Premium
MR imaging of locally advanced low rectal cancer: Relationships between imaging findings and the pathological tumor regression grade
Author(s) -
Fujii Shinya,
Nougaret Stephanie,
Escal Laure,
Azria David,
Assenat Eric,
Rouanet Philippe,
Reinhold Caroline,
Guiu Boris
Publication year - 2015
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.24783
Subject(s) - receiver operating characteristic , medicine , nuclear medicine , colorectal cancer , pathological , linear regression , radiology , cancer , mathematics , statistics
Background To evaluate the relationship between MR measurements and the pathological tumor regression grade (pTRG). Methods Two radiologists independently reviewed the pre‐ and postchemoradiation therapy (CRT) rectal MR images of 73 consecutive patients with locally advanced low rectal cancer who underwent neoadjuvant CRT and subsequent surgery and measured tumor diameter, area, signal intensity (SI). The percentage reduction rate for each parameter was calculated. The absolute SI ratio reduction rate was defined as the absolute value of the SI ratio reduction rate. The Kruskal‐Wallis test and multivariate analysis were performed to assess the correlation between each parameter and the pTRG. Receiver operating characteristic (ROC) curves were plotted for predicting favorable regression outcomes (pTRG 3–4). Results In multivariate analysis, the absolute SI ratio reduction rate was a significant predictor of pTRG for both radiologists. Area under the ROC curve (Az) values were 0.77–0.709 for diameter reduction rate, 0.757–0.694 for area, 0.652–0.648 for the SI ratio, 0.736–0.837 for the absolute SI ratio. Conclusion The absolute SI ratio reduction rate was significantly associated with pTRG and predicted favorable responses to CRT. Measurement of the diameter reduction rate is convenient and reliable in predicting favorable responses. J. Magn. Reson. Imaging 2015;42:421–426.