Fig 1.
Schematic overview of the data analysis performed in this study.
Fig 2.
(A) Semi-supervised hierarchical cluster analysis and associated heatmap of tumor samples (n = 409) and immunohistochemical markers (n = 27). (B) Bar chart showing the relative distribution of the immunohistochemical tumor types within each anatomy-based diagnosis (n tumor samples = 409), according to the results of hierarchical cluster analysis.
Fig 3.
Complex network visualizations of the hepato-pancreato-biliary tumors in the study (n = 409) based on immunohistochemical data.
Each circle (node) represents an individual tumor sample and is colored according to (A) the immunohistochemical type and (B) the anatomy-based diagnosis.
Fig 4.
(A, C, E) Comparative immunohistograms between the immunohistochemical tumor types. The X-axis shows the significant markers, according to the results of differential expression analysis. The Y-axis indicates the percentage of tumor samples in which marker expression was positive, defined as immunoreactivity in more than 10% of the tumor cells. (B, D, F) Immunohistochemical profile visualizations on the complex network. Each circle (node) represents an individual tumor sample and is colored according to the percentage of stained tumor cells for one characteristic marker of the (B) intestinal, (D) extrahepatic pancreatobiliary, and (F) intrahepatic cholangiocarcinoma immunohistochemical types.
Table 1.
Proposal of diagnostic immunohistochemical panel for the classification of the immunohistochemical tumor types.
Fig 5.
Photomicrographs (10x) of one representative tumor sample for each immunohistochemical tumor type, illustrating their characteristic morphological patterns of marker expression.
Table 2.
Clinicopathological data and results of univariate and multivariate analysis of prognostic factors for overall survival in patients with resected adenocarcinoma of the pancreato-biliary system (n = 86).
In multivariate analysis, overall survival was adjusted by pathological tumor stage (pT) and lymph node status (pN).
Fig 6.
(A-C) Kaplan-Meier plots for overall survival. The p values correspond to log rank test. In multivariate analysis, overall survival was adjusted by pathological tumor stage (pT) and lymph node status (pN).