Version 2.5.2.0

Abstract

Grant Number:	2P50CA058183-090010
Project Title:	CLASSIFICATION /PROGNOSTIC PROFILING OF DUCTAL CARCINOMA

PI Information:	Name	Email	Title
	ALLRED, D CRAIG.		PROFESSOR

Abstract: Ductal carcinoma in situ (DCIS) is in itself an essentially harmless disease, but it gives rise to about 90% of invasive breast cancers (IBCs), which are potentially lethal. DCIS is currently defined by its histological features, and its prognosis is imprecisely estimated from epidemiological evidence. While lesions within general categories look alike under the microscope, only a subset progresses to IBC, emphasizing that there are underlying biological abnormalities cansing some to remain stable and others to progress. The long-term goal of this proposal is to identify these abnormalities and begin to assess their clinical usefulness. Unique resources, exciting preliminary data, and an experienced team of investigators have been assembled to achieve this goal. The resources and data include an unrivaled bank of fresh-frozen DCIS (n = 25), as well as matched cases of normal breast and IBC for comparison, which have been evaluated for the expression of over 12,000 annotated genes using eDNA expression arrays (Clontech and Affymetrix). Analysis of this data has identified clusters of genes that may be important in the evolution of DCIS. The importance of selected genes will be confirmed and validated using another unique resource -- hundreds of samples of human DCIS with clinical follow-up, as well as other relevant types of breast tissues, assembled in tissue arrays. Aim 1 will identify genes important in the early development of DCIS by comparing expression profiles between normal breast and DCIS. The relevance of differentially expressed genes will be validated by immunohistochemistry (IHC) or RNA in situ hybridization (ISH) on tissue arrays of normal human breast and DCIS. Aim 2 is designed to identify families of biologically related DCIS through hierarchical clustering of expression array data, and to determine if they are prognostically distinct. This will again be validated by IHC/ISH on tissue arrays of DCIS from breasts with vs. without synchronous IBC. Aim 3 will compare gene expression patterns between DCIS and IBC to identify potential invasion-related genes, which will be evaluated for their prognostic significance using a case-control strategy similar to Aim 2. Finally, Aim 4 will determine whether a few of the most highly prognostic genes identified in previous aims are directly involved in causing tumor invasion progression. We will transfect the genes into human MCF 1OAT cells to determine how they change the expected proportions of DCIS and IBC arising in mouse xenografis. Understanding DCIS at a fundamental genetic level may enable more accurate assessment of risk of progression, individualization of therapy for DCIS and, most important, identification of specific defects that can be targeted therapeutically to prevent the development and progression of DCIS.

Public Health Relevance:
This Public Health Relevance is not available.

Thesaurus Terms:
breast neoplasm, method development, neoplasm /cancer classification /staging, neoplasm /cancer genetics, prognosis
gene expression, molecular oncology, tumor progression
NOD mouse, SCID mouse, heterologous transplantation, human tissue, immunocytochemistry, in situ hybridization, transfection

Institution:	BAYLOR COLLEGE OF MEDICINE
	1 BAYLOR PLAZA
	HOUSTON, TX 770303498
Fiscal Year:	2003
Department:
Project Start:	11-MAR-2003
Project End:	30-NOV-2007
ICD:	NATIONAL CANCER INSTITUTE
IRG:	ZCA1

---