Prognostic Value of CT Radiomic Features in Resectable Pancreatic Ductal Adenocarcinoma

In this work, we assess the reproducibility and prognostic value of CT-derived radiomic features for resectable pancreatic ductal adenocarcinoma (PDAC). Two radiologists contoured tumour regions on pre-operative CT of two cohorts from two institutions undergoing curative-intent surgical resection for PDAC. The first (n = 30) and second cohorts (n = 68) were used for training and validation of proposed prognostic model for overall survival (OS), respectively. Radiomic features were extracted using PyRadiomics library and those with weak inter-reader reproducibility were excluded. Through Cox regression models, significant features were identified in the training cohort and retested in the validation cohort. Significant features were then fused via Cox regression to build a single radiomic signature in the training cohort, which was validated across readers in the validation cohort. Two radiomic features derived from Sum Entropy and Cluster Tendency features were both robust to inter-reader reproducibility and prognostic of OS across cohorts and readers. The radiomic signature showed prognostic value for OS in the validation cohort with hazard ratios of 1.56 (P = 0.005) and 1.35 (P = 0.022), for the first and second reader, respectively. CT-based radiomic features were shown to be prognostic in patients with resectable PDAC. These features may help stratify patients for neoadjuvant or alternative therapies.

[1]  Amber L. Simpson,et al.  Survival Prediction in Pancreatic Ductal Adenocarcinoma by Quantitative Computed Tomography Image Analysis , 2018, Annals of Surgical Oncology.

[2]  P. Lambin,et al.  Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach , 2014, Nature Communications.

[3]  James D. Kang,et al.  Reliable Magnetic Resonance Imaging Based Grading System for Cervical Intervertebral Disc Degeneration , 2016, Asian spine journal.

[4]  Masoom A. Haider,et al.  Automated prostate cancer detection via comprehensive multi-parametric magnetic resonance imaging texture feature models , 2015, BMC Medical Imaging.

[5]  Mari Mino-Kenudson,et al.  Pancreatic ductal adenocarcinoma: long-term survival does not equal cure. , 2012, Surgery.

[6]  Eun Sun Lee,et al.  Imaging diagnosis of pancreatic cancer: a state-of-the-art review. , 2014, World journal of gastroenterology.

[7]  B. Erickson,et al.  Assessment of treatment response during chemoradiation therapy for pancreatic cancer based on quantitative radiomic analysis of daily CTs: An exploratory study , 2017, PloS one.

[8]  N. Clement,et al.  The radiographic union scale in tibial (RUST) fractures , 2016, Bone & joint research.

[9]  Farzad Khalvati,et al.  Radiomics analysis at PET/CT contributes to prognosis of recurrence and survival in lung cancer treated with stereotactic body radiotherapy , 2018, Scientific Reports.

[10]  Alejandro Munoz del Rio,et al.  CT textural analysis of hepatic metastatic colorectal cancer: pre-treatment tumor heterogeneity correlates with pathology and clinical outcomes , 2015, Abdominal Imaging.

[11]  G. Pazour,et al.  Ror2 signaling regulates Golgi structure and transport through IFT20 for tumor invasiveness , 2017, Scientific Reports.

[12]  Masoom A. Haider,et al.  Radiomics-based Prognosis Analysis for Non-Small Cell Lung Cancer , 2017, Scientific Reports.

[13]  Laura H. Tang,et al.  Pancreatic Adenocarcinoma: The Actual 5-Year Survivors , 2008, Journal of Gastrointestinal Surgery.

[14]  Guangtao Zhai,et al.  A Deep Learning-Based Radiomics Model for Prediction of Survival in Glioblastoma Multiforme , 2017, Scientific Reports.

[15]  Masoom A. Haider,et al.  A Local ROI-specific Atlas-based Segmentation of Prostate Gland and Transitional Zone in Diffusion MRI , 2016 .

[16]  S. Gallinger,et al.  Prognostic factors in resected pancreatic adenocarcinoma: analysis of actual 5-year survivors. , 2004, Journal of the American College of Surgeons.

[17]  W. Tsai,et al.  Reproducibility of radiomics for deciphering tumor phenotype with imaging , 2016, Scientific Reports.

[18]  R. Tibshirani.  The lasso method for variable selection in the Cox model. , 1997, Statistics in medicine.

[19]  Simon B. Eickhoff,et al.  One-year test–retest reliability of intrinsic connectivity network fMRI in older adults , 2012, NeuroImage.

[20]  E. Rosato,et al.  Long term survival after pancreatic resection for pancreatic adenocarcinoma , 2000, American Journal of Gastroenterology.

[21]  Masoom A. Haider,et al.  CT texture features are associated with overall survival in pancreatic ductal adenocarcinoma – a quantitative analysis , 2017, BMC Medical Imaging.

[22]  X Li,et al.  SU-D-207B-07: Development of a CT-Radiomics Based Early Response Prediction Model During Delivery of Chemoradiation Therapy for Pancreatic Cancer. , 2016, Medical physics.

[23]  Tibor Schuster,et al.  Preoperative/Neoadjuvant Therapy in Pancreatic Cancer: A Systematic Review and Meta-analysis of Response and Resection Percentages , 2010, PLoS medicine.

[24]  P. Pickhardt,et al.  CT Textural Analysis of Large Primary Renal Cell Carcinomas: Pretreatment Tumor Heterogeneity Correlates With Histologic Findings and Clinical Outcomes. , 2016, AJR. American journal of roentgenology.

[25]  M. Haider,et al.  CT texture analysis: a potential tool for prediction of survival in patients with metastatic clear cell carcinoma treated with sunitinib , 2017, Cancer Imaging.

[26]  Charles Ferté,et al.  Limits of radiomic-based entropy as a surrogate of tumor heterogeneity: ROI-area, acquisition protocol and tissue site exert substantial influence , 2017, Scientific Reports.

[27]  T. Ishiwata.  Pancreatic Ductal Adenocarcinoma: Basic and Clinical Challenges for Better Prognosis , 2013 .

[28]  David A Clausi.  An analysis of co-occurrence texture statistics as a function of grey level quantization , 2002 .

[29]  Nicola Schieda,et al.  Can Quantitative CT Texture Analysis be Used to Differentiate Fat-poor Renal Angiomyolipoma from Renal Cell Carcinoma on Unenhanced CT Images? , 2015, Radiology.

[30]  Andre Dekker,et al.  Radiomics: the process and the challenges. , 2012, Magnetic resonance imaging.

[31]  Masoom A. Haider,et al.  MAPS: A Quantitative Radiomics Approach for Prostate Cancer Detection , 2016, IEEE Transactions on Biomedical Engineering.

[32]  Andriy Fedorov,et al.  Computational Radiomics System to Decode the Radiographic Phenotype. , 2017, Cancer research.

[33]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[34]  Xiaoyu Liu,et al.  Combination of preoperative CEA and CA19-9 improves prediction outcomes in patients with resectable pancreatic adenocarcinoma: results from a large follow-up cohort , 2017, OncoTargets and therapy.

[35]  Pui-In Mak,et al.  A digital microfluidic system for loop-mediated isothermal amplification and sequence specific pathogen detection , 2017, Scientific Reports.