Chun Yang1,2,3, Ze-Kun Jiang4, Li-Heng Liu5,6,7, Meng-Su Zeng1,2,3. 1. Department of Radiology, Zhongshan Hospital, Fudan University, No. 180, Fenglin Road, Shanghai, 200032, China. 2. Shanghai Institute of Medical Imaging, Shanghai, China. 3. Department of Medical Imaging, Shanghai Medical College, Fudan University, Shanghai, China. 4. Shandong Key Laboratory of Medical Physics and Image Processing Technology, School of Physics and Electronics, Shandong Normal University, Jinan, Shandong, China. 5. Department of Radiology, Zhongshan Hospital, Fudan University, No. 180, Fenglin Road, Shanghai, 200032, China. llh9821@163.com. 6. Shanghai Institute of Medical Imaging, Shanghai, China. llh9821@163.com. 7. Department of Medical Imaging, Shanghai Medical College, Fudan University, Shanghai, China. llh9821@163.com.
Abstract
OBJECTIVE: To develop a predicting model for tumor resistance to neoadjuvant chemoradiotherapy (NCRT) in locally advanced rectal cancer (LARC) by using pre-treatment apparent diffusion coefficient (ADC) image-derived radiomics features. METHOD: A total of 89 patients with LARC were randomly assigned into training (N = 66) and testing cohorts (N = 23) at the ratio of 3:1. Radiomics features were derived from manually determined tumor region of pre-treatment ADC images. Random forest algorithm was used to determine the most relevant features and then to construct a predicting model for identifying resistant tumor. Stability and diagnostic performance of the random forest model was evaluated with the testing cohort. RESULTS: The top 10 most relevant features (entropymean, inverse variance, energymean, small area emphasis, ADCmin, ADCmean, sdGa02, small gradient emphasis, age, and size) were determined from clinical characteristics and 133 radiomics features. In the prediction of resistant tumor of the testing cohort, the random forest model constructed based on these most relevant features achieved an area under the receiver operating characteristic curve of 0.83, with the highest accuracy of 91.3%, a sensitivity of 88.9%, and a specificity of 92.8%. CONCLUSION: The random forest classifier based on radiomics features derived from pre-treatment ADC images have the potential to predict tumor resistance to NCRT in patients with LARC, and the use of predicting model may facilitate individualized management of rectal cancer.
OBJECTIVE: To develop a predicting model for tumor resistance to neoadjuvant chemoradiotherapy (NCRT) in locally advanced rectal cancer (LARC) by using pre-treatment apparent diffusion coefficient (ADC) image-derived radiomics features. METHOD: A total of 89 patients with LARC were randomly assigned into training (N = 66) and testing cohorts (N = 23) at the ratio of 3:1. Radiomics features were derived from manually determined tumor region of pre-treatment ADC images. Random forest algorithm was used to determine the most relevant features and then to construct a predicting model for identifying resistant tumor. Stability and diagnostic performance of the random forest model was evaluated with the testing cohort. RESULTS: The top 10 most relevant features (entropymean, inverse variance, energymean, small area emphasis, ADCmin, ADCmean, sdGa02, small gradient emphasis, age, and size) were determined from clinical characteristics and 133 radiomics features. In the prediction of resistant tumor of the testing cohort, the random forest model constructed based on these most relevant features achieved an area under the receiver operating characteristic curve of 0.83, with the highest accuracy of 91.3%, a sensitivity of 88.9%, and a specificity of 92.8%. CONCLUSION: The random forest classifier based on radiomics features derived from pre-treatment ADC images have the potential to predict tumor resistance to NCRT in patients with LARC, and the use of predicting model may facilitate individualized management of rectal cancer.
Entities:
Keywords:
Diffusion-weighted MRI; Drug resistance; Machine learning; Neoadjuvant therapy; Neoplasm; Rectal cancer
Authors: Claus Rödel; Ullrich Graeven; Rainer Fietkau; Werner Hohenberger; Torsten Hothorn; Dirk Arnold; Ralf-Dieter Hofheinz; Michael Ghadimi; Hendrik A Wolff; Marga Lang-Welzenbach; Hans-Rudolf Raab; Christian Wittekind; Philipp Ströbel; Ludger Staib; Martin Wilhelm; Gerhard G Grabenbauer; Hans Hoffmanns; Fritz Lindemann; Anke Schlenska-Lange; Gunnar Folprecht; Rolf Sauer; Torsten Liersch Journal: Lancet Oncol Date: 2015-07-15 Impact factor: 41.316
Authors: Julio Garcia-Aguilar; Zhenbin Chen; David D Smith; Wenyan Li; Robert D Madoff; Peter Cataldo; Jorge Marcet; Carlos Pastor Journal: Ann Surg Date: 2011-09 Impact factor: 12.969
Authors: A J Breugom; W van Gijn; E W Muller; Å Berglund; C B M van den Broek; T Fokstuen; H Gelderblom; E Kapiteijn; J W H Leer; C A M Marijnen; H Martijn; E Meershoek-Klein Kranenbarg; I D Nagtegaal; L Påhlman; C J A Punt; H Putter; A G H Roodvoets; H J T Rutten; W H Steup; B Glimelius; C J H van de Velde Journal: Ann Oncol Date: 2014-12-05 Impact factor: 32.976
Authors: Ke Nie; Liming Shi; Qin Chen; Xi Hu; Salma K Jabbour; Ning Yue; Tianye Niu; Xiaonan Sun Journal: Clin Cancer Res Date: 2016-05-16 Impact factor: 12.531
Authors: Atthaphorn Trakarnsanga; Mithat Gönen; Jinru Shia; Garrett M Nash; Larissa K Temple; José G Guillem; Philip B Paty; Karyn A Goodman; Abraham Wu; Marc Gollub; Neil Segal; Leonard Saltz; Julio Garcia-Aguilar; Martin R Weiser Journal: J Natl Cancer Inst Date: 2014-09-22 Impact factor: 13.506