Abstract 2382: Novel integration of genomic and morphological information reveals tumor tissue heterogeneity

• Published in: Cancer research (Chicago, Ill.) Vol. 76; no. 14_Supplement; p. 2382
• Main Authors: Lee, Chen-Chung, Predki, Paul, Raub, Christopher, Shibata, Darryl, Taylor, Clive, Kartalov, Emil, Anderes, Kenna
• Format: Journal Article
• Language: English
• Published: 15.07.2016
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2016-2382

Abstract Cancer treatment decisions today are predominantly based upon decades-old histological methods, which more recently have been combined with simple phenotypic or molecular tests. To advance this paradigm, we have developed a digital molecular morphology platform that integrates histological images with next-generation sequencing (NGS) or PCR data. This allows us to visualize DNA mutations and RNA expression levels across a histological image, and to correlate mutational or expression status to cell morphological features. Thus, providing a novel approach to mapping tumor heterogeneity. The basic approach is to overlay a microstructure on top of H&E-stained FFPE slides to create a large number of tissue-bottomed micro-wells, which can be as small as a single cell or contain clusters of cells. Reagents are added to the micro-wells and biochemical reactions are carried out and analyzed in place (PCR) or partially offline (NGS). In the case of NGS, barcodes are used to reference the resulting sequence back to the tissue image. Software is then used to analyze and visualize the resulting genetic information in the context of the original histological image. Our initial work has focused on FFPE cancer biopsies using targeted cancer genes and gene panels. Here we demonstrate the use of this approach to measure and visualize KRAS G12V mutational status in colorectal cancer tissues, using both PCR and NGS detection modalities. We show that with this method we are able to efficiently extract DNA from the tissue while maintaining a leak-proof physical integrity from one micro-well to another. Further, we are able to PCR-amplify DNA in the micro-wells for direct fluorescence detection, or to create libraries for sequencing. Detection of KRAS G12V correlates well with malignant morphological features in the H&E stained tissue, and appears to present a more sensitive way to detect mutations than PCR or NGS from bulk FFPE tissue alone. We are currently applying this technology to a variety of research and clinical applications, ranging from assessment of tumor heterogeneity and evolution, to prediction of patient outcome, to post-operative NGS-based characterization of surgical margins. This integration of cellular and molecular data promises to more fully enable precision medicine to guide the course of treatment and improve individual patient outcomes. Citation Format: Chen-Chung Lee, Paul Predki, Christopher Raub, Darryl Shibata, Clive Taylor, Emil Kartalov, Kenna Anderes. Novel integration of genomic and morphological information reveals tumor tissue heterogeneity. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2382.