Diffusion-weighted magnetic resonance imaging in monitoring rectal cancer response to neoadjuvant chemoradiotherapy.

PURPOSE: To prospectively monitor the response in patients with locally advanced nonmucinous rectal cancer after chemoradiotherapy (CRT) using diffusion-weighted magnetic resonance imaging. The histopathologic finding was the reference standard. METHODS AND MATERIALS: The institutional review board approved the present study. A total of 62 patients (43 men and 19 women; mean age, 64 years; range, 28-83) provided informed consent. T(2)- and diffusion-weighted magnetic resonance imaging scans (b value, 0 and 1,000 mm(2)/s) were acquired before, during (mean 12 days), and 6-8 weeks after CRT. We compared the median apparent diffusion coefficients (ADCs) between responders and nonresponders and examined the associations with the Mandard tumor regression grade (TRG). The postoperative nodal status (ypN) was evaluated. The Mann-Whitney/Wilcoxon two-sample test was used to evaluate the relationships among the pretherapy ADCs, extramural vascular invasion, early percentage of increases in ADCs, and preoperative ADCs. RESULTS: Low pretreatment ADCs (<1.0 × 10(-3)mm(2)/s) were correlated with TRG 4 scores (p = .0011) and associated to extramural vascular invasion with ypN+ (85.7% positive predictive value for ypN+). During treatment, the mean percentage of increase in tumor ADC was significantly greater in the responders than in the nonresponders (p < .0001) and a >23% ADC increase had a 96.3% negative predictive value for TRG 4. In 9 of 16 complete responders, CRT-related tumor downsizing prevented ADC evaluations. The preoperative ADCs were significantly different (p = .0012) between the patients with and without downstaging (preoperative ADC ≥1.4 × 10(-3)mm(2)/s showed a positive and negative predictive value of 78.9% and 61.8%, respectively, for response assessment). The TRG 1 and TRG 2-4 groups were not significantly different. CONCLUSION: Diffusion-weighted magnetic resonance imaging seems to be a promising tool for monitoring the response to CRT.

Restaging locally advanced rectal cancer with MR imaging after chemoradiation therapy.

In recent years, preoperative therapy has become standard procedure for locally advanced rectal cancer. Tumor shrinkage due to preoperative chemotherapy-radiation therapy (CRT) is now a reality, and pathologically complete responses are not uncommon. Some researchers are now addressing organ preservation, thus increasing the demand for both functional and morphologic radiologic evaluation of response to CRT to distinguish responding from nonresponding tumors. On magnetic resonance (MR) images, post-CRT tumor morphologic features and volume changes have a high positive predictive value but a low negative predictive value for assessing response. Preliminary results indicate that diffusion-weighted MR imaging, especially at high b values, would be effective for prediction of treatment outcome and for early detection of tumor response. Some authors have reported that the use of apparent diffusion coefficient values in combination with other MR imaging criteria significantly improves discrimination between malignant and benign lymph nodes. Sequential determination of fluorodeoxyglucose uptake at positron emission tomography/computed tomography has proved useful in differentiating responding from nonresponding tumors during and at the end of CRT. However, radionuclide techniques have limitations, such as low spatial resolution and high cost. Large studies will be needed to verify the most effective morphologic and functional imaging modalities for post-CRT restaging of rectal cancer. Supplemental material available at http://radiographics.rsna.org/lookup/suppl/doi:10.1148/rg.303095085/-/DC1.