity = 0.72). LPA exhibited a comparable magnitude of risk reduction in cancer mortality of moderate-to-vigorous physical activity regardless of equal time-length (0.87 per 30 min/day vs. 0.94 per 30 min/day, P_interaction = 0.46) or equal amount (0.74 vs. 0.94 per 150 metabolic equivalents-min/day, P_interaction = 0.11). Furthermore, replacing sedentary time by LPA of 30 min/day decreased the risk of cancer mortality by 9%.

Objectively-measured LPA conferred benefits in decreasing the risk of cancer mortality.

LPA should be considered in physical activity guidelines to decrease the risk of cancer mortality.

We used Cox proportional hazards regression models to estimate the associations of serum levels of estradiol (premenopausal women only), testosterone, and/or SHBG with DCIS risk among 182,935 women. After a median follow-up of 7.1 years, 186 and 531 DCIS cases were ascertained in premenopausal and postmenopausal women, respectively.

Total and free estradiol were positively associated with risk of DCIS among premenopausal women. The HRs for the highest versus the lowest tertiles were 1.54 (1.06–2.23) and 1.72 [95% confidence interval (CI), 1.15–2.57], respectively. Among postmenopausal women, elevated levels of free testosterone (FT), and to a lesser extent, total testosterone, were positively associated with DCIS risk. The HRs for the highest versus the lowest quartiles were 1.42 (95% CI, 1.09–1.85) and 1.16 (95% CI, 0.91–1.48), respectively. Serum SHBG levels were inversely associated with risk of DCIS among postmenopausal women (HR_{q4 vs. q1}: 0.75; 95% CI, 0.56–0.99).

This study suggests that elevated levels of estradiol are associated with increased risk of DCIS among premenopausal women, and that among postmenopausal women, elevated levels of testosterone, and particularly those of FT, are associated with increased DCIS risk, while elevated levels of SHBG are associated with reduced risk.

These findings may be helpful in developing prevention strategies aimed at reducing breast cancer risk among premenopausal and postmenopausal women.

We examined associations of alcohol and tobacco use with PD, DA, and NDA in a population-based cohort of 23,456 women screened using full-field digital mammography machines manufactured by Hologic or General Electric. MD was measured using Cumulus. Machine-specific effects were estimated using linear regression, and combined using random effects meta-analysis.

Alcohol use was positively associated with PD (P_trend = 0.01), unassociated with DA (P_trend = 0.23), and inversely associated with NDA (P_trend = 0.02) adjusting for age, body mass index, reproductive factors, physical activity, and family history of breast cancer. In contrast, tobacco use was inversely associated with PD (P_trend = 0.0008), unassociated with DA (P_trend = 0.93), and positively associated with NDA (P_trend<0.0001). These trends were stronger in normal and overweight women than in obese women.

These findings suggest that associations of alcohol and tobacco use with PD result more from their associations with NDA than DA.

PD and NDA may mediate the association of alcohol drinking, but not tobacco smoking, with increased breast cancer risk. Further studies are needed to elucidate the modifiable lifestyle factors that influence breast tissue composition, and the important role of the fatty tissues on breast health.

In the Cancer Prevention Study-II Nutrition Cohort, men diagnosed with nonmetastatic prostate cancer between baseline in 1992/1993 and 2015 were followed for mortality until 2016. Analyses of pre- and postdiagnosis intakes of red and processed meat, poultry, fish, and eggs included 9,286 and 4,882 survivors, respectively. Multivariable-adjusted RRs and 95% confidence intervals (CI) were estimated using Cox proportional hazards models.

A total of 4,682 and 2,768 deaths occurred during follow-up in pre- and postdiagnosis analyses, respectively. Both pre- and postdiagnosis intakes of total red and processed meat were positively associated with all-cause mortality (quartile 4 vs. 1: RR = 1.13; 95% CI, 1.03–1.25; P_trend = 0.02; RR = 1.22; 95% CI, 1.07–1.39; P_trend = 0.03, respectively), and both pre- and postdiagnosis poultry intakes were inversely associated with all-cause mortality (quartile 4 vs. 1 RR = 0.90; 95% CI, 0.82–0.98; P_trend = 0.04; RR = 0.84; 95% CI, 0.75–0.95; P_trend = 0.01, respectively). No associations were seen for prostate cancer–specific mortality, except that higher postdiagnosis unprocessed red meat intake was associated with lower risk.

Higher red and processed meat, and lower poultry, intakes either before or after prostate cancer diagnosis were associated with higher risk of all-cause mortality.

Our findings provide additional evidence that prostate cancer survivors should follow the nutrition guidelines limiting red and processed meat consumption to improve overall survival. Additional research on the relationship of specific meat types and mortality is needed.

A total of 96,018 American adults were identified from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. A ferric-reducing ability of plasma score was used to reflect an individual's TAC intake from diet and/or supplements. Cox regression was used to calculate hazard ratios (HR) for pancreatic cancer incidence, and competing risk regression was used to calculate subdistribution HRs for pancreatic cancer mortality. Restricted cubic spline regression was used to test nonlinearity.

A total of 393 pancreatic cancer cases and 353 pancreatic cancer–related deaths were documented. Total (diet + supplements) TAC was found to be inversely associated with pancreatic cancer incidence (HR _{quartile 4 vs. quartile 1} = 0.53; 95% confidence interval, 0.39–0.72; P_trend = 0.0002) and mortality (subdistribution HR _{quartile 4 vs. quartile 1} = 0.52; 95% confidence interval 0.38–0.72; P_trend = 0.0003) in a nonlinear dose–response manner (all P_nonlinearity < 0.01). Similar results were observed for dietary TAC. No association of supplemental TAC with pancreatic cancer incidence and mortality was found.

In the U.S. general population, dietary but not supplemental TAC level is inversely associated with risks of pancreatic cancer incidence and mortality in a nonlinear dose–response pattern.

This is the first prospective study indicating that a diet rich in antioxidants may be beneficial in decreasing pancreatic cancer incidence and mortality.

Tobacco smoking habits and other relevant information were obtained from 2,009 cases and 1,532 controls recruited in the PanGenEU study using standardized tools. Multivariate logistic regression analysis was performed to evaluate pancreatic cancer risk by smoking characteristics and interactions with other pancreatic cancer risk factors. Fractional polynomials and restricted cubic splines were used to test for nonlinearity of the dose–response relationships and to analyze their shape.

Relative to never-smokers, current smokers [OR = 1.72; 95% confidence interval (95% CI), 1.39–2.12], those inhaling into the throat (OR = 1.48; 95% CI, 1.11–1.99) or chest (OR = 1.33; 95% CI, 1.12–1.58), and those using nonfiltered cigarettes (OR = 1.69; 95% CI, 1.10–2.61), were all at an increased pancreatic cancer risk. Pancreatic cancer risk was highest in current black tobacco smokers (OR = 2.09; 95% CI, 1.31–3.41), followed by blond tobacco smokers (OR = 1.43; 95% CI, 1.01–2.04). Childhood exposure to tobacco smoke relative to parental smoking was also associated with increased pancreatic cancer risk (OR = 1.24; 95% CI, 1.03–1.49). Dose–response relationships for smoking duration, intensity, cumulative dose, and smoking cessation were nonlinear and showed different shapes by tobacco type. Effect modification by family history of pancreatic cancer and diabetes was likely.

This study reveals differences in pancreatic cancer risk by tobacco type and other habit characteristics, as well as nonlinear risk associations.

This characterization of smoking-related pancreatic cancer risk profiles may help in defining pancreatic cancer high-risk populations.

Within a nested case–control study of 500 pancreatic cancer cases diagnosed after blood collection and 1,091 matched controls enrolled in four U.S. prospective cohorts, we characterized absolute risk models that included clinical factors (e.g., body mass index, history of diabetes), germline genetic polymorphisms, and circulating biomarkers.

Model discrimination showed an area under ROC curve of 0.62 via cross-validation. Our final integrated model identified 3.7% of men and 2.6% of women who had at least 3 times greater than average risk in the ensuing 10 years. Individuals within the top risk percentile had a 4% risk of developing pancreatic cancer by age 80 years and 2% 10-year risk at age 70 years.

Risk models that include established clinical, genetic, and circulating factors improved disease discrimination over models using clinical factors alone.

Absolute risk models for pancreatic cancer may help identify individuals in the general population appropriate for disease interception.

3,986 participants from the Study of Colonoscopy Utilization, an ancillary study nested within the Prostate, Lung, Colorectal, and Ovarian Screening Trial, were included. Self-reports of polyp and adenoma were compared to medical records, and measures of sensitivity and specificity were calculated. Correlates of accurate self-report of polyp were assessed using logistic regression and weighted to account for study sampling.

The sensitivity and specificity of self-reported polyp findings were 88% and 85%, respectively, and for adenoma 11% and 99%, respectively. Among participants with a polyp, older age was associated with lower likelihood while polyp severity and non-white race were associated with increased likelihood of accurate recall. Among participants without a polyp, having multiple colonoscopies was associated with lower likelihood while family history of colorectal cancer was associated with increased likelihood of accurate recall. Among both groups, longer time since colonoscopy was associated with lower likelihood of accurate recall.

Participants recalled with reasonable accuracy whether they had a prior polyp; however, recall of histology, specifically adenoma, was much less accurate.

Identification of strategies to increase accurate self-report of colonic polyps are needed, particularly for patient–provider communications and patient reporting of results to family members.

We used reduced rank regression to derive three energy balance scores that associate lifestyle factors with combinations of prediagnostic, circulating levels of high-sensitivity C-reactive protein (hsCRP), C-peptide, and hemoglobin A_1c (HbA_1c) among 2,498 participants in the Cancer Prevention Study-II Nutrition Cohort. Among 114,989 participants, we verified 2,228 colorectal cancer cases. We assessed associations of each score with colorectal cancer incidence and by tumor molecular phenotypes using Cox proportional hazards regression.

The derived scores comprised BMI, physical activity, screen time, and 14 food groups, and explained 5.1% to 10.5% of the variation in biomarkers. The HR and 95% confidence interval (CI) for quartile 4 versus 1 of the HbA_1c+C peptide–based score and colorectal cancer was 1.30 (1.15–1.47), the hsCRP-based score was 1.35 (1.19–1.53), and the hsCRP, C-peptide, and HbA_1c-based score was 1.35 (1.19–1.52). The latter score was associated with non-CIMP tumors (HR_Q4vsQ1: 1.59; 95% CI: 1.17–2.16), but not CIMP-positive tumors (P_{heterogeneity} = 0.04).

These results further support hypotheses that systemic biomarkers of metabolic health—inflammation and abnormal glucose homeostasis—mediate part of the relationship between several energy balance–related modifiable factors and colorectal cancer risk.

Results support cancer prevention guidelines for maintaining a healthful body weight, consuming a healthful diet, and being physically active. More research is needed on these clusters of exposures with molecular phenotypes of tumors.

We developed a VAT prediction score by regression equations averaged across 100 least absolute shrinkage and selection operator models in a cross-sectional study of 1,801 older adults in the Multiethnic Cohort (MEC). The score was then used as proxy for VAT in case–control studies of postmenopausal breast (950 case–control pairs) and colorectal (831 case–control pairs) cancer in an independent sample in MEC. Abdominal MRI–derived VAT; circulating biomarkers of metabolic, hormonal, and inflammation dysfunctions; and ORs for incident cancer adjusted for BMI and other risk factors were assessed.

The final score, composed of nine biomarkers, BMI, and height, explained 11% and 15% more of the variance in VAT than BMI alone in men and women, respectively. The area under the receiver operator curve for VAT >150 cm² was 0.90 in men and 0.86 in women. The VAT score was associated with risk of breast cancer [OR (95% confidence interval [CI]) by increasing tertiles: 1.00, 1.09 (0.86–1.39), 1.48 (1.16–1.89); P_trend = 0.002] but not with colorectal cancer (P = 0.84), although an association [1.00, 0.98 (0.68–1.39), 1.24 (0.88–1.76); P_trend = 0.08] was suggested for this cancer after excluding cases that occurred within 7 years of blood draw (P_{heterogeneity} = 0.06).

The VAT score predicted risks of postmenopausal breast cancer and can be used for risk assessment in diverse populations.

These findings provide specific evidence for a role of VAT in breast cancer.

We sought to quantify and assess heterogeneity of the preferences of AML patients for treatment outcomes. An AML-specific discrete choice experiment (DCE) was developed involving multiple stakeholders. Attributes included in the DCE were event-free survival (EFS), complete remission (CR), time in the hospital, short-term side effects, and long-term side effects. Continuously coded conditional, stratified, and latent-class logistic regressions were used to model preferences of 294 patients with AML.

Most patients were white (89.4%) and in remission (95.0%). A 10% improvement in the chance of CR was the most meaningful offered benefit (P < 0.001). Patients were willing to trade up to 22 months of EFS or endure 8.7 months in the hospital or a two-step increase in long-term side effects to gain a 10% increase in chance of CR. Patients diagnosed at 60 years or older (21.6%) more strongly preferred to avoid short-term side effects (P = 0.03). Latent class analysis showed significant differences of preferences across gender and insurance status.

In this national sample of mostly AML survivors, patients preferred treatments that maximized chance at remission; however, significant preference heterogeneity for outcomes was identified. Age and gender may affect patients' preferences.

Survivor preferences for outcomes can inform patient-focused drug development and shared decision-making. Further studies are necessary to investigate the use of DCEs to guide treatment for individual patients.

Medical records were retrieved and abstracted for cases enrolled in a Los Angeles County case–control study of non-Hodgkin lymphoma (NHL). Cases were linked to the Los Angeles County cancer registry (CSP), the California state hospitalization discharge database (OSHPD), and the SEER-Medicare database. We assessed sensitivity, specificity, and positive predictive value (PPV) of cancer treatment in linked databases, compared with medical record abstraction.

We successfully retrieved medical records for 918 of 1,004 participating NHL cases and abstracted treatment for 698. We linked 59% of cases (96% of cases >65 years old) to SEER-Medicare and 96% to OSHPD. Chemotherapy was the most common treatment and best captured, with the highest sensitivity in SEER-Medicare (80%) and CSP (74%); combining all three data sources together increased sensitivity (92%), at reduced specificity (56%). Sensitivity for radiotherapy was moderate: 77% with aggregated data. Sensitivity of BMT was low in the CSP (42%), but high for the administrative databases, especially OSHPD (98%). Sensitivity for surgery reached 83% when considering all three datasets in aggregate, but PPV was 60%. In general, sensitivity and PPV for chronic lymphocytic leukemia/small lymphocytic lymphoma were low.

Chemotherapy was accurately captured by all data sources. Hospitalization data yielded the highest performance values for BMTs. Performance measures for radiotherapy and surgery were moderate.

Various administrative databases can supplement epidemiologic studies, depending on treatment type and NHL subtype of interest.

The initiative was implemented through CRCHD's National Outreach Network (NON). NON is a national network of Community Health Educators (CHE), aligned with NCI-designated Cancer Centers across the nation. In phases I and II, the CHEs focused on the dissemination of cancer-related information and implementation of evidence-based educational outreach.

In total, 3,183 pre/post surveys were obtained from male and female participants, ages 50 to 74 years, during the 347 educational events held in phase I. Results demonstrated all racial/ethnic groups had an increase in colorectal cancer–related knowledge, and each group strongly agreed that the educational event increased the likelihood that they would engage in colorectal cancer–related healthful behaviors (e.g., obtain colorectal cancer screening and increase physical activity). For phase II, Connections to Care, event participants were linked to screening. Eighty-two percent of the participants who obtained colorectal cancer screening during the 3-month follow-up period obtained their screening results.

These results suggest that culturally tailored, standardized educational messaging and data collection tools are key change agents that can serve to inform the effectiveness of educational outreach to advance awareness and knowledge of colorectal cancer.

Future initiatives should focus on large-scale national efforts to elucidate effective models of connections to care, related to colorectal cancer screening, follow-up, and treatments that are modifiable to meet community needs.

We utilized data from 13 regional United States cancer registries from 1990 to 2014 to determine secular trends in incidence and survivorship from NCGC. Data were analyzed for NHWs and the six largest Asian American subgroups: Chinese, Japanese, Filipino, Korean, Vietnamese, and South Asian (Indian/Pakistani).

There exists substantial heterogeneity in NCGC incidence between Asian subgroups, with Koreans (48.6 per 100,000 person-years) having seven-fold higher age-adjusted incidence than South Asians (7.4 per 100,000 person-years). Asians had generally earlier stages of diagnosis and higher rates of surgical resection compared with NHWs. All Asian subgroups also demonstrated higher 5-year observed survival compared with NHWs, with Koreans (41.3%) and South Asians (42.8%) having survival double that of NHWs (20.1%, P < 0.001). In multivariable regression, differences in stage of diagnosis and rates of resection partially explained the difference in survivorship between Asian subgroups.

We find substantial differences in incidence, staging, histology, treatment, and survivorship from NCGC between Asian subgroups, data which challenge our traditional perceptions about gastric cancer in Asians. Both biological heterogeneity and cultural/environmental differences may underlie these findings.

These data are relevant to the national discourse regarding the appropriate role of gastric cancer screening, and identifies high-risk racial/ethnic subgroups who many benefit from customized risk attenuation programs.