Radio-pathomic mapping model generated using annotations from five pathologists reliably distinguishes high-grade prostate cancer

Abstract. Purpose: Our study predictively maps epithelium density in magnetic resonance imaging (MRI) space while varying the ground truth labels provided by five pathologists to quantify the downstream effects of interobserver variability. Approach: Clinical imaging and postsurgical tissue from 48 recruited prospective patients were used in our study. Tissue was sliced to match the MRI orientation and whole-mount slides were stained and digitized. Data from 28 patients (n  =  33 slides) were sent to five pathologists to be annotated. Slides from the remaining 20 patients (n  =  123 slides) were annotated by one of the five pathologists. Interpathologist variability was measured using Krippendorff’s alpha. Pathologist-specific radiopathomic mapping models were trained using a partial least-squares regression using MRI values to predict epithelium density, a known marker for disease severity. An analysis of variance characterized intermodel means difference in epithelium density. A consensus model was created and evaluated using a receiver operator characteristic classifying high grade versus low grade and benign, and was statistically compared to apparent diffusion coefficient (ADC). Results: Interobserver variability ranged from low to acceptable agreement (0.31 to 0.69). There was a statistically significant difference in mean predicted epithelium density values (p  <  0.001) between the five models. The consensus model outperformed ADC (areas under the curve = 0.80 and 0.71, respectively, p  <  0.05). Conclusion: We demonstrate that radiopathomic maps of epithelium density are sensitive to the pathologist annotating the dataset; however, it is unclear if these differences are clinically significant. The consensus model produced the best maps, matched the performance of the best individual model, and outperformed ADC.

[1]  Toyonori Tsuzuki,et al.  Accuracy of gleason grading by practicing pathologists and the impact of education on improving agreement. , 2003, Human pathology.

[2]  Amy Kaczmarowski,et al.  Optimized b-value selection for the discrimination of prostate cancer grades, including the cribriform pattern, using diffusion weighted imaging , 2017, Journal of medical imaging.

[3]  C. Kim,et al.  High-b-value diffusion-weighted imaging at 3 T to detect prostate cancer: comparisons between b values of 1,000 and 2,000 s/mm2. , 2010, AJR. American journal of roentgenology.

[4]  Tapio Pahikkala,et al.  Radiomics and machine learning of multisequence multiparametric prostate MRI: Towards improved non-invasive prostate cancer characterization , 2019, PloS one.

[5]  B. Delahunt,et al.  The 2014 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma: Definition of Grading Patterns and Proposal for a New Grading System , 2015, The American journal of surgical pathology.

[6]  Aytekin Oto,et al.  Diagnosis of Prostate Cancer with Noninvasive Estimation of Prostate Tissue Composition by Using Hybrid Multidimensional MR Imaging: A Feasibility Study. , 2018, Radiology.

[7]  J. Hanley,et al.  A method of comparing the areas under receiver operating characteristic curves derived from the same cases. , 1983, Radiology.

[8]  Jing Wang,et al.  Machine learning-based analysis of MR radiomics can help to improve the diagnostic performance of PI-RADS v2 in clinically relevant prostate cancer , 2017, European Radiology.

[9]  A. Madabhushi,et al.  Radiomic features for prostate cancer detection on MRI differ between the transition and peripheral zones: Preliminary findings from a multi‐institutional study , 2017, Journal of magnetic resonance imaging : JMRI.

[10]  Michael Götz,et al.  Radiomic Machine Learning for Characterization of Prostate Lesions with MRI: Comparison to ADC Values. , 2018, Radiology.

[11]  Ezgi Mercan,et al.  Accuracy is in the eyes of the pathologist: The visual interpretive process and diagnostic accuracy with digital whole slide images , 2017, J. Biomed. Informatics.

[12]  M. G. Sabini,et al.  Multiparametric MRI Apparent Diffusion Coefficient (ADC) Accuracy in Diagnosing Clinically Significant Prostate Cancer. , 2017, In vivo.

[13]  Gaurav Pandey,et al.  Objective risk stratification of prostate cancer using machine learning and radiomics applied to multiparametric magnetic resonance images , 2019, Scientific Reports.

[14]  A. Viera,et al.  Understanding interobserver agreement: the kappa statistic. , 2005, Family medicine.

[15]  A. Jemal,et al.  Cancer statistics, 2019 , 2019, CA: a cancer journal for clinicians.

[16]  Mark McEntee,et al.  Changes in Epithelium, Stroma, and Lumen Space Correlate More Strongly with Gleason Pattern and Are Stronger Predictors of Prostate ADC Changes than Cellularity Metrics. , 2015, Radiology.

[17]  Takaaki Sano,et al.  A comparison of interobserver reproducibility of Gleason grading of prostatic carcinoma in Japan and the United States. , 2005, Archives of pathology & laboratory medicine.

[18]  Saiful Miah,et al.  Prostate imaging features that indicate benign or malignant pathology on biopsy , 2018, Translational andrology and urology.

[19]  Carole Lartizien,et al.  Prostate focal peripheral zone lesions: characterization at multiparametric MR imaging--influence of a computer-aided diagnosis system. , 2014, Radiology.

[20]  Kenneth A. Iczkowski,et al.  Gleason Probability Maps: A Radiomics Tool for Mapping Prostate Cancer Likelihood in MRI Space , 2019, Tomography.

[21]  B. van Ginneken,et al.  Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study. , 2020, The Lancet. Oncology.

[22]  Namkug Kim,et al.  Apparent diffusion coefficient: Prostate cancer versus noncancerous tissue according to anatomical region , 2008, Journal of magnetic resonance imaging : JMRI.

[23]  Andrew B Rosenkrantz,et al.  Prostate MRI can reduce overdiagnosis and overtreatment of prostate cancer. , 2015, Academic radiology.

[24]  Kenneth A. Iczkowski,et al.  Radio-pathomic Maps of Epithelium and Lumen Density Predict the Location of High-Grade Prostate Cancer , 2018, International journal of radiation oncology, biology, physics.

[25]  Rémi Souchon,et al.  Quantitative Analysis of Prostate Multiparametric MR Images for Detection of Aggressive Prostate Cancer in the Peripheral Zone: A Multiple Imager Study. , 2016, Radiology.

[26]  Ellery Wulczyn,et al.  Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancer , 2018, npj Digital Medicine.

[27]  J. Epstein,et al.  Interobserver reproducibility of Gleason grading of prostatic carcinoma: general pathologist. , 2001, Human pathology.

[28]  J. Al-Maghrabi,et al.  Gleason grading of prostate cancer in needle core biopsies: a comparison of general and urologic pathologists , 2013, Annals of Saudi medicine.

[29]  Aytekin Oto,et al.  Apparent diffusion coefficient for prostate cancer imaging: impact of B values. , 2014, AJR. American journal of roentgenology.

[30]  Tayyar A. Ozkan,et al.  Interobserver variability in Gleason histological grading of prostate cancer , 2016, Scandinavian journal of urology.

[31]  Ashutosh Tewari,et al.  Advanced Diffusion-weighted Imaging Modeling for Prostate Cancer Characterization: Correlation with Quantitative Histopathologic Tumor Tissue Composition-A Hypothesis-generating Study. , 2017, Radiology.

[32]  Martin O. Leach,et al.  Non-Invasive Prostate Cancer Characterization with Diffusion-Weighted MRI: Insight from In silico Studies of a Transgenic Mouse Model , 2017, Front. Oncol..

[33]  Arturo Brunetti,et al.  Machine learning applications in prostate cancer magnetic resonance imaging , 2019, European Radiology Experimental.

[34]  Juan Hu,et al.  Prostate cancer identification: quantitative analysis of T2-weighted MR images based on a back propagation artificial neural network model , 2015, Science China Life Sciences.

[35]  Anant Madabhushi,et al.  Enhanced multi-protocol analysis via intelligent supervised embedding (EMPrAvISE): detecting prostate cancer on multi-parametric MRI , 2011, Medical Imaging.

[36]  Guy Nir,et al.  Deep Learning-Based Gleason Grading of Prostate Cancer From Histopathology Images—Role of Multiscale Decision Aggregation and Data Augmentation , 2020, IEEE Journal of Biomedical and Health Informatics.

[37]  P. LaViolette,et al.  Progressing Bevacizumab-Induced Diffusion Restriction Is Associated with Coagulative Necrosis Surrounded by Viable Tumor and Decreased Overall Survival in Patients with Recurrent Glioblastoma , 2016, American Journal of Neuroradiology.