Objectives To evaluate new multidetector CT (MDCT) criteria as predictors of resection margin status in pancreatic body and tail adenocarcinoma (PBTA). Materials and methods This retrospective study enrolled PBTA patients who underwent pancreatectomy. Resection margin status was classified as R0 and R1/R2. Clinical and radiological features were analyzed, including tumor resectability status based on National Comprehensive Cancer Network (NCCN) criteria, tumor size, tumor-vascular contact, rim enhancement, peripancreatic fat stranding, adjacent organ invasion, and the minimum distance between the tumor and the anterior (TAS) and posterior (TPS) surfaces of the pancreas. Logistic regression analyses were performed to identify independent predictors of resection margin status, with model performance assessed using ROC curves and DeLong tests. Results A total of 318 PBTA patients were randomly assigned to a training cohort (n = 222) and a validation cohort (n = 96) in a 7:3 ratio. Among them, 115 (51.8%) and 46 (47.9%) patients underwent R1/R2 resection, respectively. Independent predictors of R1/R2 resection included tumor resectability status (odds ratio (OR), 2.79), tumor size (OR, 3.17), rim enhancement (OR, 3.28), peripancreatic fat stranding (OR, 2.38), and TAS (OR, 8.91) (all p < 0.05). Model B (including all five independent predictors) outperformed Model A (including four independent predictors without TAS) in predicting the resection margin status in both cohorts (training: AUC 0.840 vs 0.757, p = 0.001; validation: 0.810 vs 0.711, p = 0.029). Conclusion In addition to tumor resectability status, tumor size, rim enhancement, and peripancreatic fat stranding, TAS emerges as a potential predictor of resection margin status in PBTA patients. Key Points <Emphasis Type="BoldItalic">Question</Emphasis> Resection margin status is associated with overall survival in pancreatic body and tail adenocarcinoma (PBTA), and predicting resection margin status is vital for risk stratification. <Emphasis Type="BoldItalic">Findings</Emphasis> Clinical and CT-based radiological features, particularly the minimum tumor-to-anterior-surface (TAS) distance, can assist in preoperatively predicting the resection margin status in PBTA patients. <Emphasis Type="BoldItalic">Clinical relevance</Emphasis> The nomogram integrating clinical and radiological features provides a reliable tool for preoperative prediction of resection margin status, supporting individualized and optimized treatment planning for PBTA patients. Graphical Abstract

Tumor-to-anterior pancreas surface distance on CT: a potential preoperative predictor of resection margin status in pancreatic body and tail adenocarcinoma

期刊论文
发表日期：

2025年12月19日

查看全文

Yu-Yu Tu,
Jie Yin,
Si-Yao Yu,
Yu-Ping Shu,
Lin Fang,
Jing Yu,
Kui-Rong Jiang,
Fei-Yun Wu,
Qing Xu

摘要

Objectives

To evaluate new multidetector CT (MDCT) criteria as predictors of resection margin status in pancreatic body and tail adenocarcinoma (PBTA).

Materials and methods

This retrospective study enrolled PBTA patients who underwent pancreatectomy. Resection margin status was classified as R0 and R1/R2. Clinical and radiological features were analyzed, including tumor resectability status based on National Comprehensive Cancer Network (NCCN) criteria, tumor size, tumor-vascular contact, rim enhancement, peripancreatic fat stranding, adjacent organ invasion, and the minimum distance between the tumor and the anterior (TAS) and posterior (TPS) surfaces of the pancreas. Logistic regression analyses were performed to identify independent predictors of resection margin status, with model performance assessed using ROC curves and DeLong tests.

Results

A total of 318 PBTA patients were randomly assigned to a training cohort (n = 222) and a validation cohort (n = 96) in a 7:3 ratio. Among them, 115 (51.8%) and 46 (47.9%) patients underwent R1/R2 resection, respectively. Independent predictors of R1/R2 resection included tumor resectability status (odds ratio (OR), 2.79), tumor size (OR, 3.17), rim enhancement (OR, 3.28), peripancreatic fat stranding (OR, 2.38), and TAS (OR, 8.91) (all p < 0.05). Model B (including all five independent predictors) outperformed Model A (including four independent predictors without TAS) in predicting the resection margin status in both cohorts (training: AUC 0.840 vs 0.757, p = 0.001; validation: 0.810 vs 0.711, p = 0.029).

Conclusion

In addition to tumor resectability status, tumor size, rim enhancement, and peripancreatic fat stranding, TAS emerges as a potential predictor of resection margin status in PBTA patients.

Key Points

Question Resection margin status is associated with overall survival in pancreatic body and tail adenocarcinoma (PBTA), and predicting resection margin status is vital for risk stratification.

Findings Clinical and CT-based radiological features, particularly the minimum tumor-to-anterior-surface (TAS) distance, can assist in preoperatively predicting the resection margin status in PBTA patients.

Clinical relevance The nomogram integrating clinical and radiological features provides a reliable tool for preoperative prediction of resection margin status, supporting individualized and optimized treatment planning for PBTA patients.

Graphical Abstract