Preoperative prediction of breast cancer Ki-67 status via multimodal ultrasound-clinical nomogram: a single-center study

Introduction

Breast cancer represents the second most common malignancy globally, accounting for 11.6% of all cancer cases and posing the foremost threat to women’s health (1). As a clinically heterogeneous disease, it exhibits substantial variations in histomorphology, biological behavior, therapeutic response, and molecular profiles (2). Contemporary management employs multimodal strategies integrating surgical resection with tailored adjuvant therapies (chemotherapy, radiotherapy, endocrine therapy) based on tumor biology and patient status to optimize oncological outcomes while preserving quality of life. Nevertheless, delayed diagnosis remains a critical contributor to breast cancer mortality (3). Consequently, early detection is paramount for improving patient prognosis.

Conventional prognostic markers—including tumor size, lymph node involvement, and histological grade—predict recurrence risk and overall survival (4). However, molecular characterization has revealed that biomarker expression patterns critically guide personalized treatment selection and prognostic stratification (2,5). Based on expression profiles of estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and Ki-67, breast cancers are classified into four molecular subtypes: Luminal A, Luminal B, HER2-enriched, and basal-like. The 2011 St. Gallen Consensus established Ki-67 index as the definitive criterion for distinguishing Luminal B from Luminal A subtypes in ER-positive/HER2-negative disease (6).

Ki-67, a non-histone nuclear protein encoded by MKI67, regulates nucleolar organization during interphase (G1, S, G2). Its expression progressively increases throughout the cell cycle, peaking during mitosis (M phase) while remaining undetectable in quiescent cells (G0 phase) (7). Expression levels peak during mitosis (M phase), reflecting direct correlation with proliferative activity (8). These characteristics establish Ki-67 as the gold-standard indicator of cellular proliferation. As a proliferation marker, Ki-67 predicts tumor aggressiveness and therapeutic response across multiple malignancies including breast, prostate, and gastrointestinal cancers (9). Capitalizing on its biological role, Ki-67-targeted antisense oligonucleotides (ASOs) have been developed to suppress proliferation and induce apoptosis across multiple malignancies. This phenomenon has similar results in various cell lines, including IM-9, RT-4, RM-11, MB-49, 4T1, U2OS, and MHT (10,11). Clinically, Ki-67 serves as a critical prognostic biomarker owing to its specificity for proliferating cells and whole-cell-cycle detectability. Large-scale clinical evidence demonstrates progressive Ki-67 upregulation across the pathological continuum from benign lesions to ductal carcinoma in situ (DCIS) and invasive breast carcinoma. Multicenter studies confirm that elevated Ki-67 indices independently predict reduced disease-free survival and increased recurrence risk (12). Collectively, these findings provide the mechanistic rationale for Ki-67-guided personalized therapeutic strategies. Conventional Ki-67 assessment requires invasive biopsy for immunohistochemical analysis, which suffers from inherent sampling limitations and poor reproducibility due to tumor heterogeneity (13). These issues are compounded by the limited analytical validity of Ki-67, which arises from inter-observer variability, non-standardized counting methods, and ambiguous cut-off values. This underscores the need for non-invasive quantification methods.

Ultrasonography provides a cost-effective, non-invasive modality for preoperative breast lesion characterization with excellent reproducibility. Shear-wave elastography (SWE) quantifies tissue stiffness—a biomechanical property reflecting extracellular matrix remodeling associated with tumorigenesis and progression (14). Emerging evidence correlates increased SWE values with adverse prognostic factors including advanced tumor stage and lymphovascular invasion (15). This study therefore investigates multimodal ultrasound features combined with clinical parameters to develop a preoperative nomogram for non-invasive Ki-67 prediction. We present this article in accordance with the TRIPOD reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1652/rc).

Methods

Data collection and patient selection

Between March 2022 and August 2024, 173 consecutive female breast cancer patients treated at our Breast Cancer Center were initially screened. Thirty-one patients were excluded based on predefined criteria. The final cohort (n=142) underwent preoperative multiparametric ultrasound examination, with comprehensive clinical and imaging datasets enabling complete lesion characterization.

Inclusion criteria: (I) no prior neoadjuvant therapy or diagnostic biopsy. (II) Pathologically confirmed breast cancer. (III) Axillary lymph node status histologically verified. (IV) High-quality ultrasound images permitting clear feature extraction. (V) Complete clinical and imaging records.

Exclusion criteria: (I) indeterminate pathological diagnosis. (II) Receipt of neoadjuvant therapy. (III) Incomplete clinical or imaging data. (IV) Male gender.

Therefore, a complete-case analysis was employed, and no data imputation was necessary. This approach ensured a robust dataset for model development without the potential biases introduced by imputation methods.

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Ethics Committee of Yunnan Cancer Hospital (No. KYLX2023-021). The ethics committee granted a waiver of informed consent for this retrospective analysis, with all data anonymized prior to analysis.

Ultrasound protocol

Ultrasound examinations were performed using the Aixplorer V system (Supersonic Imagine, Aix-en-Provence, France) equipped with an SL15-4 linear transducer (4–15 MHz). Patients underwent standardized scanning in supine and contralateral decubitus positions. Comprehensive assessment included: lesion size, internal echogenicity, posterior acoustic features, margins, shape, hyperechoic halo, calcifications, Adler blood flow grading, SWE parameters [Emax (maximum stiffness) and Eratio (lesion-to-fat stiffness ratio)], and ipsilateral axillary lymph nodes. Two board-certified ultrasound physicians (>5 years breast imaging experience) independently performed and interpreted all examinations.

Clinical data collection

The patient’s medical records and pathological reports were retrospectively investigated, and clinical information such as age, menopausal status, body mass index (BMI), and number of pregnancies were collected. In this study, Ki-67 expression was evaluated based on core needle biopsy specimens obtained at the time of initial diagnosis. All Ki-67 staining was performed uniformly using the same automated platform and reagents. Specifically, the antibody used was the CONFIRM^TM anti-Ki-67 [30-9] Rabbit Monoclonal Primary Antibody (Ventana Medical Systems, Inc., Tucson, Arizona, USA). The staining was performed on a Ventana BenchMark series automated stainer strictly following the manufacturer’s recommended protocol, ensuring consistent staining conditions (including antibody clone, dilution, incubation time, antigen retrieval, and detection system) across all cases. The Ki-67 proliferation index was assessed using the global counting method. Pathologists systematically evaluated the entire tumor section at low power to select at least three regions representing the spectrum of staining heterogeneity. A consecutive count of no less than 1,000 tumor cells was performed at high power, and the final index was calculated as the average percentage of positive cells across these regions (16). In accordance with international consensus (including the St. Gallen International Expert Consensus and the monarchE trial), and as defined by the 2023 Chinese Society of Clinical Oncology (CSCO) Breast Cancer Guidelines, high Ki-67 expression was classified as ≥20% positively stained nuclei (17-19).

Statistical analysis

Statistical analyses were conducted using SPSS 26.0 (IBM Corp.) for regression modeling and GraphPad Prism 10 (GraphPad Software) for graphical representation. Univariate logistic regression first evaluated associations between multimodal ultrasound features/clinical variables and Ki-67 status (high vs. low expression). Variables showing univariate associations (P<0.10) underwent multivariate forward stepwise logistic regression with likelihood ratio criterion for variable selection. Statistical significance was defined as P<0.05. Significant independent predictors (P<0.05) were incorporated into a nomogram constructed with the ‘rms’ package in R 4.3.1 (R Foundation).

Model performance was assessed through:

• Calibration curves with Hosmer-Lemeshow goodness-of-fit test;
• Discriminatory capacity via area under the curve (AUC);
• Clinical net benefit by decision curve analysis (DCA).

Internal validation employed 1,000 bootstrap replicates with optimism correction using Harrell’s method.

Results

Univariate analysis demonstrated no significant associations (P≥0.1) between Ki-67 status and demographic variables (age, menopausal status, BMI, parity), lymph node involvement, lesion morphology, or Emax values (Table 1). Significant univariate predictors (P<0.10) included maximum lesion diameter, hyperechoic halo presence, Adler grade, Eratio, posterior echo reduction, and microcalcifications (Table 2).

Multivariate analysis established four independent predictors of high Ki-67 expression: hyperechoic halo presence [adjusted odds ratio (aOR) =7.934; 95% confidence interval (CI): 2.604–24.173], posterior echo reduction (aOR =0.245; 95% CI: 0.099–0.601), calcifications (aOR =3.524; 95% CI: 1.466–8.472), and Adler grade (aOR =2.334; 95% CI: 1.222–4.456) (all P<0.05; Table 2). These predictors were integrated into a clinically applicable nomogram quantifying individualized risk of high Ki-67 expression (Figure 1).

Figure 1 The nomogram was used to predict the probability of multimodal ultrasound predicting Ki-67 status in breast cancer. Calculate the probability, find out the prediction point on the highest point scale corresponding to each patient variable, and sum it. The total number of points projected onto the bottom scale represents the probability of multimodal ultrasound predicting the Ki-67 status of breast cancer.

The nomogram demonstrated moderate discrimination (AUC =0.796, 95% CI: 0.722–0.872; Figure 2) and excellent calibration (Hosmer-Lemeshow, P=0.79). Bootstrap validation (1,000 replicates) yielded apparent C-index of 0.797 (95% CI: 0.734–0.875) and optimism-corrected C-index of 0.771 (95% CI: 0.673–0.879), with calibration slope 0.92 (95% CI: 0.85–0.99) indicating minimal overfitting (Figure 3). This indicates moderate discriminative capacity with well-maintained calibration accuracy.

Figure 2 The prediction of the expression level of Ki-67 using the logistic regression analysis of certain ultrasonographic parameters. AUC, area under the curve; CI, confidence interval.

Figure 3 The calibration curve illustrates the observation and predictive diagnostic rate of multimodal ultrasound for Ki-67 status of breast cancer.

DCA demonstrated significant net clinical benefit across threshold probabilities (10–70%), supporting clinical utility for Ki-67 stratification (Figure 4).

Figure 4 The clinical decision curve showed that the nomogram model had a higher net benefit in predicting the risk of high Ki-67 expression. Mode 1: represents the nomogram of this study.

Discussion

Ultrasonography offers distinct advantages for breast cancer screening, including absence of ionizing radiation, cost-effectiveness, non-invasiveness, and excellent procedural reproducibility. Critical knowledge gaps persist regarding the association between sonographic biomarkers and key biological prognosticators in breast cancer. This study therefore investigated specific sonographic signatures—hyperechoic halo, microcalcifications, posterior acoustic alteration, maximum diameter, Adler grade, and Eratio to establish their correlation with Ki-67 status, aiming to develop a non-invasive preoperative predictor for clinical decision support. This approach provides a clinically actionable framework for personalizing therapeutic strategies, monitoring treatment response, and ultimately improving breast cancer outcomes.

This study established hyperechoic halo presence as an independent predictor of elevated Ki-67 expression, demonstrating a significant positive correlation. Conventional two-dimensional (2D) ultrasonography defines hyperechoic halo as a circumferential band of increased echogenicity enveloping the lesion periphery. Histopathologically, this feature correlates with spiculated tumor borders and peritumoral infiltration patterns. As a validated malignancy indicator (20). Pathological studies have confirmed that this feature represents the active proliferation of cancer cells, lymphocyte infiltration, and reactive hyperplasia of fibrous connective tissue around malignant tumors (21). Furthermore, subsequent studies demonstrate that an increased hyperechoic halo width significantly correlates with poor patient prognosis. These evidences establish hyperechoic halo as a quantifiable imaging biomarker of tumor invasiveness. Further analysis reveals that elevated Ki-67 expression—a marker protein of cellular proliferation—is strongly associated with aggressive tumor behavior. The imaging features of hyperechoic halo essentially reflect the proliferative active area at the edge of the tumor and the surrounding matrix reaction, which is intrinsically consistent with Ki-67-driven cell cycle progression. Thus, hyperechoic halo provides dual validation: imaging confirmation of Ki-67-associated proliferation dynamics and cross-modal evidence of tumor biology through radiology-pathology correlation.

In this study, calcification is also an independent factor affecting the expression of Ki-67, and there is a positive correlation between them. Two distinct pathomechanisms underlie breast cancer microcalcifications: dystrophic calcification: resulting from ischemic necrosis in rapidly proliferating tumors with compromised vascular supply (22). The other is cell-mediated breast cancer microcalcification formation. This pathological calcification may be related to the active secretion process of tumor cells. Some breast cells undergo epithelial-mesenchymal transition (EMT) under specific stimulation to obtain osteogenic function. Pathological calcification is formed by simulating ossification process with the participation of osteogenic related proteins such as bone morphogenetic protein (BMP) and alkaline phosphatase (ALP). Comprehensive analysis found that the mechanisms of these two calcifications are potentially associated with excessive cell proliferation. The first type of necrotic calcification directly reflects the microenvironment disorder caused by excessive tumor proliferation. The second type of cell-mediated calcification forms a cross-regulatory network with proliferation signaling pathways (such as Wnt/β-catenin) through the EMT process to promote cell invasion and metastasis (23). It has been reported that microcalcification in breast cancer is closely related to the recurrence of breast ductal carcinoma and the invasion and metastasis of breast invasive ductal carcinoma (24). In summary, we believe that microcalcification is not only an important marker of breast cancer imaging diagnosis, but also the proliferation-related pathways involved in its formation mechanism may provide a new perspective for the study of the biological function of Ki-67.

A significant inverse correlation was observed between posterior acoustic attenuation and Ki-67 expression. As one of the typical ultrasonic features of breast cancer, posterior echo attenuation is due to the abnormal deposition of collagen and collagen fibers in tumor tissue, the absorption and scattering of acoustic energy by necrotic hemorrhage and calcification (25). Well-differentiated tumors demonstrate higher prevalence of posterior attenuation compared to poorly-differentiated/invasive carcinomas. Low-grade tumors usually show slow growth rate, low metastatic tendency and usually inert clinical course. The biological behavior of this phenomenon often represents a limited demand for cell proliferation, which is consistent with the low expression of Ki-67. This further confirms that the posterior echo attenuation can be used as an imaging marker to assist in predicting the low expression of Ki-67 in tumors.

Adler blood flow grade independently predicted elevated Ki-67 expression, demonstrating significant positive correlation. Relevant studies suggest that Ki-67 plays an important role in cell proliferation and differentiation, and is positively correlated with vascular endothelial growth factor (VEGF). VEGF enhances tumor angiogenesis through dual mechanisms: neovascularization induction and vascular permeability elevation, ultimately increasing microvessel density (MVD). These processes manifest sonographically as hypervascularity (high Adler grade) and accelerated tumor growth. However, some researchers believe that HER2, as a proto-oncogene, plays an important role in inhibiting tumor apoptosis, promoting cell proliferation, and promoting the formation of malignant tumor microvessels and lymphatic vessels. While our model established Adler grade as an independent predictor, the complex interrelationships among tumor size, vascularity, and proliferation markers warrant further mechanistic investigation.

This study analyzed several SWE parameters, including Emax and Eratio. Eratio quantifies relative stiffness as lesion stiffness divided by adjacent fat stiffness. During the development of breast cancer, cancer-related growth factors in the tumor microenvironment directly or indirectly stimulate the expression of VEGF, connective tissue growth factor (CTGF) and fibroblast growth factor (FGF), and promote fibrosis, collagen fiber deposition and tumor angiogenesis (26). Lesion stiffness directly correlates with collagen content. Studies have suggested that breast cancer with poor prognosis may have higher Emax and Eratio values according to histological prognostic characteristics, immunohistochemical characteristics and subtypes (27). However, in this study, there was no significant difference between Emax and Eratio values and Ki-67 expression status. This discrepancy warrants validation in larger cohorts to clarify SWE’s role in proliferation assessment.

There are several limitations in this study. First of all, the interpretation of breast ultrasound features is largely influenced by the personal experience of ultrasound doctors. The ultrasound section images stored in the database may cause the features of other ultrasound sections to be ignored. Secondly, there are many pathological types of breast cancer, but this study only focuses on invasive breast cancer, so its representativeness is limited; finally, this study is a retrospective, single-center study with a small total sample size. Future multicenter, prospective studies with larger sample sizes are warranted to externally validate our model and improve its general applicability.

Conclusions

In summary, the validated nomogram provides clinically useful preoperative prediction of Ki-67 status, with the presence of a hyperechoic halo, posterior echo reduction, calcifications, and a high Adler grade on multimodal ultrasound identified as key predictors. The model developed in this study demonstrates good predictive performance and generalizability for estimating Ki-67 expression levels. In future clinical practice, utilizing these ultrasound features could enable a preliminary assessment of breast cancer prognosis, thereby enhancing the clinical utility of ultrasonography in the diagnosis and management of breast cancer.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1652/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1652/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1652/prf

Funding: This study was supported by scientific research foundation of Yunnan Education Department (No. 2023Y0654), Yunnan Provincial Department of Education Youth Talent Basic Research Project (No. 2024J0248), Yunnan Provincial Department of Science and Technology-Kunming Medical University Joint Special Fund for Applied Basic Research-Key Projects (Nos. 202301AY070001-009 and 202401AY070001-042), Yunnan Provincial Department of Science and Technology-Kunming Medical University Applied Basic Research Joint Special Fund-General Project (Nos. 202401AY070001-268 and 202501AY070001-108).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-1652/coif). All authors declare that this study was supported by scientific research foundation of Yunnan Education Department (No. 2023Y0654), Yunnan Provincial Department of Education Youth Talent Basic Research Project (No. 2024J0248), Yunnan Provincial Department of Science and Technology-Kunming Medical University Joint Special Fund for Applied Basic Research-Key Projects (Nos. 202301AY070001-009 and 202401AY070001-042), Yunnan Provincial Department of Science and Technology-Kunming Medical University Applied Basic Research Joint Special Fund-General Project (Nos. 202401AY070001-268 and 202501AY070001-108). The authors have no other 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 was approved by the Institutional Ethics Committee of Yunnan Cancer Hospital (No. KYLX2023-021). The ethics committee granted a waiver of informed consent for this retrospective analysis, with all data anonymized prior to analysis.

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. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
2. Noske A, Anders SI, Ettl J, et al. Risk stratification in luminal-type breast cancer: Comparison of Ki-67 with EndoPredict test results. Breast 2020;49:101-7. [Crossref] [PubMed]
3. Tian Q, Wang Y, Guo H, et al. Recent perspectives of management of breast cancer metastasis - an update. J BUON 2017;22:295-300.
4. Schaffar R, Belot A, Rachet B, et al. On the use of flexible excess hazard regression models for describing long-term breast cancer survival: a case-study using population-based cancer registry data. BMC Cancer 2019;19:107. [Crossref] [PubMed]
5. Rakha EA, Pareja FG. New Advances in Molecular Breast Cancer Pathology. Semin Cancer Biol 2021;72:102-13. [Crossref] [PubMed]
6. Goldhirsch A, Wood WC, Coates AS, et al. Strategies for subtypes--dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol 2011;22:1736-47. [Crossref] [PubMed]
7. Cuylen S, Blaukopf C, Politi AZ, et al. Ki-67 acts as a biological surfactant to disperse mitotic chromosomes. Nature 2016;535:308-12. [Crossref] [PubMed]
8. Wongmaneerung P, Chitapanarux I, Traisathit P, et al. The association between Ki-67 expression and survival in breast cancer subtypes: a cross-sectional study of Ki-67 cut-point in northern Thailand. BMC Cancer 2025;25:346. [Crossref] [PubMed]
9. Davey MG, Hynes SO, Kerin MJ, et al. Ki-67 as a Prognostic Biomarker in Invasive Breast Cancer. Cancers (Basel) 2021;13:4455. [Crossref] [PubMed]
10. Mrouj K, Andres-Sanchez N, Dubra G, et al. Ki-67 regulates global gene expression and promotes sequential stages of carcinogenesis. Proc Natl Acad Sci U S A 2021;118:e2026507118. [Crossref] [PubMed]
11. Andrés-Sánchez N, Fisher D, Krasinska L. Physiological functions and roles in cancer of the proliferation marker Ki-67. J Cell Sci 2022;135:jcs258932. [Crossref] [PubMed]
12. Kamranzadeh H, Ardekani RM, Kasaeian A, et al. Association between Ki-67 expression and clinicopathological features in prognosis of breast cancer: A retrospective cohort study. J Res Med Sci 2019;24:30. [Crossref] [PubMed]
13. Rossi C, Fraticelli S, Fanizza M, et al. Concordance of immunohistochemistry for predictive and prognostic factors in breast cancer between biopsy and surgical excision: a single-centre experience and review of the literature. Breast Cancer Res Treat 2023;198:573-82. [Crossref] [PubMed]
14. Tekcan Sanli DE, Yildirim D, Kandemirli SG, et al. Evaluation of Multiparametric Shear Wave Elastography Indices in Malignant and Benign Breast Lesions. Acad Radiol 2022;29:S50-61. [Crossref] [PubMed]
15. Kang HJ, Kim JY, Lee NK, et al. Three-dimensional versus two-dimensional shear-wave elastography: Associations of mean elasticity values with prognostic factors and tumor subtypes of breast cancer. Clin Imaging 2018;48:79-85. [Crossref] [PubMed]
16. Allison KH, Hammond MEH, Dowsett M, et al. Estrogen and Progesterone Receptor Testing in Breast Cancer: ASCO/CAP Guideline Update. J Clin Oncol 2020;38:1346-66. [Crossref] [PubMed]
17. Li J, Hao C, Wang K, et al. Chinese Society of Clinical Oncology (CSCO) Breast Cancer guidelines 2024. Transl Breast Cancer Res 2024 25;5:18.
18. Polewski MD, Nielsen GB, Gu Y, et al. A Standardized Investigational Ki-67 Immunohistochemistry Assay Used to Assess High-Risk Early Breast Cancer Patients in the monarchE Phase 3 Clinical Study Identifies a Population With Greater Risk of Disease Recurrence When Treated With Endocrine Therapy Alone. Appl Immunohistochem Mol Morphol 2022;30:237-45. [Crossref] [PubMed]
19. Nielsen TO, Leung SCY, Rimm DL, et al. Assessment of Ki67 in Breast Cancer: Updated Recommendations From the International Ki67 in Breast Cancer Working Group. J Natl Cancer Inst 2021;113:808-19. [Crossref] [PubMed]
20. Zhang XD, Kou M, Zou WY, et al. HER-2 expression is correlated with multimodal imaging features in breast cancer: a pilot study. Ann Med 2024;56:2434182. [Crossref] [PubMed]
21. Zhu Y, Jia X, Zhan W, et al. Adding contrast-enhanced ultrasound can improve the predictive ability of breast conventional ultrasound and mammography for pathological upgrade of biopsy-confirmed ductal carcinoma in situ. Eur J Radiol 2024;180:111687. [Crossref] [PubMed]
22. Wang J, Zhao L, Hu X, et al. Clinicopathological characteristics and prognostic significance of casting-type calcifications in patients with invasive breast cancer presenting with microcalcification. Sci Rep 2024;14:13351. [Crossref] [PubMed]
23. Tomecka P, Kunachowicz D, Górczyńska J, et al. Factors Determining Epithelial-Mesenchymal Transition in Cancer Progression. Int J Mol Sci 2024;25:8972. [Crossref] [PubMed]
24. Wang Q, Li B, Liu Z, et al. Prediction model of axillary lymph node status using automated breast ultrasound (ABUS) and ki-67 status in early-stage breast cancer. BMC Cancer 2022;22:929. [Crossref] [PubMed]
25. Luo H, Li J, Shi Y, et al. Stiffness in breast masses with posterior acoustic shadowing: significance of ultrasound real time shear wave elastography. BMC Med Imaging 2022;22:71. [Crossref] [PubMed]
26. Sewell-Loftin MK, Bayer SVH, Crist E, et al. Cancer-associated fibroblasts support vascular growth through mechanical force. Sci Rep 2017;7:12574. [Crossref] [PubMed]
27. Avdan Aslan A, Gültekin S, İnan MA. The Utility of Quantitative Parameters of Shear-Wave Elastography to Predict Prognostic Histologic Features of Breast Cancer. Ultrasound Q 2023;39:81-5. [Crossref] [PubMed]