signature=425615014eb25801a87f071cd91f231b,Abstract 914: Development of a prognostic and predictive ...

Abstract

Lung cancer remains the leading cause of cancer-related death in the United States. The 5 year survival for stage I-IIIA lung cancer ranges from only 15-50%. For many in the early stages of the disease, especially stage II patients, adjuvant chemotherapy (ACT) following surgical resection is standard treatment. Still, it appears that the benefits for ACT have plateaued at 4%. With these marginal survival rates and the toxicities associated with chemotherapy, there is a need to develop a predictive tool to identify those patients with early-stage disease who would most benefit from receiving ACT. Previously, we developed an E2F-based gene expression signature derived from microarrays of two non-small cell lung cancer cell lines that were transfected with small interfering RNAs targeted against several members of the E2F pathway. This signature was then filtered for genes that were altered in cancer, but not in normal lung tissues, agreeing between two publicly-available lung tumor/adjacent normal datasets. Through principal component analysis (PCA), the E2F signature was shown to be prognostic in the Molecular Classification of Lung Adenocarcinoma (MCLA) from the Director's Challenge Consortium, as well as predictive of early-stage lung adenocarcinoma patients’ response to ACT in the JBR.10 clinical trial. In order to demonstrate efficacy of the signature in a typical clinical setting, we evaluated whether the NanoString platform (a method that involves the use of a novel “barcode” system to measure gene expression) would be as effective using RNA derived from formalin-fixed, paraffin-embedded (FFPE) tumor samples as compared to fresh frozen RNA. NanoString quantification of RNA from fresh frozen samples correlated very well to FFPE samples (average r=0.9070, median r=0.9290, SD=0.0859). Further, NanoString expression for 32 pairs of both the fresh frozen and FFPE RNA were compared to gene expression microarray readings of the genes in the signature. This analysis demonstrated that the NanoString readings corresponded well to microarray expression, regardless of the method of RNA extraction for NanoString (paired NanoString-frozen vs. microarray: average r=0.5585, median r=0.5608, SD=0.0718; paired NanoString-FFPE vs. microarray: average r=0.5384, median r=0.5401, SD=0.0891). Taken together, these results demonstrate that this E2F signature may be a useful tool for predicting which patients would best respond to adjuvant chemotherapy, and NanoString analysis may be a very effective means of applying this signature to patient samples.

Citation Format: Courtney A. Kurtyka, Lu Chen, Matthew B. Schabath, Dung-Tsa Chen, William Brazelle, Eric A. Welsh, Anders E. Berglund, Steven A. Eschrich, Jhanelle E. Gray, Eric B. Haura, W. Douglas Cress. Development of a prognostic and predictive E2F signature in formalin-fixed, paraffin-embedded early-stage non-small cell lung cancer samples. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 914. doi:10.1158/1538-7445.AM2014-914