Assessment of pathological complete response to preoperative chemoradiotherapy by means of multiple mathematical models of diffusion-weighted MRI in locally advanced rectal cancer: A prospective single-center study.

Zhu, Hai‐Bin; Zhang, Xiao‐Yan; Zhou, Xiao‐Hong; Li, Xiao‐Ting; Liu, Yu‐Liang; Wang, Shuai; Sun, Ying‐Shi; Zhu, Hai-Bin; Zhang, Xiao-Yan; Zhou, Xiao-Hong; Li, Xiao-Ting; Liu, Yu-Liang; Sun, Ying-Shi

<bold>Purpose: </bold>To assess stretched-exponential, mono-exponential and intravoxel incoherent motion (IVIM) models of diffusion-weighted MRI(DWI) in predicting pathological complete response (pCR) to neoadjuvant chemoradiotherapy (CRT) in rectal cancer patients.<bold>Materials and Methods: </bold>This prospective study recruited 98 consecutive patients with locally advanced rectal cancer who underwent 3 Tesla MR examination before, during and after CRT. The apparent diffusion coefficient (ADC), IVIM-derived parameters (D, f, and D*), and stretched-exponential model-derived parameters (DDC and α) were measured. The parameters and their corresponding changes during and after CRT were compared between pCR and non-pCR. Receiver-operating characteristic curve analysis was performed to evaluate the diagnostic performance. Coefficient of variations and intraclass correlation coefficient were calculated to assess reliability and agreement.<bold>Results: </bold>Nineteen patients achieved pCR while 79 did not. The pCR group had higher ADC and α (ADC2 and α2 ), and their changes (ΔADC2 , and Δα2 ) at the endpoint than non-pCR group. α2 and ADC2 yielded similar AUCs (P = 0.339), Δα2 and ΔADC2 yielded similar AUCs (P = 0.263) ADC and α presented substantial agreement, and α presented the minimum CV (5.0-7.0%).<bold>Conclusion: </bold>ADC and α were useful for assessing pCR after CRT. α might be more useful because it demonstrated better diagnostic performance than IVIM-derived parameters and better reliability than ADC.<bold>Level Of Evidence: </bold>1 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:175-183.

Journal of Magnetic Resonance Imaging, 2017, Vol 46, Issue 1, p175

1053-1807

Academic Journal

10.1002/jmri.25567