New-Onset Diabetes and Pancreatic Ductal Adenocarcinoma: Implications for Early Recognition and Clinical Management

Abstract

Pancreatic ductal adenocarcinoma (PDAC) remains highly lethal because most patients are diagnosed after curative treatment is no longer feasible. Diabetes mellitus and PDAC have a bidirectional relationship: long-standing type 2 diabetes mellitus (T2DM) modestly increases background cancer risk, whereas new-onset diabetes (NOD), particularly after 50 years of age, may represent a paraneoplastic manifestation of occult PDAC. Population-based data suggest that approximately 0.6%–0.85% of older adults with NOD are diagnosed with PDAC within 3 years. This absolute risk is too low to justify universal imaging, but high enough to support phenotype-based triage. This review clarifies the distinction among T2DM, PDAC-associated diabetes, type 3c diabetes mellitus related to non-malignant exocrine pancreatic disease, and post-pancreatectomy dysglycemia. We also summarise epidemiological and biological evidence connecting dysglycemia with PDAC, including insulin resistance, hyperinsulinemia, impaired insulin secretion, cachexia-related metabolic change, and exocrine-endocrine crosstalk. Particular emphasis is placed on selective diagnostic escalation. Age at diabetes onset, unintended weight loss, HbA1c or glucose trajectory, insulin resistance indices, biomarkers, and risk-enrichment models, including the Enriching New-Onset Diabetes for Pancreatic Cancer (END-PAC) model, should be interpreted together rather than in isolation. Pancreas-protocol computed tomography remains the first-line diagnostic study when PDAC is suspected, while magnetic resonance imaging and endoscopic ultrasonography play complementary roles. We further discuss implications for pancreatic surgery, perioperative metabolic care, postoperative dysglycemia, and lessons from high-risk surveillance cohorts. A diabetes-centred framework may improve early recognition without promoting indiscriminate screening.