Abstract 2007: Phosphoproteomics scaled-down to tumor biopsies for future TKI treatment selection

Abstract

Abstract Introduction Mass spectrometry-based phosphoproteomics of cancer cell- and tissue lysates provides a unique approach to evaluate the cell signaling network, revealing information on aberrantly activated signaling pathways and potential drug targets. To enable phosphoproteomics-based treatment selection for improved efficacy of tyrosine kinase inhibitors (TKIs) in patients with advanced solid tumors, needle biopsies should provide reproducible profiles, representative of the individual tumor phospho-proteome. We have assessed reproducibility and intra- and inter-tumor heterogeneity by downscaling an antibody-based pTyr phosphopeptide enrichment protocol to 1 mg protein input using colorectal cancer (CRC) cell line and tissue as model, including needle biopsies from patients. Methods Phosphopeptide immunoprecipitation (IP) using P-Tyr-1000 anti-phosphotyrosine-agarose beads (PTMScan®, CST) was performed using triplicates of 1, 5 and 10 mg protein from HCT116 CRC cell line lysate and 1, 5 and 10 mg protein from 3 patient-derived tumors. In addition, multiple 14-gauge core needle biopsies were obtained from 7 CRC resection specimen for analysis at 1 mg protein input in biopsies with ≥ 50% tumor cells. After measurement by LC-MS/MS, database searching was performed for phosphopeptide identification (ID), ion-intensity-based quantification and phosphosite localization. Results Phosphoproteomics of the cell line triplicates of 1, 5 and 10 mg protein input yielded a total of 474, 574 and 689 unique phosphopeptides, and a median number of 362 (range: 272-393), 443 (237-470) and 526 (496-653) per input level, respectively. The ID-reproducibility was 43% for 1 mg and 60% for 10 mg protein replicates. In tissue, a median total of 909 (481-928) unique phosphopeptides were identified per tumor, with 361 (175-536), 629 (284-796) and 896 (398-909) phosphopeptides for the 1, 5 and 10 mg levels, respectively. For the 2-4 needle biopsies that were analyzed per tumor, a median of 208 phosphopeptides (range 140-279) were identified at a median ID reproducibility of 58% for duplo and 39% for triplo biopsies. Unsupervised clustering based on phosphopeptide intensities showed that multiple tumor biopsies from individual patients cluster together. Conclusion This scale-down study demonstrates the feasibility of label-free pTyr-phosphoproteomics at the biopsy level. Unsupervised analysis shows that tumor needle biopsies from individual patients cluster together indicating that this analysis can identify patient specific phosphoproteomic profiles. Therefore, down-scaled pTyr-based phosphoproteomics of patient tumor biopsies may enable future phosphoproteomics-based TKI treatment selection. Citation Format: Mariette Labots, Johannes C. Van der Mijn, Sander R. Piersma, Richard R. De Haas, Robin Beekhof, Thang V. Pham, Jaco C. Knol, Inge de Reus, Nicole C.T. van Grieken, Gerrit A. Meijer, Connie R. Jimenez, Henk M.W. Verheul. Phosphoproteomics scaled-down to tumor biopsies for future TKI treatment selection. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2007. doi:10.1158/1538-7445.AM2015-2007