Plasma protein biomarkers for the detection of pancreatic neuroendocrine tumors and differentiation from small intestinal neuroendocrine tumors

• DOI: 10.1111/jne.13176
• Publication Date: 2022-06-15T06:54:35Z
• Authors: Espen Thiis‐Evensen; Magnus Kjellman; Ulrich Knigge; Henning Gronbaek; Camilla Schalin‐J...; Staffan Welin; Halfdan Sorbye; Maria del Pilar S...; Roger Belusa
• Keywords: Neuroendocrine differentiation

Abstract There is an unmet need for novel biomarkers to diagnose and monitor patients with neuroendocrine neoplasms. The EXPLAIN study explores a multi‐plasma protein supervised machine learning strategy improve the diagnosis of pancreatic tumors (PanNET) differentiate them from small intestinal (SI‐NET). At time diagnosis, blood samples were collected analyzed 39 PanNET, 135 SI‐NET (World Health Organization Grade 1–2) 144 controls. Exclusion criteria other malignant diseases, chronic inflammatory reduced kidney or liver function. Prosed Oncology‐II (i.e., OLink) was used measure 92 cancer related plasma proteins. Chromogranin A separately. Median age in all groups 65–67 years similar sex distribution (females: 51%; SI‐NET, 42%; controls, 42%). Tumor grade (G1/G2): 39/61%; 46/54%. Patients metastases: 78%; 63%. classification model PanNET versus controls provided sensitivity (SEN) 0.84, specificity (SPE) 0.98, positive predictive value (PPV) 0.92 negative (NPV) 0.95, area under receiver operating characteristic curve (AUROC) 0.99; discrimination providing SEN 0.61, SPE 0.96, PPV 0.83, NPV 0.90 AUROC 0.98. These results suggest that can significantly diagnostic accuracy SI‐NET.