Elizabeth J. Sutton

Breast Cancer Subtype Intertumor Heterogeneity: MRI-Based Features Predict Results of a Genomic Assay

To investigate the association between a validated, gene‐expression‐based, aggressiveness assay, Oncotype Dx RS, and morphological and texture‐based image features extracted from magnetic resonance imaging (MRI).

Quantitative MRI radiomics in the prediction of molecular classifications of breast cancer subtypes in the TCGA/TCIA data set.

Using quantitative radiomics, we demonstrate that computer-extracted magnetic resonance (MR) image-based tumor phenotypes can be predictive of the molecular classification of invasive breast cancers. Radiomics analysis was performed on 91 MRIs of biopsy-proven invasive breast cancers from National Cancer Institute’s multi-institutional TCGA/TCIA. Immunohistochemistry molecular classification was performed including estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2, and for 84 cases, the molecular subtype (normal-like, luminal A, luminal B, HER2-enriched, and basal-like). Computerized quantitative image analysis included: three-dimensional lesion segmentation, phenotype extraction, and leave-one-case-out cross validation involving stepwise feature selection and linear discriminant analysis. The performance of the classifier model for molecular subtyping was evaluated using receiver operating characteristic analysis. The computer-extracted tumor phenotypes were able to distinguish between molecular prognostic indicators; area under the ROC curve values of 0.89, 0.69, 0.65, and 0.67 in the tasks of distinguishing between ER+ versus ER−, PR+ versus PR−, HER2+ versus HER2−, and triple-negative versus others, respectively. Statistically significant associations between tumor phenotypes and receptor status were observed. More aggressive cancers are likely to be larger in size with more heterogeneity in their contrast enhancement. Even after controlling for tumor size, a statistically significant trend was observed within each size group (P=0.04 for lesions ⩽2 cm; P=0.02 for lesions >2 to ⩽5 cm) as with the entire data set (P-value=0.006) for the relationship between enhancement texture (entropy) and molecular subtypes (normal-like, luminal A, luminal B, HER2-enriched, basal-like). In conclusion, computer-extracted image phenotypes show promise for high-throughput discrimination of breast cancer subtypes and may yield a quantitative predictive signature for advancing precision medicine.

Breast cancer molecular subtype classifier that incorporates MRI features.

To use features extracted from magnetic resonance (MR) images and a machine‐learning method to assist in differentiating breast cancer molecular subtypes.

Multicentric Cancer Detected at Breast MR Imaging and Not at Mammography: Important or Not?

PURPOSE To review the magnetic resonance (MR) imaging and pathologic features of multicentric cancer detected only at MR imaging and to evaluate its potential biologic value. MATERIALS AND METHODS This retrospective study was institutional review board approved and HIPAA compliant; informed consent was waived. A review of records from 2001 to 2011 yielded 2021 patients with newly diagnosed breast cancer who underwent biopsy after preoperative MR imaging, 285 (14%) of whom had additional cancer detected at MR imaging that was occult at mammography. In 73 patients (3.6%), MR imaging identified 87 cancers in different quadrants than the known index cancer, constituting the basis of this report. In 62 of 73 patients (85%; 95% confidence interval [CI]: 75, 92), one additional cancer was found, and in 11 of 73 (15%; 95% CI: 8, 25), multiple additional cancers were found. A χ(2) test with adjustment for multiple lesions was used to examine whether MR imaging and pathologic features differ between the index lesion and additional multicentric lesions seen only at MR imaging. RESULTS Known index cancers were more likely to be invasive than MR imaging-detected multicentric cancers (88% vs 76%, P = .023). Ductal carcinoma in situ (21 of 87 lesions [24%]; 95% CI: 15, 36) represented a minority of additional MR imaging-detected multicentric cancers. Overall, the size of MR imaging-detected multicentric invasive cancers (median, 0.6 cm; range, 0.1-6.3 cm) was smaller than that of the index cancer (median, 1.2 cm; range, 0.05-7.0 cm; P = .023), although 17 of 73 (23%) (95% CI: 14, 35) patients had larger MR imaging-detected multicentric cancers than the known index lesion, and 18 of 73 (25%) (95% CI: 15, 36) had MR imaging-detected multicentric cancers larger than 1 cm. MR imaging-detected multicentric cancers and index cancers differed in histologic characteristics, invasiveness, and grade in 27 of 73 (37%) patients (95% CI: 26, 49). In four of 73 (5%) patients (95% CI: 2, 13), MR imaging-detected multicentric cancers were potentially more biologically relevant because of the presence of unsuspected invasion or a higher grade. CONCLUSION Multicentric cancer detected only at MR imaging was invasive in 66 of 87 patients (76%), larger than 1 cm in 18 of 73 patients (25%), larger than the known index cancer in 17 of 73 patients (23%), and more biologically important in four of 73 women (5%). An unsuspected additional multicentric cancer seen only at MR imaging is likely clinically relevant disease.

Quantitative apparent diffusion coefficient measurement obtained by 3.0Tesla MRI as a potential noninvasive marker of tumor aggressiveness in breast cancer.

PURPOSE To assess the association between apparent diffusion coefficient (ADC), and histological prognostic parameters in malignant breast lesions. The ability of ADC to identify lesions with the presence of Lymphovascular invasion (LVI) in breast carcinoma was also examined. MATERIALS AND METHODS This HIPAA-compliant retrospective study consisted of 212 consecutive patients with known cancers who underwent 3.0T MRI between January 2011 and 2013. In this study, a total of 126 malignant lesions in 114 women, who had undergone DWI (b-values of 0 and 1000s/mm(2)) in addition to diagnostic MRI, were included. Patients with less than 0.8cm lesions, or those who underwent neoadjuvant chemotherapy or suboptimal DW images were excluded. Classical prognostic factors [lesion size, histopathological type and grade, lymph node (LN) status and lymphovascular invasion (LVI)], molecular prognostic markers [estrogen receptor (ER), progesterone receptor (PR) and human epidermal grow factor receptor 2 (HER2)] were reviewed and recorded. A region of interest (ROI) was drawn within the lesions to measure ADC values. Statistical analyses were performed by the Wilcoxon rank sum test (statistical significance at P<0.05). Adjusted p values from multiple comparison analysis were also calculated. RESULTS This study demonstrates an inverse correlation between ADC and LVI in malignant lesions and the ability of ADC to identify aggressiveness in lesions with positive LVI. Tumor size, grade, ER, PR, HER2 and lymph node status did not impact tumor ADC value. However, tumors with LVI showed significantly lower ADC values when compared to tumors without LVI, regardless of the enhancement type, histological grade, histological type, and LN status. CONCLUSION Our study shows that ADC could be a potential clinical adjunct in the evaluation of prognostic factors related to malignant lesion aggressiveness such as LVI.

Using computer‐extracted image phenotypes from tumors on breast magnetic resonance imaging to predict breast cancer pathologic stage

The objective of this study was to demonstrate that computer‐extracted image phenotypes (CEIPs) of biopsy‐proven breast cancer on magnetic resonance imaging (MRI) can accurately predict pathologic stage.

Incidence of Internal Mammary Lymph Nodes with Silicone Breast Implants at MR Imaging after Oncoplastic Surgery

PURPOSE To assess the incidence of benign and malignant internal mammary lymph nodes (IMLNs) at magnetic resonance (MR) imaging among women with a history of treated breast cancer and silicone implant reconstruction. MATERIALS AND METHODS The institutional review board approved this HIPAA-compliant retrospective study and waived informed consent. Women were identified who (a) had breast cancer, (b) underwent silicone implant oncoplastic surgery, and (c) underwent postoperative implant-protocol MR imaging with or without positron emission tomography (PET)/computed tomography (CT) between 2000 and 2013. The largest IMLNs were measured. A benign IMLN was pathologically proven or defined as showing 1 year of imaging stability and/or no clinical evidence of disease. Malignant IMLNs were pathologically proven. Incidence of IMLN and positive predictive value (PPV) were calculated on a per-patient level by using proportions and exact 95% confidence intervals (CIs). The Wilcoxon rank sum test was used to assess the difference in axis size. RESULTS In total, 923 women with breast cancer and silicone implants were included (median age, 46 years; range, 22-89 years). The median time between reconstructive surgery and first MR imaging examination was 49 months (range, 5-513 months). Of the 923 women, 347 (37.6%) had IMLNs at MR imaging. Median short- and long-axis measurements were 0.40 cm (range, 0.20-1.70 cm) and 0.70 cm (range, 0.30-1.90 cm), respectively. Two hundred seven of 923 patients (22.4%) had adequate follow-up; only one of the 207 IMLNs was malignant, with a PPV of 0.005 (95% CI: 0.000, 0.027). Fifty-eight of 923 patients (6.3%) had undergone PET/CT; of these, 39 (67.2%) had IMLN at MR imaging. Twelve of the 58 patients (20.7%) with adequate follow-up had fluorine 18 fluorodeoxyglucose-avid IMLN, with a median standardized uptake value of 2.30 (range, 1.20-6.10). Only one of the 12 of the fluorodeoxyglucose-avid IMLNs was malignant, with a PPV of 0.083 (95% CI: 0.002, 0.385). CONCLUSION IMLNs identified at implant-protocol breast MR imaging after oncoplastic surgery for breast cancer are overwhelmingly more likely to be benign than malignant. Imaging follow-up instead of immediate metastatic work-up may be warranted.

Background, current role, and potential applications of radiogenomics

With the genomic revolution in the early 1990s, medical research has been driven to study the basis of human disease on a genomic level and to devise precise cancer therapies tailored to the specific genetic makeup of a tumor. To match novel therapeutic concepts conceived in the era of precision medicine, diagnostic tests must be equally sufficient, multilayered, and complex to identify the relevant genetic alterations that render cancers susceptible to treatment. With significant advances in training and medical imaging techniques, image analysis and the development of high‐throughput methods to extract and correlate multiple imaging parameters with genomic data, a new direction in medical research has emerged. This novel approach has been termed radiogenomics. Radiogenomics aims to correlate imaging characteristics (ie, the imaging phenotype) with gene expression patterns, gene mutations, and other genome‐related characteristics and is designed to facilitate a deeper understanding of tumor biology and capture the intrinsic tumor heterogeneity. Ultimately, the goal of radiogenomics is to develop imaging biomarkers for outcome that incorporate both phenotypic and genotypic metrics. Due to the noninvasive nature of medical imaging and its ubiquitous use in clinical practice, the field of radiogenomics is rapidly evolving and initial results are encouraging. In this article, we briefly discuss the background and then summarize the current role and the potential of radiogenomics in brain, liver, prostate, gynecological, and breast tumors.

Intravoxel incoherent motion (IVIM) histogram biomarkers for prediction of neoadjuvant treatment response in breast cancer patients

Highlights • Intravoxel incoherent motion may predict response to breast neoadjuvant treatment.• Histogram analysis amplifies IVIM by quantifying spatial heterogeneity.• Pseudodiffusvity (Dp) showed the highest potential for response prediction.

The Potential of High Resolution Magnetic Resonance Microscopy in the Pathologic Analysis of Resected Breast and Lymph Tissue.

Pathologic evaluation of breast specimens requires a fixation and staining procedure of at least 12 hours duration, delaying diagnosis and post-operative planning. Here we introduce an MRI technique with a custom-designed radiofrequency resonator for imaging breast and lymph tissue with sufficient spatial resolution and speed to guide pathologic interpretation and offer value in clinical decision making. In this study, we demonstrate the ability to image breast and lymphatic tissue using 7.0 Tesla MRI, achieving a spatial resolution of 59 × 59 × 94 μm3 with a signal-to-noise ratio of 15–20, in an imaging time of 56 to 70 minutes. These are the first MR images to reveal characteristic pathologic features of both benign and malignant breast and lymph tissue, some of which were discernible by blinded pathologists who had no prior training in high resolution MRI interpretation.