The efficacy of strain elastography and shear-wave elastography in the evaluation of neoadjuvant chemotherapy for patients with locally advanced breast cancer

Abstract

Objective

To explore the clinical application value of ultrasound elastography (UE) technology in the evaluation of the efficacy of neoadjuvant chemotherapy (NAC) at different time-points in patients with locally advanced breast cancer.

Methods

Two group patients were performed conventional ultrasound and elastography examination before NAC (pre-NAC),after the second (post-2^nd-NAC), and the fourth cycles of NAC(post-4^th-NAC),including 40 cases of 40 breast cancer patients (age range, 25-72 years) in Strain Elastography (SE) group and 21 cases of 20 breast cancer patients (age range,31-67 years) in shear-wave elastography (SWE) group. Changes in elastographic parameters (elasticity grading and strain ratio in SE group, the maximum elastic value and the average elastic value in SWE group) related to tissue biomechanical properties were then determined and compared to clinical and pathologic tumor response after mastectomy. Receiver operating characteristic (ROC) curve analysis was performed to examine parameters at SE and SWE to predict pathologic response.

Results

The difference of the SE scores between the effective and the ineffective group were statistically significant, no matter after the second or fourth cycles of NAC(P = 0.003 and P = 0.002).The efficacy of NAC can be evaluated by SE semi-quantitative method（no matter with normal breast tissue or adipose tissue as reference）(with normal breast tissue as reference: area under the receiver operating characteristic curve(AUC): 0.763 (post-2^nd-NAC),0.836 (post-4^th-NAC) ; sensitivity:66.7% (post-2^nd-NAC) , 83.3% (post-4^th-NAC); and specificity:87.5% (post-2^nd-NAC) , 87.5% (post-4^th-NAC)；with adipose tissue as reference area under the receiver operating characteristic curve: 0.733 (post-2^nd-NAC) ,0.775 (post-4^th-NAC); sensitivity: 66.7% (post-2^nd-NAC) ,87.5% (post-4^th-NAC); and specificity: 75% (post-2^nd-NAC) _, 68.7% (post-4^th-NAC)) and SWE quantitative parameters(taking Emean as measurement parameter: area under the receiver operating characteristic curve: 0.929 (post-2^nd-NAC) ,0.929 (post-4^th-NAC); sensitivity:85.7% (post-2^nd-NAC), 78.6% (post-4^th-NAC)_; and specificity100% (post-2^nd-NAC)_, 100% (post-4^th-NAC);taking Emax as measurement parameter: area under the receiver operating characteristic curve: 0.898 (post-2^nd-NAC) , 0.908 (post-4^th-NAC); sensitivity: 85.7% (post-2^nd-NAC) _, 71.4% (post-4^th-NAC)_; and specificity100% (post-2^nd-NAC)_, 100% (post-4^th-NAC)) after the second and fourth cycles of NAC. There was no significant statistically difference between SE semi-quantitative methods and SWE quantitative parameters in the evaluation of the efficacy of NAC (P =0.082 post-2^nd-NAC and 0.324 (post-4^th-NAC)).

Conclusion

Both SE and SWE can be potentially used as an early predictor of tumor therapy response in locally advanced breast cancer patients, and there was no significant difference in their assessment effectiveness

Introduction

In recent years, the incidence of breast cancer has increased significantly, which is the most frequent malignancy among women in China [1]. About 5% − 20% of newly diagnosed breast cancer is locally advanced breast cancer (LABC). Neoadjuvant chemotherapy (NAC) has become a routine treatment for LABC. Due to the highly heterogeneity of the breast cancers, the response to NAC is different, and only a minority of patients achieve a pathologic complete response (pCR). Patients achieving pCR after NAC has low recurrence risk and prolonged disease-free survival [2]. However, only about 3% − 30% of breast cancer patients can achieve pCR after NAC treatment, while about 20% of breast cancer patients cannot benefit from NAC treatment, which leads to disease progression, loss of the best opportunity for surgical treatment[3]. Knowledge of how the tumor is responding to treatment is essential to guiding further treatment options. Early detection of unresponsive tumors and timely adjustment of treatment strategies will reduce the side effects of meaningless chemotherapy and iCmprove the prognosis of breast cancer patients.

Clinical examination combined with traditional imaging methods, including magnetic resonance imaging (MRI), conventional ultrasound and mammography, has been used to evaluate the response of tumor to NAC. However, when using these imaging techniques to evaluate the anatomical and pathological characteristics of tumors, there is a delay between the evaluation of the initial size and the reduced size of the tumors, which may prevent the radiologists from distinguishing the active tumors from the fibrotic scars.

Ultrasound elastography is a new imaging technique that can be used to assess tissue stiffness. The results of laboratory animal models showed that tumor hardness was correlated with tumor progression and chemoresistance[4, 5]. Palpation is one of the oldest diagnostic techniques. The limitation of palpation lies on its low positive detection rate, low spatial identification and strong subjectivity. The purpose of using noninvasive measurement of tissue hardness by ultrasound elastography is to overcome the above limitations [6]. SE can evaluate the hardness of breast tumor in the whole course of NAC by qualitative and semi quantitative methods [7]. SWE is a new ultrasonic technology developed in recent years. Compared with SE imaging, SWE shows an advantage in good repeatability and objectivity. Therefore, the quantitative measurement parameters of tumor hardness provided by ultrasound elastography, especially SWE, has great potential to predict the response of breast cancer patients to NAC.

The purpose of this study was to investigate and compare the potential of SE and SWE for monitoring treatment response of LABC patients undergoing NAC.

Materials And Methods

1. Research subjects

The study was approved by the institutional ethics committee. SE group：From November 2014 to December 2016, patients from the same situation who had primary breast cancer histologically diagnosed by core needle biopsy were recruited to undergo SE examination before therapy and at completion of second and fourth cycle of chemotherapy. SWE group：Breast cancer patients who received NAC in the department of breast surgery of our hospital from October 2017 to January 2019. The breast surgery is carried out after 4 cycles of chemotherapy. There are three times of conventional ultrasound and SWE examination, which carried out respectively within the week before the chemotherapy, after the second cycle of chemotherapy, and after the fourth cycle of chemotherapy yet before the operation.

Inclusion criteria were patients with breast cancer aged 18 to 80 years, had histologically confirmed large operable breast cancer (T≥2cm and N0-2), without any history of treatment for breast cancer and major organ dysfunctions. Patients who had a previous history of malignancy other than breast cancer or had received any prior chemotherapy, hormonal therapy, or radiation for present breast cancer were also excluded.

40 lesions in 40 patients in SE group were included in the study group. All the patients were female, aged between 25-72 years, with an average of (51.55 ± 11.84) years old. 21 lesions in 20 patients in SWE group were included in the study group. All the patients were female, aged between 31-67 years, with an average of (50.19 ± 10.71) years old.

2. Instruments and methods

2.1 Instruments

SE was performed with MyLab Twice color Doppler ultrasound diagnostic instrument, LA523 linear array probe with a frequency of 4-13 MHz and equipped with MyLab Desk analysis software. SWE was performed with supersonic color Doppler ultrasound diagnostic instrument equipped with Q-Box^TM quantitative measurement tool, and 4-15MHz linear array probe was used in conventional gray-scale ultrasound and SWE.

2.2 Methods

All patients were examined by two fixed ultrasound doctors. Patients took supine position or semi lateral decubitus position, with bilateral breasts fully exposed and calm breathing.

Patients in SE group: First, conventional gray-scale ultrasound was performed to determine the location of the lesion and the maximum diameter of the lesion was measured. When selecting the focus, it is necessary to avoid areas of necrosis and calcification and select areas with abundant blood flow as far as possible. When the lesion exceeds the probe range, select partial region of the lesion. Then switched to SE mode and adjusted the region of interest (ROI) of elastic imaging, so that the whole focus and surrounding normal breast tissue were all in the ROI. When the focus was too large to be completely covered, the ROI where part of the focus and the surrounding normal breast tissue co-exist were chosen. During the operation, the probe should be vertical to the body surface with slight vibration on the focus. According to the comprehensive index of the pressure and the pressure discharge frequency shown on the display screen of the diagnostic instrument, it would be suitable for the strength and frequency to reach 4. When the green spring displayed the frozen image for 5 seconds, the same focus was measured for 3 times repetitively, the mean value was calculated and recorded, trying to avoid the error of elastic measurement.

Patients in SWE group: First, gray-scale ultrasound was performed to determine the location of the lesion and the maximum diameter of the lesion was measured. When selecting the focus, it is necessary to avoid areas of necrosis and calcification and select areas with abundant blood flow as far as possible. When the lesion exceeds the probe range, select partial region of the lesion. Then SWE mode was performed and the image was recorded. In the SWE mode, the probe should be vertical to the body surface without pressure with a range of 0 ~ 180kPa.The whole focus and surrounding normal fat tissue was covered by the selected sampling frame. When necessary, the patients should hold their breath until the image was stable. Then the frozen image was recorded. The same focus was measured for 2-3 times repetitively, the mean value was calculated and recorded, trying to avoid the error of elastic measurement.

2.3 Images and data analysis 

2.3.1 Qualitative analysis and semi quantitative analysis in SE group 

Qualitative analysis was performed by grading the elasticity diagram of the focus by the 5-point method proposed by Itoh A et al[8]. 1-3 points of ultrasound elasticity grading after chemotherapy as effective treatment, and 4-5 points as ineffective treatment[9].

Semi quantitative analysis was performed by MyLab Desk analysis software. The ROI was manually selected in the focus, surrounding normal breast tissue and subcutaneous fat layer, then the strain ratio (SR)was calculated. The SR value was obtained by dividing the average value of the normal reference tissue intrinsic strain by the average value of the focus intrinsic strain. Two control groups were set up: normal breast tissue control group (SR-t group) and fat control group (SR-f group) (Figure 1).

According to the formula, the changes of SR value after the second and fourth cycle of chemotherapy of the two control groups were obtained respectively: △ SR2-t = (1-SR-t value after the second cycle of NAC / SR-t value before chemotherapy) Í100%, △ SR4-t = (1-SR-t value after the fourth cycle of NAC/ SR-t value before chemotherapy) × 100%; △SR2-f = (1-SR-f value after the second cycle of NAC/ SR-f value before chemotherapy) × 100%, △SR4-f = (1-SR-f value after the fourth cycle of NAC/ SR-f value before chemotherapy) × 100%.

All images were graded and measured by the same physician, who blinded to clinical presentation, any other type of breast imaging or their reports, and pathology findings.

2.3.2 Quantitative parameter analysis in SWE group

Q-Box^TM, a quantitative measurement tool built in the machine, was used to measure the parameters of the ROI of SWE section of the hardest part of the focus. All images were measured and analyzed by the same physician, who blinded to clinical presentation, any other type of breast imaging or their reports, and pathology findings. The whole focus and the area with abnormally increased elastic modulus at the edge of the lesion were covered by the selected sampling frame. 2mm diameter sampling frame was placed at the hardest part of the focus to calculate the maximum elastic value (Emax) and the average elastic value (Emean) of the hardest part of the focus.

According to the formula, the change in value of Yang's modulus of SWE was got after the second and fourth cycles of chemotherapy: △Emean2 = (1-Emean after the second cycle of NAC / Emean before chemotherapy) Í100%, △Emean4 = (1-Emean after the fourth cycle of NAC / Emean before chemotherapy) ×  100%; △Emax2 = (1-Emax after the second cycle of NAC/ Emax before chemotherapy) ×  100%, △Emax4 = (1-Emax after the fourth cycle of NAC/ Emax before chemotherapy) ×  100% (Figure 2-3).

2.4 Pathological analysis

All the clinical, ultrasonic and pathological data of the patients were collected, and the pathological results of the surgical specimens were sorted out. According to the Miller-Panye classification method[8], the pathological reaction of NAC was evaluated as follows: 1.The overall tumor cell density had no significant change compared with the previous level;2.The second level was that the overall tumor cell density was still high, and the cell density was reduced by less than 30%; 3.The third level was that the tumor cell density was reduced by 30% ~ 90%; 4.the fourth level was that the tumor cell density was reduced by more than 90%; 5.The fifth level was that the tumor completely disappeared, with no infiltrating cancer residue found under the microscope, but may contain Ductal carcinoma in situ, which was considered as pathologic complete remission. 1-3 was defined as ineffective treatment, 4-5 was defined as effective treatment.

2.5 Statistical methods

SPSS 24.0 statistical analysis software was used for the analysis. The measurement data were expressed as mean±standard deviation. The independent sample t test was used for the comparison between groups. Taking the pathological examination results as the gold standard, the change of SR value after the second and fourth cycles of chemotherapy was taken as the result value in the SE group to draw the ROC curve and compare the diagnostic efficacy of △ SR2 and △ SR4 in different reference tissue groups to predict the effective of NAC. In the SWE group, the change in value of Yang's modulus of SWE after the second and fourth cycles of chemotherapy was taken as the result value to draw the ROC curve and compare the diagnostic efficacy of △Emean and △Emax predicting the effective of NAC after different cycles of chemotherapy. The area under the ROC curve (AUC) was compared by Delong test. The changes in value of SR and Yang's modulus were compared in the SE group and SWE group. P < .05 was considered to indicate a significant difference.

Results

SE scores were 4–5 before NAC, and the SE scores of the effective and ineffective were (4.54 ± 0.51) and (4.63 ± 0.50), respectively. The difference between them was not statistically significant (P = 0.672).

After the second cycle of NAC, the SE scores of the effective and the ineffective were (3.33 ± 0.82) and (4.13 ± 0.72) respectively. The difference between them was statistically significant (P = 0.003). The sensitivity and specificity of NAC efficacy evaluated by SE score were 54.2% and 81.2%, respectively. After the fourth cycle of NAC, the SE scores of the effective and the ineffective were (2.46 ± 0.83) and (3.86 ± 0.89) respectively. The difference between them was statistically significant (P = 0.002). The sensitivity and specificity of NAC efficacy evaluated by SE score were 87.5% and 68.8%, respectively (Table 1).

3.3 Evaluation of breast cancer NAC efficacy by SE semi-quantitative

Taking normal breast tissue as the reference:

After the second cycle of NAC, the difference of △SR2-t between the effective and ineffective group was statistically significant (P = 0.003). The area under the ROC curve for △SR2-t was 0.763 (95% CI: 0.602 ~ 0.883). When taking △SR2-t > 35% as the cut-off value, the sensitivity and specificity were 66.7% and 87.5% respectively. After the fourth cycle of NAC, the difference of △SR4-t between the effective and ineffective group was statistically significant (P < 0.001). The AUC assessing NAC efficacy obtained from the ROC curve was 0.836 (95% CI: 0.685–0.934). When taking △SR4-t > 40% as the cut-off value, the sensitivity and specificity were 83.3% and 87.5% respectively. After the second and fourth cycles of NAC, the difference in AUC assessing NAC efficacy by △SR-t was not statistically significant (P = 0.350)(Table 2)(Figure 4A). 

Taking adipose tissue as the reference:

After the second cycle of NAC, the difference of △SR2-f between the effective and ineffective group was statistically significant (P = 0.018). The AUC assessing NAC efficacy obtained from the ROC curve was 0.733 (95% CI: 0.570 ~ 0.860). When taking △SR2-f > 25% as the cut-off value, the sensitivity and specificity were 66.7% and 75% respectively. After the fourth cycle of NAC, the difference of △SR2-t between the effective and ineffective group was statistically significant (P = 0.002). The AUC assessing NAC efficacy obtained from the ROC curve was 0.775 (95% CI: 0.615–0.891). When taking △SR4-f > 26% as the cut-off value, the sensitivity and specificity were 87.5% and 68.7% respectively. After the second and fourth cycles of NAC, the difference in AUC assessing NAC efficacy by △SR-f was not statistically significant (P = 0.412) (Table 2)(Figure 4B).

The difference between the AUC of △SR2-t and △SR2-f after the second cycle of NAC was not statistically significant (P = 0.613). The difference between the AUC of△SR4-t and △SR4-f after the fourth cycle was not statistically significant (P = 0.227).

3.4 Evaluation of breast cancer NAC efficacy by SWE quantitative parameters

Before NAC, the value of the Emean and Emax of the effective and ineffective group showed no statistical difference (P = 0.701 and 0.694). After the second cycle of NAC, both the value of Emean and Emax of the effective and the ineffective group showed statistical difference (P = 0.016 and 0.012). After the fourth cycle of chemotherapy, both the value of Emean and Emax of the effective and the ineffective group showed statistical difference (P = 0.009 and 0.015).

After the second cycle of NAC, the difference of △Emean2 between the effective and the ineffective group was statistically significant (P = 0.001). For △Emean2, the area under the ROC curve was 0.929 (95% CI: 0.728 ~ 0.995). When taking △Emean2 > 58% as the cut-off value, the sensitivity and specificity were 85.7% and 100% respectively. After the fourth cycle of NAC, the difference of △Emean4 between the two groups was statistically significant (P = 0.003). For △Emean4, the area under the ROC curve was 0.929 (95% CI: 0.728 ~ 0.995). When taking △Emean4 > 79.5% as the cut-off value, the sensitivity and specificity were 78.6% and 100% respectively. After the second and fourth cycles of NAC, the difference in AUC assessing NAC efficacy by △Emean was not statistically significant(Table 3) (Figure 5A).

After the second cycle of NAC, the difference of △Emax2 between the effective and the ineffective groups was statistically significant (P = 0.001). The AUC assessing NAC efficacy obtained from the ROC curve was 0.898 (95% CI: 0.687 ~ 0.986). When taking △Emax2 > 62.4% as the cut-off value, the sensitivity and specificity were 71.4% and 100% respectively. After the fourth period of NAC, the difference of △Emax4 between them was statistically significant (P = 0.001). The area under the ROC curve was 0.908 (95% CI: 0.701 ~ 0.989). When taking △Emax4 > 74.9% as the cut-off value, the sensitivity and specificity were 85.7% and 100% respectively. After the second and fourth cycles of NAC, the difference in AUC assessing NAC efficacy by △Emax was not statistically significant(Table 3) (Figure 5B).

After the second cycle of NAC, the difference in AUC of△Emean2 and △Emax2 was not statistically significant (P = 0.480). After the fourth cycle, the difference in AUC of them was not statistically significant (P = 0.571).

3.5 Comparison between SE semi-quantitative parameters and SWE quantitative parameters

Through the comparison of independent samples, AUC, the difference in AUC of the change of SR value △SR2-t and △Emean2 after the second cycle of NAC was not statistically significant (P = 0.082). After the fourth cycle, the difference in AUC of△SR4-t and △Emean4 was not statistically significant (P = 0.324).

Discussion

NAC is increasingly being offered to women with LABC in order to down-stage the tumor and facilitate breast conserving surgery. Due to the highly heterogeneity of the breast cancers, the response to the chemotherapy is different, and only a minority of patients achieve a pathologic complete response. Early evaluation the response of NAC is very important. It can guide clinicians to make individualized breast cancer treatment plan during the treatment, so as to avoid ineffective chemotherapy. The evaluation of breast cancer NAC efficacy is the hotspot of current research.

Ultrasonic elastic imaging, as a new technology that can reflect the hardness of the tumor, has the advantages of simplicity, noninvasive and good repeatability. At present, the evaluation methods of the chemotherapy response of breast cancer by SE technology mainly include grading and strain rate ratio method[10, 11]. Tumor hardness is closely related to the NAC response of breast cancer. Falou et al. used SE imaging to evaluate the response of 15 patients with LABC to NAC[10]. They found that the elastic strain ratio was the best predictor of NAC response.

SWE technology is a rapidly developing new imaging method to evaluate tissue hardness. SWE has been recognized as an important method to evaluate tumor characteristics and is related to tumor progression [12].

By using SE and SWE technology, our study proved that tumors of the effective group were significantly softer than that in the ineffective group before NAC and after the second and fourth cycles of NAC.

Early prediction of response to NAC in breast cancer is crucial for guiding therapy decisions.[13]. Previous studies on SWE evaluation of NAC efficacy mainly focused on the tumor hardness obtained before chemotherapy and after the completion of chemotherapy. Our study provides more information for the early prediction of the breast cancer patients’ response to NAC.

In this study, SE elasticity grading of 40 lesions in 40 patients before and after NAC can be used as an effective method to evaluate the chemotherapy efficacy of LABC. Hayashi et al. [11]showed that breast cancer patients with lower elastic score had higher clinical complete remission rate and pathological complete remission rate than those with higher elastic score, which could be used as a meathod to predict the NAC pathological response. This is consistent with the results of our study.

Elastic strain rate is a semi-quantitative evaluation of the hardness of the lesions by calculating the SR value between the lesions and the surrounding normal tissue or subcutaneous fat layer. Falou et al. [14] followed up the NAC treatment of 15 patients with LABC and found that compared with the ineffective group, the SR value of the effective group was significantly reduced, and its SR value of elastic imaging of static ROI and dynamic ROI was (81 ± 3) %. While the SR values of elastic imaging of static ROI and dynamic ROI of the ineffective group were (102 ± 2) % and (101 ± 4) % respectively. But at present, there is no unified SR critical value based on large sample to evaluate the NAC efficacy of breast cancer. Similar to the results of the above-mentioned researches, this study divided the treatment outcome into the effective and the ineffective according to the Miller-Panye classification method, and found that when taking the normal breast tissue and fat as references, the changes of SR value were more than 40% and 26% respectively, which can be used as a reliable basis for effective treatment. It also can be concluded that the strain rate ratio method was helpful for the evaluation of chemotherapy response.

In the previous studies on the strain rate ratio of breast ultrasound elastography, normal breast tissue or adipose tissue was selected as the reference[14–18]. In this study, two control groups were set up to measure the strain rate ratio respectively. The results showed that the AUC difference between the two groups was not statistically significant (P < 0.05), indicating that when assessing NAC efficacy, no matter normal breast tissue or adipose tissue was selected as reference, there was no significant difference.

In SWE group,21 lesions in 20 patients were included in the study. There was no significant difference in Emean and Emax between the effective and the ineffective groups before NAC. After the second cycle of NAC, both the Emean2 and Emax2 of the effective were larger than that in the ineffective. The difference was statistically significant. This indicated that, after the second cycle of NAC, the lesion of the effective group became softer than that in the ineffective group. After the fourth cycle of NAC, the △Emean4 and △Emax4 of the effective group were also larger than that in the ineffective group with statistically significant difference. Our study showed that there was no significant difference between △Emean and △Emax in evaluating the efficacy of chemotherapy, no matter after the second or fourth cycles of NAC. This further proves that SWE has great potential value in the early prediction of chemotherapy efficacy.

In the comparison of SE semi-quantitative parameters, the AUC of △SR in normal breast tissue control group was larger than that in the adipose tissue control group, but there was no significant difference between two control groups in the evaluation of the efficacy of NAC. In the comparison of quantitative parameters of SWE, the change of elastic mean value of SWE was greater than the change of maximum value with no significant difference between them. When comparing SR value change (when using normal breast tissue as control) with the change of elastic mean value of SWE, whether after the second or fourth cycle of NAC, the difference in AUC between them was not statistically significant. This revealed that there was no significant difference in the evaluation efficacy of chemotherapy response between SE and SWE. Previous studies have found that both SE and SWE can improve diagnostic performance in differentiating breast lesions when combined with conventional US [19, 20]. However, no significant differences were found between two elastography methods in the evaluation of the efficacy of NAC.

There are still some limitations in this study: First, It is of great significance for the early evaluation of NAC efficacy. However, the sample size of this study was too small, further research is needed; Second, the heterogeneity of breast cancer subtypes was not considered; Third, he elastic imaging parameters after the first cycle of chemotherapy were not analyzed; Fourth, some postoperative pathology showed obvious fibrosis or hyaline degeneration in the original focus, which led to the higher hardness of the lesions after NAC. Finally, patients in our study did not have two kinds of ultrasound elastography at the same time, so only independent samples were compared.

Conclusion

Both SE and SWE technology can be used to evaluate NAC efficacy of patients with LABC at different time-points, and no significant difference was found in evaluation efficacy between them.

Declarations

i.Ethics approval and consent to participate

The study was approved by the institutional ethics committee and all methods were performed in accordance with the relevant guidelines and regulations. Informed consent was obtained from all subjects.

Shanghai Jiaotong University School of Medicin,Renji Hospital Ethics Committee Approval Letter NO.[2014]97K.

ii. Consent for publication

Not applicable

iii. Availability of data and materials

The datasets generated during and/or analysed during the current study are available in the [The efficacy of strain elastography and shear-wave elastography in the evaluation of neoadjuvant chemotherapy for patients with locally advanced breast cancer] repository, https://doi.org/10.6084/m9.figshare.19419782.v1

iiii. Competing interests

The authors declare that they have no competing interests. 

iiiii. Funding

The study was supported by the National Natural Science Foundation of China [Grant No. 81801697] and the Natural Science Foundation of Shanghai under Grant [number 14411968200].

iiiiii. Authors' contributions

LJ and HL mainly designed the research scheme. YW and WY conducted data collection and statistical analysis. CW collected ultrasound images，and was a major contributor in writing the manuscript. All authors have read and approved the final manuscript.

iiiiiii. Acknowledgements

Thanks to LW, YJ and XL for their help in the process of image collection.

References

1. Wei W, H Zeng, R Zheng et al. Cancer registration in China and its role in cancer prevention and control. Lancet Oncol 2020; 21: e342-e349.
2. Dekker TJ, A Charehbili, VT Smit et al. Disorganised stroma determined on pre-treatment breast cancer biopsies is associated with poor response to neoadjuvant chemotherapy: Results from the NEOZOTAC trial. Mol Oncol 2015; 9: 1120–1128.
3. van der Hage JA, CJ van de Velde, JP Julien et al. Preoperative chemotherapy in primary operable breast cancer: results from the European Organization for Research and Treatment of Cancer trial 10902. J Clin Oncol 2001; 19: 4224–4237.
4. Ricci P, E Maggini, E Mancuso et al. Clinical application of breast elastography: state of the art. Eur J Radiol 2014; 83: 429–437.
5. Cosgrove DO, WA Berg, CJ Doré et al. Shear wave elastography for breast masses is highly reproducible. Eur Radiol 2012; 22: 1023–1032.
6. Yang H, Xu Y, Zhao Y et al. The role of tissue elasticity in the differential diagnosis of benign and malignant breast lesions using shear wave elastography. BMC Cancer. 2020 Sep 29;20(1):930.
7. Sinha D, Sharma S, Kundaragi NG et al. Added value of strain elastography in the characterisation of breast lesions: A prospective study. Ultrasound. 2020 Aug;28(3):164–173.
8. Itoh A, E Ueno, E Tohno et al. Breast disease: clinical application of US elastography for diagnosis. Radiology 2006; 239: 341–350.
9. Youk JH, EJ Son, HM Gweon et al. Quantitative Lesion-to-Fat Elasticity Ratio Measured by Shear-Wave Elastography for Breast Mass: Which Area Should Be Selected as the Fat Reference? PLoS One 2015; 10: e0138074.
10. Ogston KN, ID Miller, S Payne et al. A new histological grading system to assess response of breast cancers to primary chemotherapy: prognostic significance and survival. Breast 2003; 12: 320–327.
11. Hayashi M, Y Yamamoto, M Ibusuki et al. Evaluation of tumor stiffness by elastography is predictive for pathologic complete response to neoadjuvant chemotherapy in patients with breast cancer. Ann Surg Oncol 2012; 19: 3042–3049.
12. Schrader J, TT Gordon-Walker, RL Aucott et al. Matrix stiffness modulates proliferation, chemotherapeutic response, and dormancy in hepatocellular carcinoma cells. Hepatology 2011; 53: 1192–1205.
13. Rousseau C, A Devillers, C Sagan et al. Monitoring of early response to neoadjuvant chemotherapy in stage II and III breast cancer by [18F]fluorodeoxyglucose positron emission tomography. J Clin Oncol 2006; 24: 5366–5372.
14. Falou O, A Sadeghi-Naini, S Prematilake et al. Evaluation of neoadjuvant chemotherapy response in women with locally advanced breast cancer using ultrasound elastography. Transl Oncol 2013; 6: 17–24.
15. Stachs A, S Hartmann, J Stubert et al. Differentiating between malignant and benign breast masses: factors limiting sonoelastographic strain ratio. Ultraschall Med 2013; 34: 131–136.
16. Zhou J, C Zhou, W Zhan et al. Elastography ultrasound for breast lesions: fat-to-lesion strain ratio vs gland-to-lesion strain ratio. Eur Radiol 2014; 24: 3171–3177.
17. Barr RG. Breast Elastography: How to Perform and Integrate Into a "Best-Practice" Patient Treatment Algorithm. J Ultrasound Med. 2020 Jan;39(1):7–17.
18. Kokubu Y, Yamada K, Tanabe M et al. Evaluating the usefulness of breast strain elastography for intraductal lesions. J Med Ultrason (2001). 2021 Jan;48(1):63–70.
19. Jia W, Luo T, Dong Y et al. Breast Elasticity Imaging Techniques: Comparison of Strain Elastography and Shear-Wave Elastography in the Same Population. Ultrasound Med Biol. 2021 Jan;47(1):104–113.
20. Singla V, Prakash A, Prabhakar N et al. Does Shear Wave Elastography Score Over Strain Elastography in Breast Masses or Vice Versa? Curr Probl Diagn Radiol. 2020 Mar-Apr;49(2):96–101.