Author(s)

  • Tilman Werner (Presenting Author) | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Klara-Luisa Budau | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Miguel Cosenza Contreras | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Frank Hause | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Johanna Thiery | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Annika Topitsch | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Gaia Gentile | Charles River Germany | Am Flughafen 12, 79108, Freiburg, Germany
  • Konrad Kurowski | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Niko Pinter | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Julia Schüler | Charles River Germany | Am Flughafen 12, 79108, Freiburg, Germany
  • Martin Werner | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Carlie Sigel | Memorial Sloan Kettering Cancer Center | 1275 York Avenue , NY 10065, New York, United States
  • Laura Tang | Memorial Sloan Kettering Cancer Center | 1275 York Avenue , NY 10065, New York, United States
  • Peter Bronsert | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany
  • Oliver Schilling | Institut für Klinische Pathologie | Breisacher Straße 115a, 79106, Freiburg, Germany

Abstract

Intrahepatic cholangiocarcinoma (ICC) is associated with poor patient survival due to frequent recurrences and the lack of reliable prognostic markers. In this study, we analyzed 80 treatment-naïve ICC tumors, 77 adjacent non-malignant tissues, and 9 patient-derived xenografts (PDX) via mass spectrometry-based proteomics to identify proteomic profiles linked to patient outcomes.
We found two distinct subclusters with significantly different times to recurrence (TTR). The first cluster, characterized by a high extracellular matrix (ECM) content and increased proteolytic processing, had a mean TTR of 1,125 days. The second cluster, associated with increased mRNA and protein turnover, had a mean TTR of 306 days. A four-protein classifier trained on our data successfully stratified these ICC clusters in the Dong et al. dataset (2022), revealing similar associations between proteomic signatures and clinical outcomes. Furthermore, a mixed-species proteomic analysis of various PDX models underscored the role of tumor-stroma interactions in ICC. Notably, we also identified translation regulator EIF4A1 as enriched in the protein-turnover cluster. Inhibition of EIF4A1 with the novel drug eFT226 significantly reduced tumor growth in an ICC PDX model.
Overall, our study establishes two prognostic proteomic clusters, validates their relevance across datasets, and introduces EIF4A1 inhibition as a potential strategy to mitigate ICC progression.