Skip to Main Content

An official website of the United States government

Pancreatic cancer risk: associations with 1) Meat-derived carcinogen intake (HCAs and BaP) & 2) Nucleotide excision repair (NER) (DNA repair)

Principal Investigator
Name
Kristin Anderson
Degrees
-
Institution
University of Minnesota
Position Title
-
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
2006-0303
Initial CDAS Request Approval
Feb 12, 2007
Title
Pancreatic cancer risk: associations with 1) Meat-derived carcinogen intake (HCAs and BaP) & 2) Nucleotide excision repair (NER) (DNA repair)
Summary
Pancreatic cancer is rapidly fatal in most cases. Identifying and modifying risk factors provides a means to reduce mortality from this cancer. Aside from cigarette smoking, few risk factors for the disease are known. Meat intake has been associated with pancreatic cancer in many epidemiologic studies, but data are inconsistent. Inconsistencies may stem, in part, from failure to account for meat preparation and doneness levels in addition to intake. Heterocyclic amines (HCAs) are suspected human carcinogens formed in meat and fish cooked at high temperatures. When activated by Phase I and II enzymes, HCAs form bulky DNA-adducts that can cause mutations and cancer. Repair of bulky DNA adducts, via the nucleotide excision repair (NER) pathway, can reduce mutagenesis and cancer. Thus, individuals with low versus high NER may be at increased cancer risk.We hypothesize that pancreatic cancer risk is linked with levels of dietary HCA exposure, metabolic activation of these carcinogens and the level of activity in pathways that repair HCA DNA-adducts. We propose to estimate the risk of pancreatic cancer in relation to: 1) dietary meat carcinogen levels (using the PLCO questionnaire) 2) Internal dose of HCAs (using albumin adduct levels of PhIP, the most abundant HCA in the diet) and 3) Nucleotide Excision Repair level as measured in lymphocytes isolated from cryopreserved whole blood or from lymphoblastoid cell lines. The NCI developed questionnaires to estimate intake of several meat carcinogens including benzo(a)pyrene (BaP) (another possible pancreatic carcinogen) and the HCAs - PhIP, MeIQx, and DiMeIQx. A mutagenic activity-index can also be derived from these measures. Using this questionnaire, pancreatic cancer has been positively associated with various measures of well-done meat intake in a population-based case-control study and a large cohort study, but further investigation is warranted. Aside from our major aims, our proposal would allow for comparison of questionnaire-derived HCA data with a measure of internal dose of an HCA. We realize that there may be interactions between HCA exposure - measured by questionnaire or biomarker - and NER level. However, we lack power, at present, to consider interactions. However, assessment of main effects is a reasonable and appropriate next step and may help to identify a modifiable cause for human pancreatic cancer.
Aims

Hypotheses: Pancreatic cancer risk increases with: 1) Levels of meat-borne carcinogens (measured by questionnaire) 2) Levels of PhIP-albumin adducts (a biomarker of internal dose) and 3) Low (vs. high) nucleotide excision repair (NER). Specific Aims: 1) Identify pancreatic cancer cases and controls within the PLCO study (see B.8). A case-cohort design will be used to study pancreatic cancer cases and controls. Incident cases of exocrine pancreas cancer will be eligible for these analyses (as defined in B.8.). We will identify 100 cases of incident adenocarcinoma of the pancreas, each of whom has a plasma sample and a cryopreserved whole blood sample (or lymphoblastoid cell line). Those with biological samples will be used for PhIP-Alb and NER assays. Controls will be subjects that are cancer free at baseline and at blood collection. All pancreas cancer cases, and cancer free controls (1:3 ratio) that completed the diet questionnaire, will be included for the meat analysis. 2) Derive intake levels of well-done barbecued pan-fried meat, HCAs and BaP and a mutagenic activity index from PLCO dietary questionnaires that ascertain total meat intake, cooking methods and doneness preferences. Assessment of the meat intake and doneness preferences, carcinogen variables and mutagenic index will be carried out as described by Sinha and Rothman, 1999 and as analyzed previously by our group in collaboration with Dr. Sinha (Anderson, 2002; 2005, Attached as PDFs). 3) Measure PhIP-albumin adduct (PhIP-Alb) levels in plasma samples from 200 subjects (100 cases and 100 controls). We will extract PhIP-Alb adducts from 0.5 mls of frozen stored plasma. The adducted proteins will be hydrolyzed and levels of PhIP determined with a liquid chromatography-tandem mass spectrometry method. 4) Evaluate the association of pancreatic cancer with dietary questionnaire assessments and with PhIP-albumin adduct levels. Relative risks (RRs) and their 95% confidence intervals (CI) can be estimated using kernel-smoothed life-table models with cross-validated smoothing parameters and bootstrapped confidence intervals. Relative risks for the various measures such as, types of meat intake (in daily grams), derived carcinogen variables, and PhIP-Alb levels will be modeled as smooth continuous variables and, if warranted by this analysis, ordinal categories of RR will be described. These models will be adjusted for potential confounding using propensity scores estimated from complete dietary and demographic information. 5) Culture lymphocytes from cryopreserved lymphocytes or lymphoblastoid cell lines and perform measurements of nucleotide excision repair activity on cell cultures from 200 subjects (100 cases and 100 controls) plus quality control samples (10%). The NER assay utilizes cultured lymphocytes, which can be recovered from cryopreserved or immortalized lymphocytes. Dr. Gross' laboratory has developed a protocol and has successfully cultured lymphocytes from pilot study PLCO samples. A flow cytometry-based alkaline unwinding assay has been developed, which specifically measures activity in the nucleotide excision repair pathway and provides quantitative measurements of lymphocyte NER activity. 6) Evaluate the relationship between nucleotide excision repair activity and the risk of pancreatic cancer. It is hypothesized that pre-diagnostic NER activity is lower among individuals that later develop pancreatic cancer compared to cancer-free controls. The degree of association between NER activity and pancreatic cancer will be estimated using both classical methods and logistic regression techniques.

Collaborators

Kristin Anderson (Division of Epidemiology & CH, U of MN)
Myron Gross (Laboratory Medicine & Pathology, U of MN)
Rashmi Sinha (NCI)
Bharat Thyagarajan (Laboratory Medicine & Pathology, U of MN)
Timothy Sielaff (Virginia Piper Cancer Institute)
Timothy Church (Division of Environmental Health, U of MN)

Related Publications