Scholarlink

Background <p>Pancreatic cancer (PC) is one of the most aggressive malignancies, with a poor prognosis and a high rate of postoperative recurrence. Patients with early recurrence typically do not benefit from surgery. Therefore, identifying high-risk patients for early tumor recurrence before surgery is critical for developing personalized treatment strategies. This study is to investigate the clinical value of a multi-scale model, constructed using CT and pathological images, in predicting early recurrence risk following curative surgery for pancreatic cancer.</p> Methods <p>A total of 121 patients with pancreatic ductal adenocarcinoma (PDAC) who underwent radical surgery at our hospital between February 2018 and October 2022 were included in this study. A retrospective analysis was conducted on the clinical, CT imaging, and histopathological data of all patients, with follow-up until December 2023. Based on clinical factors, radiomics features, and pathomics features, we developed clinical, radiomics, pathomics, and integrated model. A nomogram was generated to visually analyze the integrated model. The optimal cutoff values were determined using X-tile software, and all patients were stratified into high-risk and low-risk groups. Kaplan-Meier survival curves were plotted, and Log-rank tests were performed to assess the significance of survival differences between the groups. The predictive performance of the four models was evaluated using the concordance index (C-index).</p> Results <p>Multivariate Cox regression analyses identified postoperative CA19-9 levels and perineural invasion as independent predictors of early recurrence. The integrated model demonstrated superior predictive performance compared to other individual models, with C-index values of 0.752 (95% CI: 0.681–0.814) in the training cohort and 0.740 (95% CI: 0.680–0.793) in the validation cohort.</p> Conclusion <p>The multi-scale integrated model incorporating clinical parameters, radiomics features, and pathomics signatures demonstrates superior predictive accuracy for early recurrence in pancreatic cancer patients, enabling precise risk stratification and supporting personalized adjuvant treatment planning.</p> Clinical trial number <p>Not applicable.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Postoperative early recurrence prediction of pancreatic cancer based on CT and pathomics

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月28日

查看全文
  • Wei Cao,
  • Jing-yu Li,
  • Min Wang,
  • Wu-yang Zhang,
  • Bin Wang,
  • Dan Shi,
  • Meng-ru Li,
  • Wei Wei

摘要

Background

Pancreatic cancer (PC) is one of the most aggressive malignancies, with a poor prognosis and a high rate of postoperative recurrence. Patients with early recurrence typically do not benefit from surgery. Therefore, identifying high-risk patients for early tumor recurrence before surgery is critical for developing personalized treatment strategies. This study is to investigate the clinical value of a multi-scale model, constructed using CT and pathological images, in predicting early recurrence risk following curative surgery for pancreatic cancer.

Methods

A total of 121 patients with pancreatic ductal adenocarcinoma (PDAC) who underwent radical surgery at our hospital between February 2018 and October 2022 were included in this study. A retrospective analysis was conducted on the clinical, CT imaging, and histopathological data of all patients, with follow-up until December 2023. Based on clinical factors, radiomics features, and pathomics features, we developed clinical, radiomics, pathomics, and integrated model. A nomogram was generated to visually analyze the integrated model. The optimal cutoff values were determined using X-tile software, and all patients were stratified into high-risk and low-risk groups. Kaplan-Meier survival curves were plotted, and Log-rank tests were performed to assess the significance of survival differences between the groups. The predictive performance of the four models was evaluated using the concordance index (C-index).

Results

Multivariate Cox regression analyses identified postoperative CA19-9 levels and perineural invasion as independent predictors of early recurrence. The integrated model demonstrated superior predictive performance compared to other individual models, with C-index values of 0.752 (95% CI: 0.681–0.814) in the training cohort and 0.740 (95% CI: 0.680–0.793) in the validation cohort.

Conclusion

The multi-scale integrated model incorporating clinical parameters, radiomics features, and pathomics signatures demonstrates superior predictive accuracy for early recurrence in pancreatic cancer patients, enabling precise risk stratification and supporting personalized adjuvant treatment planning.

Clinical trial number

Not applicable.