Abstract 922: Evaluating markers of cisplatin sensitivity and survival in small cell lung cancer

Abstract

Abstract Background: Molecular characterization of small cell lung cancer (SCLC) can lead to improved outcome through the identification of tumor subtypes with differential clinical outcome and response to treatment. We characterized SCLC cell lines and patient samples by a combination of a genome-wide and a custom expression panel to assess expression changes associated with cisplatin sensitivity, in the case of cell lines, and overall survival, in terms of SCLC patients.Method: SCLC cell lines (H69, H128, H146, H526, H187, H209, D53, DMS153, and DMS114) and 79 archival samples of pulmonary neuroendocrine tumor samples were employed. Sensitivity to cisplatin, and PARP inhibitor, veliparib was determined by MTS assay. In order to identify expression profiles characteristic of sensitive and insensitive cell lines, gene expression for untreated and treated cells was obtained from Illumina microarray gene expression and NanoString platforms. Comparisons of expression were performed between treated cell lines and controls using a combination of ANOVA and empirical variance distributions to assess sensitivity. Prognostic impact of differentially expressed genes in cell lines was assessed in patients using gene expression obtained by NanoString and protein expression by immunohistochemistry (IHC). Kaplan Meier Method, Logrank test, and Cox proportional hazard model were employed for testing differences in survival. Additionally, a quantile survival analyses was performed to identify patient subgroups associated with early and/or late changes in expression.Results: Unsupervised analysis of the Illumina probesets identified a 23 gene panel with differential expression between cell lines with differing platinum sensitivity: Sensitive - GLS, UBEC2, HACL1, MSI2 and LOC100129585; insensitive - CENPE, CRYGS, FAM83D, FLJ44342, GNA12, LOC88523, LRDD, N4BP2L2, SLC35A3, SPC25; shared - AURKA, CENPA, DLGAP5, HMMR, KIF20B, LOC100129585, LOC100131735, RBMX, SFRS3. Unsupervised analysis of expression data from Illumina and nCounter NanoString showed considerable concordance in cluster patterns. The Cox regression models of gene expression on overall survival identified LOC100131735 as a predictive marker and, SLC35A3, SPC25, DLGAP5 and UBE2C genes as prognostic biomarkers. Concomitantly high expression of SLC35A3 and GLS genes characterized SCLC with prolonged survival (Logrank p=0.025) while a trend toward reduced risk of death (HR: 0.798; 95%CI, 0.375-1.700, p=0.559) was observed with high GLS protein expression.Conclusions: Our use of both cell line and patient SCLC data has provided evidence of markers characteristics of cisplatin insensitivity that also show a relation with poor overall survival in patients. Additional studies to fully understand the mechanisms underlying such markers are currently underway. Supported through the Georgia Cancer Coalition Distinguished Scholar Award and NIH 5K23CA164015 grant to TK Owonikoko Citation Format: Taofeek Kunle Owonikoko, Guojing Zhang, Gabriel L. Sica, Zhengjia Chen, Jeffrey M. Switchenko, Sungjin Kim, Anthony A. Gal, Suresh S. Ramalingam, Xingming Deng, Michael R. Rossi, Jeanne Kowalski, Fadlo R. Khuri. Evaluating markers of cisplatin sensitivity and survival in small cell lung cancer. [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 922. doi:10.1158/1538-7445.AM2014-922