Radiomics features based on automatic segmented MRI images: Prognostic biomarkers for triple-negative breast cancer treated with neoadjuvant chemotherapy

A Multiparametric MRI-based Radiomics Model for Stratifying Postoperative Recurrence in Luminal B Breast Cancer

This study aims to develop an MRI-based radiomics model to assess the likelihood of recurrence in luminal B breast cancer. The study analyzed medical images and clinical data from 244 patients with luminal B breast cancer. Of 244 patients, 35 had experienced recurrence and 209 had not. The patients were randomly divided into the training set (51.5 ± 12.5 years old; n = 171) and the test set (51.7 ± 11.3 years old; n = 73) in a ratio of 7:3. The study employed univariate and multivariate Cox regression along with the least absolute shrinkage and selection operator (LASSO) regression methods to select radiomics features and calculate a risk score. A combined model was constructed by integrating the risk score with the clinical and pathological characteristics. The study identified two radiomics features (GLSZM and GLRLM) from DCE-MRI that were used to calculate a risk score. The AUCs were 0.860 and 0.868 in the training set and 0.816 and 0.714 in the testing set for 3- and 5-year recurrence risk, respectively. The combined model incorporating the risk score, pN, and endocrine therapy showed improved predictive power, with AUCs of 0.857 and 0.912 in the training set and 0.943 and 0.945 in the testing set for 3- and 5-year recurrence risk, respectively. The calibration curve of the combined model showed good consistency between predicted and measured values. Our study developed an MRI-based radiomics model that integrates clinical and radiomics features to assess the likelihood of recurrence in luminal B breast cancer. The model shows promise for improving clinical risk stratification and treatment decision-making.

Prediction of Receptor Status in Radiomics: Recent Advances in Breast Cancer Research

Breast cancer is a multifactorial heterogeneous disease and the leading cause of cancer-related deaths in women; its diagnosis and treatment require clinical sensitivity and a comprehensive disciplinary research approach. The expression of different receptors on tumor cells not only provides the basis for molecular typing of breast cancer but also has a decisive role in the diagnosis, treatment, and prognosis of breast cancer. To date, immunohistochemistry (IHC), which uses invasive histological sampling, has been extensively used in clinical practice to analyze the status of receptors and to make an accurate diagnosis of breast cancer. As an invasive assay, IHC can provide important biological information on tumors at a single point in time, but cannot predict future changes (due to treatment or tumor mutations) without additional invasive procedures. These issues highlight the need to develop a non-invasive method for predicting receptor status. The emerging field of radiomics may offer a non-invasive approach to identification of receptor status without requiring biopsy. In this paper, we present a review of the latest research results in radiomics for predicting the status of breast cancer receptors, with potential important clinical applications.

Pretreatment Sarcopenia and MRI-Based Radiomics to Predict the Response of Neoadjuvant Chemotherapy in Triple-Negative Breast Cancer

The association between sarcopenia and the effectiveness of neoadjuvant chemotherapy (NAC) in triple-negative breast cancer (TNBC) remains uncertain. This study aims to examine the potential of sarcopenia as a predictive factor for the response to NAC in TNBC, and to assess whether its combination with MRI radiomic signatures can improve the predictive accuracy. We collected clinical and pathological information, as well as pretreatment breast MRI and abdominal CT images, of 121 patients with TNBC who underwent NAC at our hospital between January 2012 and September 2021. The presence of pretreatment sarcopenia was assessed using the L3 skeletal muscle index. Clinical models were constructed based on independent risk factors identified by univariate regression analysis. Radiomics data were extracted on breast MRI images and the radiomics prediction models were constructed. We integrated independent risk factors and radiomic features to build the combined models. The results of this study demonstrated that sarcopenia is an independent predictive factor for NAC efficacy in TNBC. The combination of sarcopenia and MRI radiomic signatures can further improve predictive performance.

Exploring Radiomics Features Based on H&E Images as Potential Biomarkers for Evaluating Muscle Atrophy: A Preliminary Study

Radiomics features have been widely used as novel biomarkers in the diagnosis of various diseases, but whether radiomics features derived from hematoxylin and eosin (H&E) images can evaluate muscle atrophy has not been studied. Therefore, this study aims to establish a new biomarker based on H&E images using radiomics methods to quantitatively analyze H&E images, which is crucial for improving the accuracy of muscle atrophy assessment. Firstly, a weightless muscle atrophy model was established by laying macaques in bed, and H&E images of the shank muscle fibers of the control and bed rest (BR) macaques were collected. Muscle fibers were accurately segmented by designing a semi-supervised segmentation framework based on contrastive learning. Then, 77 radiomics features were extracted from the segmented muscle fibers, and a stable subset of features was selected through the LASSO method. Finally, the correlation between radiomics features and muscle atrophy was analyzed using a support vector machine (SVM) classifier. The semi-supervised segmentation results show that the proposed method had an average Spearman's and intra-class correlation coefficient (ICC) of 88% and 86% compared to manually extracted features, respectively. Radiomics analysis showed that the AUC of the muscle atrophy evaluation model based on H&E images was 96.87%. For individual features, GLSZM_SZE outperformed other features in terms of AUC (91.5%) and ACC (84.4%). In summary, the feature extraction based on the semi-supervised segmentation method is feasible and reliable for subsequent radiomics research. Texture features have greater advantages in evaluating muscle atrophy compared to other features. This study provides important biomarkers for accurate diagnosis of muscle atrophy.

A CT-based radiomics nomogram involving the cystic fluid area for differentiating appendiceal mucinous neoplasms from appendicitis with intraluminal fluid

OBJECTIVE

To develop and validate a radiomics nomogram based on computed tomography (CT) to distinguish appendiceal mucinous neoplasms (AMNs) from appendicitis with intraluminal fluid (AWIF).

METHOD

A total of 211 patients from two medical institutions were retrospectively analysed, of which 109 were pathologically confirmed as having appendicitis with concomitant CT signs of intraluminal fluid and 102 as having AMN. All patients were randomly assigned to a training (147 patients) or validation cohort (64 patients) at a 7:3 ratio. Radiomics features of the cystic fluid area of the appendiceal lesions were extracted from nonenhanced CT images using 3D Slicer software. Minimum redundancy maximum relevance and least absolute shrinkage and selection operator regression methods were employed to screen the radiomics features and develop a radiomics model. Combined radiomics nomogram and clinical-CT models were further developed based on the corresponding features selected after multivariate analysis. Lastly, receiver operating characteristic curves, and decision curve analysis (DCA) were used to assess the models' performances in the training and validation cohorts.

RESULTS

A total of 851 radiomics features were acquired from the nonenhanced CT images. Subsequently, a radiomics model consisting of eight selected features was developed. The combined radiomics nomogram model comprised rad-score, age, and mural calcification, while the clinical-CT model contained age and mural calcification. The combined model achieved area under the curves (AUCs) of 0.945 (95% confidence interval [CI]: 0.895, 0.976) and 0.933 (95% CI: 0.841, 0.980) in the training and validation cohorts, respectively, which were larger than those obtained by the radiomics (training cohort: AUC, 0.915 [95% CI: 0.865, 0.964]; validation cohort: AUC, 0.912 [95% CI: 0.843, 0.981]) and clinical-CT models (training cohort: AUC, 0.884 [95% CI: 0.820, 0.931]; validation cohort: AUC, 0.767 [95% CI: 0.644, 0.863]). Finally, DCA showed that the clinical utility of the combined model was superior to that of the clinical CT and radiomics models.

CONCLUSION

Our combined radiomics nomogram model constituting radiomics, clinical, and CT features exhibited good performance for differentiating AMN from AWIF, indicating its potential application in clinical decision-making.

Conditional generative adversarial network driven radiomic prediction of mutation status based on magnetic resonance imaging of breast cancer

BACKGROUND

Breast Cancer (BC) is a highly heterogeneous and complex disease. Personalized treatment options require the integration of multi-omic data and consideration of phenotypic variability. Radiogenomics aims to merge medical images with genomic measurements but encounter challenges due to unpaired data consisting of imaging, genomic, or clinical outcome data. In this study, we propose the utilization of a well-trained conditional generative adversarial network (cGAN) to address the unpaired data issue in radiogenomic analysis of BC. The generated images will then be used to predict the mutations status of key driver genes and BC subtypes.

METHODS

We integrated the paired MRI and multi-omic (mRNA gene expression, DNA methylation, and copy number variation) profiles of 61 BC patients from The Cancer Imaging Archive (TCIA) and The Cancer Genome Atlas (TCGA). To facilitate this integration, we employed a Bayesian Tensor Factorization approach to factorize the multi-omic data into 17 latent features. Subsequently, a cGAN model was trained based on the matched side-view patient MRIs and their corresponding latent features to predict MRIs for BC patients who lack MRIs. Model performance was evaluated by calculating the distance between real and generated images using the Fréchet Inception Distance (FID) metric. BC subtype and mutation status of driver genes were obtained from the cBioPortal platform, where 3 genes were selected based on the number of mutated patients. A convolutional neural network (CNN) was constructed and trained using the generated MRIs for mutation status prediction. Receiver operating characteristic area under curve (ROC-AUC) and precision-recall area under curve (PR-AUC) were used to evaluate the performance of the CNN models for mutation status prediction. Precision, recall and F1 score were used to evaluate the performance of the CNN model in subtype classification.

RESULTS

The FID of the images from the well-trained cGAN model based on the test set is 1.31. The CNN for TP53, PIK3CA, and CDH1 mutation prediction yielded ROC-AUC values 0.9508, 0.7515, and 0.8136 and PR-AUC are 0.9009, 0.7184, and 0.5007, respectively for the three genes. Multi-class subtype prediction achieved precision, recall and F1 scores of 0.8444, 0.8435 and 0.8336 respectively. The source code and related data implemented the algorithms can be found in the project GitHub at https://github.com/mattthuang/BC_RadiogenomicGAN .

CONCLUSION

Our study establishes cGAN as a viable tool for generating synthetic BC MRIs for mutation status prediction and subtype classification to better characterize the heterogeneity of BC in patients. The synthetic images also have the potential to significantly augment existing MRI data and circumvent issues surrounding data sharing and patient privacy for future BC machine learning studies.

Nomograms predict survival in elderly women with triple-negative breast cancer: A SEER population-based study

BACKGROUND

The effective treatment of breast cancer in elderly patients remains a major challenge.

OBJECTIVE

To construct a nomogram affecting the overall survival of triple-negative breast cancer (TNBC) and establish a survival risk prediction model.

METHODS

A total of 5317 TPBC patients with negative expression of estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2) who were diagnosed and received systematic treatment from 2010 to 2015 were collected from the American Cancer Surveillance, Epidemiology and End Results (SEER) database. They were randomly divided into training set (n= 3721) and validation set (n= 1596). Univariate and multivariate Cox regression analysis were used to identify prognostic features, and a nomogram was established to predict the probability of 1-year, 3-year and 5-year OS and BCSS. We used consistency index (C-index), calibration curve, area under the curve (AUC) and decision curve analysis (DCA) to evaluate the predictive performance and clinical utility of the nomogram.

RESULTS

The C-indices of the nomograms for OS and BCSS in the training cohort were 0.797 and 0.825, respectively, whereas those in the validation cohort were 0.795 and 0.818, respectively. The receiver operating characteristic (ROC) curves had higher sensitivity at all specificity values as compared with the Tumor Node Metastasis (TNM) system. The calibration plot revealed a satisfactory relationship between survival rates and predicted outcomes in both the training and validation cohorts. DCA demonstrated that the nomogram had clinical utility when compared with the TNM staging system.

CONCLUSION

This study provides information on population-based clinical characteristics and prognostic factors for patients with triple-negative breast cancer, and constructs a reliable and accurate prognostic nomogram.

A Review of AI-Based Radiomics and Computational Pathology Approaches in Triple-Negative Breast Cancer: Current Applications and Perspectives

Breast cancer is one of the most common and deadly cancers worldwide. Approximately, 20% of all breast cancers are characterized as triple negative (TNBC). TNBC typically is associated with a poorer prognosis relative to other breast cancer subtypes. Due to its aggressiveness and lack of response to hormonal therapy, conventional cytotoxic chemotherapy is the usual treatment; however, this treatment is not always effective, and an important percentage of patients develop recurrence. More recently, immunotherapy has started to be used on some populations with TNBC showing promising results. Unfortunately, immunotherapy is only applicable to a minority of patients and responses in metastatic TNBC have overall been modest in comparison to other cancer types. This situation evidences the need for developing effective biomarkers that help to stratify and personalize patient management. Thanks to recent advances in artificial intelligence (AI), there has been an increasing interest in its use for medical applications aiming at supporting clinical decision making. Several works have used AI in combination with diagnostic medical imaging, more specifically radiology and digitized histopathological tissue samples, aiming to extract disease-specific information that is difficult to quantify by the human eye. These works have demonstrated that analysis of such images in the context of TNBC has great potential for (1) risk-stratifying patients to identify those patients who are more likely to experience disease recurrence or die from the disease and (2) predicting pathologic complete response. In this manuscript, we present an overview on AI and its integration with radiology and histopathological images for developing prognostic and predictive approaches for TNBC. We present state of the art approaches in the literature and discuss the opportunities and challenges with developing AI algorithms regarding further development and clinical deployment, including identifying those patients who may benefit from certain treatments (e.g., adjuvant chemotherapy) from those who may not and thereby should be directed toward other therapies, discovering potential differences between populations, and identifying disease subtypes.

Artificial intelligence in breast imaging: potentials and challenges

Breast cancer, which is the most common type of malignant tumor among humans, is a leading cause of death in females. Standard treatment strategies, including neoadjuvant chemotherapy, surgery, postoperative chemotherapy, targeted therapy, endocrine therapy, and radiotherapy, are tailored for individual patients. Such personalized therapies have tremendously reduced the threat of breast cancer in females. Furthermore, early imaging screening plays an important role in reducing the treatment cycle and improving breast cancer prognosis. The recent innovative revolution in artificial intelligence (AI) has aided radiologists in the early and accurate diagnosis of breast cancer. In this review, we introduce the necessity of incorporating AI into breast imaging and the applications of AI in mammography, ultrasonography, magnetic resonance imaging, and positron emission tomography/computed tomography based on published articles since 1994. Moreover, the challenges of AI in breast imaging are discussed.

Triple-Negative Breast Cancer: Histopathologic Features, Genomics, and Treatment

Triple-negative breast cancer (TNBC) is a heterogeneous and aggressive group of tumors that are defined by the absence of estrogen and progesterone receptors and lack of ERBB2 (formerly HER2 or HER2/neu) overexpression. TNBC accounts for 8%-13% of breast cancers. In addition, it accounts for a higher proportion of breast cancers in younger women compared with those in older women, and it disproportionately affects non-Hispanic Black women. TNBC has high metastatic potential, and the risk of recurrence is highest during the 5 years after it is diagnosed. TNBC exhibits benign morphologic imaging features more frequently than do other breast cancer subtypes. Mammography can be suboptimal for early detection of TNBC owing to factors that include the fast growth of this cancer, increased mammographic density in young women, and lack of the typical features of malignancy at imaging. US is superior to mammography for TNBC detection, but benign-appearing features can lead to misdiagnosis. Breast MRI is the most sensitive modality for TNBC detection. Most cases of TNBC are treated with neoadjuvant chemotherapy, followed by surgery and radiation. MRI is the modality of choice for evaluating the response to neoadjuvant chemotherapy. Survival rates for individuals with TNBC are lower than those for persons with hormone receptor-positive and human epidermal growth factor receptor 2-positive cancers. The 5-year survival rates for patients with localized, regional, and distant disease at diagnosis are 91.3%, 65.8%, and 12.0%, respectively. The early success of immunotherapy has raised hope regarding the development of personalized strategies to treat TNBC. Imaging and tumor biomarkers are likely to play a crucial role in the prediction of TNBC treatment response and TNBC patient survival in the future. ©RSNA, 2023 Quiz questions for this article are available in the supplemental material.

Radiomics Based on DCE-MRI for Predicting Response to Neoadjuvant Therapy in Breast Cancer

RATIONALE AND OBJECTIVES

To compare the value of radiomics and diameter% based on pre- and early-treatment dynamic enhanced MR (DCE-MRI) of the breast in predicting response to neoadjuvant therapy (NAT) in breast cancer and to construct a tool for early noninvasive prediction of NAT outcomes.

MATERIALS AND METHODS

Retrospective analysis of clinical and imaging data of 142 patients with primary invasive breast cancer who underwent DCE-MRI before and after two cycles of NAT at our institution. Enroled patients were randomly assigned in a 7:3 ratio to the training group and the test group. Patients were divided into pathological complete response (pCR) and non-pathological complete response groups based on surgical pathology findings after NAT. The maximum diameter relative regression values (Diameter%) before and after treatment were calculated and the conventional imaging Diameter% model was constructed. Based on pre- and early-NAT DCE-MRI, the optimal features of pre-NAT, early-NAT, and delta radiomics were screened using redundancy analysis, least absolute shrinkage, and selection operator methods to construct the corresponding radiomics model and calculate the Radscores. Indicators that were statistically significant in the univariate analysis of clinical data were further screened by stepwise regression and combined with Radscores to construct the fusion model. All models were evaluated and compared.

RESULTS

In the test set, the area under the curve (AUC) of the delta radiomics model (0.87) was higher than that of the pre-NAT, early-NAT radiomics models (0.57, 0.78) and the Diameter% model (0.83). The fusion model had the best efficacy in predicting pCR after NAT, with AUCs of 0.91 in the training and test sets. And its nomogram plot showed that Radscore of early-NAT radiomics had the greatest weight. In the test set, the fusion model and Delta radiomics model improved the efficacy of predicting pCR by 35.56% and 14.19%, respectively, compared to the Diameter% model (P = 0 and .039). Clinical decision curves showed the highest overall clinical benefit for the fusion model.

CONCLUSION

Radiomics, especially delta and early-NAT radiomics, may be potential biomarkers for early noninvasive prediction of NAT outcomes. And a fusion model constructed from meaningful clinicopathological indicators combined with radiomics can effectively predict NAT response.

Predicting pathological complete response based on weakly and semi-supervised joint learning from breast cancer MRI

Neoadjuvant chemotherapy (NAC) is the standard treatment for breast cancer patients. Patients achieving complete pathological response (pCR) after NAC usually have a good prognosis. However, automatic pCR prediction has been a challenging problem due to lacking well annotations in 3D MRI. Thus far, unifying different annotation information to predict the tumor's early response to NAC has not been systematically addressed. This paper proposes a weakly and semi-supervised joint learning method that integrates attentional features from multi-parametric MRI with radiomic features for predicting pCR to NAC in breast cancer patients. The attention-based multi-instance learning (MIL) is first developed to generate informative MRI bag-level features and mine key instances. The mean-teacher framework is then employed to segment tumor regions in a semi-supervised setting for extracting radiomic features. We perform experiments on 442 patients' data and show that our method achieves an AUC value of 0.85 in pCR prediction, which is superior to comparative methods. It is also shown that learning from multi-parametric MRI outperforms that of single-parameter MRI in pCR prediction.

Predicting IDH Mutation Status in Low-Grade Gliomas Based on Optimal Radiomic Features Combined with Multi-Sequence Magnetic Resonance Imaging

The IDH somatic mutation status is an important basis for the diagnosis and classification of gliomas. We proposed a "6-Step" general radiomics model to noninvasively predict the IDH mutation status by simultaneously tuning combined multi-sequence MRI and optimizing the full radiomics processing pipeline. Radiomic features (n = 3776) were extracted from multi-sequence MRI (T1, T2, FLAIR, and T1Gd) in low-grade gliomas (LGGs), and a total of 45,360 radiomics pipeline were investigated according to different settings. The predictive ability of the general radiomics model was evaluated with regards to accuracy, stability, and efficiency. Based on numerous experiments, we finally reached an optimal pipeline for classifying IDH mutation status, namely the T2+FLAIR combined multi-sequence with the wavelet image filter, mean data normalization, PCC dimension reduction, RFE feature selection, and SVM classifier. The mean and standard deviation of AUC, accuracy, sensitivity, and specificity were 0.873 ± 0.05, 0.876 ± 0.09, 0.875 ± 0.11, and 0.877 ± 0.15, respectively. Furthermore, 14 radiomic features that best distinguished the IDH mutation status of the T2+FLAIR multi-sequence were analyzed, and the gray level co-occurrence matrix (GLCM) features were shown to be of high importance. Apart from the promising prediction of the molecular subtypes, this study also provided a general tool for radiomics investigation.

Integration of radiogenomic features for early prediction of pathological complete response in patients with triple-negative breast cancer and identification of potential therapeutic targets

Background

We established a radiogenomic model to predict pathological complete response (pCR) in triple-negative breast cancer (TNBC) and explored the association between high-frequency mutations and drug resistance.

Methods

From April 2018 to September 2019, 112 patients who had received neoadjuvant chemotherapy were included. We randomly split the study population into training and validation sets (2:1 ratio). Contrast-enhanced magnetic resonance imaging scans were obtained at baseline and after two cycles of treatment and were used to extract quantitative radiomic features and to construct two radiomics-only models using a light gradient boosting machine. By incorporating the variant allele frequency features obtained from baseline core tissues, a radiogenomic model was constructed to predict pCR. Additionally, we explored the association between recurrent mutations and drug resistance.

Results

The two radiomics-only models showed similar performance with AUCs of 0.71 and 0.73 (p = 0.55). The radiogenomic model had a higher predictive ability than the radiomics-only model in the validation set (p = 0.04), with a corresponding AUC of 0.87 (0.73–0.91).

Two highly frequent mutations were selected after comparing the mutation sites of pCR and non-pCR populations. The MED23 mutation p.P394H caused epirubicin resistance in vitro (p < 0.01). The expression levels of γ-H2A.X, p-ATM and p-CHK2 in MED23 p.P394H cells were significantly lower than those in wild type cells (p < 0.01). In the HR repair system, the GFP positivity rate of MED23 p.P394H cells was higher than that in wild-type cells (p < 0.01).

Conclusions

The proposed radiogenomic model has the potential to accurately predict pCR in TNBC patients. Epirubicin resistance after MED23 p.P394H mutation might be affected by HR repair through regulation of the p-ATM-γ-H2A.X-p-CHK2 pathway.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12967-022-03452-1.

Radiomics in Triple Negative Breast Cancer: New Horizons in an Aggressive Subtype of the Disease

In the last decade, the analysis of the medical images has evolved significantly, applications and tools capable to extract quantitative characteristics of the images beyond the discrimination capacity of the investigator's eye being developed. The applications of this new research field, called radiomics, presented an exponential growth with direct implications in the diagnosis and prediction of response to therapy. Triple negative breast cancer (TNBC) is an aggressive breast cancer subtype with a severe prognosis, despite the aggressive multimodal treatments applied according to the guidelines. Radiomics has already proven the ability to differentiate TNBC from fibroadenoma. Radiomics features extracted from digital mammography may also distinguish between TNBC and non-TNBC. Recent research has identified three distinct subtypes of TNBC using IRM breast images voxel-level radiomics features (size/shape related features, texture features, sharpness). The correlation of these TNBC subtypes with the clinical response to neoadjuvant therapy may lead to the identification of biomarkers in order to guide the clinical decision. Furthermore, the variation of some radiomics features in the neoadjuvant settings provides a tool for the rapid evaluation of treatment efficacy. The association of radiomics features with already identified biomarkers can generate complex predictive and prognostic models. Standardization of image acquisition and also of radiomics feature extraction is required to validate this method in clinical practice.

Breast cancer diagnosis in an early stage using novel deep learning with hybrid optimization technique

Breast cancer is one of the primary causes of death that is occurred in females around the world. So, the recognition and categorization of initial phase breast cancer are necessary to help the patients to have suitable action. However, mammography images provide very low sensitivity and efficiency while detecting breast cancer. Moreover, Magnetic Resonance Imaging (MRI) provides high sensitivity than mammography for predicting breast cancer. In this research, a novel Back Propagation Boosting Recurrent Wienmed model (BPBRW) with Hybrid Krill Herd African Buffalo Optimization (HKH-ABO) mechanism is developed for detecting breast cancer in an earlier stage using breast MRI images. Initially, the MRI breast images are trained to the system, and an innovative Wienmed filter is established for preprocessing the MRI noisy image content. Moreover, the projected BPBRW with HKH-ABO mechanism categorizes the breast cancer tumor as benign and malignant. Additionally, this model is simulated using Python, and the performance of the current research work is evaluated with prevailing works. Hence, the comparative graph shows that the current research model produces improved accuracy of 99.6% with a 0.12% lower error rate.