Patient population: All patients with biopsy-proven rectal cancer treated in our institution between July 2013 and July 2018 were considered for elegibility.
Inclusion criteria were: 1) patients with LARC staged with MRI; 2) availability of MRI staging including DWI performed before and after CRT (about 6-8 weeks after chemoradiotherapy); 3) CRT followed by surgery; 4) availability of pathological reports of surgical specimens.
Exclusion criteria were 1) mucinous adenocarcinoma; 2) presence of remote metastases or unresectable primary tumor.
A total of 66 patients (39 males,
27 females,
age range 28-83,
mean age 63 years) were enrolled in this study.
Treatment protocol: total prescribed dose of 45 Gy for Planning Target Volume 1 (PTV1,
consisting of Clinical Target Volume - CTV1 including tumor,
mesorectum,
internal iliac and presacral lymph nodes and a margin of 1 cm in all directions,
taking into account the individual organ motion) delivered in 25 fractions of 1.8 Gy per day,
for 5 days a week,
over 5 weeks,
boosted with a dose of 10 Gy for PTV2 (CTV2 including the macroscopic tumor and the corresponding mesorectum + a margin of 0.5–1 cm in the lateral–lateral and anterior–posterior direction and a margin of 1.5–2 cm in superior–inferior direction) delivered in 10 fractions of 1 Gy twice a week over 5 weeks; concurrent chemotherapy (oral administration of capecitabine at 825 mg/m2,
for 5 days a week).
Imaging protocol: All pre-CRT and post-CRT MRI examinations were performed using a 1.5-Tesla scanner (Sonata Siemens,
Erlangen,
Germany),
with a phased-array body surface coil.
N-butyl scopolamine bromide (20 mg) was administered by intramuscular injection,
if not contraindicated,
before MRI,
to reduce intestinal motion artifacts.
-Axial Fast low-angle shot (FLASH) 2D T1- weighted images of the whole pelvis
-Axial Turbo Spin-Echo (TSE) T2-weighted images of the whole pelvis.
-Thin-section (3-mm) Sagittal T2-weighted TSE images covering the entire length of the rectum,
to plan axial scans.
-Thin-section (3-mm) Axial images through the rectal cancer,
perpendicular to the long axis of the tumor.
- Oblique Coronal T2- weighted TSE images along the long axis of the anal canal,
only for low rectal tumors.
-DWI acquired using three increasing b values (50–400 to 800 s/mm2) with echo planar imaging sequence with respiratory triggering.
-ADC maps derived automatically on a voxel- by-voxel basis by using commercially available software (Syngo; Siemens Medical Solutions).
Table 1 summarizes MR sequences parameters.
Image analysis: two sets of MR images,
performed before and after CRT,
were prospectively analyzed,
evaluating:
-
For quantitative analysis,
tumor's ADC values measured pre-CRT and post-CRT,
manually drawing three non-overlapping Region Of Interest (ROI,
size 9,8 mm2) on the ADC map.
T2 weighted-images and DWI were observed to assist in correctly identifying the tumor.
The ROIs were delineated excluding cancer margins,
distortion artifacts and macroscopically visible necrotic or cystic portions; when no residual tumor was visible,
ROIs were traced on what was considered to be the normal residual rectal wall,
as much as possible in the same area initially used in pre-CRT MR examination.
The values of all subreadings of each time point examination were averaged and the mean ADC value pre-CRT and post-CRT was respectively calculated for each lesion.
DADC (ADC post- ADC pre) was also calculated.
-
For qualitative analysis,
Magnetic Resonance - TRG (MR-TRG) was visually assessed,
using histopathological TRG,
scored according to Mandard's score system,
as reference standard.
Definition of Mandard's score system and of MR-TRG are detailed in Table 2.
According to histopathological analysis of the surgical specimen,
patients with TRG1-2 scores were classified as good responders to CRT,
whereas patients with TRG 3-5 were considered as poor-responders to CRT.
Statistical analysis:
Patients were divided into good response group (TRG 1-2) and poor response group (TRG 3-5) as well as complete response group (CR,
ypTRG = 1) and non-complete response group (non-CR,
ypTRG 2-5),
respectively.
Student’s t test,
with calculation of the relative p value,
was used to compare mean ADCpre-CRT,
ADCpost-CRT,
ΔADC values between CR and non-CR,
as well as between good responders and poor responders patients.
A p value <0.05 was considered statistically significant.
The cut-off ADC values of the receiver-operating characteristics (ROC) curve was employed to establish the diagnostic performance of ADC to assess tumor response.
The area under the ROC curve (Az) was calculated and considered indicative of diagnostic accuracy,
using 95% confidence intervals to express the statistical precision of the results.
Optimal cut-off values,
sensitivity,
specificity,
positive predictive value (PPV),
and negative predictive value (NPV) were calculated.
Spearman's Rank correlation test was used to assess the correlation between mrTRG and TRG.
A p value <0.05 was considered statistically significant.