Identification of tissue biomarkers of prognostic significance in pancreatic cancer

Dingyuan Hu

Research output: ThesisDoctoral Thesis (compilation)

450 Downloads (Pure)


Background: Pancreatic cancer is the third leading cause of cancer-related mortality. Lack of early detection strategies and therapeutic resistance are main contributors to the poor prognosis. Unfortunately, there are no tissue biomarkers available for the prognosis of pancreatic cancer in routine clinical use.Aim: To identify and validate novel tissue biomarkers for the prognosis of pancreatic cancer.Methods: A mass spectrometry-based proteomic approach was applied to formalin-fixed paraffin-embedded specimens from surgically resected pancreatic cancer in 9 patients with short survival (<12 months) and 10 patients with long survival (>45 months). The dysregulated biomarkers were further verified by targeted proteomics, parallel reaction monitoring. Finally, we evaluated prognostic candidates (CLCA1, galectin 4, P4HA2, PRTN3 and fibronectin) by tissue microarray and immunohistochemistry in a larger cohort of patients with pancreatic cancer who underwent surgical resection (n=144). Bioinformatic analysis was exploited to assess pathways and networks linked to the prognosis. Kaplan-Meier and Cox proportional hazards modeling were used to explore the association between biomarkers and survival.Results/Conclusion: A total of 24 and 147 proteins were significantly upregulated in patients with short survival and long survival, respectively. Bioinformatic analysis linked proteins representing “activated stroma factors” and “basal tumor factors” to poor prognosis and highlighted TCF1 and CTNNB1 as possible upstream regulators. By targeted proteomics, seven proteins were verified to be upregulated in patients with short survival (MMP9, CLIC3, MMP8, PRTN3, P4HA2, THBS1 and FN1), while 18 proteins were upregulated in patients with long survival, including EPCAM, galectin 4, VIL1, CLCA1 and TPPP3 (I). By immunohistochemical validation, we found that low CLCA1 expression correlated significantly with shorter disease-free survival (II). Furthermore, galectin 4 expression significantly correlated with disease recurrence within 1 year of surgery and with overall survival at 1- and 3-year (III). Besides, a low P4HA2 and high PRTN3 expression pattern correlated with shorter disease-free survival and overall survival (IV). Finally, high stromal FN1 expression was associated with aggressive tumor characteristics in patients with resected pancreatic cancer, although it was not associated with survival (V).
Translated title of the contributionIdentifiering av vävnadsbiomarkörer av prognostisk betydelse vid pankreascancer
Original languageEnglish
Awarding Institution
  • Department of Clinical Sciences, Lund
  • Andersson, Roland, Supervisor
  • Ansari, Daniel, Assistant supervisor
  • Marko-Varga, György, Assistant supervisor
  • Yi, Jiang, Assistant supervisor, External person
Award date2019 May 13
Place of PublicationLund
ISBN (Print)978-91-7619-777-6
Publication statusPublished - 2019

Bibliographical note

Defence details
Date: 2019-05-13
Time: 13:00
Place: Föreläsningssal 1, Centralblocket, Entrégatan 7, Skånes Universitetssjukhus i Lund
External reviewer(s)
Name: Sund, Malin
Title: professor
Affiliation: Umeå University, Umeå, Sverige

Subject classification (UKÄ)

  • Surgery
  • Cancer and Oncology

Free keywords

  • pancreatic cancer
  • proteomics
  • mass spectrometry
  • Biomarkers
  • prognosis
  • CLCA1


Dive into the research topics of 'Identification of tissue biomarkers of prognostic significance in pancreatic cancer'. Together they form a unique fingerprint.

Cite this