Fifteen patients (median age 56 years, range 32-69 years; 13 men and 2 women) with LARC were invited to participate in our study between June 2006 and December 2007. Each patient had histologically proven rectal adenocarcinoma of stage T3-T4 and was determined to be node-positive by endorectal ultrasound and pelvic MRI. Patients with a history of pelvic irradiation or chemotherapy, any other malignancy, or distant metastases were excluded (Table 1). The clinical and histopathological classification and stage according to the International Union Against Cancer TNM system[13] were recorded. Tumor regression grading was evaluated according to the criteria of Dworak et al[14] (grade 0, no regression; grade 1, minor regression, dominant tumor mass with obvious fibrosis in 25% or less of the tumor mass; grade 2, moderate regression, 26%-50% of the tumor mass; grade 3, good regression, more than 50% tumor regression; and grade 4, total regression, no viable tumor cells, only fibrotic mass). A pCR was defined as the absence of viable tumor cells in the primary tumor and lymph nodes (ypT0N0). T downstaging was defined as the downstaging of the tumor from cT3 to ypT0-2 or from cT4 to ypT0-3. Good regression was defined as TRG 3-4, and poor regression was defined as TRG 0-2.

All patients received neoadjuvant concurrent CRT. Radiotherapy (RT) was delivered with a linear accelerator using 6- and 15-MV photons and a three-field technique (posterior-anterior and right and left laterals). Every patient underwent a planning computed tomography (CT) scan in the treatment position (prone position) using a belly board. Three-dimensional conformal RT was used for all patients based on the planning CT, with a total dose of 45 Gy at 1.8 Gy per fraction per day, Monday-Friday. Neoadjuvant chemotherapy was delivered concurrently with RT. Starting on day 1 of RT, patients received capecitabine 625 mg/m² orally, bid (Monday-Friday), and oxaliplatin 50 mg/m² weekly for five consecutive weeks. Surgical resection was scheduled for 5-8 wk after the completion of neoadjuvant treatment.

Each enrolled patient was examined by DW-MRI at six scheduled times. The initial DW-MRI scan was performed 7 d prior to the start of RT. DW-MRI scans were then taken once weekly during the course of neoadjuvant treatment.

DWI was performed on a 1.5 T magnetic resonance machine (1.5 T Signa Twin Speeder with Excite, GE, United States) using a phased-array body coil. Before DW-MRI, standard T2-weighted fast spin echo sequence and T1-weighted spin echo sequence images were used for clinical staging. DWI echo planar images were acquired in the transverse plane using a GRE-EPI sequence (TR/TE 3000/min; field of view 22 cm²; matrix size 128× 128; slice thickness 4 mm; intersection gap 1 mm). DW-MR images and ADC maps were obtained using b values of 0 and 1000 s/mm² applied in the x, y, and z directions. Patients did not undergo bowel preparation, receive anti-spasmodic medication, or undergo rectal distention before the MR examination. For the image analysis, the data were transferred to a Workstation (AW4.0, GE Medical Systems) and analyzed using the Functool dynamic analysis tool (GE Medical Systems). ADC values were calculated based on the ADC maps. The ADC map of the largest tumor extension in the transverse T2-weighted images was used for the analysis. Regions of interest (ROIs) were drawn manually along the edge of the tumor with a b value of 1000 s/mm² on the selected ADC maps by an experienced radiologist (Zhang S, with 10 years of experience in clinical MRI), who did not participate in the treatment of the patients or the evaluation of the therapeutic effect.

Statistical analysis was performed using SPSS 12.0 statistical software. Paired comparisons were performed using the Wilcoxon test. Spearman’s correlation was used to assess the significance of differences between groups. A P value < 0.05 was considered statistically significant.

After neoadjuvant treatment, pCR was observed in 2 patients. Downstaging of the tumor was observed in eight patients. The tumor regression grades after neoadjuvant treatment were grade 0-2 in 6 patients and grade 3-4 in 9 patients.

Diffusion data from 15 patients were obtained prior to and at constant intervals once weekly during the course of neoadjuvant treatment. The observed ADC values are shown in Table 2. A total of 88 ADC values were obtained in our study, and 2 ADC values were excluded due to measurement errors. Sample T2-weighted and diffusion-weighted images prior to treatment are shown in Figure 1. Sample ADC maps from the images taken weekly during the course of neoadjuvant treatment are shown in Figure 2.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Sample T2-weighted and diffusion-weighted images from a patient with locally advanced rectal cancer prior to treatment. A: T2-weighted transaxial image through the pelvis prior to treatment; B: Corresponding apparent diffusion coefficient map of the patient obtained from the diffusion-weighted images. A region of interest was drawn around the tumor.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Apparent diffusion coefficient maps of one patient obtained at constant intervals once weekly during the course of neoadjuvant treatment from the diffusion-weighted images. A-E: Axial apparent diffusion coefficient maps obtained during treatment (weeks 1-5). A region of interest was drawn around the tumor.

The mean tumor ADC value slightly increased from 0.749 × 10^-3 mm²/s (95%CI: 0.641 × 10^-3-0.858 × 10^-3 mm²/s) prior to treatment to 0.772 × 10^-3 mm²/s (95%CI: 0.627 × 10^-3-0.918 × 10^-3 mm²/s) after the 1^st week of treatment. There was also a significant increase at the 2^nd week to 0.884 × 10^-3 mm²/s 95%CI: 0.775 × 10^-3-0.994 × 10^-3 mm²/s). Subsequently, the mean ADC decreased to 0.800 × 10^-3 mm²/s (95%CI: 0.675 × 10^-3-0.925 × 10^-3 mm²/s) at the 3^rd week and 0.766 × 10^-3 mm²/s (95%CI: 0.659 × 10^-3-0.872 × 10^-3 mm²/s) at the 4^th week. ADC increased again at the 5^th week to 0.839 × 10^-3 mm²/s (95%CI: 0.702 × 10^-3-0.976 × 10^-3 mm²/s). We also observed a significant increase in the mean ADC value at the 2^nd (P = 0.004) and 5^th week (P = 0.033) during treatment relative to the values prior to treatment. The mean observed ADC values and P values are shown in Table 3.

We compared the tumor ADC values of the responder and non-responder groups to predict the treatment response based on T downstage and TRG criteria.

Downstaging of the tumor was observed for 8 of the 15 patients (53.3%). The ADC values at the 5^th week during treatment increased for 6/8 patients with T downstaging and increased for 5/7 patients without T downstaging relative to the mean tumor ADC values before treatment. The mean observed ADC values for patients with and without T downstage are shown in Table 4. The difference between these two groups with respect to the mean ADC values measured at the six timepoints did not reach significance.

For the eight patients with tumor downstaging, there was a significant increase (P = 0.011) in the mean tumor ADC at the 2^nd week of treatment relative to the ADC before treatment, whereas for the seven patients without tumor downstaging, there was no significant change in the ADC at any timepoint during treatment relative to the ADC values before treatment (Figure 3A, Table 3).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Mean tumor apparent diffusion coefficient values in patients at six measurement points. A: With and without T downstaging; B: With good regressin and poor regression. Cycle: Outlier; Star: Extreme value.

Good regression (TRG 3-4) was observed in 9 of the 15 patients (60%), and poor regression (TRG 0-2) was observed in 6 patients (40%). The ADC values at the 5^th week during treatment were increased in 7/9 patients with good regression and 5/6 patients with poor regression relative to the mean tumor ADC values before treatment. The mean observed ADC values in patients with good regression and poor regression are shown in Table 4. Before treatment and at the 3^rd, 4^th, and 5^th week during treatment, significant differences in the mean ADC values between the two groups were obtained.

For the nine patients with good regression, there was a significant increase (P = 0.004) in the mean tumor ADC value at the 2^nd week of treatment relative to the ADC values before treatment, whereas for the six patients with poor regression, no significant change in ADC was observed at any timepoint during treatment (Figure 3B, Table 3). The two patients with pCR demonstrated lower ADC (0.659 × 10^-3 mm²/s and 0.813 × 10^-3 mm²/s) before treatment but significantly increased tumor ADC (0.825 × 10^-3 mm²/s and 1.133 × 10^-3 mm²/s) at the 2^nd week of treatment.

Neoadjuvant (chemo)radiation followed by surgery has become the standard treatment for locally advanced rectal cancer (LARC). However, approximately 20%-30% of patients do not benefit from this neoadjuvant treatment due to radioresistance of the tumor. Functional non-invasive diffusion-weighted magnetic resonance imaging (DW-MRI) studies are increasingly used to predict response to cancer therapy, but definitive evidence is limited, especially for patients with rectal cancer treated with neoadjuvant chemoradiation therapy (CRT).

DW-MRI is a non-invasive functional MRI technique. To date, published data on the value of DW-MRI as a predictive tool for assessing responses to anti-cancer treatment in patients with rectal cancer are scarce and conflicting.

The authors found that CRT induced a significant increase in the mean apparent diffusion coefficient (ADC) value of LARC. The pretreatment tumor ADC was negatively correlated with tumor regression after CRT for the treatment of LARC, and the ADC value at the 2^nd week of therapy was significantly correlated with the tumor response.

The results of this study suggest that DW-MRI may be a valuable clinical tool to help predict or assess the responses of rectal tumors to neoadjuvant concurrent CRT at an early timepoint.

DW-MRI is a non-invasive functional MRI technique that provides information by measuring water proton mobility in tissues. ADC values can be calculated from DWI measurements according to the impediment to free diffusion of water molecules in a single voxel due to restricting barriers such as membranes, macromolecules, and fibers inside different tissue compartments.

This is an interesting study that investigates the use of DW-MRI as a predictor of the tumor response in 15 patients with rectal cancer undergoing CRT therapy by measuring the tumor ADC. This is an emerging field in which new knowledge is needed, and this study, despite its limits, provides novel information that may help to settle the current debate about the utility of DW-MRI as a predictive tool for the response to anti-cancer treatment.