Abstract 4534: Label free molecular imaging of tumor sections for two and three dimensional tissue classification and pathway mapping

Abstract

Abstract Mass spectrometry imaging allows for the presence and abundance of a wide range of bio-molecules to be measured directly from tissue sections without the use of labels or tags. This results in the ability to compare the spatial distribution of hundreds of metabolites, lipids, peptides, and proteins within the same section in a single experiment. Desorption electrospray ionization (DESI) mass spectrometry imaging can differentiate and classify tissue types, and associated disease states, by measuring the unique characteristic molecular fingerprints of the tissues1,2. Furthermore, using histologically annotated databases of molecular profiles, unknown tissues can be classified using the mass spectrometry output coupled to machine learning approaches. A major benefit of DESI is that the analyzed sections are left largely unchanged by the process; the sections can subsequently be stained and the results validated by immunohistochemistry or conventional H&E approaches. During DESI fingerprinting analyses, a full mass spectrum on a cell by cell resolution of 20-50µm pixel size is collected. Chemical changes on very small scales can be distinguished and explored, allowing the chemistry of tumor margins and interface zones to be understood through mapping results onto metabolic pathways. As the data in a DESI experiment is collected and stored, multiple interrogations of the data are possible; as new discoveries are made, previously collected data can be retrospectively mined for additional insights. In this study, the three-dimensional chemical heterogeneity of drug dosed spheroids and tumor xenografts was measured with DESI and compared with MRI and PET imaging modalities and with FT-IR spectroscopic techniques. We demonstrate the power of this multi-analytical approach in glioma models and in colorectal and prostrate tumor samples to uncover metabolic and lipidomic changes during disease progression and to reveal the inherent heterogeneity within a tumor. 1. 1. Calligaris, D.Caragacianu, D.Liu, et al; PNAS, 2014, 111, 15185-151892. Guenther, S., Muirhead, L.J., et al, Cancer Res, 2015, 75, 1828-1837 Citation Format: Emrys A. Jones, Dipa Gurung, Fiona Henderson, Matthew Gentry, Danielle McDougall, Yasmin Shanneik, James Langridge, Adam McMahon, Zoltan Takats. Label free molecular imaging of tumor sections for two and three dimensional tissue classification and pathway mapping [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4534.