Proliferation status assessment in rectal cancer with continuous-time random walk and virtual magnetic resonance elastography.

Rationale And Objectives: To evaluate the diagnostic performance of continuous-time random walk (CTRW), virtual magnetic resonance elastography (vMRE), and diffusion-weighted imaging (DWI) in predicting proliferation status in rectal cancer.

Methods: A cohort of 90 patients with rectal cancer was enrolled. Parameters from CTRW (α, β, and D), vMRE (μ), and DWI (ADC) were compared between high- (n = 52) and low-proliferation (n = 38) groups. Logistic regression (LR), area under the receiver operating characteristic curve (AUC), and DeLong tests were performed for multiparameter analysis, diagnostic efficiency evaluation, and comparison, respectively.

Results: The high-proliferation group showed significantly lower α, β, D, and ADC values, and higher μ values than the low-proliferation group (all P < 0.05). Multivariate analysis identified α, β, D, and ADC as independent predictive markers for proliferation status in rectal cancer (P < 0.001). The joint diagnosis integrating these independent factors yielded an AUC of 0.952 (95% CI: 0.886 ∼ 0.986), significantly surpassing the predictive accuracy of individual parameters such as α, β, D, μ, and ADC (AUCs = 0.864, 0.817, 0.829, 0.792, and 0.802; Z = 2.830, 3.312, 3.059, 3.697, and 3.873; all P < 0.05, respectively). The bootstrap (1000 samples)-based internal validation showed that the joint diagnosis performed well, with an AUC of 0.937 (95% CI: 0.923 ∼ 0.947).

Conclusion: CTRW, vMRE, and DWI effectively assess proliferation status in rectal cancer, and the joint diagnosis integrating α, β, D, and ADC demonstrating superior diagnostic performance.