Islet–exocrine crosstalk in diabetes: Distinguishing whole‐pancreas pathology from causal communication

Aims To critically evaluate how communication between the endocrine and exocrine pancreas contributes to diabetes and to distinguish whole‐pancreas involvement from mechanistically demonstrated crosstalk. Methods We conducted a narrative review of anatomical, imaging, histological, molecular and experimental evidence on endocrine–exocrine communication, prioritising recent primary human and mechanistic studies. Results Islets and exocrine tissue communicate through local vascular, neural, stromal and extracellular matrix routes. Diabetes is associated with distinct whole‐pancreas phenotypes: type 1 diabetes (T1D) shows reduced pancreatic volume, acinar loss, exocrine immune infiltration and periductal or perivascular fibrosis, whereas type 2 diabetes (T2D) more often shows acinar‐to‐ductal metaplasia, lobular or parenchymal fibrosis, angiopathy and increased pancreatic fat fraction on imaging, although histological adiposity is inconsistent. The strongest mechanistic evidence supports insulin‐mediated acinar protection, acinar protease effects on β‐cells and transfer of mutant carboxyl ester lipase in maturity‐onset diabetes of the young type 8 (MODY8), while stromal, matrix and immune signals represent important but less fully validated amplifying pathways. In T1D and T2D, crosstalk most plausibly acts as a stage‐dependent amplifier rather than a universal initiating event. Conclusions Diabetes should be studied as a multicellular pancreatic disorder without interpreting associated pathology as evidence of causation. Longitudinal human studies and multicellular models that allow selective perturbation are required to identify clinically actionable interactions.