Multimodal liquid biopsy surveillance after pancreatectomy for pancreatic ductal adenocarcinoma: evidence and a confirmation‑gated, trial‑ready framework.

Relapse after curative‑intent pancreatectomy for pancreatic ductal adenocarcinoma (PDAC) is common and may become clinically apparent only after a delay in conventional imaging and CA19‑9 monitoring. Liquid biopsy offers serial access to tumor‑derived signals, but post‑operative minimal residual disease (MRD) is often near the assay limit of detection, raising risks of false positives, over‑imaging, and premature treatment escalation outside trials. We performed a scoping review of clinical studies evaluating circulating tumor DNA (ctDNA), circulating tumor cells (CTCs), and extracellular vehicles (EVs) in resected or curatively treated PDAC, focusing on recurrence prediction, lead time, and practical implementation. Across cohorts, tumor‑informed ctDNA and longitudinal CTC kinetics show consistent prognostic value and may precede radiographic recurrence by weeks to months, although sensitivity varies with tumor shedding patterns and assay design. EV markers and signatures provide complementary biology and may be informative in low‑shedding states, but reproducibility is limited by pre‑analytical and platform heterogeneity, necessitating conservative interpretation. We propose a pragmatic, confirmation‑gated trigger algorithm for "liquid‑positive, imaging‑negative" scenarios: borderline signals are considered indeterminate and prompt repeat sampling and/or orthogonal confirmation; imaging is expedited when ctDNA converts on serial draws or when concordant positivity is observed across two analytes (ctDNA/CTC/EV); and systemic therapy changes should be avoided outside prospective utility‑driven trials. Finally, we outline trial designs (ctDNA‑positive escalation, ctDNA‑negative de‑escalation, and strategy comparisons) incorporating survival, decision‑curve, and health‑economic endpoints to establish clinical utility.