Show simple item record

dc.contributor.authorZardab, Men_US
dc.date.accessioned2023-07-24T13:30:20Z
dc.date.issued2023
dc.identifier.urihttps://qmro.qmul.ac.uk/xmlui/handle/123456789/89739
dc.description.abstractPancreatic ductal adenocarcinoma (PDAC) has poor overall survival due to infrequency, late presentation and lack of diagnostic biomarkers or effective therapies. Prediction models harnessing machine learning may help in early diagnosis. AHNAK2 is associated with oncogenesis in multiple cancers, has potentially a poor prognostic biomarker potential; however its role is unclear. A case-control study was performed on a prospective cohort of PDAC (N=344) and benign pancreatic disease patients (N=360) recruited between 2008 and 2021 from the Barts Pancreatic Tissue Bank (BPTB) with a prediction model trained using routinely-collected clinical variables. Enzyme linked immunosorbent assays (ELISA) were performed on PDAC patients (N=30) and matches healthy controls (N=30) recruited from the BPTB measuring AHNAK2 concentration in plasma. Immunohistochemistry (IHC) of PDAC patient samples (N=14) was performed. In vitro experiments on multiple PDAC cell lines were performed to explore the mechanistic role of AHNAK2. Increased odds of PDAC were observed for age >75 years (Odds ratio (OR) 10.1 [95% confidence interval, 5.6-18.2]; p<0.001 Fisher’s exact test), jaundice and weight loss (OR 3.12 [1.86-5.21] and 3.06 [2.03-4.6] p<0.001). Reduced sodium (OR 4.2 [1.6-10.6]; p=0.006), red blood cell count (OR 0.44 [0.30-0.63]; p<0.001), Creatinine (OR 3.1 [1.4-6.6]; p=0.008) and raised Bilirubin (OR 4.8 [2.9-8.1]; p<0.001), Alkaline phosphatase (ALP) (OR 4.8 [2.9-8.1]; p<0.001) and Alanine Transferase (ALT) (OR 2.6 [1.5-4.4]; p=0.002), were associated with PDAC. Potentially 96.8% sensitivity can be reached with the prediction model (AUROC=0.90 [0.85, 0.94]; Spiegelhalter’s z=-1.8, p=0.07) using routine clinical parameters leading to earlier detection of 84.7% of the PDAC patients using machine learning. AHNAK2 levels discriminated between PDAC and healthy controls (p<0.001, χ2 test) with an AUROC of 0.903 [0.822, 0.985] performing better than CA 19-9. High tissue AHNAK2 expression was associated with stage III-IV disease (p=0.036, Wilcoxon rank sum test). AHNAK2 co-localised with Cortactin, an essential protein in cellular protrusion and invasion in cancer, in multiple cell lines which was attenuated in hypoxia and on Fibronectin and Collagen I substrates. Knockdown of AHNAK2 reduced invasion of Capan-2 and PS1 co-culture spheroid models significantly (p<0.001, Wilcoxon rank sum test). Further validation work is required for the clinical model for earlier detection of PDAC in primary care. AHNAK2 plasma levels could potentially to be used as a diagnostic biomarker in PDAC. There is strong evidence of AHNAK2s role in oncogenesis in PDAC, particularly in the formation of cellular protrusions and invasion in a hypoxia. Therefore, AHNAK2 could be a potential therapeutic target.en_US
dc.language.isoenen_US
dc.titleInvestigating the role of AHNAK2 in Pancreatic Ductal Adenocarcinomaen_US
pubs.notesNot knownen_US
rioxxterms.funderDefault funderen_US
rioxxterms.identifier.projectDefault projecten_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

  • Theses [4235]
    Theses Awarded by Queen Mary University of London

Show simple item record