Molecular markers, pathology, and ultrasound features of invasive breast cancer

Multi-center study: ultrasound-based deep learning features for predicting Ki-67 expression in breast cancer

Applying deep learning algorithms to mine ultrasound features of breast cancer and construct a machine learning model that accurately predicts Ki-67 expression level. This multi-center retrospective study analyzed clinical and ultrasound data from 929 breast cancer patients. We integrated deep features from the tumor and peritumoral areas to build a fusion model for predicting Ki-67 expression. The model underwent performance validation on both internal and external test datasets. Its accuracy as well as clinical usefulness were evaluated by diverse statistical metrics. In the ultrasound depth feature model for the tumor area, the Support Vector Machine (SVM) algorithm achieved the highest performance, with an accuracy of 0.782, ROAUC of 0.771 (95% CI 0.704-0.838), sensitivity of 0.905, specificity of 0.543, and F1 score of 0.846. In the depth feature model for the peritumoral area, the Light Gradient Boosting Machine (LightGBM) algorithm demonstrated superior performance, achieving an accuracy of 0.728, ROAUC of 0.623 (95% CI 0.545-0.702), sensitivity of 0.892, specificity of 0.407, and F1 score of 0.813. The SVM algorithm exhibited superior performance in both internal and external test sets when validated the fusion model integrating depth features from tumor and peritumoral area. Internal test set validation in clinical application indicated significantly lower disease-free survival in the high Ki-67 expression group compared to the low expression group (P = 0.005). Through comprehensive analysis of breast cancer ultrasound images and the application of machine learning techniques, we developed a highly accurate model for predicting Ki-67 expression levels.

Construction of a nomogram prediction model for the pathological complete response after neoadjuvant chemotherapy in breast cancer: a study based on ultrasound and clinicopathological features

Objective

To explore the application value of ultrasound in evaluating the efficacy of neoadjuvant chemotherapy (NAC) for breast cancer and construct a nomogram prediction model for pathological complete response (pCR) following different cycles of NAC based on ultrasound and clinicopathological features, and further investigate the optimal prediction cycle.

Methods

A total of 249 breast cancer patients who received NAC were recruited. Ultrasound assessment was performed before NAC and after two cycles of NAC (NAC2), four cycles of NAC (NAC4), and six cycles of NAC (NAC6). All patients underwent surgical resection after NAC6 and the samples were sent for histopathological and immunohistochemical examination. Clinical efficacy was determined according to the Response Evaluation Criteria in Solid Tumors (RECIST). Pathological efficacy was determined according to the Miller-Payne evaluation system (MP); grade 5 was classified as pCR group, while Grades 1-4 were classified as the non-pCR group (npCR). The patients were randomly divided into the training set and the validation set at a ratio of 7:3. The ultrasound and clinicopathological features of the training set were compared, and a nomogram prediction model was constructed based on these features. Finally, the ROC curve, calibration curve, and DCA were used for verification.

Result

Among the 249 patients, 71 (28.5%) achieved pCR, whereas the remaining 178 (71.5%) exhibited npCR. The maximum tumor diameter measured by ultrasound after NAC6 was 1.20 (0.70, 2.10) cm, which was significantly positively correlated with the maximum tumor diameter measured by pathology after surgical resection (r=0.626, P<0.05). In the training set, multivariate logistic regression analysis revealed that tumor size, posterior echo, RECIST evaluation, and PR status were significantly correlated with pCR after NAC2, NAC4, and NAC6 (P<0.05). These indicators were incorporated into static and dynamic nomogram models, demonstrating high predictive performance, calibration, and clinical value in both the training and validation sets.

Conclusion

Regardless of the cycle of NAC, patients with a small tumor, no posterior shadow, a valid RECIST, and a negative PR were more likely to achieve pCR. Evaluation after NAC2 can provide early predictive value in clinical practice.

HER-2 expression is correlated with multimodal imaging features in breast cancer: a pilot study

OBJECTIVE

A pilot study to evaluate the correlation between multimodal imaging features and the expression of the human epidermal growth factor receptor type 2 (HER-2) in breast cancer to provide a basis for clinical treatment and prognosis evaluation.

METHODS

We included a total of 62 patients with breast cancer admitted to the Affiliated Hospital of Hebei University between 2018 and 2022. All of them underwent the relevant investigations, including ultrasound, mammography, and enhanced magnetic resonance imaging (MRI), in the hospital within one month before surgery or biopsy. HER-2 expression level was divided into negative and positive by immunohistochemistry(IHC). Using SPSS 24.0 statistical software to analyze the differences in imaging features between the HER-2 positive and the HER-2 negative groups.

RESULTS

There was a statistically significant difference between the HER-2 positive and the HER-2 negative groups (p = 0.005) in the hyperechoic halo sign around the lesion detected by ultrasonography as well as in the apparent diffusion coefficient (ADC) on MRI (p = 0.047). The sensitivity and specificity of the hyperechoic halo sign in predicting HER-2 positivity was 48.3% and 84.8% respectively, and the area under the curve (AUC) for the ADC value to predict HER-2 expression was 0.533. When b was equal to 800 and the ADC value (cutoff value) was 0.000888 mm2/s, the sensitivity and specificity were 65.5% and 51.5%, respectively.

CONCLUSION

A combination of multimodal imaging features and HER-2 gene expression can provide more valuable information for clinical diagnosis and therapeutic schedule in breast cancer.

A Model Combining Conventional Ultrasound Characteristics, Strain Elastography and Clinicopathological Features to Predict Ki-67 Expression in Small Breast Cancer

To establish a predictive model incorporating conventional ultrasound, strain elastography and clinicopathological features for Ki-67 expression in small breast cancer (SBC) which defined as maximum diameter less than2 cm. In this retrospective study, 165 SBC patients from our hospital were allocated to a high Ki-67 group (n = 104) and a low Ki-67 group (n = 61). Multivariate regression analysis was performed to identify independent indicators for developing predictive models. The area under the receiver operating characteristic (AUC) curve was also determined to establish the diagnostic performance of different predictive models. The corresponding sensitivities and specificities of different models at the cutoff value were compared. Conventional ultrasound parameters (spiculated margin, absence of posterior shadowing and Adler grade 2-3), strain elastic scores and clinicopathological information (HER2 positive) were significantly correlated with high expression of Ki-67 in SBC (all p < .05). Model 2, which incorporated conventional ultrasound features and strain elastic scores, yielded good diagnostic performance (AUC = 0.774) with better sensitivity than model 1, which only incorporated ultrasound characteristics (78.85%vs. 55.77%, p = .000), with specificities of 77.05% and 62.30% (p = .035), respectively. Model 3, which incorporated conventional ultrasound, strain elastography and clinicopathological features, yielded better performance (AUC = 0.853) than model 1 (AUC = 0.694) and model 2 (AUC = 0.774), and the specificity was higher than model 1 (86.89% vs. 77.05%, p = .001). The predictive model combining conventional ultrasound, strain elastic scores and clinicopathological features could improve the predictive performance of Ki-67 expression in SBC.

Intra- and peritumoral radiomics features based on multicenter automatic breast volume scanner for noninvasive and preoperative prediction of HER2 status in breast cancer: a model ensemble research

The aim to investigate the predictive efficacy of automatic breast volume scanner (ABVS), clinical and serological features alone or in combination at model level for predicting HER2 status. The model weighted combination method was developed to identify HER2 status compared with single data source model method and feature combination method. 271 patients with invasive breast cancer were included in the retrospective study, of which 174 patients in our center were randomized into the training and validation sets, and 97 patients in the external center were as the test set. Radiomics features extracted from the ABVS-based tumor, peritumoral 3 mm region, and peritumoral 5 mm region and clinical features were used to construct the four types of the optimal single data source models, Tumor, R3mm, R5mm, and Clinical model, respectively. Then, the model weighted combination and feature combination methods were performed to optimize the combination models. The proposed weighted combination models in predicting HER2 status achieved better performance both in validation set and test set. For the validation set, the single data source model, the feature combination model, and the weighted combination model achieved the highest area under the curve (AUC) of 0.803 (95% confidence interval [CI] 0.660-947), 0.739 (CI 0.556,0.921), and 0.826 (95% CI 0.689,0.962), respectively; with the sensitivity and specificity were 100%, 62.5%; 81.8%, 66.7%; 90.9%,75.0%; respectively. For the test set, the single data source model, the feature combination model, and the weighted combination model attained the best AUC of 0.695 (95% CI 0.583, 0.807), 0.668 (95% CI 0.555,0.782), and 0.700 (95% CI 0.590,0.811), respectively; with the sensitivity and specificity were 86.1%, 41.9%; 61.1%, 71.0%; 86.1%, 41.9%; respectively. The model weighted combination was a better method to construct a combination model. The optimized weighted combination models composed of ABVS-based intratumoral and peritumoral radiomics features and clinical features may be potential biomarkers for the noninvasive and preoperative prediction of HER2 status in breast cancer.

Association between ultrasound BI-RADS signs and molecular typing of invasive breast cancer

Objective

To explore the correlation between ultrasound images and molecular typing of invasive breast cancer, so as to analyze the predictive value of preoperative ultrasound for invasive breast cancer.

Methods

302 invasive breast cancer patients were enrolled in Heping Hospital affiliated to Changzhi Medical College in Shanxi, China during 2020 to 2022. All patients accepted ultrasonic and pathological examination, and all pathological tissues received molecular typing with immunohistochemical (IHC) staining. The relevance between different molecular typings and ultrasonic image, pathology were evaluated.

Results

Univariate analysis: among the four molecular typings, there were significant differences in tumor size, shape, margin, lymph node and histological grade (P<0.05). 1. Size: Luminal A tumor was smaller (69.4%), Basal -like type tumors are mostly larger (60.9%); 2. Shape: Basal-like type is more likely to show regular shape (45.7%); 3. Margin: Luminal A and Luminal B mostly are not circumscribed (79.6%, 74.8%), Basal -like type shows circumscribed(52.2%); 4. Lymph nodes: Luminal A type tends to be normal (87.8%), Luminal B type,Her-2+ type and Basal-like type tend to be abnormal (35.6%,36.4% and 39.1%). There was no significant difference in mass orientation, echo pattern, rear echo and calcification (P>0.05). Multivariate analysis: Basal-like breast cancer mostly showed regular shape, circumscribed margin and abnormal lymph nodes (P<0.05).

Conclusion

There are differences in the ultrasound manifestations of different molecular typings of breast cancer, and ultrasound features can be used as a potential imaging index to provide important information for the precise diagnosis and treatment of breast cancer.

Noninvasive Prediction of Ki-67 Expression in Hepatocellular Carcinoma Using Machine Learning-Based Ultrasomics: A Multicenter Study

OBJECTIVES

To investigate the ability of ultrasomics to predict Ki-67 expression in hepatocellular carcinoma (HCC).

METHODS

A total of 244 patients from three hospitals were retrospectively recruited (training dataset, n = 168; test dataset, n = 43; and validation dataset, n = 33). Lesion segmentation of the ultrasound images was performed manually by two radiologists. In total, 1409 ultrasomics features were extracted. Feature selection was conducted using the intra-class correlation coefficient, variance threshold, mutual information, and recursive feature elimination plus eXtreme Gradient Boosting. The support vector machine was combined with the learning curve and grid search parameter tuning to construct the clinical, ultrasomics, and combined models. The predictive performance of the models was assessed using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity and accuracy.

RESULTS

The ultrasomics model performed well on the training, test, and validation datasets. The AUC (95% confidence interval [CI]) for these datasets were 0.955 (0.912-0.981), 0.861 (0.721-0.947), and 0.665 (0.480-0.819), respectively. The combination of ultrasomics and clinical features significantly improved model performance on all three datasets. The AUC (95% CI), sensitivity, specificity, and accuracy were 0.986 (0.955-0.998), 0.973, 0.840, and 0.869 on the training dataset; 0.871 (0.734-0.954), 0.750, 0.829, and 0.814 on the test dataset; and 0.742 (0.560-0.878), 0.714, 0.808, and 0.788 on the validation dataset, respectively.

CONCLUSIONS

Ultrasomics was proved to be a potential noninvasive method to predict Ki-67 expression in HCC.

The comparison of efficacy and safety evaluation of vacuum-assisted Elite 10-G system and the traditional BARD 14-G core needle in breast diagnosis: an open-label, parallel, randomized controlled trial

BACKGROUND

Vacuum-assisted biopsy (VAB) and core needle biopsy (CNB) are both widely used methods in diagnosing breast lesions. We aimed to determine whether the Elite 10-gauge VAB achieves higher accuracy than the BARD spring-actuated 14-gauge CNB.

MATERIALS AND METHODS

This was a phase 3, open-label, parallel, randomized controlled trial (NCT04612439). In total, 1470 patients with ultrasound (US)-visible breast lesions requiring breast biopsy were enrolled from April to July 2021 and randomized at a 1 : 1 ratio to undergo VAB or CNB. All patients underwent surgical excision after needle biopsy. The primary outcome was accuracy, defined as the proportion of patients who had a consistent qualitative diagnosis between the biopsy and surgical pathology results. The underestimation rate, false-negative rate and safety evaluations were the secondary endpoints.

RESULTS

A total of 730 and 732 patients were evaluable for endpoints in the VAB and CNB groups, respectively. The accuracy of VAB surpassed that of CNB in the whole population (94.8 vs. 91.1%, P =0.009). The overall malignant underestimation rate was significantly lower in the VAB group than in the CNB group (21.4 vs. 30.9%, P =0.035). Additionally, significantly more false-negative events were noted in the CNB group (4.9 vs. 7.8%, P =0.037). In patients who presented with accompanying calcification, the accuracy of VAB surpassed that of CNB (93.2 vs. 88.3%, P =0.022). The potential superiority of VAB was indicated in patients with heterogeneous echo on US.

CONCLUSIONS

In general, the 10-G VAB procedure is a reasonable alternative to the 14-G CNB procedure with higher accuracy. We recommend the use of VAB for lesions with accompanying calcification or heterogeneous echo on US.

Deep learning radiomics model based on breast ultrasound video to predict HER2 expression status

Purpose

The detection of human epidermal growth factor receptor 2 (HER2) expression status is essential to determining the chemotherapy regimen for breast cancer patients and to improving their prognosis. We developed a deep learning radiomics (DLR) model combining time-frequency domain features of ultrasound (US) video of breast lesions with clinical parameters for predicting HER2 expression status.

Patients and Methods

Data for this research was obtained from 807 breast cancer patients who visited from February 2019 to July 2020. Ultimately, 445 patients were included in the study. Pre-operative breast ultrasound examination videos were collected and split into a training set and a test set. Building a training set of DLR models combining time-frequency domain features and clinical features of ultrasound video of breast lesions based on the training set data to predict HER2 expression status. Test the performance of the model using test set data. The final models integrated with different classifiers are compared, and the best performing model is finally selected.

Results

The best diagnostic performance in predicting HER2 expression status is provided by an Extreme Gradient Boosting (XGBoost)-based time-frequency domain feature classifier combined with a logistic regression (LR)-based clinical parameter classifier of clinical parameters combined DLR, particularly with a high specificity of 0.917. The area under the receiver operating characteristic curve (AUC) for the test cohort was 0.810.

Conclusion

Our study provides a non-invasive imaging biomarker to predict HER2 expression status in breast cancer patients.

Breast Cancer—How Can Imaging Help?

Breast cancer is the most common malignant disease among women, causing death and suffering worldwide. It is known that, for the improvement of the survival rate and the psychological impact it has on patients, early detection is crucial. For this to happen, the imaging techniques should be used at their full potential. We selected and examined 44 articles that had as subject the use of a specific imaging method in breast cancer management (mammography, ultrasound, MRI, ultrasound-guided biopsy, PET-CT). After analyzing their data, we summarized and concluded which are the best ways to use each one of the mentioned techniques for a good outcome. We created a simplified algorithm with easy steps that can be followed by radiologists when facing this type of neoplasia.