Integrative translational biomarkers to guide therapy in pancreatic ductal adenocarcinoma: from rare molecular subsets to resectable and metastatic disease.

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal malignancies and, despite the therapeutic advances, 5-year survival rates remain drastically low. The major challenge in PDAC management is the marked biological heterogeneity of the disease and the lack of robust biomarkers to guide treatment selection. This thesis aimed to identify translational biomarkers capable of improving patient stratification and supporting precision medicine approaches across different molecular and clinical settings of PDAC. In this work, we firstly explored rare molecular subsets of pancreatic cancer with potential therapeutic vulnerabilities. A narrative review summarized emerging targeted strategies in uncommon histological and molecular subtypes, including KRAS wild-type (wt) PDAC. In an original study, early-onset KRAS-wt PDAC was shown to enrich for actionable genomic alterations, supporting broader use of comprehensive genomic profiling in younger patients. In resectable PDAC, biomarkers analysis from phase 2 nITRO perioperative trial demonstrated that circulating inflammatory biomarkers, particularly TNF-α, were associated with treatment response, surgical eligibility, and survival outcomes. Germline alterations in DNA damage repair genes were linked to enhanced sensitivity to platinum-based therapy, whereas DPYD c.496A>G variant was associated with severe treatment-related gastrointestinal toxicity. In metastatic PDAC, translational analyses from the randomized phase 2 NAPAN trial identified IL-8 as a robust independent prognostic biomarker associated with shorter survival outcomes. In parallel, KRAS-mutant (KRASMUT) circulating tumor DNA (ctDNA) detection was associated with poorer outcomes, while longitudinal KRASMUT ctDNA dynamics reflected radiological tumor load changes during treatment. Overall, this thesis demonstrates that the integration of genomic, inflammatory, and circulating biomarkers can improve therapeutic decision-making in rarer molecular subsets and in different clinical stages of PDAC. These findings support the transition from empirical treatment approaches toward a more personalized and biologically informed management of the disease.