A nomogram to predict disease-free survival in patients with residual triple-negative breast cancer after neoadjuvant chemotherapy based on clinicopathological and sonographic features

Introduction

Breast cancer is one of the leading causes of cancer-related deaths among women worldwide, with a 5-year recurrence rate ranging from 10% to 15% (1,2). Triple-negative breast cancer (TNBC) constitutes 10–20% of all breast cancer cases (3). Compared to other subtypes, TNBC is associated with a higher grade, larger tumor size, and a higher rate of lymph node positivity (4-6). Consequently, optimizing the prognosis of patients with TNBC in clinical practice remains a matter of considerable importance.

Currently, treatment options for TNBC include neoadjuvant chemotherapy (NAC), surgical intervention, radiotherapy, and other modalities. Given the absence of specific therapeutic targets for TNBC, NAC plays a pivotal role in comprehensive treatment strategies. It is primarily utilized to reduce the size of the primary tumor to facilitate surgical resection and to assess tumor sensitivity to chemotherapeutic agents, thereby guiding the selection of postoperative chemotherapy regimens (7). Pathological complete response (pCR) has been identified as a key predictor of disease-free survival (DFS) in breast cancer patients (8-10). However, due to the marked heterogeneity of TNBC, a subset of patients fails to achieve pCR following NAC (11). Importantly, not all patients with residual disease after NAC will experience tumor recurrence or distant metastasis.

Thus, identifying predictive factors associated with DFS, such as age, gender, genetic factors, lifestyle, and environmental influences, can aid in the identification and screening of high-risk populations. In clinical practice, doctors can develop targeted treatment strategies for high-risk individuals and enhance postoperative monitoring.

However, there are no effective methods in clinical practice to predict the occurrence of adverse events in patients who fail to achieve pCR currently. It is reported that the presence of lymphovascular space invasion (LVSI) and extranodal extension increases the risk of recurrence in TNBC patients without pCR (12). Kennedy et al. (13) also found that pathologic lymph node positivity and LVSI predicted worse prognosis. However, previous studies had small sample sizes and only included clinical and pathological information, lacking consideration of imaging features. Currently, ultrasound has become an important means of routine screening for breast cancer, and the evaluation of ultrasound images has been widely used in the research on the diagnosis and prognosis of breast cancer (14,15). Ko et al. (16) found that circumscribed margins and posterior echogenic enhancement were closely related to the triple-negative subtype and its aggressive features. It is also reported that the vertical direction of lesions on pre-operative ultrasound was a predictive factor for axillary lymph node metastasis and an adverse prognosis (17).

This study integrates clinicopathological characteristics and sonographic features to pinpoint predictive factors for poor prognosis in TNBC patients failing to achieve pCR post-NAC, aiming to develop a nomogram model that offers guidance for clinical treatment strategies. We present this article in accordance with the TRIPOD reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2182/rc).

Methods

Patients

A total of 254 women diagnosed with TNBC at Guangdong Provincial People’s Hospital between January 2013 and December 2022 were retrospectively enrolled. Our exclusion criteria were as follows: (I) non-primary tumor sites (n=4); (II) no NAC prior to surgery (n=14); (III) achieving pCR after NAC (n=119); (IV) incomplete or missing clinical-pathological features and preoperative ultrasound images (n=14). Finally, 103 TNBC patients were included in the analysis. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study’s protocol was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (No. KY2023-1127-02). The requirement for informed consent in retrospective cohorts was waived.

Data collection

Clinicopathological information was obtained from the electronic medical records of patients. We collected data including age, menopausal status, clinical T stage, number of lesions, lesion location, the regimen and duration of NAC. Pathological data included pathological type, estrogen receptor (ER) status, progesterone receptor (PR) status, human epidermal growth factor receptor 2 (HER2) status, Ki-67 expression level, and axillary lymph node involvement. The pathological types were primarily categorized into invasive ductal carcinoma and other types, the latter including metaplastic carcinoma, invasive mixed-type carcinoma, and invasive lobular carcinoma. Given that these three pathological types are rare and represent a relatively small proportion of the cohort, we combined them in accordance with established practices in prior prognostic studies to minimize the risk of unstable statistical estimates (18). Axillary lymph node involvement was evaluated by sentinel lymph node biopsy and/or axillary lymph node dissection. Tumor staging was performed according to the 8th edition of the American Joint Committee on Cancer (AJCC) Tumor-Node-Metastasis (TNM) staging manual.

When the nuclear staining in tumor cells was less than 1%, the status of ER and PR was considered negative (19). HER2 negativity was defined by an immunohistochemical HER2 score of 0 or 1+, or the absence of HER2 amplification in fluorescence in situ hybridization (20,21). TNBC was characterized by the simultaneous negative expression of ER, PR, and HER2. The Ki-67 expression level was categorized as high or low expression using a cutoff value of 20%.

In this study, we chose DFS as the primary endpoint. Consistent with the design of pivotal breast cancer trials, DFS was defined as the interval from surgery to the occurrence of local-regional recurrence, distant metastasis, or contralateral breast cancer (22,23). All breast cancer patients included in the study were systematically followed up according to established protocols, with follow-up data obtained from electronic medical records.

Assessment of sonographic images

Sonographic images were retrospectively collected from the image archiving server. All patients underwent ultrasound examination using Toshiba APLIO 500 TUS-A500, APLIO 400 TUS-A400, Hitachi HI VISION Asendus, Hitachi HI VISION Preirus and Mindray Resona8. Each breast tumor mass was evaluated by two ultrasound physicians, with at least 3 years of experience in breast imaging, based on the following criteria: shape, orientation, margin, calcification, echo pattern, size, peripheral halo, posterior acoustic features, blood flow distribution, blood flow grade, and skin edema. Hyperechoic halo was defined as a thick, echogenic rim that partially or completely surrounds a lesion and must be differentiated from echogenic pseudocapsule (24). Its features were poorly defined margins, greater thickness, and heterogeneous appearance. It can be seen around any shape of mass or non-mass lesion. In contrast, the pseudocapsule was typically thin, uniform, and most commonly appeared around oval or lobulated masses. For patients with multiple lesions, the largest lesion was selected for assessment. The evaluation of all lesions was conducted in accordance with the latest fifth edition of the American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS^®) Atlas, which provides a standardized approach for categorizing breast lesions based on their imaging characteristics. Assessments were performed independently, and the two physicians were blinded to the pathological findings and prognosis. In the event of disagreement, a consensus-based interpretation will be reached.

Statistical analysis

Statistical analyses were performed using R software (version 4.2.3). Inter-observer agreement for sonographic features was assessed using Cohen’s kappa statistic, with a 95% confidence interval (CI) calculated to evaluate the precision of the estimate.

Missing data were present in 4.8% of the cohort, primarily affecting blood flow distribution, blood flow grade, and Ki-67. To maximize statistical power and minimize bias, we avoided complete-case analysis. Instead, missing values were handled using multiple imputation by chained equations (MICE). We created five imputed datasets assuming that data were missing at random (MAR). All multivariable regression analyses were performed on the pooled imputed datasets based on Rubin’s rules.

Univariate Cox proportional hazards regression analysis was initially conducted to identify potential predictors of DFS. Variables with P<0.1 in univariate analysis were subsequently included in the multivariate Cox proportional hazards model. The proportional hazards assumption was verified using Schoenfeld residuals. Backwards elimination method was employed to identify independent prognostic factors while checking for potential collinearity. To address the potential risk of overfitting, Firth’s Penalized Cox Regression was performed as a sensitivity analysis.

Hazard ratios (HRs) with 95% CIs were calculated. The overall model performance was assessed using Wald test, log-rank test, and likelihood ratio test. Model discrimination was evaluated using time-dependent receiver operating characteristic (ROC) curves and area under the curve (AUC) values at different time points (2-5 years) using the timeROC package (version 0.4). Model calibration was assessed using calibration curves.

Survival curves were estimated using the Kaplan-Meier method and compared using the log-rank test. Stratified analyses were performed based on axillary lymph node metastasis status, pathological type, and peripheral halo sign. Additionally, patients were stratified according to the number of positive independent predictors present (0, 1, or 2), and survival differences between these groups were analyzed.

A nomogram was constructed using the rms package (version 6.7.1) to predict 2- to 5-year DFS probabilities. Survival analyses were performed using the survival package (version 3.7.0) and survminer package (version 0.5.0). Statistical significance was set at P<0.05 for all analyses unless otherwise specified.

Results

Patient characteristics

A total of 103 patients were included in the study, and all patients underwent preoperative ultrasound examination. The median age of the study subjects was 49.1 years, with a standard deviation of 10.9 years; 51 (49.5%) of them were postmenopausal at diagnosis. The pathological type of most tumors was invasive ductal carcinoma. In the study, 44 (42.7%) patients were found to have axillary lymph node metastasis, a condition that significantly affects the prognosis. According to research, the 5-year survival rate for patients with axillary lymph node metastasis can range from 60% to 70% with early detection and active treatment. The median follow-up duration was 35.5 [interquartile range: 28.3–42.2] months. Among these patients, 15 (14.6%) experienced DFS events; specifically, five, six, and four patients had local recurrence, distant metastasis, and both local recurrence and distant metastasis, respectively. The clinicopathological information and sonographic characteristics of these patients are detailed in Table S1, which includes data on the correlation between clinicopathology and sonographic features, as demonstrated in similar studies. The inter-observer agreement between the two physicians was substantial, with a Cohen’s kappa statistic of 0.801 (95% CI: 0.687–0.915).

Clinicopathological and ultrasound factors associated with DFS

Univariate survival analysis demonstrated that the pathological type of non-invasive ductal carcinoma (P=0.021) and involvement of axillary lymph node (P=0.054) were significantly associated with poorer DFS outcomes (Table 1). Additionally, in terms of sonographic characteristics, presence of hyperechoic halo (P=0.059) was associated with recurrence and metastasis (Table 2).

Variables significantly associated with the DFS event in the univariate analysis were further incorporated into the multivariable Cox regression analysis. The results indicated that the pathological type of non-invasive ductal carcinoma (HR =7.741, 95% CI: 1.928–31.081, P=0.004), involvement of axillary lymph node (HR =3.455, 95% CI: 1.152–10.359, P=0.027), and the presence of hyperechoic halo (HR =4.43, 95% CI: 1.164–16.852, P=0.029) were identified as independent predictors of poor prognosis (Table 3). To validate the proportional hazards assumption of the multivariate Cox regression model, Schoenfeld residual tests were performed for the three independent prognostic factors. All factors yielded individual test results with P>0.05, and no evident time-dependent trends were observed in the Beta(t) curves (Figure S1). Thus, the model satisfied the proportional hazards assumption and was statistically valid. Additionally, multivariate Cox regression analysis with a backward elimination approach was performed. No variables were excluded on account of collinearity, and all retained variables yielded stable parameter estimates in the final model (Table S2). Thus, the three aforementioned variables were identified as independent prognostic factors for DFS in the study cohort, and are therefore suitable for incorporation into a DFS prediction nomogram.

To validate the robustness of the multivariate Cox model against potential overfitting, Firth’s Penalized Cox Regression was conducted. The results were highly consistent with the original model. Detailed comparisons between the standard multivariate Cox regression and Firth’s Penalized Cox Regression are presented in Table S3.

Prognostic stratification was performed based on different influencing factors. As shown in Figure 1A-1C, in the Kaplan-Meier survival analysis, DFS significantly differed among patients based on the presence or absence of invasive ductal carcinoma (P=0.011), axillary lymph node involvement (P=0.044), and a hyperechoic halo (P=0.045). The involvement of axillary lymph nodes, as a critical prognostic factor, has been shown to correlate with overall survival (OS) rates, with studies indicating that the lymph node ratio may be a more accurate predictor of outcomes than the absolute number of positive nodes. Furthermore, patients with multiple risk factors exhibited poorer survival outcomes. The 5-year DFS rate for patients without any risk factors was 100.0%, whereas the rates for patients with one and two risk factors were 75.3% and 53.9%, respectively (P<0.01) (Figure 1D).

Figure 1 Stratified analyses were performed to estimate disease-free survival in various subgroups, comparing (A) invasive ductal carcinoma and other pathological types, (B) axillary lymph node involvement and non-involvement, (C) hyperechoic halo and non-halo, (D) number of factors significant in multivariable analysis. IDC, invasive ductal carcinoma; LN, lymph node.

Nomogram construction and validation

Based on the three prognostic factors identified by the multivariate Cox analysis, a nomogram model was developed to predict the 2-, 3-, 4-, and 5-year DFS rates. As illustrated in Figure 2A by summing the specific scores of the prognostic variables, the corresponding survival probability for each patient can be determined.

Figure 2 Constructed nomogram and its ROC curves and calibration curves. (A) Nomogram of predicting 2-, 3-, 4-, and 5-year DFS for TNBC patients who failed to achieve pCR after NAC. (B) ROC curves of the nomogram and the corresponding AUC. (C) Calibration curves of the nomogram. The X-axis represents the model predicted probability, while the Y-axis represents the actual probability. AUC, area under the curve; CI, confidence interval; DFS, disease-free survival; IDC, invasive ductal carcinoma; NAC, neoadjuvant chemotherapy; pCR, pathological complete response; ROC, receiver operating characteristic; TNBC, triple-negative breast cancer.

To evaluate the established predictive model, ROC curves were constructed for 2- to 5-year DFS rates. In this study, AUCs for predicting 2-, 3-, 4-, and 5-year DFS rates were 0.767, 0.786, 0.785, and 0.739, respectively (Figure 2B). Figure 2C presented the calibration curves for the predicted DFS probabilities at 2, 3, 4, and 5 years after treatment, demonstrating excellent consistency between the predicted and actual survival probabilities.

Discussion

The prognosis of TNBC has historically been poor, but recent advances in treatment options, such as trophoblast cell surface antigen 2-targeted antibody-drug conjugates, have begun to change the clinical management landscape by improving survival rates and offering new hope for patients (25). Previous studies have shown that achieving pCR after NAC is a key predictor of DFS in TNBC patients (10). However, in clinical practice, achieving pCR is not universal among TNBC patients, and failure to achieve it does not inevitably imply a risk of tumor recurrence or metastasis. Therefore, identifying predictive factors associated with poor prognosis and accurately identifying high-risk groups among TNBC patients who do not achieve pCR is crucial for the development of personalized treatment strategies. Nonetheless, prior research has seldom combined clinicopathological data with sonographic features to explore the predictors of unfavorable outcomes in this patient group. Our study revealed that the non-invasive ductal carcinoma subtype, axillary lymph node involvement, and the appearance of a hyperechoic halo on ultrasound images emerged as independent factors indicating poor survival rates. The nomogram model based on these factors can provide a reliable prediction of survival outcomes for TNBC patients who did not achieve pCR.

Multivariate analysis has shown that the presence of positive axillary lymph nodes is an independent predictor of DFS in patients with TNBC, as supported by similar findings in other cancer types. This factor has been widely acknowledged as a negative prognostic indicator in breast cancer and has also been confirmed in specific analyses targeting triple-negative subtypes (26-28). Kennedy et al. (13) found that in TNBC patients with residual disease after NAC, pathologic nodal positivity was a significant predictor of distant metastasis (HR 3.08, 95% CI: 1.54–6.14, P=0.001). Ilie et al. (29) analyzed 111 TNBC patients with residual lesions before surgery, and the results showed that lymph node involvement was significantly associated with poorer DFS and lower OS (P<0.001). Furthermore, several researchers have pointed out that a higher number of axillary lymph node metastases correlates with a poorer prognosis, as the number of positive lymph nodes influences the pathological staging of breast cancer and serves as a critical predictor of survival outcomes (30-32). A recent study (33) also demonstrated that the residual lymph node burden after NAC has a stronger predictive impact on the mortality rate of TNBC subtypes, with an absolute difference of 64% in 5-year OS rates between ypN0 and ypN3 patients. In the ER+/HER2− subtype, the absolute 5-year OS rate difference between ypN0 and ypN3 patients is only 25%. Therefore, for TNBC patients who still have residual lymph node metastases after NAC, clinicians should consider the specific characteristics of their subtype and develop individualized treatment strategies.

Our study identified the presence of a hyperechoic halo on ultrasound as an independent predictor of poorer prognosis in TNBC patients with residual lesions following NAC. This finding aligns with the results of previous investigations. For instance, Hashimoto et al. (34) conducted a study involving 187 patients with invasive breast cancer and demonstrated that those with a higher halo ratio (≥0.42) had an increased risk of lymph node metastasis, reduced DFS and OS rates, and an unfavorable prognosis. A hyperechoic halo is a characteristic ultrasound finding in breast cancer, defined as an abnormally hyperechoic structure at the interface between the hypoechoic edges of a lesion and surrounding glandular or adipose tissue (35). Multiple studies have indicated that breast cancer cell infiltration into adjacent adipose tissue induces reactive proliferation and inflammatory exudation in the surrounding connective tissue, leading to irregular tissue interfaces that manifest as hyperechoic halos on sonograms (36,37). Chen et al. (38) reported that hyperechoic halo can predict axillary lymph node tumor burden in early-stage breast cancer, while other research has linked the extent of the hyperechoic halo to the degree of tumor lymphovascular invasion (39,40). Collectively, these observations suggest that the presence of this ultrasound sign may reflect the aggressive biological behavior of breast cancer and indicate a poorer prognosis in affected patients.

Additionally, by stratifying patients based on the number of significant prognostic factors identified in the multivariate analysis, we found that those with a higher number of risk factors exhibited poorer DFS—a finding consistent with prior research. Kennedy et al. (13) found that the 5-year distant metastasis-free survival rates for TNBC patients with 1 or 2 risk factors were 54.9% and 27.5%, respectively, compared to 76.5% for patients without any risk factors (P<0.001). It is also reported that TNBC patients who failed to achieve pCR following NAC, those with multiple risk factors exhibited a significantly higher risk of locoregional recurrence compared to those with a single risk factor (12). This finding is of great significance for guiding the individualized treatment plans and monitoring strategies for TNBC patients.

In this study, the model achieved an AUC of 0.739–0.786 for predicting 2- to 5-year DFS, which is comparable to previously reported results in TNBC. Zhang et al. (31) developed a predictive model for progression-free survival, achieving a C-index of 0.88. However, their study did not specifically focus on the subgroup of patients with residual disease after NAC. In contrast, the present study concentrates on TNBC patients who did not achieve pCR after NAC, and identifies hyperechoic halo as an independent prognostic factor, thereby providing a novel imaging-based biomarker for residual TNBC. Sheng et al. (32) constructed a DFS prediction model with a C-index of 0.69 but excluded patients who received NAC, and found mammographic features to lack prognostic significance. The current model demonstrates superior predictive performance avoiding the application limitations of mammography in dense breasts, and significantly enhancing the clinical accessibility of the model. Ma et al. (27) developed a rapid recurrence risk prediction model for TNBC patients who received NAC. The C-index of the training set and validation set was 0.938 and 0.910 respectively, demonstrating excellent predictive accuracy. However, their model relies on specialized pathological assessments such as stromal tumor-infiltrating lymphocytes expression levels, which are not routinely standardized across institutions. In contrast, the proposed model does not require specialized testing, making it more feasible and scalable, particularly in resource-limited healthcare environments. In summary, our findings offer a practical and accessible tool for the clinical management of TNBC patients with residual disease after NAC.

There are several limitations in this study. First, our study is limited by the relatively small sample size and the low number of DFS events (n=15), resulting in an events-per-variable (EPV) ratio of 5. To address the potential risk of overfitting associated with low EPV, we performed Firth’s penalized Cox regression analysis. The results confirmed that pathological type of non-invasive ductal carcinoma, lymph node involvement, and the presence of a hyperechoic halo on ultrasound images remained statistically significant predictors even after penalizing for small-sample bias, supporting the robustness of our findings. Second, as a retrospective single-center study, this research was constrained by the lack of external validation. The performance of our model may be influenced by institutional-specific clinical practices, ultrasound imaging protocols, and patient population characteristics. Thus, the generalizability of the model to other centers remains to be verified. To address this issue, we have established collaborations with two tertiary hospitals and plan to conduct a prospective multi-center study. We will optimize the model parameters based on the validation results to ensure its clinical utility across different medical centers. Third, non-invasive ductal carcinoma was identified as a strong independent predictor of poor prognosis in this study. However, it is important to note that the ‘other’ pathological type group comprises three rare subtypes, each with very limited sample sizes. The combined HR may be influenced by subtype heterogeneity and constraints related to small sample sizes, and its reliability needs to be verified. Therefore, future studies with larger cohorts focusing on rare pathological types in TNBC are warranted to validate these findings. Fourth, although we have defined hyperechoic halo clearly and the inter-observer agreement in this study was substantial, the experience levels of different physicians may still affect the interpretation results, especially in primary medical institutions, which may reduce the repeatability of the model. This subjective variation is a limitation of ultrasound assessment. In the future, it is necessary to further improve this through standardized training and multi-center consistency verification. Additionally, the median follow-up time in this study was 35.5 months, and a prospective study with extended follow-up is necessary to further validate the effectiveness of this model. Finally, our study is limited by the absence of genomic and transcriptomic data, potentially undermining the accuracy and interpretability of the predictive model. Future investigations could explore the relationship between genomic and sonographic features to enhance the model’s performance.

Conclusions

In conclusion, our findings indicate that the pathological subtype of non-invasive ductal carcinoma, involvement of axillary lymph nodes, and the presence of hyperechoic halo on ultrasound imaging are all associated with poorer DFS in patients with TNBC who failed to achieve pCR following NAC. A nomogram based on these three variables can serve as an effective tool for predicting the survival outcomes of TNBC patients and may provide guidance for treatment strategies in high-risk populations. Future prospective studies are necessary to validate the predictive performance of this model.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2182/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2182/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2182/prf

Funding: This work was supported by the National Center for Inheritance and Innovation of Traditional Chinese Medicine Research Special Project (No. 2022QN18), and the Guangzhou Science and Technology Project (No. 2024A04J3915).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-aw-2182/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study’s protocol was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (No. KY2023-1127-02). The requirement for informed consent in retrospective cohorts was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
2. Colleoni M, Sun Z, Price KN, et al. Annual Hazard Rates of Recurrence for Breast Cancer During 24 Years of Follow-Up: Results From the International Breast Cancer Study Group Trials I to V. J Clin Oncol 2016;34:927-35. [Crossref] [PubMed]
3. Billar JA, Dueck AC, Stucky CC, et al. Triple-negative breast cancers: unique clinical presentations and outcomes. Ann Surg Oncol 2010;17:384-90. [Crossref] [PubMed]
4. Kumar P, Aggarwal R. An overview of triple-negative breast cancer. Arch Gynecol Obstet 2016;293:247-69. [Crossref] [PubMed]
5. Dent R, Trudeau M, Pritchard KI, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res 2007;13:4429-34. [Crossref] [PubMed]
6. Foulkes WD, Smith IE, Reis-Filho JS. Triple-negative breast cancer. N Engl J Med 2010;363:1938-48. [Crossref] [PubMed]
7. Derks MGM, van de Velde CJH. Neoadjuvant chemotherapy in breast cancer: more than just downsizing. Lancet Oncol 2018;19:2-3. [Crossref] [PubMed]
8. Kong X, Moran MS, Zhang N, et al. Meta-analysis confirms achieving pathological complete response after neoadjuvant chemotherapy predicts favourable prognosis for breast cancer patients. Eur J Cancer 2011;47:2084-90. [Crossref] [PubMed]
9. Liedtke C, Mazouni C, Hess KR, et al. Response to neoadjuvant therapy and long-term survival in patients with triple-negative breast cancer. J Clin Oncol 2008;26:1275-81. [Crossref] [PubMed]
10. Symmans WF, Wei C, Gould R, et al. Long-Term Prognostic Risk After Neoadjuvant Chemotherapy Associated With Residual Cancer Burden and Breast Cancer Subtype. J Clin Oncol 2017;35:1049-60. [Crossref] [PubMed]
11. Bianchini G, De Angelis C, Licata L, et al. Treatment landscape of triple-negative breast cancer - expanded options, evolving needs. Nat Rev Clin Oncol 2022;19:91-113. [Crossref] [PubMed]
12. Gabani P, Merfeld E, Srivastava AJ, et al. Predictors of Locoregional Recurrence After Failure to Achieve Pathologic Complete Response to Neoadjuvant Chemotherapy in Triple-Negative Breast Cancer. J Natl Compr Canc Netw 2019;17:348-56. [Crossref] [PubMed]
13. Kennedy WR, Tricarico C, Gabani P, et al. Predictors of Distant Metastases in Triple-Negative Breast Cancer Without Pathologic Complete Response After Neoadjuvant Chemotherapy. J Natl Compr Canc Netw 2020;18:288-96. [Crossref] [PubMed]
14. Park VY, Choi JS, Han K, et al. Imaging Features and Diagnostic Performance of US in Nonmass Lesions with Varying Clinical Indications. Radiology 2025;317:e243398. [Crossref] [PubMed]
15. Liu Y, Wang Y, Huang J, et al. Deep learning-based prediction of axillary pathological complete response in patients with breast cancer using longitudinal multiregional ultrasound. EBioMedicine 2025;119:105896. [Crossref] [PubMed]
16. Ko ES, Lee BH, Kim HA, et al. Triple-negative breast cancer: correlation between imaging and pathological findings. Eur Radiol 2010;20:1111-7. [Crossref] [PubMed]
17. Wang H, Zhan W, Chen W, et al. Sonography with vertical orientation feature predicts worse disease outcome in triple negative breast cancer. Breast 2020;49:33-40. [Crossref] [PubMed]
18. Lyu M, Yi S, Li C, et al. Multimodal prediction based on ultrasound for response to neoadjuvant chemotherapy in triple negative breast cancer. NPJ Precis Oncol 2025;9:259. [Crossref] [PubMed]
19. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College Of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. J Clin Oncol 2010;28:2784-95. [Crossref] [PubMed]
20. Bauer KR, Brown M, Cress RD, et al. Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a population-based study from the California cancer Registry. Cancer 2007;109:1721-8. [Crossref] [PubMed]
21. Aksoy S, Dizdar O, Harputluoglu H, et al. Demographic, clinical, and pathological characteristics of Turkish triple-negative breast cancer patients: single center experience. Ann Oncol 2007;18:1904-6. [Crossref] [PubMed]
22. Whelan TJ, Olivotto IA, Parulekar WR, et al. Regional Nodal Irradiation in Early-Stage Breast Cancer. N Engl J Med 2015;373:307-16. [Crossref] [PubMed]
23. Fisher B, Anderson S, Bryant J, et al. Twenty-year follow-up of a randomized trial comparing total mastectomy, lumpectomy, and lumpectomy plus irradiation for the treatment of invasive breast cancer. N Engl J Med 2002;347:1233-41. [Crossref] [PubMed]
24. Song HP, Shu R. Interpretation of updates in the sixth edition of ACR BI-RADS ultrasound. Chin J Ultrasonogr 2025;34:369-76.
25. Tolaney SM, Cardillo TM, Chou CC, et al. The Mode of Action and Clinical Outcomes of Sacituzumab Govitecan in Solid Tumors. Clin Cancer Res 2025;31:1390-9. [Crossref] [PubMed]
26. Yu Y, Tan Y, Xie C, et al. Development and Validation of a Preoperative Magnetic Resonance Imaging Radiomics-Based Signature to Predict Axillary Lymph Node Metastasis and Disease-Free Survival in Patients With Early-Stage Breast Cancer. JAMA Netw Open 2020;3:e2028086. [Crossref] [PubMed]
27. Ma T, Liu XY, Cai SL, et al. Development and validation of a nomogram for predicting rapid relapse in triple-negative breast cancer patients treated with neoadjuvant chemotherapy. Front Cell Dev Biol 2024;12:1417366. [Crossref] [PubMed]
28. Wang H, Xia GF, Zhang ZR, et al. Development and validation of a prognostic nomogram for breast cancer patients who underwent chemoradiotherapy and surgery: a retrospective cohort study based on the SEER database and two Chinese cohorts. Am J Cancer Res 2023;13:5065-81.
29. Ilie SM, Briot N, Constatin G, et al. Pathologic and immunohistochemical prognostic markers in residual triple-negative breast cancer after neoadjuvant chemotherapy. Front Oncol 2023;13:1309890. [Crossref] [PubMed]
30. Yu F, Hang J, Deng J, et al. Radiomics features on ultrasound imaging for the prediction of disease-free survival in triple negative breast cancer: a multi-institutional study. Br J Radiol 2021;94:20210188. [Crossref] [PubMed]
31. Zhang L, Zhang X, Han P, et al. Nomograms predicting recurrence in patients with triple negative breast cancer based on ultrasound and clinicopathological features. Br J Radiol 2022;95:20220305. [Crossref] [PubMed]
32. Sheng DL, Shen XG, Shi ZT, et al. Survival outcome assessment for triple-negative breast cancer: a nomogram analysis based on integrated clinicopathological, sonographic, and mammographic characteristics. Eur Radiol 2022;32:6575-87. [Crossref] [PubMed]
33. Thai CHNC. Differential prognostic value of residual nodal burden in breast cancer subtypes. Breast Cancer Res Treat 2025;209:315-22. [Crossref] [PubMed]
34. Hashimoto H, Suzuki M, Oshida M, et al. Quantitative ultrasound as a predictor of node metastases and prognosis in patients with breast cancer. Breast Cancer 2000;7:241-6. [Crossref] [PubMed]
35. Zhang L, Li J, Xiao Y, et al. Identifying ultrasound and clinical features of breast cancer molecular subtypes by ensemble decision. Sci Rep 2015;5:11085. [Crossref] [PubMed]
36. Kuba MG, Giess CS, Wieczorek TJ, et al. Hyperechoic malignancies of the breast: Underlying pathologic features correlating with this unusual appearance on ultrasound. Breast J 2020;26:643-52. [Crossref] [PubMed]
37. Park SY, Park JY, Park JW, et al. Unexpected hyperechoic lesions of the breast and their correlations with pathology: a pictorial essay. Ultrasonography 2022;41:597-609. [Crossref] [PubMed]
38. Chen Y, Xie Y, Li B, et al. Automated Breast Ultrasound (ABUS)-based radiomics nomogram: an individualized tool for predicting axillary lymph node tumor burden in patients with early breast cancer. BMC Cancer 2023;23:340. [Crossref] [PubMed]
39. Durmus T, Stöckel J, Slowinski T, et al. The hyperechoic zone around breast lesions - an indirect parameter of malignancy. Ultraschall Med 2014;35:547-53. [Crossref] [PubMed]
40. Bach A, Hameister C, Slowinski T, et al. Can acoustic structural quantification be used to characterize the ultrasound echotexture of the peripheral zone of breast lesions? Clin Hemorheol Microcirc 2019;72:189-200. [Crossref] [PubMed]