Skip to Content
U.S. National Institutes of Health
Last Updated: 11/21/12

Biological Breast Cancer Classification by qRT-PCR

Matthew J. Ellis, M.D., Ph.D.
Washington University, St. Louis, MO

Dr. Ellis' program will refine and validate diagnostic signatures that identify five subtypes of breast tumors. The profiles will be refined by selecting the set of 100 genes that defines all subtypes. qRT-PCR assays will be developed to measure the profile in formalin fixed, paraffin embedded (FFPE) tissues. The ability of the signatures to identify two of the subsets, LumA and Lum B, among ER+, node negative breast cancer patients who will not benefit from chemotherapy will be evaluated. The signature will also be evaluated to determine if it can predict which patients will respond to specific chemotherapies. The predictor will be validated in three CLIA-approved clinical laboratories at University of North Carolina (UNC), University of Utah and Washington University.

Collaborators:

  • The project includes investigators from Washington University, University of Utah, UNC, University of British Columbia (UBC) and CALGB.
  • Statistical support will be provided by the CALGB Data Center and by statisticians at the individual institutions.

Projects:

  • Develop a qRT-PCR assay for the selected 100 genes from the previously developed diagnostic signature.
  • Develop the bioinformatics and database support for the project.
  • Evaluate the performance of the qRT-PCR assay on specimens from the collaborating institutions including a training set of tissue from UBC.
  • Validate the qRT-PCR assay in independent test sets of specimens from UBC and from the NCI Cooperative Breast Cancer Tissue Resource (CBCTR).
  • Evaluate the efficacy of chemotherapy in the breast cancer subtypes by analyzing specimens from two clinical trials, CALGB 9344 and SWOG 8814, using the qRT-PCR assay.
  • Refine, standardize and then evaluate the assay in the clinical labs at the University of Utah, Washington University and UNC as the final phase of the project.

Featured Publications:

Parker JS et al (2009) Supervised Risk Predictor of Breast Cancer Based on Intrinsic Subtypes. J Clin Oncol. PMID: 19204204

This paper from the Ellis SPECS project describes the PAM50 assay, which can be run in a routine clinical laboratory setting to distinguish the “ intrinsic” subtypes of breast cancer. This signature will add useful information to the established diagnostic categories of breast cancer and help avoid either under- or over-treatment.