Deep Learning Model Based on Dual-Modal Ultrasound and Molecular Data for Predicting Response to Neoadjuvant Chemotherapy in Breast Cancer

The development and validation of a risk stratification system for assessing axillary status after neoadjuvant therapy in node-positive breast cancer: a multicenter, prospective, observational study

OBJECTIVE

It is not clear which procedure is most optimal for axilla after neoadjuvant therapy (NAT) in node-positive breast cancer patients. Accurately identifying patients with axillary pathologic complete response (pCR) is crucial to minimize the overtreatment of axilla. This study was designed to develop a risk stratification model for axillary pCR.

METHODS

In this multicenter, prospective, observational study, node-positive breast cancer patients who received NAT followed by axillary lymph node dissection (ALND) were enrolled between June 2021 and April 2024. We assessed the performance of breast shear wave elastography (SWE) utilizing virtual touch imaging quantification in determining axillary status across ultrasound (US) nodal stages following NAT. A predictive model incorporating axilla US nodal stage and breast SWE was developed using multivariate logistic regression analysis. Last, a simplified risk score was developed based on the calculated prediction probability from this model and validated in the external test cohort.

RESULTS

The axillary pCR rates were 52.53% in the training cohort ( n = 257) and 51.79% in the external test cohorts ( n = 195). Approximately 21.67% of US N0 cases were false negatives; 42.35% of US N1 cases were false positives. With SWE, the false negative rate was 11.53% in US N0 patients and false positive rate was 22.22% in US N1 patients. The model based on dual-modality US demonstrated strong discriminatory ability (AUC, 0.93), precise calibration (slope of calibration curve, 0.99), and practical clinical utility (probability threshold, 4.5-94.5%); the percentages of accuracy, sensitivity, and specificity were 87.94%, 88.52%, and 87.41%, respectively. Patients scoring 1 demonstrated a low axillary non-pCR rate (5.21%-6.97%), potentially reducing unnecessary ALND rate (17.12%-24.10%).

CONCLUSIONS

The risk stratification model integrating axilla US and breast SWE demonstrated good performance for assessing axillary status after NAT in node-positive breast cancer and might provide guidance for less aggressive management for specific individuals.

Harnessing Artificial Intelligence to Enhance Global Breast Cancer Care: A Scoping Review of Applications, Outcomes, and Challenges

BACKGROUND

In recent years, Artificial Intelligence (AI) has shown transformative potential in advancing breast cancer care globally. This scoping review seeks to provide a comprehensive overview of AI applications in breast cancer care, examining how they could reshape diagnosis, treatment, and management on a worldwide scale and discussing both the benefits and challenges associated with their adoption.

METHODS

In accordance with PRISMA-ScR and ensuing guidelines on scoping reviews, PubMed, Web of Science, Cochrane Library, and Embase were systematically searched from inception to end of May 2024. Keywords included "Artificial Intelligence" and "Breast Cancer". Original studies were included based on their focus on AI applications in breast cancer care and narrative synthesis was employed for data extraction and interpretation, with the findings organized into coherent themes.

RESULTS

Finally, 84 articles were included. The majority were conducted in developed countries (n = 54). The majority of publications were in the last 10 years (n = 83). The six main themes for AI applications were AI for breast cancer screening (n = 32), AI for image detection of nodal status (n = 7), AI-assisted histopathology (n = 8), AI in assessing post-neoadjuvant chemotherapy (NACT) response (n = 23), AI in breast cancer margin assessment (n = 5), and AI as a clinical decision support tool (n = 9). AI has been used as clinical decision support tools to augment treatment decisions for breast cancer and in multidisciplinary tumor board settings. Overall, AI applications demonstrated improved accuracy and efficiency; however, most articles did not report patient-centric clinical outcomes.

CONCLUSIONS

AI applications in breast cancer care show promise in enhancing diagnostic accuracy and treatment planning. However, persistent challenges in AI adoption, such as data quality, algorithm transparency, and resource disparities, must be addressed to advance the field.

Elastography-based AI model can predict axillary status after neoadjuvant chemotherapy in breast cancer with nodal involvement: a prospective, multicenter, diagnostic study

OBJECTIVE

To develop a model for accurate prediction of axillary lymph node (LN) status after neoadjuvant chemotherapy (NAC) in breast cancer patients with nodal involvement.

METHODS

Between October 2018 and February 2024, 671 breast cancer patients with biopsy-proven LN metastasis who received NAC followed by axillary LN dissection were enrolled in this prospective, multicenter study. Preoperative ultrasound (US) images, including B-mode ultrasound (BUS) and shear wave elastography (SWE), were obtained. The included patients were randomly divided at a ratio of 8:2 into a training set and an independent test set, with five-fold cross-validation applied to the training set. The authors first identified clinicopathological characteristics and conventional US features significantly associated with the axillary LN response and developed corresponding prediction models. The authors then constructed deep learning radiomics (DLR) models based on BUS and SWE data. Models performances were compared, and a combination model was developed using significant clinicopathological data and interpreted US features with the SWE-based DLR model. Discrimination, calibration and clinical utility of this model were analyzed using the receiver operating characteristic curve, calibration curve, and decision curve, respectively.

RESULTS

Axillary pathologic complete response (pCR) was achieved in 52.41% of patients. In the test cohort, the clinicopathologic model had an accuracy of 71.30%, while radiologists' diagnoses ranged from 64.26 to 71.11%, indicating limited to moderate predictive ability for the axillary response to NAC. The SWE-based DLR model, with an accuracy of 80.81%, significantly outperformed the BUS-based DLR model, which scored 59.57%. The combination DLR model boasted an accuracy of 88.70% and a false-negative rate of 8.82%. It demonstrated strong discriminatory ability (AUC, 0.95), precise calibration ( P -value obtained by Hosmer-Lemeshow goodness-of-fit test, 0.68), and practical clinical utility (probability threshold, 2.5-97.5%).

CONCLUSIONS

The combination SWE-based DLR model can predict the axillary status after NAC in patients with node-positive breast cancer, and thus, may inform clinical decision-making to help avoid unnecessary axillary LN dissection.

Enhancing detection of high-level axillary lymph node metastasis after neoadjuvant therapy in breast cancer patients with nodal involvement: a combined approach of axilla ultrasound and breast elastography

PURPOSE

To develop a combined approach using shear wave elastography (SWE) and conventional ultrasound (US) to determine the extent of positive axillary lymph nodes (LNs) following neoadjuvant therapy (NAT) in breast cancer patients with nodal involvement.

METHODS

This prospective, multicenter study was registered on the Chinese Clinical Trial Registry (ChiCTR2400085035). From October 2018 to February 2024, a total of 303 breast cancer patients with biopsy-proven positive LN were enrolled. The conventional US features of axillary LNs and SWE characteristics of breast lesions after NAT were analyzed. The diagnostic performances of axilla US, breast SWE, and their combination in detecting residual metastasis in axillary level III after NAT were assessed.

RESULTS

Pathologically positive LN(s) in axilla level III were detected in 13.75% of cases following NAT. The kappa value for the axilla level with positive LN confirmed by surgical pathology and detected by US is 0.39 (p < 0.001). The AUC of conventional axilla US to determine the status of axilla level III LNs after NAT was 0.67, with a sensitivity of 51.52%, a specificity of 74.36%. The breast SWE displayed moderate performance for detecting residual metastasis in axilla level III following NAT, with an AUC of 0.79, sensitivity of 84.85%, and specificity of 74.36%. Compared to axilla US and breast SWE alone, the combination of axilla US with breast SWE achieved a stronger discriminatory ability (AUC, 0.86 vs 0.67 vs 0.79, p < 0.05, Delong's test) and precise calibration (X2 = 13.90, p = 0.085, HL test), with an improved sensitivity of 93.94% and a comparable specificity of 75.64%%.

CONCLUSIONS

SWE outperformed conventional US in identifying the axilla levels with nodal metastasis following NAT in patients with initially diagnosed positive axilla. Furthermore, combining breast SWE with axilla US showed good diagnostic performance for detecting residual metastasis in axilla level III after NAT.

Ultrasound-Based Deep Learning Radiomics Nomogram for Tumor and Axillary Lymph Node Status Prediction After Neoadjuvant Chemotherapy

RATIONALE AND OBJECTIVES

This study aims to explore the feasibility of the deep learning radiomics nomogram (DLRN) for predicting tumor status and axillary lymph node metastasis (ALNM) after neoadjuvant chemotherapy (NAC) in patients with breast cancer. Additionally, we employ a Cox regression model for survival analysis to validate the effectiveness of the fusion algorithm.

MATERIALS AND METHODS

A total of 243 patients who underwent NAC were retrospectively included between October 2014 and July 2022. The DLRN integrated clinical characteristics as well as radiomics and deep transfer learning features extracted from ultrasound (US) images. The diagnostic performance of DLRN was evaluated by constructing ROC curves, and the clinical usefulness of models was assessed using decision curve analysis (DCA). A survival model was developed to validate the effectiveness of the fusion algorithm.

RESULTS

In the training cohort, the DLRN yielded area under the receiver operating characteristic curve values of 0.984 and 0.985 for the tumor and LNM, while 0.892 and 0.870, respectively, in the test cohort. The consistency indices (C-index) of the nomogram were 0.761 and 0.731, respectively, in the training and test cohorts. The Kaplan-Meier survival curves showed that patients in the high-risk group had significantly poorer overall survival than patients in the low-risk group (P < 0.05).

CONCLUSION

The US-based DLRN model could hold promise as clinical guidance for predicting the status of tumors and LNM after NAC in patients with breast cancer. This fusion model can also predict the prognosis of patients, which could help clinicians make better clinical decisions.

Machine learning-based discrimination of benign and malignant breast lesions on US: The contribution of shear-wave elastography

PURPOSE

To build and validate a combined radiomics and machine learning (ML) approach using B-mode US and SWE images to differentiate benign from malignant solid breast lesions (BLs) and compare its performance with that of an expert radiologist.

METHODS

Patients with at least one BI-RADS 2-6 BL who performed breast US integrated with SWE were retrospectively included. B-mode US and SWE images were manually segmented to extract radiomics features. A multi-step feature selection process was performed and a predictive model built using the Logistic Regression algorithm. The diagnostic accuracy was evaluated with the AUC and Matthews Correlation Coefficient (MCC) metrics. The performance of the ML classifier was compared to that of an expert radiologist.

RESULTS

427 Bls were included and divided into a training (286 BLs, of which 127 benign and 159 malignant) and a test set (141 BLs, of which 59 benign and 82 malignant). Of 1098 features extracted from B-mode US and SWE images, 13 were finally selected. The ML classifier showed an AUC of 0.768 and 0.746, and an MCC of 0.403 and 0.423 in the training and test sets, respectively. The performance was higher than that of the expert radiologist assessing only B-mode US images, but significantly lower when SWE images were also provided.

CONCLUSION

A ML approach based on B-mode US and SWE images may represent a potential tool in the characterization of BLs. SWE still gives its most relevant contribution in the clinical setting rather than included in a radiomics pipeline.

Delta Radiomics Based on Longitudinal Dual-modal Ultrasound Can Early Predict Response to Neoadjuvant Chemotherapy in Breast Cancer Patients

RATIONALE AND OBJECTIVES

To develop a monitoring model using radiomics analysis based on longitudinal B-mode ultrasound (BUS) and shear wave elastography (SWE) to early predict pathological response to neoadjuvant chemotherapy (NAC) in breast cancer patients.

MATERIALS AND METHODS

In this prospective study, 112 breast cancer patients who received NAC between September 2016 and March 2022 were included. The BUS and SWE data of breast cancer were obtained prior to treatment as well as after two and four cycles of NAC. Radiomics features were extracted followed by measuring the changes in radiomics features compared to baseline after the second and fourth cycles of NAC (△R [C2], △R [C4]), respectively. The delta radiomics signatures were established using a support vector machine classifier.

RESULTS

The area under receiver operating characteristic curve (AUC) values of △RBUS (C2) and △RBUS (C4) for predicting the response to NAC were 0.83 and 0.84, while those of △RSWE (C2) and △RSWE (C4) were 0.88 and 0.90, respectively. △RSWE exhibited significantly superior performance to △RBUS for predicting NAC response (Delong test, p < 0.01). No significant differences were observed in the performances between △R (C2) and △R (C4) based on BUS or SWE data. The longitudinal dual-modal ultrasound radiomics (LDUR) model had an excellent discrimination, good calibration and clinical usefulness, with the AUC, sensitivity and specificity of 0.97, 95.52% and 91.11%, respectively.

CONCLUSION

The LDUR model achieved excellent performance in predicting the pathological response to chemotherapy during the early stages of NAC for breast cancer.

Elasticity Values as a Predictive Modality for Response to Neoadjuvant Chemotherapy in Breast Cancer

Shear-wave elastography (SWE) is an effective tool in discriminating malignant lesions of breast and axillary lymph node metastasis in patients with breast cancer. However, the association between the baseline elasticity value of breast cancer and the treatment response of neoadjuvant chemotherapy is yet to be elucidated. Baseline SWE measured mean stiffness (E-mean) and maximum stiffness (E-max) in 830 patients who underwent neoadjuvant chemotherapy and surgery from January 2012 to December 2022. Association of elasticity values with breast pCR (defined as ypTis/T0), pCR (defined as ypTis/T0, N0), and tumor-infiltrating lymphocytes (TILs) was analyzed. Of 830 patients, 356 (42.9%) achieved breast pCR, and 324 (39.0%) achieved pCR. The patients with low elasticity values had higher breast pCR and pCR rates than those with high elasticity values. A low E-mean (adjusted odds ratio (OR): 0.620; 95% confidence interval (CI): 0.437 to 0.878; p = 0.007) and low E-max (adjusted OR: 0.701; 95% CI: 0.494 to 0.996; p = 0.047) were independent predictive factors for breast pCR. Low elasticity values were significantly correlated with high TILs. Pretreatment elasticity values measured using SWE were significantly associated with treatment response and inversely correlated with TILs, particularly in HR+HER2- breast cancer and TNBC.