Grant Number:	7R01CA100555-05 Interpret this number
Primary Investigator:	Chiu, Brian
Organization:	University Of Chicago
Project Title:	Lymphoma Defined Cytogenetically for Epidemiologic Study
Fiscal Year:	2009

DESCRIPTION (provided by applicant): It is important that analytic epidemiologic studies of non-Hodgkin's lymphoma (NHL) focus on risk factors according to NHL subtypes because: 1) NHL comprises a heterogeneous group of malignancies that vary in etiology; and 2) the patterns of change in NHL incidence vary across different subtypes. At this time, NHL cannot be subdivided into standard pathologic categories for epidemiologic studies. We propose to define NHL according to rearrangements of the BCL2, BCL6, and C-MYC genes because cytogenetic studies have shown that recurring chromosomal abnormalities leading to the deregulation of these genes are important in the pathogenesis of certain NHL subtypes. The central innovative hypothesis is that risk factors differ for discrete subsets of NHL defined by genetic abnormalities, The rationale is that cases with specific, nonrandom chromosomal abnormalities may be more closely related etiologically than NHL cases as a whole. The specific aims are to investigate whether the associations with cigarette smoking, farming and pesticide exposure, family history of hematopoietic cancer, and dietary intake of animal fat differs for BCL2 rearrangement-positive NHL, BCL6 rearrangement-positive NHL and C-MYC rearrangement-positive NHL. We estimate that 540 tumor blocks will be available (and of which, 530 will yield chromosomal information) from NHL cases in two existing population-based, case-control studies conducted in Nebraska, and there will be data on 1967 controls for comparison. Fluorescence in-situ hybridization (FISH) technique will be used to determine the rearrangements of the BCL2, BCL6, and C-MYC genes. Subtypes of NHL will be defined by the presence or absence of the specific genetic abnormalities. Logistic regression model will be used to calculate the odds ratios for rearrangement defined subsets of NHL associated with exposures. Polytomous logistic regression will be used to compare associations across discrete subsets of NHL defined by rearrangements. The results will be significant because they could provide new insights about the etiology of lymphomagenesis. Because the study population is well-characterized and extensive data on exposures have been collected, the current proposal is extremely cost-effective to address the etiology of NHL, one of the most common malignancies in this country. The long-term objective is to improve understanding of the disease process which may ultimately lead to appropriate preventive measures in the general population.