The liver is the most common metastatic site of colorectal cancer (CRC). Approximately 20% of newly diagnosed CRC will have synchronous colorectal liver metastasis (CRLM) which portends a poor prognosis, with 5-year overall survival 13%. Curative-intent surgery to resect CRLM can be effective in providing durable, long-term remission in 20-25% of patients. A smaller group of patients will have operable recurrent intrahepatic disease, while the remainder will recur with inoperable metastasis. Prognostic assessment of CRLM patient outcomes relies on clinicopathological parameters but the current models are poorly effective. To identify candidate prognostic biomarkers we mapped the proteomic landscape of 42 CRLM consisting of 17 metachronous and 25 synchronous specimens.
We obtained fresh-frozen CRLM tumour tissue from patients that underwent liver resection at Royal North Shore Hospital 2007-2017. We conducted LC-MS using a HF-X Orbitrap MS operating in DIA mode over 120 minute acquisition time. Data was searched using DIA-NN (1.8.1) in library-free mode with match between runs enabled. For peptide identification, a reverse-decoy approach was used with 1%FDR. Using this method 7450 proteins were quantified in all samples. Unsupervised hierarchical clustering of metachronous specimens indicated high inter-lesion similarity for some patients, while other patients’ lesions showed discordant protein expression. Unsupervised hierarchical clustering of all specimens revealed three groups. Group 1 was enriched with extracellular matrix proteins, immune effectors and O-glycan processing proteins. Group 2 were defined by a proliferation signature of ribosome biogenesis, DNA replication and mitochondrial gene expression. Group 3 was enriched with liver metabolism related proteins. Survival analysis showed that patients with Group 2 CRLM signature experienced longer 12 month survival compared with other groups. Ongoing work includes immunohistochemistry validation and feature reduction to select the fewest effective markers for specimen classification.