Title |
T2-weighted signal intensity-selected volumetry for prediction of pathological complete response after preoperative chemoradiotherapy in locally advanced rectal cancer
|
---|---|
Published in |
European Radiology, June 2018
|
DOI | 10.1007/s00330-018-5520-1 |
Pubmed ID | |
Authors |
Sungwon Kim, Kyunghwa Han, Nieun Seo, Hye Jin Kim, Myeong-Jin Kim, Woong Sub Koom, Joong Bae Ahn, Joon Seok Lim |
Abstract |
To evaluate the diagnostic value of signal intensity (SI)-selected volumetry findings in T2-weighted magnetic resonance imaging (MRI) as a potential biomarker for predicting pathological complete response (pCR) to preoperative chemoradiotherapy (CRT) in patients with rectal cancer. Forty consecutive patients with pCR after preoperative CRT were compared with 80 age- and sex-matched non-pCR patients in a case-control study. SI-selected tumor volume was measured on post-CRT T2-weighted MRI, which included voxels of the treated tumor exceeding the SI (obturator internus muscle SI + [ischiorectal fossa fat SI - obturator internus muscle SI] × 0.2). Three blinded readers independently rated five-point pCR confidence scores and compared the diagnostic outcome with SI-selected volumetry findings. The SI-selected volumetry protocol was validated in 30 additional rectal cancer patients. The area under the receiver-operating characteristic curve (AUC) of SI-selected volumetry for pCR prediction was 0.831, with an optimal cutoff value of 649.6 mm3 (sensitivity 0.850, specificity 0.725). The AUC of the SI-selected tumor volume was significantly greater than the pooled AUC of readers (0.707, p < 0.001). At this cutoff, the validation trial yielded an accuracy of 0.87. SI-selected volumetry in post-CRT T2-weighted MRI can help predict pCR after preoperative CRT in patients with rectal cancer. • Fibrosis and viable tumor MRI signal intensities (SIs) are difficult to distinguish. • T2 SI-selected volumetry yields high diagnostic performance for assessing pathological complete response. • T2 SI-selected volumetry is significantly more accurate than readers and non-SI-selected volumetry. • Post-chemoradiation therapy T2-weighted MRI SI-selected volumetry facilitates prediction of pathological complete response. |
X Demographics
Geographical breakdown
Country | Count | As % |
---|---|---|
United States | 1 | 100% |
Demographic breakdown
Type | Count | As % |
---|---|---|
Scientists | 1 | 100% |
Mendeley readers
Geographical breakdown
Country | Count | As % |
---|---|---|
Unknown | 27 | 100% |
Demographic breakdown
Readers by professional status | Count | As % |
---|---|---|
Researcher | 5 | 19% |
Student > Ph. D. Student | 5 | 19% |
Student > Bachelor | 3 | 11% |
Student > Doctoral Student | 3 | 11% |
Lecturer > Senior Lecturer | 1 | 4% |
Other | 2 | 7% |
Unknown | 8 | 30% |
Readers by discipline | Count | As % |
---|---|---|
Medicine and Dentistry | 11 | 41% |
Neuroscience | 2 | 7% |
Chemistry | 1 | 4% |
Nursing and Health Professions | 1 | 4% |
Unknown | 12 | 44% |