ABSTRACT
INTRODUCTION: The aim of this study was to determine if correlations exist between quantitative parameters from dynamic contrast-enhanced (DCE) and diffusion-weighted (DW) MRI with National Comprehensive Cancer Network (NCCN) risk group, Gleason score (GS), maximum tumour diameter (MTD), pre-treatment prostate-specific antigen (PSA), clinical T stage and MRI prostate volume in prostate cancer. METHOD: We retrospectively reviewed 3T multiparametric MRI reports on biopsy-proven prostate cancer patients performed during radiation treatment evaluation or an active surveillance protocol. DCE-MRI parameters included K(trans) (influx volume transfer coefficient), Kep (efflux reflux rate constant) and iAUC (initial area under the curve). Average DCE and apparent diffusion coefficient (ADC) values were recorded for regions of interest on DW-MRI. Relationships between MRI metrics and risk group, GS, MTD, PSA, clinical T stage and MRI prostate volume were examined using analysis of variance. Central and peripheral tumours were also analysed separately in a sub-analysis. Statistical significance was defined as P < 0.0125. RESULTS: Of 58 patients, 29%, 52% and 19% had low (L), intermediate (I), or high (H) NCCN risk disease, respectively. K(trans) significantly correlated with PSA. For central tumours, K(trans) significantly correlated with MTD and PSA, and Kep significantly correlated with PSA. For peripheral tumours, iAUC was significantly different when stratified by L/I/H risk and GS, and ADC score with L/I/H risk, GS, and clinical T stage. CONCLUSIONS: DCE- and DW-MRI metrics correlate with some risk stratification factors in prostate cancer. Further work is required to determine if MRI metrics are complementary or independent prognostic factors.
Subject(s)
Biomarkers, Tumor/blood , Diffusion Magnetic Resonance Imaging/methods , Image Interpretation, Computer-Assisted/methods , Multimodal Imaging/methods , Prostate-Specific Antigen/blood , Prostatic Neoplasms/diagnosis , Adult , Aged , Aged, 80 and over , Contrast Media , Humans , Male , Middle Aged , Prognosis , Prostatic Neoplasms/blood , Reproducibility of Results , Risk Assessment , Sensitivity and Specificity , Statistics as Topic , Tumor BurdenABSTRACT
BACKGROUND: Significant interest exists in establishing radiologic imaging as a valid biomarker for assessing the response of cancer to a variety of treatments. To address this problem, we have chosen to study patients with metastatic colorectal carcinoma to learn whether statistical learning theory can improve the performance of radiologists using CT in predicting patient treatment response to therapy compared with the more traditional RECIST (Response Evaluation Criteria in Solid Tumors) standard. RESULTS: Predictions of survival after 8 months in 38 patients with metastatic colorectal carcinoma using the Support Vector Machine (SVM) technique improved 30% when using additional information compared to WHO (World Health Organization) or RECIST measurements alone. With both Logistic Regression (LR) and SVM, there was no significant difference in performance between WHO and RECIST. The SVM and LR techniques also demonstrated that one radiologist consistently outperformed another. CONCLUSIONS: This preliminary research study has demonstrated that SLT algorithms, properly used in a clinical setting, have the potential to address questions and criticisms associated with both RECIST and WHO scoring methods. We also propose that tumor heterogeneity, shape, etc. obtained from CT and/or MRI scans be added to the SLT feature vector for processing.
Subject(s)
Carcinoma/diagnostic imaging , Carcinoma/secondary , Colorectal Neoplasms/diagnostic imaging , Tomography, X-Ray Computed , Area Under Curve , Biomarkers, Tumor , Carcinoma/drug therapy , Carcinoma/mortality , Colorectal Neoplasms/drug therapy , Colorectal Neoplasms/mortality , Colorectal Neoplasms/pathology , Humans , Logistic Models , Odds Ratio , ROC Curve , Software , Survival AnalysisABSTRACT
To establish radiologic imaging as a valid biomarker for assessing the response of cancer to different treatments. We study patients with metastatic colorectal carcinoma to learn whether Statistical Learning Theory (SLT) improves the performance of radiologists using Computer Tomography (CT) in predicting patient treatment response to therapy compared with traditional Response Evaluation Criteria in Solid Tumours (RECIST) standard. Preliminary research demonstrated that SLT algorithms can address questions and criticisms associated with both RECIST and World Health Organization (WHO) scoring methods. We add tumour heterogeneity, shape, etc., obtained from CT or MRI scans the feature vector for processing.