The purpose of this study was to preliminarily explore the feasibility of pre-treatment intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) for predicting the efficacy of neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced rectal cancer (LARC). Fifty-four
patients underwent preoperative IVIM-DWI before (pre-nCRT) and after nCRT (post-nCRT). The values of pre-nCRT and post-nCRT IVIM-DWI parameters (apparent diffusion coefficient standard (ADC), true diffusion coefficient (D), pseudodiffusion coefficient (D*) and perfusion fraction (f),
together with the percentage changes (Δ% parametric value) induced by nCRT, were compared between pathological complete response (pCR) and non-pCR groups. After nCRT, the ADC and Dvalues for LARC increased significantly (all P <0.05). The TRG score revealed positive
correlations with pre f (r = 0.357, P = 0.020), postD (r =0.551,P <0.001), and Δ%D (r =0.605, P <0.001). The pCR group (n =10) had higher preD *, pref, postD, Δ%ADC, and Δ%D values than the non-pCR group
(n =32) (allP <0.05). The GR group (n =15) exhibited higher postD, %ADC, and %D values than the PR group (n =27) (allP <0.05). Based on ROC analysis, Δ%D had a higher area under the curve value than Δ%ADC (P =0.009) in discriminating
the pCR and non-pCR groups. IVIM-DWI may be helpful in identifying pCR to nCRT for LARC and is more accurate than traditional DWI.
A Deep Learning Model to Predict the Response to Neoadjuvant Chemoradiotherapy by the Pretreatment Apparent Diffusion Coefficient Images of Locally Advanced Rectal Cancer
