Oral Presentation HUPO 2019 - 18th Human Proteome Organization World Congress

Proteogenomics of colorectal cancer liver metastases: complementing precision oncology with phenotypic data (#172)

Bernhard Blank-Landeshammer 1 , Vincent R Richard 2 , Georgia Mitsa 2 , Laxmikanth Kollipara 1 , Ingo Feldmann 1 , Maud Marques 3 , Karen Gambaro 3 , Suzan McNamara 3 , Alan Spatz 4 , René P. Zahedi 2 4 , Albert Sickmann 1 , Gerald Batist 4 , Christoph H Borchers 2 4 5 6
  1. Leibniz-Institut für Analytische Wissenschaften – ISAS - e.V., Dortmund, Germany
  2. Segal Cancer Proteomics Centre, Montreal, QC, Canada
  3. Exactis Innovation and Segal Cancer Centre, Montreal, QC, Canada
  4. Gerald Bronfman Department of Oncology, Jewish General Hospital, Montreal, Canada
  5. UVic Proteomics Centre, Victoria, Canada
  6. Dept. of Biochemistry and Microbiology, University of Victoria, Victoria, BC, Canada

In colorectal cancer (CRC), KRAS mutational status is an important precision oncology marker determining whether patients are eligible for targeted anti-EGFR treatment. Nevertheless, response rates to anti-EGFR treatment are only 20-30%, while certain patients respond despite having KRAS-mutations. We describe a proteogenomic analysis of CRC liver metastases (mCRC), representing an ideal setting for the analysis of therapeutic resistance. Deep proteomic profiling of KRASwildtype and KRASG12V mCRC enabled the identification of >9,000 proteins. We detected considerable changes in protein expression, both in line and in discordance with whole exosome sequencing (WES) and RNAseq data, including numerous proteins with a known role in progression and resistance of CRC tumors. Interestingly, ERBB2 was massively upregulated in the KRASwildtype tumor, despite the absence of mutations in both tumors. We identified protein evidence for a number of predicted somatic mutations, among those KRASG12V. For 8 proteins, we developed targeted parallel reaction monitoring (PRM) assays that enabled the precise and sensitive absolute quantification of both, the mutated and the canonical variants. Using these PRM assays, we phenotyped individual mCRC tumors and paired healthy tissues, by determining the actual mutation rates on the protein level. Our data demonstrates that total KRAS protein expression varies between tumors (0.47–1.01 fmol/µg total protein) and healthy tissues (0.13–0.64 fmol/µg). In some patients, KRAS protein expression is comparable between tumor and control tissue, while in others KRAS is significantly upregulated in the tumor.  In most KRASG12V-positive tumors, the G12V-mutation level was 42-100%, while one patient’s tumor had only 10% KRASG12V but 90% KRASwildtype. This might be an example for a potential false-negative exclusion from targeted treatment: Although the patient was ineligible for targeted anti-EGFR treatment based on hotspot sequencing and instead received chemotherapy, a (combined) anti-EGFR therapy might have been a therapeutic option – potentially missed because of lacking phenotypic information.