Related Experiment Video
Updated: May 28, 2025
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
Advanced diffusion-weighted imaging biomarkers for non-invasive assessment of tumor microenvironment in rectal
Jie Yuan1, Yiqun Sun2,3, Kun Liu4
1Department of Radiology, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese Medicine, Shanghai, China.
Purpose:
To explore the heterogeneity of the rectal cancer microenvironment (TME) using restricted spectrum imaging (RSI) and investigate its association with tumor stroma and Ki67 as key histopathologic indicators.
Materials And Methods:
In this prospective study, 66 patients with rectal cancer underwent pretreatment MRI with RSI. The optimal model format was determined by Bayesian Information Criterion (BIC). RSI3-derived parameters (RSI3-C1, RSI3-C2, RSI3-C3) and ADC values were measured and correlated with stroma status, Ki67 expression, and clinicopathological features. The diagnostic performance of these quantitative imaging biomarkers was assessed using receiver operating characteristic (ROC) analysis.
Results:
The three-compartment RSI model (RSI3) was optimal for characterizing rectal cancer (△BIC = 0). RSI3-C1, RSI3-C2, and RSI3-C3 showed significant differences between low-stroma and high-stroma groups (P < 0.05). RSI3-C2 exhibited the highest accuracy in characterizing stroma status (AUC = 0.800, sensitivity = 79.2%, specificity = 71.4%). All RSI3 parameters and ADC values differed significantly between low-Ki67 and high-Ki67 groups (P < 0.05). RSI3-C1 demonstrated the highest accuracy in characterizing Ki67 status (AUC = 0.824, sensitivity = 90.0%, specificity = 69.4%). Significant differences were observed in RSI3-C3 and ADC values for tumor differentiation (P < 0.05). RSI3-C3 showed the highest accuracy in characterizing differentiation status (AUC = 0.721, sensitivity = 66.7%, specificity = 83.3%).
Conclusion:
RSI3-derived parameters show potential as non-invasive biomarkers for evaluating TME in rectal cancer. This innovative approach may improve decision-making, leading to better patient outcomes in rectal cancer management.
