Introduction: RON and Met, members of the Met family of tyrosine kinases, are implicated as mediators of tumor progression and metastasis in cancer. Over-expression of each receptor is prognostic of poor survival in resected and metastatic cancers, and expression of RON/Met in preclinical models and early phase clinical trials predicts response to RON/Met specific inhibitors. RON is expressed as a number of alternate splice variants/transcripts, complicating the quantification of the receptor by standard clinical methods. Inter/intrapatient tumor heterogeneity suggests that an expedient, reliable, medium throughput oncogene protein expression profiling will provide vital information to better personalize cancer care, with emphasis on serial biopsies to assess acquired treatment resistance mechanisms. To date, clinical quantification of protein in formalin fixed paraffin embedded (FFPE) tissues is limited to immunohistochemistry (IHC), which is semi-quantitative at best. Moreover, IHC of multiple proteins of interest is laborious, time consuming, wasteful of scarce tissue, and costly. Other protein quantification methods (ELISA, ECL) would require non-standard tissue processing for analysis. We sought to develop a quantitative mass spectrometric (MS) assay for RON utilizing Liquid Tissue – Selected Reaction Monitoring (SRM), with subsequent multiplex quantification of RON, Met, and other previously validated proteins in a panel of gastroesophageal cancer (GEC) cell lines and tissues.
Methods: Using trypsin digestion mapping of recombinant RON, we identified unique peptide sequences, and built quantitative MS assays which could be multiplexed into a single SRM analysis of 1μg of tumor protein. Assays were preclinically validated on 10 different formalin fixed (FF) cell lines. The final assay was validated and the N-terminal RON SRM demonstrated an LOD/LOQ of 62/125. Alternate peptides were chosen to quantify differences in RON splice variants/transcripts.
We then tested the RON MS assay using a panel of FF GEC cell lines previously characterized by immunoblot (IB) and IHC FFPE pellet. In addition to RON, we multiplexed SRM quantification of Met, EGFR, HER2, HER3, IGF1R, and cSRC. We evaluated 15 GEC lines including three AGS lines: wild type (AGS-WT), scrambled shRNA (AGS-SC) and RON shRNA knockdown (AGS-KD) to assess ‘post-treatment’ changes in oncogene expression profiles. We then evaluated 20 GEC human cancer tissues and 5 paraneoplastic normal tissues using laser capture microdissection of the target material from a single unstained 10μm thick section per sample.
Results: In the initial analysis, 4/10 cell lines (HCC827, Colo205, HT29, A431) expressed N-terminal RON (∼250 amol/μg cell protein). Validation of the RON SRM assay on GEC cell lines revealed very high concordance when compared to IB and IHC measurement. The AGS-WT/SC cells showed comparable levels of N-terminal RON (284/323 amol/ug cell protein), while RON was not detected in AGS-KD cells, as expected. Correlation of IB with RON intracellular/extracellular MS assay data will be presented. Measurement of RON in the GEC tissues correlated well with IHC. RON expression was seen in 75% of GEC tissues, and was lower/undetectable in adjacent normal tissues. Multiplex oncogene quantification of all cell lines and tissues, along with expression profile changes in the AGS RON KD line compared to AGS-WT/SC will be presented.
Conclusions: Taken together, these data demonstrate a sensitive, accurate, and quantitative assay to measure RON and its variants in FF cells. Multiplexed oncogene expression of these tumors was feasible and expedient using limited tissue, and is a novel clinically applicable approach for tumor characterization for baseline and post-treatment assessment.