A Novel Radiomics-Integrated Panel for Preoperative Stratification of Pancreatic Neuroendocrine Tumors (PNETs).

Document Type

Article

Publication Date

5-1-2026

Publication Title

Cancers

Abstract

Simple Summary: Pancreatic neuroendocrine tumors (PNETs) are clinically heterogeneous neoplasms whose preoperative risk stratification is limited by the unavailability of histologic grade before surgery. We tested whether quantitative features extracted from routine preoperative CT can support preoperative risk stratification in a two-center retrospective cohort of 44 patients. We constructed a small panel of biologically informed hybrid signatures combining lesion radiomic primitives with clinical variables. We additionally constructed Δ-radiomic features that use each patient's contralateral pancreas as an internal control. After ComBat batch correction across the two contributing centers, the Δ-radiomic signature B2 (ΔBusyness × Ki-67) was the most consistent progression-associated signature on univariable Cox regression, and a parsimonious classifier built on Family B Δ-signatures exceeded the clinical baseline for progression discrimination while maintaining acceptable calibration. The findings support radiomics as a candidate adjunct to preoperative risk assessment in PNETs and motivate prospective external validation. Background. Preoperative risk stratification of pancreatic neuroendocrine tumors (PNETs) is constrained by the unavailability of histologic grade before resection. We hypothesized that a panel of biologically informed CT-radiomic signatures, combined with patient-level Δ-radiomics referenced to the contralateral pancreas, would support preoperative discrimination of progression and grade in a two-center pilot cohort. Methods. Forty-four patients with histologically confirmed PNET who underwent contrast-enhanced preoperative CT and surgical resection at two academic centers were analyzed. Lesion and contralateral non-tumor-bearing pancreatic parenchyma regions of interest were revised in 3D Slicer by a board-certified pancreatic surgeon and verified intraoperatively against surgical pathology. PyRadiomics v3.0 features were extracted with IBSI-concordant settings. Parametric ComBat batch correction was applied across the two centers (biological-covariate balance verified beforehand), and Δ-radiomic features (lesion combat–pancreas combat) were computed for the 106 intensity/texture primitives. We constructed a panel of biology-informed hybrid signatures partitioned into a preoperative lesion-only family (Family A; seven signatures) and a preoperative Δ-radiomic family (Family B; three signatures). Candidate features were filtered through correlation clustering, baseline-adjusted likelihood-ratio testing with Benjamini–Hochberg FDR control, and 100-bootstrap stability selection. Three predictor blocks were compared per target with three classifiers each (Logistic Regression, Random Forest, Gradient Boosting): M0 (five-variable clinical baseline), MA (M0 + Family A), and MB (M0 + Family B). Discrimination was reported as AUC with bootstrap 95% CI; calibration was assessed using the Brier score and TRIPOD-recommended calibration intercept and slope; and cross-center generalization was evaluated with leave-one-center-out (LOCO) cross-validation. Univariable Cox regression with bootstrap and permutation inference was used for progression-free survival (PFS). Results. The cohort had 16 progression events and eight deaths (median follow-up was 38 months, IQR 14–59). Prespecified clinical–radiomic and Δ-radiomic signatures were associated with progression-free survival, including B2 = ΔBusyness × Ki-67 (HR 0.38, 95% CI 0.19–0.76, p = 0.006). For progression prediction, the Δ-radiomic model achieved the strongest discrimination, with a nested cross-validation AUC of 0.85 and leave-one-center-out AUC of 0.87. For higher-grade disease, radiomic models also demonstrated high discrimination, with AUCs up to 0.93. Conclusions. Radiomics-derived shape and texture features, especially when combined with clinical markers, may noninvasively identify aggressive PNET phenotypes and support preoperative risk stratification. Prospective validation in larger multicenter cohorts is warranted.

First Page

1663

Volume

18

Issue

10

Publisher

MDPI

Rights

© 2026 by the authors

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Link to Full Text

Share

COinS