MicroRNAs and Molecular Imaging: Diagnosis and Personalized Treatment of Pancreatobiliary Cancer

Accurate diagnostic and predictive biomarkers are critically needed in pancreatobiliary cancer to guide appropriate treatment decisions. Reliable pretreatment diagnosis is essential to ensure correct patient selection for systemic therapy, avoid unnecessary treatment or surgery in patients with benign disease, and prevent delays in initiating therapy for malignancies. However, current diagnostic modalities remain limited. Clinical presentation is often non-specific, carbohydrate antigen 19-9 (CA19-9) lacks sufficient discriminatory power, and radiological assessment is challenging due to overlapping imaging features. Although endoscopic procedures may aid diagnosis, they carry procedural risks and frequently fail to yield representative tissue samples. Consequently, a proportion of patients undergoing surgery for presumed malignancy are ultimately found to have benign disease. Moreover, predictive tumor biomarkers to guide the selection of optimal systemic therapy are currently lacking. Together, these limitations highlight the need for improved, minimally invasive diagnostic and predictive strategies. To address these challenges, this thesis explored novel minimally invasive biomarkers and imaging approaches to enhance diagnosis and treatment stratification in pancreatobiliary cancer. The first chapters investigated microRNAs (miRNAs) as diagnostic biomarkers using blood- and bile-based liquid biopsy. Owing to their role in cancer biology, stability in bodily fluids, and tissue-specific expression, miRNAs represent a promising biomarker class; accordingly, both circulating cell-free miRNAs and extracellular vesicle (EV)–associated miRNAs were analyzed to capture complementary molecular information. Although 18F–prostate-specific membrane antigen (PSMA) PET/CT is an established molecular imaging modality routinely used for staging and restaging of prostate cancer, several case reports have described unexpected PSMA tracer uptake in pancreatic ductal adenocarcinoma (PDAC), with subsequent histopathological confirmation. These observations suggest diagnostic potential beyond prostate cancer and provided the rationale for evaluating PSMA PET/CT as a novel imaging approach for PDAC detection. Finally, predictive biomarkers for systemic therapy were investigated to address drug resistance and interpatient variability in treatment response, including the analysis of miRNAs in patients with advanced disease and the evaluation of human equilibrative nucleoside transporter 1 (hENT-1) as a potential biomarker to optimize patient selection and improve individualized therapeutic benefit. In conclusion, by exploring minimally invasive biomarkers and advanced molecular imaging techniques, this work aims to improve diagnostic accuracy, enable better treatment stratification, and support more personalized clinical decision-making in pancreatobiliary cancer. Collectively, the findings presented herein provide a foundation for future translational research and may contribute to the development of clinically applicable strategies to improve outcomes for patients with pancreatobiliary malignancies.