441 IDENTIFICATION OF A MULTIPLE PEPTIDE SIGNATURE BY IMAGING MASS SPECTROMETRY WHICH ACCURATELY PREDICTS MORTALITY IN RENAL CELL CARCINOMA

Abstract

INTRODUCTION AND OBJECTIVES: Imaging mass spectrometry (IMS) offers substantial promise for biomarker identification given its ability to obtain detailed peptide expression data while retaining spatial information in situ. Here, we sought to determine whether IMS could be used to identify peptide signatures associated with increased mortality risk after nephrectomy for renal cell carcinoma (RCC). METHODS: We constructed a tissue microarray with 2 matched tumor and normal cores from nephrectomy specimens of 35 patients with clear cell RCC. After deparaffinization and antigen retrieval, trypsin digestion was performed directly on the tissue. Samples were analyzed utilizing an AutoFlex Speed matrix assisted laser desorption ionization (MALDI) time-of-flight mass spectrometer. For this study, only the tumor cores were analyzed. Additionally, this approach requires dichotomous survival analysis, and patients were categorized as short (overall survival 24 months; n 14 patients) or long survivors ( 24 months; n 21 patients). Data analysis was performed with ClinProTools 2.2 and FlexImaging 2.1 software. Peptide peaks differentially expressed in short vs. long survivors were used to develop a peptide signature associated with mortality risk after nephrectomy. RESULTS: A signature consisting of 22 peptides was developed that accurately discriminated between shortand long-term survivors. An average of 11 different spots were assayed within each core, and when each spot was assessed individually, the peptide signature predicted survival (short vs. long) with an accuracy of 70%. Classifying each core in its entirety demonstrated an overall predictive accuracy of 88.5% (Figure). CONCLUSIONS: MALDI IMS was able to identify and map specific peptides that accurately stratified patients with RCC by survival. While there are currently no prognostic biomarkers utilized in the care of patients with RCC, this approach offers substantial promise by simultaneously assessing and localizing a vast range of protein expression patterns while maintaining the spatial orientation within tissue. Further work is required to validate the accuracy of this pattern of peptide expression and to characterize and validate these differentially expressed peptides.