Use of MRI for Personalized Treatment of More Aggressive Tumors

HomeRadiologyVol. 295, No. 3 PreviousNext Reviews and CommentaryFree AccessEditorialUse of MRI for Personalized Treatment of More Aggressive TumorsRiham H. El Khouli , Michael A. JacobsRiham H. El Khouli , Michael A. JacobsAuthor AffiliationsFrom the Department of Radiology, Nuclear Medicine and Molecular Imaging, University of Kentucky College of Medicine, 800 Rose St, Lexington, KY 40506-9983 (R.H.E.K.); and Department of Radiology, Johns Hopkins University School of Medicine, Baltimore, Md (M.A.J.).Address correspondence to R.H.E.K. (e-mail: [email protected]).Riham H. El Khouli Michael A. JacobsPublished Online:Mar 31 2020https://doi.org/10.1148/radiol.2020200678MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In See also the article by Kim et al in this issue.Dr El Khouli is an assistant professor of radiology and the medical director of research at the University of Kentucky. Her research has focused on optimization of the advanced MRI technique for better understanding and characterization of breast lesions. As a director of the theranostics program at the University of Kentucky, her current research focuses on optimization of advanced molecular imaging techniques to improve understanding and utilization of different theranostics models. She is listed as a coinvestigator on multiple National Institutes of Health grants and clinical trials.Download as PowerPointOpen in Image Viewer Dr Jacobs is a professor of radiology and oncology at the Johns Hopkins University School of Medicine and an American Board of Radiology diagnostic medical physicist. His research interests include developing novel radiologic methods to detect, monitor, and treat many different diseases. These new methods will help in characterization of disease and determination of treatment response by using multiparametric imaging, multiparametric radiomics, and deep learning. He is currently on the editorial boards of Radiology and Medical Physics.Download as PowerPointOpen in Image Viewer For more than 10 years, dynamic contrast material–enhanced (DCE) breast MRI has been a key modality in the care of patients diagnosed with breast cancer. DCE MRI is used to detect occult ipsilateral and contralateral breast cancer and improves assessment of disease extent with superior diagnostic accuracy compared with mammography or US (1). The high sensitivity of MRI reflects tumor vascularity and allows functional information about the breast cancer to be obtained. Tumor vascularity in turn is a marker of tumor aggressiveness and is a predictor of treatment response and prognosis in different cancers (2,3). In clinical practice, tumor vascularity is reflected in the shape of the time–signal intensity or kinetic curves. A persistent curve is defined as an increase in signal intensity and is most commonly seen in benign lesions. A plateau curve is defined as signal intensity that remains constant after the initial peak enhancement and is seen in benign and malignant lesions. A washout curve is defined as a decrease in the late phase of MRI enhancement and is common in malignant lesions. To improve lesion classification, radiology reports indicate the worst kinetic curve observed (washout is better than plateau enhancement, which in turn is better than persistent enhancement) in the lesion. In the research setting, pharmacokinetic modeling analyzes quantitative data obtained with DCE MRI for breast cancer characterization, treatment response assessment, and prognostic prediction (2–4).In recent years, intratumoral heterogeneity, defined as the coexistence of subpopulations of cancer cells that differ in their genetic, morphologic, and behavioral variability within a primary tumor and between a primary tumor and its metastasis, has drawn increasing attention. Tumor heterogeneity reflects clonal diversity of the cancer cells. The clonal expansion of aggressive tumor cells leads to resistance to chemotherapy and radiation therapy and poor prognosis in patients. In patients who are not responding to treatment, it is clinically infeasible to reperform biopsy of the primary tumor, to perform biopsy of the metastases, or both. Therefore, the identification of MRI biomarkers that are associated with high tumor heterogeneity and worse prognosis may allow for modification of the treatment regimen and address this unmet need.In this issue of Radiology (5), Kim et al describe the relationship between higher kinetic heterogeneity and peak enhancement parameters extracted from preoperative DCE MRI computer-aided analysis software. Kim et al found that greater kinetic heterogeneity and peak enhancement were associated with worse distant metastasis–free survival in women with invasive breast cancer. Moreover, among the 276 women included in the study, 28 (10.1%) developed distant metastasis with median follow-up of 79 months (>5 years). The authors also reported that women with distant metastasis at initial staging had primary tumors with higher kinetic heterogeneity than did women with no distant metastasis. This could have implications for identification of intrinsic features of primary breast cancer that predict recurrence and metastatic-free survival of patients, tailoring treatment according to expected tumor aggressiveness and prognostic features (6,7).Kim et al did not find a difference in the predominant kinetic curve type and worst kinetic curve type between the metastatic and nonmetastatic groups. However, they did find a difference in delayed enhancement profile and mean value of different contrast profiles (ie, washout and plateau enhancement), as well as in the persistent curve component between the metastatic and nonmetastatic groups. This lack of difference might be due to the categorization of continuous data, the selection of the single worst curve type, and the semiquantitative analysis of changes in signal intensity.Prior studies have shown that early peak enhancement and delayed washout pattern at MRI are the most important prognostic indicators for both malignancy and more aggressive nature of tumors (1,4,8,9). Kim and colleagues add another potential value of these two DCE MRI parameters, offering greater potential in predicting distant metastasis–free survival at initial staging.The authors provide a multivariable predictive model consisting of different features of tumor grade (low [grades 1 and 2] vs high [grade 3]), lymphovascular invasion, peak enhancement, and kinetic heterogeneity. The model determined that greater kinetic heterogeneity, higher peak enhancement, higher histologic grade, and the presence of lymphovascular invasion were predictive of worse distant metastasis–free survival in women with invasive breast cancer. Of note, the kinetic heterogeneity (entropy) feature was the most significant in this model.Kim et al also describe a DCE heterogeneity feature assessed using commercially available computer-aided detection software widely used in clinical practice (5), adding practical value and easy application to clinical settings without having to buy sophisticated software or hardware.Although the study is limited by the small number of women and, more importantly, the percentage of women in the study sample who were diagnosed with distant metastatic recurrence (28 of 276 [10.1%]), Kim and colleagues managed to identify the prognostic value of DCE heterogeneity, potential threshold in peak enhancement (213%), and kinetic heterogeneity (0.66) to predict distant metastasis recurrence. The identification of these imaging biomarkers from breast MRI at the time of diagnosis is a strength of this study. Validation of their results using a separate data set and comparison with other predictive markers, such as the 21-gene array to predict recurrence of the cancer (OncotypeDX), is very important (10).This study fits perfectly with the personalized medicine concept of tailoring management according to the individual patient and tumor characteristics rather than using a one-solution-fits-all approach, especially in a subgroup requiring more aggressive treatment, more frequent imaging, or both. A prospective study with a larger sample size would give more insight regarding the value and applicability of DCE MRI–derived kinetic heterogeneity of breast cancer lesions. In summary, breast MRI shows great potential in predicting more aggressive tumors and worse patient prognosis, as well as their relationship to tumor phenotypes and genomic mutations.Disclosures of Conflicts of Interest: R.H.E.K. disclosed no relevant relationships. M.A.J. disclosed no relevant relationships.References1. El Khouli RH, Macura KJ, Kamel IR, Jacobs MA, Bluemke DA. 3-T dynamic contrast-enhanced MRI of the breast: pharmacokinetic parameters versus conventional kinetic curve analysis. AJR Am J Roentgenol 2011;197(6):1498–1505. Crossref, Medline, Google Scholar2. Preim U, Glaßer S, Preim B, Fischbach F, Ricke J. Computer-aided diagnosis in breast DCE-MRI--quantification of the heterogeneity of breast lesions. Eur J Radiol 2012;81(7):1532–1538. Crossref, Medline, Google Scholar3. Alic L, van Vliet M, van Dijke CF, Eggermont AM, Veenland JF, Niessen WJ. Heterogeneity in DCE-MRI parametric maps: a biomarker for treatment response? Phys Med Biol 2011;56(6):1601–1616. Crossref, Medline, Google Scholar4. Kim JJ, Kim JY, Kang HJ, et al. Computer-aided Diagnosis-generated Kinetic Features of Breast Cancer at Preoperative MR Imaging: Association with Disease-free Survival of Patients with Primary Operable Invasive Breast Cancer. Radiology 2017;284(1):45–54. Link, Google Scholar5. Kim JY, Kim JJ, Hwangbo L, et al. Kinetic Heterogeneity of Breast Cancer Determined Using Computer-aided Diagnosis of Preoperative MRI Scans: Relationship to Distant Metastasis-Free Survival. Radiology 2020;295:517–526. Link, Google Scholar6. Sutton EJ, Oh JH, Dashevsky BZ, et al. Breast cancer subtype intertumor heterogeneity: MRI-based features predict results of a genomic assay. J Magn Reson Imaging 2015;42(5):1398–1406. Crossref, Medline, Google Scholar7. Parekh VS, Jacobs MA. Integrated radiomic framework for breast cancer and tumor biology using advanced machine learning and multiparametric MRI. NPJ Breast Cancer 2017;3(1):43. Crossref, Medline, Google Scholar8. Nam SY, Ko ES, Lim Y, et al. Preoperative dynamic breast magnetic resonance imaging kinetic features using computer-aided diagnosis: Association with survival outcome and tumor aggressiveness in patients with invasive breast cancer. PLoS One 2018;13(4):e0195756. Crossref, Medline, Google Scholar9. Trimboli RM, Codari M, Khouri Chalouhi K, et al. Correlation between voxel-wise enhancement parameters on DCE-MRI and pathological prognostic factors in invasive breast cancers. Radiol Med (Torino) 2018;123(2):91–97. Crossref, Medline, Google Scholar10. Dowsett M, Cuzick J, Wale C, et al. Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and node-positive postmenopausal patients with breast cancer treated with anastrozole or tamoxifen: a TransATAC study. J Clin Oncol 2010;28(11):1829–1834. Crossref, Medline, Google ScholarArticle HistoryReceived: Feb 24 2020Revision requested: Feb 28 2020Revision received: Mar 10 2020Accepted: Mar 20 2020Published online: Mar 31 2020Published in print: June 2020 FiguresReferencesRelatedDetailsCited ByAdvances in the development of therapeutic strategies against COVID-19 and perspectives in the drug design for emerging SARS-CoV-2 variantsJialingYin, ChengchengLi, ChunhongYe, ZhihuiRuan, YicongLiang, YongkuiLi, JianguoWu, ZhenLuo2022 | Computational and Structural Biotechnology Journal, Vol. 20Accompanying This ArticleKinetic Heterogeneity of Breast Cancer Determined Using Computer-aided Diagnosis of Preoperative MRI Scans: Relationship to Distant Metastasis-Free SurvivalMar 31 2020RadiologyRecommended Articles PET Molecular Imaging as a Tool for Precision OncologyRadiology2020Volume: 296Issue: 2pp. 379-380Kinetic Heterogeneity of Breast Cancer Determined Using Computer-aided Diagnosis of Preoperative MRI Scans: Relationship to Distant Metastasis-Free SurvivalRadiology2020Volume: 295Issue: 3pp. 517-526Ultrafast Breast MRI to Predict Pathologic Response after Neoadjuvant TherapyRadiology2022Volume: 305Issue: 3pp. 575-577PET of Fibroblast-Activation Protein for Breast Cancer Diagnosis and StagingRadiology2021Volume: 302Issue: 1pp. 48-49Computer-aided Diagnosis–generated Kinetic Features of Breast Cancer at Preoperative MR Imaging: Association with Disease-free Survival of Patients with Primary Operable Invasive Breast CancerRadiology2017Volume: 284Issue: 1pp. 45-54See More RSNA Education Exhibits Staging Breast Cancer: A Case Based ReviewDigital Posters2022Imaging Biomarkers in Targeted Therapies: From Quantitative Imaging to RadiomicsDigital Posters2019Everything but BRCA: Imaging Hereditary Breast Cancer GenesDigital Posters2019 RSNA Case Collection Inflammatory breast cancerRSNA Case Collection2020Malignancy on abbreviated screening breast MRIRSNA Case Collection2020Metastatic Axillary Lymphadenopathy RSNA Case Collection2021 Vol. 295, No. 3 Metrics Altmetric Score PDF download