Breast Cancer Lesions Research Articles

Patterns of Immunohistochemical Expression of P53, BCL2, PTEN, and HER2/neu Tumor Markers in Specific Breast Cancer Lesions.

Objective This study aimed to associate the expression of P53, BCL2, PTEN, and HER2/neu tumor markers in specific breast cancer lesions. Methods This study analyzed the immunohistochemical expression of P53, BCL2, PTEN, and HER2/neu tumor markers for 306 patients who presented with lesions. Tissue blocks and patients' identification data were retrieved from the department of pathology, AL Madinah Almonwarah hospital, Al Madinah, UAE. Results Of the 306 patients, 104 had benign lesions and 202 had malignancy (including 194 females and 6 males). Most females were presented with invasive ductal carcinoma (IDC), followed by infiltrating ductal carcinoma, and invasive lobular carcinoma (ILC), representing 70%, 23.2%, and 3.7%, respectively. Positive P53, BCL2, PTEN, and HER2 were identified in 20.8%, 11.9%, 91%, and 18.3%, respectively. Conclusion: The expression of P53, BCL2, PTEN, and HER2/neu tumor markers among Saudi patients with breast cancer is relatively similar in many parts of the world.

Calcification, Posterior Acoustic, and Blood Flow: Ultrasonic Characteristics of Triple-Negative Breast Cancer.

Previous studies suggest that triple-negative breast cancer (TNBC) may have unique imaging characteristics, however, studies focused on the imaging characteristics of TNBC are still limited. The aim of the present study is to analyze the ultrasonic characteristics of TNBC and to provide more reliable information on imaging diagnosis of TNBC. This retrospective study was performed including 162 TNBC patients with 184 TNBC lesions. 174 non-TNBC cases with 196 lesions were used as the control group. The median size of TNBC lesions and non-TNBC lesions were 23 mm × 16 mm and 21 mm × 15 mm, respectively. The shape of most breast cancer lesions was irregular. However, 15.30% (28/183) TNBC lesions and 16.84% (33/196) non-TNBC lesions were oval-shaped. Most breast cancer lesions (79.78% TNBC & 85.71% non-TNBC) were ill-defined. In comparison to non-TNBC, the distinctive ultrasonic characteristics of TNBC were summarized as three features: calcifications, posterior acoustic, and blood flow. Microcalcifications was less common in non-TNBC. The remarkable posterior acoustic characteristics on TNBC were no posterior acoustic features (136, 73.91%). Avascular pattern (21.74%) was also more common in TNBC. The other feature of TNBC was markedly hypoechoic lesions (23.91%). The above-mentioned differences between TNBC and non-TNBC were significant. 93.48% TBNC and 94.39% non–TNBC lesions were in BI-RADS-US category of 4A-5. The results indicate that TNBC has some distinctive ultrasound characteristics. Ultrasound is a useful adjunct in early detection of breast cancer. A combination of ultrasound with mammography is excellent for detecting breast cancer.

AACR Cancer Progress Report 2022: Decoding Cancer Complexity, Integrating Science, and Transforming Patient Outcomes.

AACR Cancer Progress Report 2022: Decoding Cancer Complexity, Integrating Science, and Transforming Patient Outcomes.

Breast lesion detection using an anchor-free network from ultrasound images with segmentation-based enhancement

The survival rate of breast cancer patients is closely related to the pathological stage of cancer. The earlier the pathological stage, the higher the survival rate. Breast ultrasound is a commonly used breast cancer screening or diagnosis method, with simple operation, no ionizing radiation, and real-time imaging. However, ultrasound also has the disadvantages of high noise, strong artifacts, low contrast between tissue structures, which affect the effective screening of breast cancer. Therefore, we propose a deep learning based breast ultrasound detection system to assist doctors in the diagnosis of breast cancer. The system implements the automatic localization of breast cancer lesions and the diagnosis of benign and malignant lesions. The method consists of two steps: 1. Contrast enhancement of breast ultrasound images using segmentation-based enhancement methods. 2. An anchor-free network was used to detect and classify breast lesions. Our proposed method achieves a mean average precision (mAP) of 0.902 on the datasets used in our experiment. In detecting benign and malignant tumors, precision is 0.917 and 0.888, and recall is 0.980 and 0.963, respectively. Our proposed method outperforms other image enhancement methods and an anchor-based detection method. We propose a breast ultrasound image detection system for breast cancer detection. The system can locate and diagnose benign and malignant breast lesions. The test results on single dataset and mixed dataset show that the proposed method has good performance.

Percutaneous CT-Guided Bone Lesion Biopsy for Confirmation of Bone Metastases in Patients with Breast Cancer.

We aimed to determine diagnostic accuracy of CT-guided bone lesion biopsy for the confirmation of bone metastases in patients with breast cancer and assessment of hormone receptor status in metastatic tissue. A total of 56 female patients with breast cancer that underwent CT-guided biopsy of suspected bone metastasis were enrolled in this retrospective study. Three different techniques were employed to obtain samples from various sites of skeleton. Collectively, 11 true negative and 3 false negative findings were revealed. The sensitivity of CT-guided biopsy for diagnosing bone metastases was 93.6%, specificity was 100% and accuracy was 94.8%. Discordance in progesterone receptor status and complete concordance in estrogen receptor status was observed. Based on our single-center experience, bone metastasis biopsy should be routinely performed in patients with breast cancer and suspicious bone lesions, due to the impact on further treatment.

An Application of Generalized Linear Models to Fine Needle Aspiration in Breast Cancer

Breast cancer is currently the most dangerous cancer for women worldwide. Doctors routinely employ biopsies, diagnostic mammograms, and other techniques to detect and diagnose breast cancer. The Fine Needle Aspiration, also referred to as fine needle biopsy, is a technique for assessing tumors that involves inserting a needle into a mass to retrieve alive cells. However, the current breast biopsy test is time-consuming and unable to detect early breast cancer. Applying the statistical tools to fine-needle aspiration is helpful in developing its feasibility and reducing test time, thereby reducing the cost of service as well as waiting time. In this study, the diagnostic model was fitted with a generalized linear model as the framework and Least Absolute Shrinkage and Selection Operator regression as the essential methods. Amongst cellular level features, which are variables in the model, some features were identified that play an essential role in the models, including texture, smoothness, concave points, and fractal dimension. The high accuracy (>0.9) obtained from the model in data testing supported that Generalized-Linear-Models-based machine prediction can effectively assist physicians in their clinical diagnosis. In addition, essential features in the model could be considered to have some association with the hidden lesion of breast cancer.

Combined 18F-FDG and 18F-Alfatide II PET May Predict Luminal B (HER2 Negative) Subtype and Nonluminal Subtype of Invasive Breast Cancer.

Noninvasive PET molecular imaging using radiopharmaceuticals is important to classify breast cancer in the clinic. The aim of this study was to investigate the combination of 18F-FDG and 18F-Alfatide II for predicting molecular subtypes of invasive breast cancer. Forty-four female patients with clinically suspected breast cancer were recruited and underwent 18F-FDG and 18F-Alfatide II PET/CT within a week. Tracer uptake in breast lesions was assessed using the maximum standardized uptake value (SUVmax), mean standardized uptake value (SUVmean), and SUVmax ratio of 18F-FDG to 18F-Alfatide II (FAR). Invasive breast cancer lesions were further classified as luminal A subtype, luminal B subtype, human epidermal growth factor receptor-2 (HER2) overexpressing subtype, and triple negative subtype according to the expression of the estrogen receptor (ER), progesterone receptor (PR), HER2, and Ki-67. Among 44 patients, 35 patients were pathologically diagnosed with invasive breast cancer. The SUVmax and SUVmean of 18F-FDG were significantly higher in the ER-negative group than those in the ER-positive group, as well as in the PR-negative group than those in the PR-positive group. However, the SUVmax and SUVmean of 18F-Alfatide II were higher in the ER-positive group and the PR-positive group. By combining 18F-FDG and 18F-Alfatide II, the FAR was lower in the ER-positive group and the PR-positive group. The HER2 overexpressing subtype showed the highest SUVmax and SUVmean for 18F-FDG while the luminal B (HER2 negative) subtype revealed the lowest values. The luminal B (HER2 negative) subtype showed the highest 18F-Alfatide II SUVmax, while the triple negative subtype showed the lowest 18F-Alfatide II SUVmax. The FAR was the lowest in the luminal B (HER2 negative) subtype and much higher in the HER2 overexpressing and triple negative subtypes. FAR less than 1 predicted the luminal B (HER2 negative) subtype with high specificity (93.1%) and NPV (90%). FAR greater than 3 predicted the HER2 overexpressing subtype and triple negative subtype (namely, the nonluminal subtype) with very high specificity (100%) and PPV (100%). In summary, FAR, the combined PET parameter of 18F-FDG and 18F-Alfatide II, can be used to predict molecular subtypes of invasive breast cancer, especially for the luminal B (HER2 negative) subtype and the nonluminal subtype.

Evaluating DNA Methylation in Random Fine Needle Aspirates from the Breast to Inform Cancer Risk

Introduction. Critical regulatory genes are functionally silenced by DNA hypermethylation in breast cancer and premalignant lesions. The objective of this study was to examine whether DNA methylation assessed in random fine needle aspirates (rFNA) can be used to inform breast cancer risk. Methods. In 20 women with invasive breast cancer scheduled for surgery at Johns Hopkins Hospital, cumulative methylation status was assessed in a comprehensive manner. rFNA was performed on tumors, adjacent normal tissues, and all remaining quadrants. Pathology review was conducted on blocks from all excised tissue. The cumulative methylation index (CMI) for 12 genes was assessed by a highly sensitive QM-MSP assay in 280 aspirates and tissue from 11 incidental premalignant lesions. Mann–Whitney and Kruskal Wallis tests were used to compare median CMI by patient, location, and tumor characteristics. Results. The median age of participants was 49 years (interquartile range [IQR]: 44–58). DNA methylation was detectable at high levels in all tumor aspirates (median CMI = 252, IQR: 75–111). Methylation was zero or low in aspirates from adjacent tissue (median CMI = 11, IQR: 0–13), and other quadrants (median CMI = 2, IQR: 1–5). Nineteen incidental lesions were identified in 13 women (4 malignant and 15 premalignant). Median CMI levels were not significantly different in aspirates from quadrants ( p = 0.43 ) or adjacent tissue ( p = 0.93 ) in which 11 methylated incidental lesions were identified. Conclusions. The diagnostic accuracy of methylation based on rFNA alone to detect premalignant lesions or at-risk quadrants is poor and therefore should not be used to evaluate cancer risk. A more targeted approach needs to be evaluated.

Role of dynamic and parametric whole-body FDG PET/CT imaging in molecular characterization of primary breast cancer: a single institution experience.

The aim of this pilot study was to assess the role of dynamic whole-body PET and parametric imaging in the biological characterization of primary breast cancer. In total 24 histologically proven primary breast cancer lesions in 21 consecutive patients were retrospectively analyzed. Each patient underwent 18F-fluoro-deoxyglucose whole-body dynamic PET-CT before any treatment. Dynamic PET images were acquired in the list mode for a total duration of 70 min. The reconstructed parametric imaging generated Patlak plot-based 'Slope' and 'Intercept' images, from which parametric indices ki and DV were obtained. The standard uptake value (SUV) metric was also obtained by summing the last few frames of the dynamic study. ki, distribution volume (DV) and SUV were correlated with the histological tumor grade, biomarkers [hormone receptors and human epidermal growth factor receptor 2 (HER-2) neu expression] and molecular subtypes (A, B and C) as well as with tumor size, regional nodal metastases and distant metastases. The mean ki was found to be significantly higher in grade III than II lesions (P = 0.005), HER-2 neu positive status (P = 0.04) and molecular subtype B (P = 0.04) as well as in greater than T1 lesions(P = 0.0003 and P = 0.04, respectively) and node-positive lesions (P = 0.009). Though mean ki was not found to be significant for the hormone receptors status (P = 0.08), it showed the best correlation compared to the other parameters (P = 0.8 for DV and P = 0.1 for SUV). Spearman's correlation test, area under the curve (AUC) and mismatch percentage also revealed ki to predict tumor grade (AUC, 0.95; r = 0.7; P = 0.0001), HER-2 neu status and molecular subtypes (AUC, 0.81; r = 0.49 and P = 0.01) along with the hormone receptors status (AUC, 0.83; r = 0.32; P = 0.1). The mean DV failed to show any association with any of the biological or anatomical staging parameters. Though ki was found to be comparable to that of SUV in almost all the assessed parameters, it appeared to be better for predicting hormone receptors status even though both parameters were not statistically significant. Our initial observation in a small cohort of breast cancer patients suggests that ki is promising in stratifying primary breast cancer lesions according to the tumor grade and biological characteristics.

Predicting axillary lymph node metastasis in breast cancer using the similarity of quantitative dual-energy CT parameters between the primary lesion and axillary lymph node

PurposeTo evaluate the similarity of quantitative dual-energy computed tomography (DECT) parameters between the primary breast cancer lesion and axillary lymph node (LN) for predicting LN metastasis.Materials and methodsThis retrospective study included patients with breast cancer who underwent contrast-enhanced DECT between July 2019 and April 2021. Relationships between LN metastasis and simple DECT parameters, similarity of DECT parameters, and pathological and morphological features were analyzed. ROC curve analysis was used to evaluate diagnostic ability.ResultsOverall, 137 LNs (39 metastases and 98 non-metastases) were evaluated. Significant differences were observed in some pathological (nuclear grade, estrogen receptor status, and Ki67 index) and morphological characteristics (shortest and longest diameters of the LN, longest-to-shortest diameter ratio, and hilum), most simple DECT parameters, and all DECT similarity parameters between the LN metastasis and non-metastasis groups (all, P < 0.001–0.004). The shortest diameter of the LN (odds ratio 2.22; 95% confidence interval 1.47, 3.35; P < 0.001) and the similarity parameter of 40-keV attenuation (odds ratio, 2.00; 95% confidence interval 1.13, 3.53; P = 0.017) were independently associated with LN metastasis compared to simple DECT parameters of 40-keV attenuation (odds ratio 1.01; 95% confidence interval 0.99, 1.03; P =0.35). The AUC value of the similarity parameters for predicting metastatic LN was 0.78–0.81, even in cohorts with small LNs (shortest diameter < 5 mm) (AUC value 0.73–0.78).ConclusionThe similarity of the delayed-phase DECT parameters could be a more useful tool for predicting LN metastasis than simple DECT parameters in breast cancer, regardless of LN size.

A machine learning model based on ultrasound image features to assess the risk of sentinel lymph node metastasis in breast cancer patients: Applications of scikit-learn and SHAP.

BackgroundThis study aimed to determine an optimal machine learning (ML) model for evaluating the preoperative diagnostic value of ultrasound signs of breast cancer lesions for sentinel lymph node (SLN) status.MethodThis study retrospectively analyzed the ultrasound images and postoperative pathological findings of lesions in 952 breast cancer patients. Firstly, the univariate analysis of the relationship between the ultrasonographic features of breast cancer morphological features and SLN metastasis. Then, based on the ultrasound signs of breast cancer lesions, we screened ten ML models: support vector machine (SVM), extreme gradient boosting (XGBoost), random forest (RF), linear discriminant analysis (LDA), logistic regression (LR), naive bayesian model (NB), k-nearest neighbors (KNN), multilayer perceptron (MLP), long short-term memory (LSTM), and convolutional neural network (CNN). The diagnostic performance of the model was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), Kappa value, accuracy, F1-score, sensitivity, and specificity. Then we constructed a clinical prediction model which was based on the ML algorithm with the best diagnostic performance. Finally, we used SHapley Additive exPlanation (SHAP) to visualize and analyze the diagnostic process of the ML model.ResultsOf 952 patients with breast cancer, 394 (41.4%) had SLN metastasis, and 558 (58.6%) had no metastasis. Univariate analysis found that the shape, orientation, margin, posterior features, calculations, architectural distortion, duct changes and suspicious lymph node of breast cancer lesions in ultrasound signs were associated with SLN metastasis. Among the 10 ML algorithms, XGBoost had the best comprehensive diagnostic performance for SLN metastasis, with Average-AUC of 0.952, Average-Kappa of 0.763, and Average-Accuracy of 0.891. The AUC of the XGBoost model in the validation cohort was 0.916, the accuracy was 0.846, the sensitivity was 0.870, the specificity was 0.862, and the F1-score was 0.826. The diagnostic performance of the XGBoost model was significantly higher than that of experienced radiologists in some cases (P<0.001). Using SHAP to visualize the interpretation of the ML model screen, it was found that the ultrasonic detection of suspicious lymph nodes, microcalcifications in the primary tumor, burrs on the edge of the primary tumor, and distortion of the tissue structure around the lesion contributed greatly to the diagnostic performance of the XGBoost model.ConclusionsThe XGBoost model based on the ultrasound signs of the primary breast tumor and its surrounding tissues and lymph nodes has a high diagnostic performance for predicting SLN metastasis. Visual explanation using SHAP made it an effective tool for guiding clinical courses preoperatively.

How Dual-Energy Contrast-Enhanced Spectral Mammography Can Provide Useful Clinical Information About Prognostic Factors in Breast Cancer Patients: A Systematic Review of Literature.

IntroductionIn the past decade, a new technique derived from full-field digital mammography has been developed, named contrast-enhanced spectral mammography (CESM). The aim of this study was to define the association between CESM findings and usual prognostic factors, such as estrogen receptors, progesterone receptors, HER2, and Ki67, in order to offer an updated overview of the state of the art for the early differential diagnosis of breast cancer and following personalized treatments.Materials and MethodsAccording to the PRISMA guidelines, two electronic databases (PubMed and Scopus) were investigated, using the following keywords: breast cancer AND (CESM OR contrast enhanced spectral mammography OR contrast enhanced dual energy mammography) AND (receptors OR prognostic factors OR HER2 OR progesterone OR estrogen OR Ki67). The search was concluded in August 2021. No restriction was applied to publication dates.ResultsWe obtained 28 articles from the research in PubMed and 114 articles from Scopus. After the removal of six replicas that were counted only once, out of 136 articles, 37 articles were reviews. Eight articles alone have tackled the relation between CESM imaging and ER, PR, HER2, and Ki67. When comparing radiological characterization of the lesions obtained by either CESM or contrast-enhanced MRI, they have a similar association with the proliferation of tumoral cells, as expressed by Ki-67. In CESM-enhanced lesions, the expression was found to be 100% for ER and 77.4% for PR, while moderate or high HER2 positivity was found in lesions with non-mass enhancement and with mass closely associated with a non-mass enhancement component. Conversely, the non-enhancing breast cancer lesions were not associated with any prognostic factor, such as ER, PR, HER2, and Ki67, which may be associated with the probability of showing enhancement. Radiomics on CESM images has the potential for non-invasive characterization of potentially heterogeneous tumors with different hormone receptor status.ConclusionsCESM enhancement is associated with the proliferation of tumoral cells, as well as to the expression of estrogen and progesterone receptors. As CESM is a relatively young imaging technique, a few related works were found; this may be due to the “off-label” modality. In the next few years, the role of CESM in breast cancer diagnostics will be more thoroughly investigated.

Novel lncRNAs with diagnostic or prognostic value screened out from breast cancer via bioinformatics analyses.

BackgroundRecent studies have shown that long non-coding RNAs (lncRNAs) may play key regulatory roles in many malignant tumors. This study investigated the use of novel lncRNA biomarkers in the diagnosis and prognosis of breast cancer.Materials and MethodsThe database subsets of The Cancer Genome Atlas (TCGA) by RNA-seq for comparing analysis of tissue samples between breast cancer and normal control groups were downloaded. Additionally, anticoagulant peripheral blood samples were collected and used in this cohort study. The extracellular vesicles (EVs) from the plasma were extracted and sequenced, then analyzed to determine the expressive profiles of the lncRNAs, and the cancer-related differentially expressed lncRNAs were screened out. The expressive profiles and associated downstream-mRNAs were assessed using bioinformatics (such as weighted correlation network analysis (WGCNA), Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genome (KEGG) enrichments, Receiver-Operating Characteristic (ROC) curve and survival analysis, etc.) to investigate the diagnostic and prognostic values of these EV lncRNAs and their effectors.ResultsIn this study, 41 breast cancer-related lncRNAs were screen out from two datasets of tissue and fresh collected plasma samples of breast cancer via the transcriptomic and bioinformatics techniques. A total of 19 gene modules were identified with WGCNA analysis, of which five modules were significantly correlated with the clinical stage of breast cancer, including 28 lncRNA candidates. The ROC curves of these lncRNAs revealed that the area under the curve (AUC) of all candidates were great than 70%. However, eight lncRNAs had an AUC >70%, indicating that the combined one has a good diagnostic value. In addition, the results of survival analysis suggested that two lncRNAs with low expressive levels may indicate the poor prognosis of breast cancer. By tissue sample verification, C15orf54, AL157935.1, LINC01117, and SNHG3 were determined to have good diagnostic ability in breast cancer lesions, however, there was no significant difference in the plasma EVs of patients. Moreover, survival analysis data also showed that AL355974.2 may serve as an independent prognostic factor and as a protective factor.ConclusionA total of five lncRNAs found in this study could be developed as biomarkers for breast cancer patients, including four diagnostic markers (C15orf54, AL157935.1, LINC01117, and SNHG3) and a potential prognostic marker (AL355974.2).

Therapeutic peculiarities in diseases of the nipple skin

Die Mamillenregion ist durch besondere anatomische Verhältnisse charakterisiert und lässt sich aus dermatologischer Perspektive in Brusthaut, Haut des Brustwarzenhofes (Areola) und Haut der Brustwarze (Papilla mammae) unterteilen. Im klinischen Zusammenhang sind die Brustwarzen häufig während der Stillzeit durch mechanische Beanspruchung, Milieuänderung mit Mazeration durch den Milchfluss sowie durch mikrobielle Erreger alteriert. Zudem besteht hier die Gefahr der Entwicklung einer Mastitis puerperalis. Außerhalb der Schwangerschaft und Stillzeit finden sich gelegentlich an der Mamillenhaut Ekzemerkrankungen, häufig bei atopischer Disposition (atopisches Mamillenekzem) oder als irritatives Kontaktekzem („joggers nipple“). Seltener werden allergische Kontaktekzeme auf Konservierungsstoffe von Topika oder Metallen (Piercings) beobachtet. Auch im Rahmen einer Skabiesinfestation wird eine Beteiligung der Mamillen, insbesondere bei Frauen, regelmäßig beobachtet. Von großer klinischer Bedeutung sind seltene, präinvasive Läsionen eines Mammakarzinoms oder der Morbus Paget der Mamille vom extramammären Typ. Durch die besonderen anatomischen Gegebenheiten ist es naheliegend, dass bei der Anwendung von Topika sich auch spezifische Penetrationsbedingungen ableiten. Experimentelle Untersuchungen an Humanhaut ex vivo legen nahe, dass in Abhängigkeit von der Molmasse und der Löslichkeit des Arzneistoffs sowie des eingesetzten Vehikelsystems eine deutliche Zunahme der kutanen Bioverfügbarkeit, insbesondere an der Brustwarze selbst durch den transpapillären Diffusionsweg, auftreten kann. Dies sollte insbesondere bei der topischen Anwendung von Arzneistoffen mit bekanntem dosisabhängigem Nebenwirkungspotenzial (z. B. Glukokortikoiden) beachtet werden. Allerdings fehlt dafür bisher eine klinische Evidenz.

Tumor volume doubling time estimated from digital breast tomosynthesis mammograms distinguishes invasive breast cancers from benign lesions.

The aim of this study was to determine whether lesion size metrics on consecutive screening mammograms could predict malignant invasive carcinoma versus benign lesion outcome. We retrospectively reviewed suspicious screen-detected lesions confirmed by biopsy to be invasive breast cancers or benign that were visible on current and in-retrospect prior screening mammograms performed with digital breast tomosynthesis from 2017 to 2020. Four experienced radiologists recorded mammogram dates, breast density, lesion type, lesion diameter, and morphology on current and prior exams. We used logistic regression models to evaluate the association of invasive breast cancer outcome with lesion size metrics such as maximum dimension, average dimension, volume, and tumor volume doubling time (TVDT). Twenty-eight patients with invasive ductal carcinoma or invasive lobular carcinoma and 40 patients with benign lesions were identified. The mean TVDT was significantly shorter for invasive breast cancers compared to benign lesions (0.84 vs. 2.5 years; p = 0.0025). Patients with a TVDT of less than 1 year were shown to have an odds ratio of invasive cancer of 6.33 (95% confidence interval, 2.18-18.43). Logistic regression adjusted for age, lesion maximum dimension, and lesion volume demonstrated that shorter TVDT was the size variable significantly associated with invasive cancer outcome. Invasive breast cancers detected on current and in-retrospect prior screening mammograms are associated with shorter TVDT compared to benign lesions. If confirmed to be sufficiently predictive of benignity in larger studies, lesions visible on mammograms which in comparison to prior exams have longer TVDTs could potentially avoid additional imaging and/or biopsy. • We propose tumor volume doubling time as a measure to distinguish benign from invasive breast cancer lesions. • Logistic regression results summarized the utility of the odds ratio in retrospective clinical mammography data.

Joint Transformer and Multi-scale CNN for DCE-MRI Breast Cancer Segmentation

Automatic segmentation of breast cancer lesions in dynamic contrast-enhanced magnetic resonance imaging is challenged by low accuracy of delineation of the infiltration area, variable structure and shapes, large intensity heterogeneity changes, and low boundary contrast. This study constructed a two-stage breast cancer image segmentation framework and proposes a novel breast cancer lesion segmentation model (TR-IMUnet). The benchmark U-Net network model enables a rough delineation of the breast area in the acquired images and eliminates the influence of unrelated tissues (chest muscle, fat, and heart) on breast tumor segmentation. Based on the extracted results of the region of interest, the rectified linear unit (ReLU) function of the encoding–decoding structure in the model was replaced by an improved ReLU function to reserve and adjust the data dynamically according to input information. The segmentation accuracy of breast cancer lesions was improved by embedding a multi-scale fusion block and a transformer module in the coding path of the model, thereby obtaining multi-scale and global attention information. The experimental results showed that the breast tumor segmentation indexes Dice coefficient (Dice), Intersection over Union (IoU), Sensitivity (SEN), and Positive Predictive Value (PPV) increased by 4.27, 5.21, 3.37, and 3.68%, respectively, relative to the U-Net reference model. The proposed model improves the segmentation results of breast cancer lesions and reduces small area mis-segmentation and calcification segmentation.

Active contour model of breast cancer DCE‐MRI segmentation with an extreme learning machine and a fuzzy C‐means cluster

Due to the low contrast, blurred boundary and intensity inhomogeneity of the images, accurate segmentation of breast cancer lesions with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) still has great challenges. This paper proposed an improved active contour model (ACM) for segmenting breast cancer lesions in DCE-MRI images. First, based on the extreme learning machine (ELM) method, a robust function is proposed that combines image intensities and time-domain features to enhance the difference between the lesions and other tissues. Second, an edge-stop function (ESF) is introduced by combining the image intensity, time-domain feature, and Hessian shape index to detect the irregular and blurred boundaries. At the boundary of breast cancer lesions, the energy function of ACM is minimized and the evolution of the contour curve completes, so the accurate lesion region of breast cancer can be segmented. The mean Dice similar coefficient (DICE), Jaccard similarity (JC) and Hausdorff distance (HD) of the segmentation of the proposed model in 50 samples are 85.88±6.62%, 75.72±9.68% and 11.62±4.72 mm, respectively. The results segmented by the proposed ACM are more similar to the manual segmentation than the compared models.

Breast-specific Gamma Imaging (BSGI) as a Complementary Imaging Tool for BI-RADS 0 and 4a Lesions on Mammography or Ultrasonography

Background: Mammography (MMG) and ultrasonography (US) have been used as standard imaging modalities for the diagnosis of breast cancer. However, several drawbacks have been attributed to these modalities. Breast-specific gamma imaging (BSGI), as a nuclear medicine imaging technique, has been introduced as a supplementary tool for diagnosing breast cancer. Objectives: This study aimed to determine whether the addition of BSGI to MMG or US interpretations could improve the diagnostic accuracy and reduce the need for further examinations or unnecessary biopsies of breast lesions. Patients and Methods: This retrospective study was conducted on 548 patients with 638 breast lesions from February 2013 to December 2018. The performance of BSGI, MMG, and US was examined for identifying breast cancer and high-risk lesions. Subgroups were classified by adding the results of BSGI for Breast Imaging-Reporting and Data System (BI-RADS) 0 and 4a lesions on MMG and BI-RADS 4a lesions on US. The diagnostic performance of each subgroup was then compared. The sensitivity, specificity, positive predictive value, and negative predictive value were also calculated. The diagnostic accuracy was determined by measuring the area under the receiver operating characteristic curve (AUC). Besides, factors associated with false-positive and false-negative results of BSGI were extracted. Results: The BSGI showed a sensitivity of 88.26% for breast cancer diagnosis, which was comparable to the sensitivity of MMG (87.95%) and lower than that of US (97.83%). The specificity and AUC of BSGI (81.62% and 0.85, respectively) were superior to those of MMG (66.83% and 0.77, respectively) and US (15.20% and 0.57, respectively). In the subgroup analysis of MMG, the sensitivity, positive predictive value, and AUC of MMG0+BSGI and MMG4a+BSGI increased significantly compared to MMG alone. In the MMG4a+BSGI group, the specificity also significantly increased. In the US subgroups, the specificity and AUC of US4a+BSGI increased significantly compared to US alone. Based on the results, a low Ki-67 index was associated with a false-negative result of BSGI. Conclusion: The addition of BSGI to MMG or US could improve the diagnostic performance, especially for BI-RADS 0 and 4a lesions. Additionally, the concomitant use of BSGI with MMG or US might reduce the need for an additional examination or unnecessary biopsy.

Analysis of Cross-Combinations of Feature Selection and Machine-Learning Classification Methods Based on [18F]F-FDG PET/CT Radiomic Features for Metabolic Response Prediction of Metastatic Breast Cancer Lesions.

Simple SummaryBreast cancer is a leading cause of morbidity and mortality worldwide. The metastatic disease is largely responsible for cancer patient deaths, and its treatment implies usually different therapies. In this context, the prediction of response to treatment or depiction of treatment-resistant phenotypes is essential in clinical practice, especially in the new era of precision medicine. In this study, we used several combinations of feature selection methods and machine-learning classifiers to construct predictive models of the metabolic response to the treatment of metastatic breast cancer lesions. These models were based on clinical variables and radiomic features extracted from 2-deoxy-2-[18F]fluoro-D-glucose positron emission tomography/computed tomography ([18F]F-FDG PET/CT) images, obtained prior to the treatment. Our main goal was to know if this prediction was feasible and to identify those combinations with better predictive performance. We found that several combinations were successful to predict the metabolic response to treatment, of which the least absolute shrinkage and selection operator (Lasso) + support vector machines (SVM) had the best mean performance in terms of area under the curve, in both training and validation cohorts. Model performances depended largely on the selected combinations.Background: This study aimed to identify optimal combinations between feature selection methods and machine-learning classifiers for predicting the metabolic response of individual metastatic breast cancer lesions, based on clinical variables and radiomic features extracted from pretreatment [18F]F-FDG PET/CT images. Methods: A total of 48 patients with confirmed metastatic breast cancer, who received different treatments, were included. All patients had an [18F]F-FDG PET/CT scan before and after the treatment. From 228 metastatic lesions identified, 127 were categorized as responders (complete or partial metabolic response) and 101 as non-responders (stable or progressive metabolic response), by using the percentage changes in SULpeak (peak standardized uptake values normalized for body lean body mass). The lesion pool was divided into training (n = 182) and testing cohorts (n = 46); for each lesion, 101 image features from both PET and CT were extracted (202 features per lesion). These features, along with clinical and pathological information, allowed the prediction model’s construction by using seven popular feature selection methods in cross-combination with another seven machine-learning (ML) classifiers. The performance of the different models was investigated with the receiver-operating characteristic curve (ROC) analysis, using the area under the curve (AUC) and accuracy (ACC) metrics. Results: The combinations, least absolute shrinkage and selection operator (Lasso) + support vector machines (SVM), or random forest (RF) had the highest AUC in the cross-validation, with 0.93 ± 0.06 and 0.92 ± 0.03, respectively, whereas Lasso + neural network (NN) or SVM, and mutual information (MI) + RF, had the higher AUC and ACC in the validation cohort, with 0.90/0.72, 0.86/0.76, and 87/85, respectively. On average, the models with Lasso and models with SVM had the best mean performance for both AUC and ACC in both training and validation cohorts. Conclusions: Image features obtained from a pretreatment [18F]F-FDG PET/CT along with clinical vaiables could predict the metabolic response of metastatic breast cancer lesions, by their incorporation into predictive models, whose performance depends on the selected combination between feature selection and ML classifier methods.

PERCIST-like response assessment with FDG PET based on automatic segmentation of all lesions in metastatic breast cancer.

e13057 Background: In metastatic breast cancer (MBC), treatment response is often assessed with FDG PET per PERCIST which evaluates changes in SULpeak of the single hottest tumor lesion identified on the baseline and follow-up images. PERCIST therefore does not consider tumor heterogeneity. This work aims to compare responses determined with the automatic segmentation of all lesions to response determined manually per PERCIST. Methods: 10 MBC patients (61±14 y/o) undergoing either chemo- or hormonotherapy were randomly selected from the prospective EPICURE study (NCT03958136). A baseline and two follow-up FDG PET were acquired at pre-, early- (1 month) and mid-treatment times for each patient. All metastatic lesions on all images were manually segmented by experts. Using the Advanced Normalization Tools (ANTs) image registration method, we wrapped baseline lesion segmentations to automatically obtain the follow-up ones. These registered segmentations were compared to the ones done manually using standard biomarkers: SULpeak, lesion size and Total Lesion Glycolysis (TLG). Differences between baseline and follow-up images were visually represented by coloring the follow-up segmentations: in green for responsive lesions (decreasing SULpeak) and in red for progressive ones (increasing SULpeak). Two expert physicians were then asked to evaluate treatment response while seeing these colored segmentations. They assessed the FDG PET images in pairs, evaluating for each patient the baseline and one of the corresponding follow-ups in a blinded manner: either the early- or the mid-treatment follow-up. Evaluations were then compared: i) early- vs mid-treatment response and ii) follow-up response vs patient’s clinical outcome. Results: Biomarkers extracted from the registered segmentations were similar to the ones extracted from the manual segmentations, with a Lin correlation coefficient of 0.92, 0.87 and 0.95 for the SULpeak, lesion size and TLG respectively. These findings were obtained within ̃10min, whereas the manual segmentation of the three PET images for any given patient took ̃1h. With the use of colored segmentations, early follow-up evaluations were predictive of mid-treatment response in 65% of the cases. The blinded physicians agreed with the clinical outcomes 85% and 95% of the time for the early- and mid-treatment images respectively. Conclusions: With segmentations automatically derived from ANTs registration, we managed to extract biomarkers that are comparable to the ones obtained with manual segmentations; both segmentations carried similar information. ANTs fast registration and biomarkers computation can make it a useful tool in clinical routine. In addition, lesion coloring helped evaluate treatment response and early-treatment follow-up images were shown to be predictive of mid-treatment response. Clinical trial information: NCT03958136.