This study suggests that SWE can characterize ECM in the tumor microenvironment and serve as a promising modality for evaluating axillary LNs after NAC. The performance of SWE on breast tumors is more effective than that on axillary LNs for determining the axilla status in patients with initially node-positive breast cancer after NAC.

In our study, NAC successfully eradicated nodal disease in 51.41% of the patients included in our analysis. However, accurately identifying patients who are likely to achieve a nodal pCR after NAC remains a challenge. As the most recommended imaging modality for the assessment of residual disease in axillary LNs [4], the accuracy of conventional US in the evaluation of axillary LNs after NAC is insufficient, and there is a great variability in its diagnostic performance [19,20,21,22]. And several studies have reported that UE can improve the performance of US in diagnosing axillary LNs [9, 23]. In a meta-analysis evaluating eight imaging modalities for the detection of metastasis in axillary LNs in breast cancer patients, UE was found to be the most effective method for preoperative detection [23]. In the context of assessing axillary status after NAC, only a few studies have focused on the role of UE [12]. And the pathological basis for using UE to evaluate axillary LNs after NAC is unclear, particularly regarding the preferred approach for assessing axillary status—whether UE should be performed on breast tumors or axillary LNs.

In our study, there were stronger positive correlations between SWV and CVF compared to TCD for both breast tumors and axillary LNs after NAC. Similarly, Chamming's F reported that there was a very significant correlation between elasticity and fibrosis in breast cancer prior to treatment, but no significant correlation with viable cellular tissue [17]. These findings suggest that collagen deposition may contribute more to tumor stiffness than tumor cell presence. In the context of breast lesions, the presence of necrotic tissue can indicate advanced disease or aggressive tumor behavior. Similarly, the presence of necrotic tissue in axillary LNs can suggest advanced disease and may be associated with the spread of malignant tumors to the LNs. Based on our study findings, however, necrosis is rarely observed in breast lesions or axillary LNs after NAC. This implies that the presence of necrosis has little impact on the detection of residual metastasis in axillary LNs. The results of this study concluded that collagen deposition was identified as the primary factor contributing to tissue stiffness in breast cancer patients after NAC.

This study demonstrated that the CVF of breast tumors in axillary residual metastasis group was found to be significantly higher than that in axillary pCR group. These findings indicate the presence of a greater cancer-associated collagen composition in breast tumors with positive LNs after NAC. Correspondingly, this study found that the breast lesions in axillary residual metastasis group exhibited higher stiffness compared to those in axillary pCR group. This study suggests that SWE has the potential to evaluate ECM characteristics in breast cancer after NAC, and the breast SWE can reflect the differences in collagen deposition within breast tumors between patients with negative and positive LNs. Abnormal collagen composition within the tumor microenviroment can contribute to increased interstitial pressure, which in turn can lead to the collapse of tumor vessels [24, 25]. This collapse reduces tumor perfusion, thus limiting the delivery of chemotherapy drugs to the tumor site. Consequently, tumors with higher stiffness, as indicated by SWE, may indicate resistance to chemotherapy. This study further proves that breast SWE can characterize the collagen composition within breast tumors after NAC and have the potential to act as predictors for axillary responses to NAC.

Similar to breast tumors, this study found that the SWV values of axillary LNs in axillary residual metastasis group were significantly higher than those in axillary pCR group. The presence of metastatic foci within LNs can lead to an increase in their stiffness due to various factors, including the increased of matrix and cytoskeletal stiffness, abnormal cell proliferation, microcalcification of malignant lesions, and deposition of other abnormal tissues in the stroma [26]. Several studies have demonstrated that axilla SWE can be used to evaluate metastasis in LNs both in vivo [27] and in vitro [28]. In this study, we found that the SWE performed on axillary LNs were not as effective as that performed on breast tumors for the detection of axillary residual metastasis after NAC. As the pathological characteristic contributing most to stiffness after NAC, collagen composition in axillary LNs is also less effective than in breast tumors for assessing axillary status. Pathological analysis revealed that the difference in CVF of axillary LNs was smaller compared to that of breast tumors when distinguishing between the positive and negative LN groups after NAC. Similarly, a smaller difference in the SWV of axillary LNs than breast tumors was also observed between the two axillary response groups. This implies that axilla SWE carries a higher risk in false negative and false positive results than breast SWE for evaluating axillary LNs.

Upon retrospective review of false negative cases, there was only minimal collagen deposition in axillary LNs, even in those with residual metastasis. Unlike primary breast malignancies, metastases in axillary LNs may rarely induce a desmoplastic reaction [11]. Thus, we speculate that axilla UE may lead to underestimation in the diagnosis of axillary LNs. In addition, we observed that LNs in a significant portion of false positive cases were located in a deep position (vertical distance from epidermis > 2.5 cm). It has been suggested that the perpendicular depth of LN location may affect the signal stability of UE [29]. Therefore, it can be inferred that UE of LNs in deep axillary location may lead to overestimation of elasticity, thereby resulting in false positive findings. In addition, compared to the breast, the potential impact of artifacts and a more complex acquisition process in the axillary region might also contribute to the lower performance of axilla SWE.

In the context of NAC for breast cancer, this study suggested that SWE can be utilized to assess collagen deposition in breast tumors and axillary LNs. These findings provided pathological evidence supporting the use of SWE for diagnosing axillary LNs after NAC in patients with node-positive breast cancer. More importantly, the study proved that breast SWE is the preferred approach for evaluating the axillary status after NAC, as it demonstrates inherent advantages over axilla SWE, as confirmed by pathological analysis. Nevertheless, this study has some limitations that should be acknowledged. First, this study is limited by its single-center design. Second, it is important to note that we did not conduct a comparison between SWE and the standard evaluation for LNs (conventional axilla US). Further research is needed to comprehensively assess the effectiveness of SWE in this context. Third, due to the absence of clip placement within biopsied LNs, it was challenging to determine the pathological status of index LNs before treatment. Further, it is known that NAC impacts tumor extracellular collagen in a complex and subtype-specific manner [30], but we did not conduct a subgroup analysis based on the specific molecular type of breast cancer in our study. Finally, there is no specific pairing process implemented to directly correlate the LNs observed using SWE with their corresponding pathology analysis. The axillary US assessment focused on the LNs with the most suspicious features, and it is likely that the pathologically analyzed LNs include those that have been identified and analyzed during the imaging process.