Objectives <p>This study aimed to investigate the value of contrast-enhanced computed tomography (CECT)-derived extracellular volume (ECV) fraction for preoperative evaluation of tumor-stroma ratio (TSR) in colorectal cancer (CRC).</p> Materials and methods <p>We retrospectively analyzed CRC patients who underwent CECT between July 2021 and July 2023. Patients were randomly allocated into a training cohort (<i>n</i> = 188) and a validation cohort (<i>n</i> = 78). Based on histopathological TSR score, patients were stratified into high-TSR (stroma proportion &gt; 50%) and low-TSR (stroma proportion ≤ 50%) groups. Clinical characteristics and CECT parameters were compared and analyzed. Logistic regression analysis was conducted on the training cohort to identify independent predictors for high TSR. Predictive performance was assessed using receiver operating characteristic (ROC) curves, with area under the curve (AUC) differences compared using the Delong test.</p> Results <p>In the training cohort, the high-TSR group showed significantly lower non-contrast CT value (CTpre) but higher equilibrium phase CT value (CT EP), the difference between venous phase CT value and CTpre (D-value), arterial phase CT enhancement rate (CER), and ECV fraction (all <i>P</i> &lt; 0.05). The ECV fraction demonstrated the highest predictive performance (AUC = 0.808, 95% confidence interval [CI]: 0.745–0.872). CT EP and ECV fraction were identified as independent predictors for high TSR. A model combining ECV fraction and CT EP yielded AUCs of 0.811 (95% CI: 0.748–0.874) and 0.733 (95% CI: 0.623–0.843) in the training and validation cohorts, respectively, with no significant differences compared to the ECV fraction alone (all <i>P</i> &gt; 0.05).</p> Conclusions <p>The ECV fraction serves as a reliable CECT biomarker for preoperative high TSR prediction in CRC, showing potential for noninvasive stroma evaluation and guiding individualized therapeutic strategies.</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

Contrast-enhanced CT-derived extracellular volume fraction in preoperative prediction of tumor-stroma ratio in colorectal cancer

  • Yuting Zhang,
  • Hongyu Zhang,
  • Daming Li,
  • Long Yuan,
  • Liangna Deng,
  • Tao Han,
  • Mengyuan Jing,
  • Chunyu Zhang,
  • Bin Zhang,
  • Junlin Zhou

摘要

Objectives

This study aimed to investigate the value of contrast-enhanced computed tomography (CECT)-derived extracellular volume (ECV) fraction for preoperative evaluation of tumor-stroma ratio (TSR) in colorectal cancer (CRC).

Materials and methods

We retrospectively analyzed CRC patients who underwent CECT between July 2021 and July 2023. Patients were randomly allocated into a training cohort (n = 188) and a validation cohort (n = 78). Based on histopathological TSR score, patients were stratified into high-TSR (stroma proportion > 50%) and low-TSR (stroma proportion ≤ 50%) groups. Clinical characteristics and CECT parameters were compared and analyzed. Logistic regression analysis was conducted on the training cohort to identify independent predictors for high TSR. Predictive performance was assessed using receiver operating characteristic (ROC) curves, with area under the curve (AUC) differences compared using the Delong test.

Results

In the training cohort, the high-TSR group showed significantly lower non-contrast CT value (CTpre) but higher equilibrium phase CT value (CT EP), the difference between venous phase CT value and CTpre (D-value), arterial phase CT enhancement rate (CER), and ECV fraction (all P < 0.05). The ECV fraction demonstrated the highest predictive performance (AUC = 0.808, 95% confidence interval [CI]: 0.745–0.872). CT EP and ECV fraction were identified as independent predictors for high TSR. A model combining ECV fraction and CT EP yielded AUCs of 0.811 (95% CI: 0.748–0.874) and 0.733 (95% CI: 0.623–0.843) in the training and validation cohorts, respectively, with no significant differences compared to the ECV fraction alone (all P > 0.05).

Conclusions

The ECV fraction serves as a reliable CECT biomarker for preoperative high TSR prediction in CRC, showing potential for noninvasive stroma evaluation and guiding individualized therapeutic strategies.