skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions

Abstract

Purpose: The purpose of this study is to investigate the relationship between computed tomographic (CT) texture features of primary lesions and metastasis-free survival for rectal cancer patients; and to develop a datamining prediction model using texture features. Methods: A total of 220 rectal cancer patients treated with neoadjuvant chemo-radiotherapy (CRT) were enrolled in this study. All patients underwent CT scans before CRT. The primary lesions on the CT images were delineated by two experienced oncologists. The CT images were filtered by Laplacian of Gaussian (LoG) filters with different filter values (1.0–2.5: from fine to coarse). Both filtered and unfiltered images were analyzed using Gray-level Co-occurrence Matrix (GLCM) texture analysis with different directions (transversal, sagittal, and coronal). Totally, 270 texture features with different species, directions and filter values were extracted. Texture features were examined with Student’s t-test for selecting predictive features. Principal Component Analysis (PCA) was performed upon the selected features to reduce the feature collinearity. Artificial neural network (ANN) and logistic regression were applied to establish metastasis prediction models. Results: Forty-six of 220 patients developed metastasis with a follow-up time of more than 2 years. Sixtyseven texture features were significantly different in t-test (p<0.05) between patients with and without metastasis,more » and 12 of them were extremely significant (p<0.001). The Area-under-the-curve (AUC) of ANN was 0.72, and the concordance index (CI) of logistic regression was 0.71. The predictability of ANN was slightly better than logistic regression. Conclusion: CT texture features of primary lesions are related to metastasisfree survival of rectal cancer patients. Both ANN and logistic regression based models can be developed for prediction.« less

Authors:
; ; ; ; ; ;  [1]
  1. Fudan University Shanghai Cancer Center, Shanghai (China)
Publication Date:
OSTI Identifier:
22499355
Resource Type:
Journal Article
Resource Relation:
Journal Name: Medical Physics; Journal Volume: 42; Journal Issue: 6; Other Information: (c) 2015 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
60 APPLIED LIFE SCIENCES; COMPUTERIZED TOMOGRAPHY; IMAGE PROCESSING; IMAGES; MEDICAL PERSONNEL; METASTASES; NEOPLASMS; NEURAL NETWORKS; PATIENTS; RADIOTHERAPY; RECTUM

Citation Formats

Zhong, H, Wang, J, Shen, L, Hu, W, Wan, J, Zhou, Z, and Zhang, Z. SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions. United States: N. p., 2015. Web. doi:10.1118/1.4924342.
Zhong, H, Wang, J, Shen, L, Hu, W, Wan, J, Zhou, Z, & Zhang, Z. SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions. United States. doi:10.1118/1.4924342.
Zhong, H, Wang, J, Shen, L, Hu, W, Wan, J, Zhou, Z, and Zhang, Z. Mon . "SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions". United States. doi:10.1118/1.4924342.
@article{osti_22499355,
title = {SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions},
author = {Zhong, H and Wang, J and Shen, L and Hu, W and Wan, J and Zhou, Z and Zhang, Z},
abstractNote = {Purpose: The purpose of this study is to investigate the relationship between computed tomographic (CT) texture features of primary lesions and metastasis-free survival for rectal cancer patients; and to develop a datamining prediction model using texture features. Methods: A total of 220 rectal cancer patients treated with neoadjuvant chemo-radiotherapy (CRT) were enrolled in this study. All patients underwent CT scans before CRT. The primary lesions on the CT images were delineated by two experienced oncologists. The CT images were filtered by Laplacian of Gaussian (LoG) filters with different filter values (1.0–2.5: from fine to coarse). Both filtered and unfiltered images were analyzed using Gray-level Co-occurrence Matrix (GLCM) texture analysis with different directions (transversal, sagittal, and coronal). Totally, 270 texture features with different species, directions and filter values were extracted. Texture features were examined with Student’s t-test for selecting predictive features. Principal Component Analysis (PCA) was performed upon the selected features to reduce the feature collinearity. Artificial neural network (ANN) and logistic regression were applied to establish metastasis prediction models. Results: Forty-six of 220 patients developed metastasis with a follow-up time of more than 2 years. Sixtyseven texture features were significantly different in t-test (p<0.05) between patients with and without metastasis, and 12 of them were extremely significant (p<0.001). The Area-under-the-curve (AUC) of ANN was 0.72, and the concordance index (CI) of logistic regression was 0.71. The predictability of ANN was slightly better than logistic regression. Conclusion: CT texture features of primary lesions are related to metastasisfree survival of rectal cancer patients. Both ANN and logistic regression based models can be developed for prediction.},
doi = {10.1118/1.4924342},
journal = {Medical Physics},
number = 6,
volume = 42,
place = {United States},
year = {Mon Jun 15 00:00:00 EDT 2015},
month = {Mon Jun 15 00:00:00 EDT 2015}
}
  • Purpose: To evaluate the role of mid-treatment and post-treatment FDG-PET/CT in predicting progression-free survival (PFS) and distant metastasis (DM) of anal cancer patients treated with chemoradiotherapy (CRT). Methods: 17 anal cancer patients treated with CRT were retrospectively studied. The median prescription dose was 56 Gy (range, 50–62.5 Gy). All patients underwent FDG-PET/CT scans before and after CRT. 16 of the 17 patients had an additional FDG-PET/CT image at 3–5 weeks into the treatment (denoted as mid-treatment FDG-PET/CT). 750 features were extracted from these three sets of scans, which included both traditional PET/CT measures (SUVmax, SUVpeak, tumor diameters, etc.) and spatialtemporalmore » PET/CT features (comprehensively quantify a tumor’s FDG uptake intensity and distribution, spatial variation (texture), geometric property and their temporal changes relative to baseline). 26 clinical parameters (age, gender, TNM stage, histology, GTV dose, etc.) were also analyzed. Advanced analytics including methods to select an optimal set of predictors and a model selection engine, which identifies the most accurate machine learning algorithm for predictive analysis was developed. Results: Comparing baseline + mid-treatment PET/CT set to baseline + posttreatment PET/CT set, 14 predictors were selected from each feature group. Same three clinical parameters (tumor size, T stage and whether 5-FU was held during any cycle of chemotherapy) and two traditional measures (pre- CRT SUVmin and SUVmedian) were selected by both predictor groups. Different mix of spatial-temporal PET/CT features was selected. Using the 14 predictors and Naive Bayes, mid-treatment PET/CT set achieved 87.5% accuracy (2 PFS patients misclassified, all local recurrence and DM patients correctly classified). Post-treatment PET/CT set achieved 94.0% accuracy (all PFS and DM patients correctly predicted, 1 local recurrence patient misclassified) with logistic regression, neural network or support vector machine model. Conclusion: Applying radiomics approach to either midtreatment or post-treatment PET/CT could achieve high accuracy in predicting anal cancer treatment outcomes. This work was supported in part by the National Cancer Institute Grant R01CA172638.« less
  • Purpose: To explore the feasibility of pretreatment test for iso-NTCP DGART and to compare the pretreatment test results with post-treatment evaluations. Methods: NTCP here refers to late rectal wall toxicity only and is calculated with the ring rectal wall DVH. Simulation for one time iso- NTCP DGART starts after half of the total dose was done for 10 patients to investigate if TCP gains could be achieved. Six patients were treated using a 12-fraction 4.3Gy technique and four using 16-fraction 3.63Gy technique. For each of the 12-fraction cases a VMAT plan was generated in Pinnacle3™ using the daily CT obtainedmore » prior to the 6th fraction. A pretreatment simulation was performed using only the first 6 daily CTs. The idea is to add the 6 original plan delivered doses with 6 DGART plan delivered doses by deformable dose accumulation (DDA) on each of the first 6 CTs, resulting in 6 rectal wall doses (RWDs) and NTCPs. The 95% confidence interval (95%CI) for the 6 NTCPs were computed.The posttreatment evaluation was done by: a) copy the DGART plan to 6 CTs for fraction 7–12 and calculate the 6 actual DGART delivered fractional doses; b) sum the 6 actual DGART doses with the 6 original plan delivered doses by DDA on each of the 12 CTs resulting in 12 post-treatment RWDs and NTCPs; c) boxplot the 12 post-treatment NTCPs. Results: Target dose gain is 0.76–1.93 Gy. The 95%CI widths of the pretreatment tests NTCPs were 1.1–2.7%. For 5 patients, the planned NTCP fell within the 95%CI. For 4 patients, the planned NTCP was lower than the 95%CI lines. Post-treatment results show that for 7 patients, the upper quartile was within the 95%CI; for 2 patients, the upper quartile were higher than the 95%CI. Conclusion: The pretreatment test yields conservative prediction of the actual delivered NTCP.« less
  • Purpose: To determine whether volumes based on contours of the peritoneal space can be used instead of individual small bowel loops to predict for grade ≥3 acute small bowel toxicity in patients with rectal cancer treated with neoadjuvant chemoradiation therapy. Methods and Materials: A standardized contouring method was developed for the peritoneal space and retrospectively applied to the radiation treatment plans of 67 patients treated with neoadjuvant chemoradiation therapy for rectal cancer. Dose-volume histogram (DVH) data were extracted and analyzed against patient toxicity. Receiver operating characteristic analysis and logistic regression were carried out for both contouring methods. Results: Grade ≥3more » small bowel toxicity occurred in 16% (11/67) of patients in the study. A highly significant dose-volume relationship between small bowel irradiation and acute small bowel toxicity was supported by the use of both small bowel loop and peritoneal space contouring techniques. Receiver operating characteristic analysis demonstrated that, for both contouring methods, the greatest sensitivity for predicting toxicity was associated with the volume receiving between 15 and 25 Gy. Conclusion: DVH analysis of peritoneal space volumes accurately predicts grade ≥3 small bowel toxicity in patients with rectal cancer receiving neoadjuvant chemoradiation therapy, suggesting that the contours of the peritoneal space provide a reasonable surrogate for the contours of individual small bowel loops. The study finds that a small bowel V15 less than 275 cc and a peritoneal space V15 less than 830 cc are associated with a less than 10% risk of grade ≥3 acute toxicity.« less
  • Purpose: The current tumor, node, metastasis system needs refinement to improve its ability to predict survival of patients with non-small-cell lung cancer (NSCLC) treated with (chemo)radiation. In this study, we investigated the prognostic value of tumor volume and N status, assessed by using fluorodeoxyglucose-positron emission tomography (PET). Patients and Methods: Clinical data from 270 consecutive patients with inoperable NSCLC Stages I-IIIB treated radically with (chemo)radiation were collected retrospectively. Diagnostic imaging was performed using either integrated PET-computed tomography or computed tomography and PET separately. The Kaplan-Meier method, as well as Cox regression, was used to analyze data. Results: Univariate survival analysismore » showed that number of positive lymph node stations (PLNSs), as well as N stage on PET, was associated significantly with survival. The final multivariate Cox model consisted of number of PLNSs, gross tumor volume (i.e., volume of the primary tumor plus lymph nodes), sex, World Health Organization performance status, and equivalent radiation dose corrected for time; N stage was no longer significant. Conclusions: Number of PLNSs, assessed by means of fluorodeoxyglucose-PET, was a significant factor for survival of patients with inoperable NSCLC treated with (chemo)radiation. Risk stratification for this group of patients should be based on gross tumor volume, number of PLNSs, sex, World Health Organization performance status, and equivalent radiation dose corrected for time.« less
  • Purpose: Rectal cancer is often clinically resistant to radiotherapy (RT) and identifying molecular markers to define the biologic basis for this phenomenon would be valuable. The nuclear factor {kappa}-light chain-enhancer of activated B cells (NF-{kappa}B) is a potential anti-apoptotic transcription factor that has been associated with resistance to RT in model systems. The present study was designed to evaluate NF-{kappa}B activation in patients with rectal cancer undergoing chemoradiotherapy to determine whether NF-{kappa}B activity correlates with the outcome in rectal cancer patients. Methods and Materials: A total of 22 patients underwent biopsy at multiple points in a prospective study and themore » data from another 50 were analyzed retrospectively. The pretreatment tumor tissue was analyzed for multiple NF-{kappa}B subunits by immunohistochemistry. Serial tumor biopsy cores were analyzed for NF-{kappa}B-regulated gene expression using reverse transcriptase polymerase chain reaction and for NF-{kappa}B subunit nuclear localization using immunohistochemistry. Results: Several NF-{kappa}B target genes (Bcl-2, cellular inhibitor of apoptosis protein [cIAP]2, interleukin-8, and tumor necrosis factor receptor-associated-1) were significantly upregulated by a single fraction of RT at 24 h, demonstrating for the first time that NF-{kappa}B is activated by RT in human rectal tumors. The baseline NF-{kappa}B p50 nuclear expression did not correlate with the pathologic response to RT. However, an increasing baseline p50 level was prognostic for overall survival (hazard ratio, 2.15; p = .040). Conclusion: NF-{kappa}B nuclear expression at baseline in rectal cancer was prognostic for overall survival but not predictive of the response to RT. Larger patient numbers are needed to assess the effect of NF-{kappa}B target gene upregulation on the response to RT. Our results suggest that NF-{kappa}B might play an important role in tumor metastasis but not to the resistance to chemoradiotherapy.« less