Quantitative Phosphoproteomics Reveals Subtypes And Kinase Targets In Diffuse-Type Gastric Cancer

The diffuse-type gastric cancer(DGC)constitutes a poor-prognosis subgroup of gastric cancer with no known effective molecularly targeted therapies.The need to elucidate DGC-specific molecular characteristics for identifying new actionable targets is crucial.A total of 83 patients with their nearby tissues were collected and used for phosphorylation peptide enrichment and quantitative mass spectrometry analysis.Impressively,DGC could be subtyped into 3 major subtypes(Ph1-3)based on the phosphorylation profiles using consensus clustering.Ph1 has the best survival apparently.Both Ph2 and Ph3 have bad survival,while Ph2 tends to be more sensitive to chemotherapy and Ph3 shows insensitive to chemotherapy.Moreover,the activities of 16 kinases were significantly associated with overall survival of DGC patients,which might be the potential therapeutic kinase targets.In addition,most of the studies analysis phosphoproteomic data superficially and did not focused on how to use the phosphoproteomics to nominate kinase candidates for clinical intervention.We reanalysed the phosphoproteomic data of High-grade serous ovarian cancer(HGSOC)from the Clinical Proteomic Tumor Analysis Consortium(CPTAC).The results show that HGSOC could be classified into 5 major subtypes that are associated with different overall survival and unique kinase activity patterns.Then,we ident ified 29 kinases whose increased activities in tumours were associated with poor survival.The main significance of the study could be summarized to the following points:(1)We provided a new phosphoproteomic encyclopedia of 83 DGC tumors and the corresponding nearby tissues.Based on the phosphorylated sites,we predicted the kinase patterns of DGC,which shows that many pathways such as MTOR signaling were dysregulated;(2)We revealed the subtypes of DGC with the phosphorylated residues and confirms the correlation of the subtypes with clinical outcome and chemotherapy;(3)Furthermore,we reported a bioinformatics analysis workflow to distil information and knowledge from large-scale phosphoproteimics data of clinical samples.Our work detailed the process of how to subtype cancer with phosphorylation data to be associated with clinical outcome and nominate actionable kinase targets and inhibitors for clinical intervention.In summary,our results indicate that molecular characteristics of DGC and ovarian cancer reflected in phosphoproteomic patterns might dictate clinical outcomes.The phosphoproteomics could be a vital technique to assist in the prediction of prognosis and the treatment of cancer for kinase inhibiton therapy.