Proteomic analysis and exploratory immune cell deconvolution of murine colorectal and pancreatic tumors

The CT26 colorectal cancer and KPC pancreatic cancer models are widely used syngeneic murine tumor models with different characteristics which influence their response to cancer treatments. Despite their extensive use, their proteomic landscapes remain insufficiently characterized. We applied label-free quantitative proteomics to KPC and CT26 tumors across three replicate experiments to identify biological differences between the models, explore immune cell deconvolution from bulk proteomic data, and assess reproducibility across experiments and preprocessing strategies. CT26 and KPC tumors showed distinct global proteomic profiles as KPC tumors were enriched in proteins associated with extracellular matrix remodeling, cytoskeletal organization, adhesion, and metabolic adaptation, whereas CT26 tumors were enriched in proteins related to proliferation, RNA processing, and translation. Immune cell deconvolution indicated qualitative differences in immune cell composition, where CT26 tumors had a higher level of CD4 T cells and bone marrow-derived macrophages and lower level of bone marrow-derived dendritic cells compared with KPC tumors. However, results were sensitive to missing-value handling and limited by availability of reference samples and lack of orthogonal validation. Across experiments and analytical workflows, the main differences between CT26 and KPC were reproducible, whereas experiment-specific effects were more variable. While further research is needed to assess the clinical relevance of our findings, they provide a proteomic framework for understanding biological differences between CT26 and KPC tumors and highlight the importance of reproducibility in proteomics-based tumor profiling.