Demirkan, Gokhan

Abstract Background: Understanding heterogeneity within individual breast tumors is key to the ability to predict therapeutic outcome. Molecular heterogeneity is commonly evaluated based on genomic features, including mRNA abundance, gene copy number events, and somatic mutations. The expression profile and activation state of key proteins is widely recognized as another key element in defining tumor heterogeneity. We have taken advantage of NanoString 3D Biology™ technology (for research use only) and curated nCounter Vantage 3DTM Solid Tumor Assay to interrogate a survey panel of HER2-positive breast tumors with the ultimate goal of determining key relationships between multiple genomic and proteomic profiles in individual tumors. Methods: We analyzed samples from 24 HER2+ breast cancer patients using NanoString technology to quantify the expression profile for over 25 total and phospho signaling proteins, including PI3K/MAPK/EGFR/HER2, 770 mRNA corresponding to 13 canonical cancer pathways, and 104 somatic mutations and small INDELS that are commonly associated with cancer, including 8 known PIK3CA mutations. These analyses were carried out in a matched fresh frozen and FFPE samples on the nCounter paltform. Data were analyzed by nSolver to identify genotype specific expression profiles across the 24 samples. Results: In our proof-of-concept data set, we successfully demonstrate that NanoString’s 3D biology Technology shows concordance across both FFPE and fresh frozen sample types for DNA, RNA, and protein. NanoString analysis also showed high concordance to gold-standard techniques used to assess genotype and RNA expression profiles. The combination of digital DNA, RNA, and protein data from our HER2+ breast cancer samples yielded potentially actionable data based on mapping of mutational status as the driver of key differences in protein expression and mRNA abundance of the signaling targets profiled. This work sheds new light on HER2+ breast cancer biology and the interplay between genomic and proteomic profiles while setting the stage for future studies that further probe the differences observed in this sample set. Conclusions: Simultaneous analysis of mutational status (SNV) and expression at the level of both mRNA and protein promises to enable a more detailed view of the relationship between genotype and the biological and clinical behavior of key tumor types. The NanoString Vantage 3DTM Solid Tumor platform provides a rapid, reliable, and economic means of assessing these analytes simultaneously. The application of these analytes to models that make clinically actionable predictions will require additional analyses of large sample cohorts, but such analysis is quite feasible using a variety of sample types. Acknowledgements: Supported in part by grants from the Breast Cancer Research Foundation and the 26.2 with Donna Foundation. Citation Format: Sarayna Chumsri, Daniel J. Serie, Brian M. Necela, Jennifer M. Kachergus, Bianca C. Axenfeld, Gokhan Demirkan, Gavin Meredith, P. Martin Ross, Anisha Kharkia, Erin Piazza, Afshin Mashadi-Hossein, Sarah Warren, Sarah A. McLaughlin, Joseph Beechem, Gary Geiss, E. Aubrey Thompson. Simultaneous analysis of the mutational landscape and RNA and protein expression profile of HER2-positive breast cancer using 3D BiologyTM [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3377. doi:10.1158/1538-7445.AM2017-3377

Abstract It is widely appreciated that early detection is the key to making inroads in the fight against cancers. Despite the availability of endoscopic screening, compliance is poor. The diagnosis of CRC is often made in advanced stages, and the consequence is a dismal 50% five year survival across all CRC. There is urgent need for biomarkers with performance that can meet the need of early diagnosis. Like most cancers, colorectal cancers (CRCs) express tumor antigens that can induce humoral immune response. The identification of the autoantibodies against tumor antigens may facilitate early diagnosis of CRC. The goal for this pilot project was to demonstrate the feasibility of discovering autoantibody biomarkers for colorectal cancer using Nucleic Acid Programmable Protein Arrays (NAPPA). Autoantibodies in serum samples collected from 40 patients with DNA mismatch repair proficient colorectal cancer who had died of their cancer within 18 months of their blood draw and from 40 age-gender matched healthy controls were profiled on NAPPA arrays with randomly selected 2000 human protein. The fact that p53, a tumor antigen that has been previously reported to be related to various cancers including CRC, was one of the top candidates based on the statistical analysis demonstrates the validity of our discovery platform. Besides p53, thirty-one candidate antigens were identified to have sensitivity greater than or equal to 25% at 95% specificity and fifty-one candidate antigens had an area under the ROC curve (AUC) equal to or greater than 0.6. Validation of the combined 83 unique genes selected based on these two different data analysis methods in an independent sample set in a blind fashion is on-going. This pilot study using late stage cancer samples demonstrates the feasibility and power of an immunoproteomics study using protein arrays. This approach may rapidly lead to identification of markers that could have relevance in monitoring disease progression as well as in early cancer detection. Citation Format: Gokhan Demirkan, Leah Soderberg, Garrick Wallstrom, Emily Szeto, Kristi Barker, Jie Wang, Joshua LaBaer, Noralane M. Lindor, Ji Qiu. Proteome scale identification of autoantibody biomarkers in colorectal 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 887. doi:10.1158/1538-7445.AM2014-887