Radiomic Analysis of Pharmacokinetic Heterogeneity Within Tumor Based on the Unsupervised Decomposition of Dynamic Contrast‐Enhanced MRI for Predicting Histological Characteristics of Breast Cancer

Abstract

Background Breast tumor heterogeneity is associated with histological characteristics. However, pharmacokinetic (PK) heterogeneity within tumor might merit further exploration. Purpose To enhance the predictive power of molecular subtypes, Ki‐67, and tumor grade by analyzing PK heterogeneity within tumor based on dynamic contrast‐enhanced magnetic resonance imaging (DCE‐MRI). Study Type Retrospective. Population Two hundred and eight biopsy‐proven breast cancer patients, randomly divided into a training cohort (N = 144) and a testing cohort (N = 64). Field Strength/Sequence T1‐weighted DCE‐MRI at 3.0 T. Assessment A convex analysis of mixtures‐compartmental modeling decomposition method was used to estimate the PK parameter (i.e., the volume transfer constant Ktrans) in tumor subregions with distinct physiological kinetic patterns, including fast‐flow kinetics, slow‐flow kinetics, and plasma input. Radiomic features based on the PK parameter were calculated from each tumor subregion. Statistical Tests The training cohort was used to build random forest classifiers based on the optimal features determined by the 5‐fold cross‐validation method. The performance was assessed on the testing cohort using the area under the receiver operating characteristic curve (AUC). The AUCs derived from the tumor subregion‐based PK parameter were compared with those of the original images of the entire tumor using the DeLong test. A P‐value of <0.05 was considered statistically significant. Results The tumor subregion‐based PK parameter, which yielded the highest AUCs of 0.8782, 0.7568, 0.7019, 0.7963, 0.8080, and 0.7375 for luminal A, luminal B, basal‐like, human epidermal growth factor receptor 2, Ki‐67, and tumor grade, respectively, obtained better diagnostic performance than the original images in the entire tumor (highest AUCs = 0.8612, 0.6191, 0.5593, 0.7704, 0.7494, and 0.6261, respectively). In particular, statistically significant improvement in the diagnostic performance was obtained for luminal B. Data Conclusion Radiomic analysis of PK heterogeneity within tumor can enhance the predictive performance of radiomic models compared with that of the entire tumor. Level of Evidence 4 Technical Efficacy Stage 3

Radiomics Characterization of Tissues in an Animal Brain Tumor Model Imaged Using Dynamic Contrast Enhanced (DCE) MRI

“…Physiological and mechanical properties of tumor have been shown to characterize heterogeneity in gene expression, and phenotype for Glioblastoma (GBM) 47,53,55,64,71,[125][126][127][128][129][130][131] . Many studies [31][32][33][34][35][36][37][38][39][40][41][42][132][133][134][135][136][137] Zhang et al 41,137 have recently shown the association between the subregional PK-based radiomics information of breast tumor and its histological characteristics. They have also shown that compared to the entire tumor region, subregional PK-based radiomics information can enhance the predictive performance of the radiomic models 41,137 .…”

“…Many studies [31][32][33][34][35][36][37][38][39][40][41][42][132][133][134][135][136][137] Zhang et al 41,137 have recently shown the association between the subregional PK-based radiomics information of breast tumor and its histological characteristics. They have also shown that compared to the entire tumor region, subregional PK-based radiomics information can enhance the predictive performance of the radiomic models 41,137 . This highlights the importance of the inclusion of local information in characterization of tumor heterogeneity and strongly supports the dynamic convolutional radiomics analysis used in this study.…”

“…Various radiomics analyses and adaptive models (AMs) have been developed to analyze multiparametric MR images of glioblastoma (GBM) to predict outcomes of clinical interest, such as recurrence and survival [7][8][9][10][11][12][13][14][15][16][17] , response to treatment 7,8,13,14,[17][18][19] , molecular mutation status 16,[20][21][22][23][24] , and subclinical peritumoral infiltration 18, [24][25][26][27][28][29] . Radiomics has shown considerable success in the prediction of noninvasive biomarkers of outcome 5,30 , quantification and tissue characterization in the field of dynamic contrast enhanced (DCE) MRI [31][32][33][34][35][36][37][38][39][40][41][42] , and association of imaging biomarkers to biological mechanisms 5,43,44 . However, little progress has been made toward modeling and ...…”

New Cluster Analysis Method for Quantitative Dynamic Contrast‐Enhanced MRI Assessing Tumor Heterogeneity Induced by a Tumor‐Microenvironmental Ameliorator (E7130) Treatment to a Breast Cancer Mouse Model

“…8 Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), and pharmacokinetic (PK) modeling, in particular, can identify tumor subregions that have distinct patterns of contrast uptake 9 and can reveal tissue perfusion and permeability. 10 In clinical practice, a hypervascular area typically shows a high volume transfer constant (K trans ) value indicative of increased tumor perfusion, even though K trans is defined as a volume transfer constant from tumor capillaries to the extravascular extracellular space (EES). 11,12 It has been shown in preclinical studies that areas near the tumor rim show high K trans values, whereas those near the core show low K trans values.…”

Editorial for “Radiomic Analysis of Pharmacokinetic Heterogeneity Within Tumor Based on the Unsupervised Decomposition of DCE‐MRI for Predicting Histological Characteristics of Breast Cancer”

“…In this issue of JMRI, Zhang et al 7 employed a radiomic pipeline to evaluate the predictive ability of pharmacokinetic parameters for breast cancer molecular subtypes, Ki‐67 expression level, and tumor grade. Specifically, they compared the volume transfer constant K trans features from tumor subregions, obtained by a CAM decomposition method, to those of the entire tumor, and found enhanced predictive performance with the habitat imaging approach.…”