Conventional US combined with acoustic radiation force impulse (ARFI) elastography for prediction of triple-negative breast cancer and the risk of lymphatic metastasis

Abstract

PURPOSE: To evaluate the combination of conventional ultrasound (US) and acoustic radiation force impulse (ARFI) elastography in predicting triple-negative breast cancer and the likelihood of lymphatic metastasis.

MATERIALS AND METHODS: A total of 178 women presenting from May 2013 to September 2015 with pathologically proven triple-negative (n = 60) or hormone receptor (HR)-positive/human epidermal growth factor receptor 2 (HER2)-negative breast cancer (n = 118) were reviewed. Conventional US and ARFI imaging including Virtual touch tissue image (VTI) and Virtual Touch tissue Quantification (VTQ), were performed for each breast nodule. Parameters collected included patient age, lymph node involvement, histological grade, tumor subtype, appearance on conventional US, shear wave velocity (SWV) value, VTI score and the boundary on VTI. The ARFI findings were compared with the pathological findings.

RESULTS: Younger age (P < 0.001), higher histological grade (P < 0.001), lymphatic metastasis (P < 0.001), more nodes involved (P = 0.001), larger tumor size (P = 0.008), regular shape (P = 0.032), high VTI score (P = 0.006), unclear VTI boundary (P = 0.033), SWV ≥3.51 m/s (P = 0.015) were significantly associated with triple-negative breast cancer. High VTI score (P = 0.004) and a high SWV ≥4.15 m/s (P = 0.002) were significantly associated with lymphatic metastasis.

CONCLUSION: Conventional US combined with ARFI may be used to predict triple-negative breast cancer. Those cancers with higher SWV and the high VTI score have a higher likelihood of lymphatic metastasis.

Keywords

Conventional US acoustic radiation force impulse (ARFI)triple-negative breast cancer lymphatic metastasis

Abbreviations

ARFI

acoustic radiation force impulse

HER2

human epidermal growth factor receptor 2

estrogen receptor

progesterone receptors

hormone receptor

BIRADS

Breast Imaging Reporting and Data System

ROI

region of interest

SWV

shear wave velocity

ultrasound

VTI

virtual touch tissue imaging

VTQ

virtual touch tissue quantification

NHG

Nottingham Histology Grade

ROC

the receiver operator curve

1 Introduction

Triple-negative breast cancer refers to the subtype of breast cancer that lacks the human epidermal growth factor receptor 2 [HER2], estrogen receptor(ER) negative and progesterone receptors (PR) negative [16]. Although triple-negative breast cancers account for only 15% – 20% of all breast cancers [27], they are currently receiving a large amount of research attention, which is appropriate given their disappointing prognosis, unique biology, aggressive and early metastasis, and poor chemotherapy response when compared with endocrine-sensitive and HER2-positive breast cancers. It has been reported that triple-negative breast cancer occurs more frequently in younger women with a family history of breast cancer [8].

Breast Ultrasonography (US) is the preferred imaging method for the identification of breast lesions. US features that can be used to predict biological behavior include tumor size, shape, margin, boundary, internal echo, the ratio between tall and wide and the internal blood flew signals. It is widely used to screen for breast lesions because of its simple approach, non-invasiveness and avoidance of radiation exposure. However, conventional US suffers from low specifity [11] and final results depend heavily on the operator’s experience [31].

Over the past 10 years, US elastography has been introduced to clinical practice for the diagnosis of breast cancer and many studies have investigated the use of elastrography. Some studies [12, 18] have shown that US elastography is highly reproducible for assessing the elastographic features of breast masses within and across observers compared with conventional US. It offers a new view to assess different aspects of the mass, because lesions that have an abnormal internal structure often show different elasticity characteristics.

Acoustic radiation force impulse (ARFI) elastography is one of the newly developed elastrography techniques [9 , 24]. It is based on the principle that localized tissue displacement is less in harder tissues than in softer ones, and the tissue stiffness can be evaluated both qualitatively and quantitatively. The technique includes Virtual Touch tissue imaging (VTI; Siemens Medical Solutions, Mountain View, California, USA) and Virtual Touch tissue quantification (VTQ; Siemens Medical Solutions) [9]. VTI can evaluate tissue stiffness qualitatively, repeating the displacement process along multiple imaging lines [20]. VTQ on the other hand can evaluate tissue stiffness quantitatively and is performed by measuring the time peak displacement in a transverse direction at each lateral location and the corresponding shear-wave velocity (SWV) within the tissue [21]. Usually, stiffer tissues show a greater shear velocity [34]. The technique can overcome many limitations of previous techniques, such as high operator dependence and poor reproducibilit y [6]. It is an important complement to conventional US with studies having suggested that conventional US combined with shear wave elastography shows greater potential in the differentiation of malignant from benign breast lesions.

Triple-negative breast cancer deserves increasing attention giving to its unique biologic behaviors and the poor prognosis to women, then it would be meaningful if ARFI elastography combined with conventional US have some value in diagnosing triple-negative breast cancer. Our retrospective study aimed at evaluating the value of conventional US combined with ARFI in predicting the triple-negative breast cancer. We assumed that differences in the expression of immunological receptors might affect the internal structure of the tumor, and that would possibly be reflected in the US image.

2 Materials and methods

2.1 Patient selection

This research involved retrospectively reviewing the conventional US images and corresponding ARFI images of breast cancer patients. The study was approved by the institutional review board of our hospital and the requirement to obtain informed consent was waived, also the manuscript is in accordance with the ethical guidelines of Clinical Hemorheology and Microcirculation [3].

The study enrolled 236 patients with pathologically confirmed breast cancers who presented to our hospital between May 2014 and September 2015. There are 134 women who had more than one lesion, in which case we selected the largest lesion to be included in our research. The inclusion criteria were as follows: (1) all lesions were at least 5 mm in the maximal diameter; (2) patients had undergone conventional US and ARFI examinations, and imaging data were complete; (3) all breast cancers were pathologically confirmed and FISH analyses had been performed. The exclusion criteria were as follows: (1) patients who had received prior neoadjuvant chemotherapy (n = 5); (2) maximum diameter of the tumor <5 mm (n = 15); (3) obvious cystic lesions (n = 10); (4) less than five VTQ measurement results on the same lesion (n = 7); (5) incomplete or missing imaging information (n = 21).

Therefore, 58 lesions were excluded from our study, we ultimately analyzed 178 breast lesions, including 60 triple-negative breast cancers and 118 hormone receptor (HR)-positive/HER-2 negative breast cancers (Fig. 1). Of the 178 patients, the mean age for the women was 57.57±11.69 years (range, 26–88 years), and the diameter of the lesions ranged from 5 to 76 mm (mean, 22.32 mm ± 11.52).

Fig.1

Flowchart of the selection of the breast lesions.

2.2 Conventional ultrasound and ARFI examination

The conventional US and ARFI were performed on every patient using the same S2000 US scanner (Siemens Medical Solutions) equipped with a linear 9L4 transducer (frequency range, 4–9 MHz). All the examinations were performed by a experienced radiologist in breast US and breast ARFI. Patients were examined in the supine position with their breast and armpit fully exposed. Conventional US was performed firstly including B mode US and color Doppler US and with the breast being scanned in a clockwise direction, then the supraclavicular lymph nodes and lastly the axillary lymph nodes.

Afterwards, ARFI elastography were applied. we chose the biggest and clearest section of the pathologically proven invasive cancer, keeping the probe still without compression, and switched to the elastography mode. The VTI button was pressed and a shear wave was excited with VTI images being generated [9, 16], which were then saved. Once the ARFI image was stable, the region of interest (ROI) was selected in the center of the lesion but avoiding areas of calcification (except for diffuse calcification in the lesion) and any cystic spaces. The SWV of the lesions were measured seven times without movement or compression. The VTQ images were then saved and the SWV value was objectively measured and recorded. An area of normal breast tissue with approximately the same depth of glandular layer was measured using the same principle. For each patient, the system automatically adjusted the depth, focal zone placement, and overall gain.

2.3 Image interpretation

The images obtained were reviewed independently in the same settings by two experienced radiologists with consensus.

We recorded the conventional US image characteristics with reference to the ACR-BI-RADS [19] lexicon and the lymph node status. For each lesion, we specified the size (5 mm ≤ diameter <10 mm, 10 mm ≤ diameter <20 mm, or 20 mm ≤ diameter), shape (regular or irregular), margin (smooth or rough), boundary (clear or unclear), internal echo (even or uneven), calcification (microcalcification [<1 mm in diameter], macrocalcification [≥ 1 mm in diameter with shadowing], or no calcification) and Doppler vascularity (detected or undetected).

On the VTI image we used the color gradient from bright (soft) to dark (hard) to represent the grayscale value that provided information on stiffness, so that the lesions could be visually confirmed. According to the grayscale value shown in the lesion and the surrounding normal tissue, VTI images were divided into five grades (1, 2, 3, 4a, or 4b), referencing the work of Mitsuhiro Tozaki et al. [25]. We then recorded the VTI score and the boundary definition of the mass (poorly defined or well-defined) on the VTI image, with higher grades normally indicating stiffer tissue.

In order to minimize the chance for error and bias in the VTQ results, we removed the highest and lowest measurements and the remaining measurements were averaged as the final SWV - results expressed in meters/second. The results for lesions that failed to produce an SWV, shown as “X.XX m/sec”, were replaced by 0 m/sec or 8.4 m/sec, with 0 m/sec corresponding to the cystic portion and 8.4 m/sec corresponding to the solid portion [32].

2.4 Clinicopathological data

The medical records of the patients were reviewed and data on age and distribution of lesions were compiled. Pathological reports from excisional biopsies, breast-conserving surgery or mastectomy specimens were also reviewed to determine the tumor type of the invasive tumor, its histological grade, whether the lymph node has transferred and the number of the transferred lymph node status.

The breast cancer histological grade was recorded as grade I, II or III as per the Nottingham Histology Grade (NHG) [13]. The histological diagnoses were made by one pathologist with 10 years of working experience in pathologic diagnosis. We used the routine paraffin sections as the reference standard and obtained immunohistochemical staining results by FISH analyses. The receptors we recorded included ER, PR, and HER2. The status of each receptor was considered to be negative if the expression was <10% and positive if the expression was ≥10%. The results for HER2 expression on immunohistochemical analysis were scored as negative, 1+, 2+ or 3+, according to the manufacturer’s recommendations. Specimens with a score of negative or 1+, were treated as HER2 negative [1]. We collected all the triple-negative tumors, and selected the HR-positive/HER2-negative breast cancers for comparison.

2.5 Statistical analysis

SPSS 17.0 (SPSS, version 17.0; SPSS, Chicago, Illinois, USA) was used for statistical analysis with P < 0.05 considered to indicate a significant difference. To compare the imaging findings between triple-negative breast cancer and ER-positive/PR-positive/HER2-negative breast cancer, we used the chi-squared and Fisher exact tests. The correlation of ARFI findings in patients with triple-negative breast cancer and lymphatic metastasis was analyzed with multivariate analysis. We determined the best cut-off value of the SWV for the prediction of the triple-negative histological subtype and for the risk of lymphatic metastasis in the patients with triple-negative breast cancer by analysis of the receiver operator curve (ROC). The Student t test was used to inspect that there was no tumor size-related bias or age-related bias between the two groups.

3 Results

In total 178 patients, all of whom were women, with a mean age 57.6±11.7 years (range 26–88) were involved in the study. The largest lesion only was studied in women who had multiple lesions resulting in 178 breast lesions being analyzed. All of lesions were invasive ductal breast carcinomas, among which there were 60 triple-negative breast cancers and 118 HR-positive/HER2-negative breast cancers.

3.1 Patient characteristics

The basic characteristics of the patients with breast cancer lesions are shown in Table 1. These indicate that women of a younger age (45.57±11.1 VS 56.42±11.8, P < 0.001) may have a higher risk of suffering from the triple-negative breast cancer. Triple-negative breast cancers were more likely to have a high histological tumor grade (50% [30 of 60] VS 22% [26 of 118], P < 0.001). The triple-negative breast cancer patients were more likely to have early lymph node metastasis (43% [26 of 60] VS 25% [29 of 118], P < 0.001) and to have more lymph nodes involved (4.46±0.783 VS 2.37±0.298, P = 0.001). In contrast, the distribution of lesions did not present a significantly association with triple-negative breast cancer compared with the non-triple-negative breast cancer (P = 0.216).

Table 1
Basic characteristics of patients and their breast cancer lesions

Triple-Negative Breast ER-positive/PR-positive/HER2-negative

Basic Characteristic of Lesions Cancer (n = 60) Breast Cancer (n = 118) P value

Age(y)^# # 45.57±11.1 56.42±11.8 <0.001^*

Distribution of lesion 0.216

No. of patients with single lesion 44 (73%) 84 (71%)

No. patients with multiple lesions 16 (27%) 34 (29%)

Histologic tumor grade <0.001^*

Grade 1 1 (2%) 6 (5%)

Grade 2 29 (48%) 86 (73%)

Grade 3 30 (50%) 26 (22%)

Lymph node metastasis <0.001^*

Yes 26 (43%) 29 (25%)

No 34 (57%) 89 (75%)

Involvement numbers^# # 4.46±0.783 2.37±0.298 0.001^*

	Triple-Negative Breast	ER-positive/PR-positive/HER2-negative
Age(y)^# #	45.57±11.1	56.42±11.8	<0.001^*
Distribution of lesion			0.216
No. of patients with single lesion	44 (73%)	84 (71%)
No. patients with multiple lesions	16 (27%)	34 (29%)
Histologic tumor grade			<0.001^*
Grade 1	1 (2%)	6 (5%)
Grade 2	29 (48%)	86 (73%)
Grade 3	30 (50%)	26 (22%)
Lymph node metastasis			<0.001^*
Yes	26 (43%)	29 (25%)
No	34 (57%)	89 (75%)
Involvement numbers^# #	4.46±0.783	2.37±0.298	0.001^*

ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2. Numbers in parentheses are percentages. ^*Indicate a significant difference. ^# #Data are means±standard deviations.

3.2 Conventional US findings

The comparison of the conventional US findings for triple-negative and the ER-positive/PR-positive/HER2-negative breast cancer are listed in Table 2. On conventional US, significant differences were found in the tumor size (P = 0.008) and shape (P = 0.032), with tumors of larger size (67% [40 of 60] VS 44% [52 of 118]) and regular shape (58% [35 of 60] VS 16% [19 of 118]) being significantly associated with triple-negative breast cancer. In contrast, tumor boundary (P = 0.09), margins (P = 0.09), the ratio between depth and width (P = 0.170), internal echo (P = 0.069), calcification (P = 0.118) and blood flew signals (P = 0.382) showed no significant difference between the two subtypes.

Table 2
The comparison of the lesion type on the conventional US between the triple-negative breast cancer and the ER-positive/PR-positive/HER2-negative breast cancer

Triple-Negative Breast Cancer ER-positive/PR-positive/HER2-negative

Lesion type on US imaging (n = 60) Breast Cancer(n = 118) P value

Tumor size 0.008^*

5 mm ≤ d <10 mm 3 (5%) 12 (10%)

10 mm ≤ d <20 mm 17 (28%) 54 (46%)

20 mm ≤ d 40 (67%) 52 (44%)

Shape 0.032^*

Irregular 25 (42%) 99 (84%)

Regular 35 (58%) 19 (16%)

Boundary 0.090

Poorly defined 45 (75%) 92 (78%)

Well defined 15 (25%) 26 (22%)

Margin 0.090

Smooth 8 (13%) 18 (15%)

Rough 52 (87%) 100 (85%)

Ratio between tall and wide 0.170

≥1 22 (36%) 51 (43%)

<1 38 (64%) 67 (57%)

Internal echo 0.069

Evenly 5 (8%) 13 (11%)

Unevenly 55 (92%) 105 (89%)

Calcification 0.118

No Calcification 21 (35%) 36 (31%)

Microcalcification(d≥1 mm) 19 (32%) 55 (47%)

Macrocalcification(<1 mm) 20 (33%) 27 (23%)

Blood flew signals 0.382

Detected 41 (68%) 86 (73%)

Undetected 19 (32%) 32 (27%)

	Triple-Negative Breast Cancer	ER-positive/PR-positive/HER2-negative
Tumor size			0.008^*
5 mm ≤ d <10 mm	3 (5%)	12 (10%)
10 mm ≤ d <20 mm	17 (28%)	54 (46%)
20 mm ≤ d	40 (67%)	52 (44%)
Shape			0.032^*
Irregular	25 (42%)	99 (84%)
Regular	35 (58%)	19 (16%)
Boundary			0.090
Poorly defined	45 (75%)	92 (78%)
Well defined	15 (25%)	26 (22%)
Margin			0.090
Smooth	8 (13%)	18 (15%)
Rough	52 (87%)	100 (85%)
Ratio between tall and wide			0.170
≥1	22 (36%)	51 (43%)
<1	38 (64%)	67 (57%)
Internal echo			0.069
Evenly	5 (8%)	13 (11%)
Unevenly	55 (92%)	105 (89%)
Calcification			0.118
No Calcification	21 (35%)	36 (31%)
Microcalcification(d≥1 mm)	19 (32%)	55 (47%)
Macrocalcification(<1 mm)	20 (33%)	27 (23%)
Blood flew signals			0.382
Detected	41 (68%)	86 (73%)
Undetected	19 (32%)	32 (27%)

ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; d, the maximal diameter of the lesion. Numbers in parentheses are percentages. ^*Indicate a significant difference

3.3 ARFI findings

The comparison of the ARFI findings between the triple-negative and ER-positive/PR-positive/HER2-negative breast cancer is shown in Table 3 and Fig. 4. It shows that the higher VTI score (27% [16 of 60] VS 14% [17 of 118], P = 0.006), a poorly defined VTI boundary (73% [44 of 60] VS 57% [67 of 118], P = 0.033) and SWV ≥3.51 m/s (63% [38 of 60] VS 43% [51 of 118], P < 0.01) are more common in triple-negative breast cancers (Figs. 2 and 3). The best cut-off value of SWV for the prediction of the triple-negative breast cancer was found to be 3.51 m/s by analysis of the ROC curve.

Table 3
The comparison of the lesion type on ARFI between the triple-negative breast cancer and the ER-positive/PR-positive/HER2-negative breast cancer

Triple-Negative ER-positive/PR-positive/

Lesions Characteristic Breast Cancer HER2-negative Breast P value

on ARFI (n = 60) Cancer(n = 118)

VTI grade 0.006^*

Grade 2 2 (3%) 1 (1%)

Grade 3 2 (3%) 30 (25%)

Grade 4 4a 40 (67%) 70 (59%)

4b 16 (27%) 17 (14%)

VTI boundary 0.033^*

Poorly defined 44 (73%) 67 (57%)

Well defined 16 (27%) 51 (43%)

SWV(m/s) <0.01^*

≥3.51 38 (63%) 51 (43%)

<3.51 22 (37%) 67 (57%)

	Triple-Negative	ER-positive/PR-positive/
VTI grade				0.006^*
Grade 2		2 (3%)	1 (1%)
Grade 3		2 (3%)	30 (25%)
Grade 4	4a	40 (67%)	70 (59%)
	4b	16 (27%)	17 (14%)
VTI boundary				0.033^*
Poorly defined		44 (73%)	67 (57%)
Well defined		16 (27%)	51 (43%)
SWV(m/s)				<0.01^*
≥3.51		38 (63%)	51 (43%)
<3.51		22 (37%)	67 (57%)

ARFI, acoustic radiation force impulse; ER, estrogen receptor; PR, progesterone receptor; HER2, human epidermal growth factor receptor 2; VTI,Virtual Touch tissue imaging; SWV, shear-wave velocity. Numbers in parentheses are percentages. ^*Indicate a significant difference.

Fig.2

Images in a 45-year-old woman with triple-negative breast cancer. (a) At conventional US, a 21-mm lesion in the left breast (arrows in a) appears to have relatively regular shape, well defined boundary, smooth margin, no calcification. At elastography, (b) VTI grade of 4b (arrows), (c) SWV of “X.XX” m/sec (ie, 8.4 m/sec) at VTQ is assigned. (d) Histologic specimen (hematoxylin-eosin stain; original magnification, 400) shows that this lesion is finally confirmed to be a poorly differentiated breast invasive ducal carcinoma.

Fig.3

Images in a 57-year-old woman with ER-positive/PR-positive/HER2-negative breast cancer. (a) At conventional US, a 12-mm lesion in the right breast (arrows in a) appears to have relatively irregular shape, poorly defined boundary, rough margin, micro-calcification, poorly defined margin. At elastography, (b) VTI grade of 2 (arrows), (c) SWV of 2.49 m/sec (ie, 8.4 m/sec) at VTQ is assigned. (d) Histologic specimen (hematoxylin-eosin stain; original magnification, 400) shows that this lesion is finally confirmed to be a well differentiated breast invasive ducal carcinoma.

Fig.4

ARFI findings of the mass lesion corresponded to Table 3. The column diagram shows that (A) the higher VTI score, (B) a poorly defined VTI boundary, and (C) SWV ≥3.51 m/s are more common in triple-negative breast cancers.

3.4 Correlation of ARFI findings in patients with triple-negative breast cancer and lymphatic metastasis

Correlation of the ARFI findings in patients with triple-negative breast cancer and lymphatic metastasis is shown in Table 4 and Fig. 5. When the triple-negative breast cancers only were analyzed, significant differences were found for the highest VTI score (46% [12 of 26] VS 12% [4 of 34], P = 0.004) and an SWV ≥4.15 m/s (69% [18 of 26] VS 32% [11 of 34], P = 0.002), between patients with and without lymphatic metastasis, while the boundary on VTI was not found to be of significance (P = 0.227). The best cut-off value of SWV for the presence of lymphatic metastasis in patients with triple-negative breast cancer was found to be 4.15 m/s by analysis of the ROC curve.

Table 4
Correlation of ARFI findings in patients with triple-negative breast cancer with the lymphatic metastasis

Lymphatic metastasis

ARFI Findings Yes (26) No (34) P value

VTI score 0.004^*

Score 2 0 (0) 2 (6%)

Score3 0 (0) 2 (6%)

Score4 4a 14 (54%) 26 (76%)

4b 12 (46%) 4 (12%)

Boundary on VTI 0.227

Clear 9 (35%) 7 (21%)

Uclear 17 (65%) 27 (79%)

SWV(m/s) 0.002^*

≥4.15 18 (69%) 11 (32%)

<4.15 8 (31%) 23 (68%)

		Lymphatic metastasis
VTI score				0.004^*
Score 2		0 (0)	2 (6%)
Score3		0 (0)	2 (6%)
Score4	4a	14 (54%)	26 (76%)
	4b	12 (46%)	4 (12%)
Boundary on VTI				0.227
Clear		9 (35%)	7 (21%)
Uclear		17 (65%)	27 (79%)
SWV(m/s)				0.002^*
≥4.15		18 (69%)	11 (32%)
<4.15		8 (31%)	23 (68%)

Numbers in parentheses are percentages. ^*Indicate a significant difference.

Fig.5

Correlation of ARFI findings in patients with triple-negative breast cancer and lymphatic metastasis corresponded to Table 4. The column diagram shows that patients with (A) the highest VTI score (i.e.grade 4b) and (C)SWV ≥ 4.15 m/s tend to have higher risk of lymphatic metastasis; but (B) the boundary on VTI was not found to be of significance.

4 Discussion

Although accounting for only a relatively small proportion of breast cancers, triple-negative breast cancers are aggressive in nature and responsible for a large proportion of breast cancer deaths. From our study findings, it appears that younger patient, single nodule, higher histological tumor grade, earlier lymph node metastasis and more numbers of lymph nodes involved are significantly associated with triple-negative breast cancer. Previous studies [15, 28] have demonstrated that positive lymph node status, lymphovascular invasion and increasing stage (adjuvant vs neoadjuvant) were risk factors for triple-negative breast cancers. Bae et al. [5, 8] discovered that triple-negative invasive tumors affected younger women with a family history of breast cancer and that triple-negative invasive tumors were generally larger in size, of higher histological grade and had a worse prognosis than other breast cancers, owing to their aggressive nature and the absence of effective targeted therapy, which are consistent with ours.

Conventional US is an important method for the detection and characterization of breast masses in the routine examination of breast disease and is established as the first-line guidance process for percutaneous biopsy. Therefore accurate US diagnosis of breast lesions is important for the patients. The researches of Choi et al. [10] presented that triple-negative breast cancer has its unique biological behavior, many triple-negative breast cancers may simulate the benign lesions compared to the ER-positive/PR-positive/HER2-negative breast cancer and might be misinterpreted as the benign. Several additional studies [26, 29] had shown that triple-negative tumors often manifest as circumscribed, benign-looking lesions with an homogeneous internal echo, although the exact reason for this association is not yet fully understood. Our study finding indicates that among all the characteristics the increasing tumor size and the regular shape on conventional US are significantly associated with the triple-negative breast cancers. We also found the internal echo may have corresponding tendencies, but it did not reach a level of statistical significance, which were partly consistent with the previous study. Anders et al. [2, 22] have described factors such as the nuclear grade, tumor circumscription, inflammatory cell reaction, intraductal growth pattern and tubule formation, and have concluded that HR negativity implies an increased proliferation rate, which may explain the larger size of triple-negative breast cancers.

With the development of US techniques, as a noninvasive technology, ARFI allows quantitative and the semiquantitative measurement of the stiffness of lesions in the form of SWV values [11, 20]. Different disease subgroups may have different levels of stiffness, which can express different elasticity patterns. In our study, it has been shown that the highest VTI score, a poorly defined VTI boundary and an SWV ≥3.51 m/s are statistically significant in the differentiation of triple-negative breast cancer from ER-positive/PR-positive/HER2-negative breast cancer. Similar to our results, Kojima et al. [17] reported that 88% of triple-negative breast cancers appeared as hard masses, with elasticity scores of 4 or 5. Boisserie-Lacroix et al. [7] reported quantitative SWV results of 17 triple-negative cases that showed relatively high mean elasticity values ranging from 50 to 232 kPa in the perilesional area. Wojcinski et al. [30] considered that the lack of expression of ER, PR and HER2 results in a genetic mutation, which may tend to exert stronger desmoplastic effects on peritumoral tissues. The stiffness of the peritumoral tissues were affected, meaning that the VTI boundary appeared poorly defined. The numbers of mitoses increased, reflecting enhanced cellularity and an excessive desmoplastic reaction, therefore architectural distortions occurred and a higher mean stiffness showed on a macroscopic level.

Triple-negative breast cancer accompanied by lymphatic metastasis may indicate the poorer prognosis [4]. When we came to analyze the ARFI findings in patients with triple-negative breast cancer and lymphatic metastasis, we found that the VTI score and the SWV value were also related to metastasis in triple-negative tumors, although the boundary on VTI was not significant. It indicated that in triple-negative tumors with a higher VTI score and larger SWV value may be an indicator of a higher risk of lymphatic metastasis. We have learned from previous studies that increased stiffness is associated with nodal metastases [14, 33], but few large-scale investigations have looked at the which ARFI findings correlate with lymphatic metastasis in patients with triple-negative breast cancer. As shown in many studies, triple-negative breast cancers are highly vascular and show lymphatic invasion [2 , 5], which may contribute to earlier and more lymphatic metastasis; Researches of Sarvazyan, Garra et al. also have made the assumption that cancer cells in the fibrous stroma infiltrate the lesion caused by nearby construction activity, thereby reducing adhesion between the cells and leading more readily to lymphatic metastasis [23]. however, these assumptions still need further validation.

There are also some limitations in our study. First, in our study, a result of X.XX was recorded as 8.4 m/sec in a solid portion and 0 m/sec in a cystic lesion, as in the previous study from Xu et al. [9 , 32]. Handling the data in this manner can partly avoid selection bias, but may ultimately lead to a higher mean SWV value, meaning that an information bias appears. Second, we did not assess interobserver or intraobserver variability with regard to performing elastography and interpreting the data. Subjectivity may exist in the observers, particularly in the evaluation of the VTI and this will need to be evaluated in further studies. Finally, the performance of ARFI also has its limitations, as the findings can depend on many factors, including breast size, breast density, lesion depth, and the proximity of a lesion to the nipple, making it difficult to achieve a consistent optimal image quality in all cases. Furthermore, during performance of this test in our study, we noted that lesions in the different sections may show different elasticity characteristics and further research needs to be done to explain this phenomenon.

5 Conclusion

Conventional US combined with acoustic radiation force impulse (ARFI) elastography may be able to predict triple-negative breast cancer. The triple-negative breast cancer has a higher SWV value and a higher VTI score, which may indicate a high probability of lymph node metastasis and may indicate a poorer prognosis for the patient. Thus, the tecnique can provide much useful information for the clinical treatment. It would be meaningful for the patients with triple-negative breast cancer to improve the long-term survival rate by earlier clinical intervention.

Footnotes

Acknowledgments

This work was supported by National Natural Science Foundation of China (81471673) and the Research Project of Science and Technology Committee of Shanghai Municipality (124119a3201).

References

Allred

D.C.

, Harvey

J.M.

, Berardo

and Clark

G.M.

, Prognostic and predictive factors in breast cancer by immunohistochemical analysis, Mod Pathol 11(2) (1998), 155–168.

Anders

C.K.

and Carey

L.A.

, Biology, metastatic patterns, and treatment of patients with triple-negative breast cancer, Clin Breast Cancer 9(Suppl 2) (2009), S73–S81.

Anonymous, Ethical guidelines for publication in Clinical Hemorheology and Microcirculation: Update 2016, Clin Hemorheol Microcirc 63(1) (2016), 1–2.

Avery-Kiejda

K.A.

, Braye

S.G.

, Mathe

, Forbes

J.F.

and Scott

R.J.

, Decreased expression of key tumour suppressor microRNAs is associated with lymph node metastases in triple negative breast cancer, BMC Cancer 14 (2014), 51. Published online 2014 January 31.

Bae

M.S.

, Moon

H.G.

, Han

, Noh

D.Y.

, Ryu

H.S.

, Park

I.A.

, Chang

J.M.

, Cho

and Moon

W.K.

, Early stage Triple-negative Breast cancer: Imaging and Clinical-Pathologic Factors Associated with Recurrence1, Radiology 278(2) (2016), 356–364.

Berg

W.A.

, Cosgrove

D.O.

, Doré

C.J.

et al., Shear-wave elastography improves the specificity of breast US: The BE1 multinational study of 939 masses, Radiology 262(2) (2012), 435–449.

Boisserie-Lacroix

, Mac Grogan

, Debled

, Ferron

, Asad-Syed

, Brouste

, Mathoulin-Pelissier

and Hurtevent-Labrot

, Radiological features of triple-negative breast cancers (73 cases), Diagn Interv Imaging 93(3) (2012), 183–190.

Carey

, Winer

, Viale

, Cameron

and Gianni

, Triple-negative breast cancer: Disease entity or title of convenience? Nat Rev Clin Oncol 7(12) (2010), 683–692.

Chen

B.D.

, Xu

H.X.

, Zhang

Y.F.

, Liu

B.J.

, Guo

L.H.

, Li

D.D.

, Zhao

C.K.

, Li

X.L.

, Wang

and Zhao

S.S.

, The diagnostic performances of strain elastography (SE), acoustic radiation force impulse (ARFI) and point shear-wave speed (pSWS) for non-calcified thyroid nodules, Clin Hemorheol Microcirc (2016).

10.

Choi

Y.J.

, Seong

M.H.

, Choi

S.H.

, Kook

S.H.

, Kwag

H.J.

, Park

Y.L.

and Park

C.H.

, Ultrasound and clinicopathological characteristics of triple receptor-negative breast cancers, J Breast Cancer 14(2) (2011), 119–123.

11.

Corsetti

, Ferrari

, Ghirardi

et al., Role of ultrasonography in detecting mammographically occult breast carcinoma in women with dense breasts, Radiol Med (Torino) 111(3) (2006), 440–448.

12.

Cosgrove

D.O.

, Berg

W.A.

, Doré

C.J.

et al., Shear wave elastography for breast masses is highly reproducible, Eur Radiol 22(5) (2012), 1023–1032.

13.

Elston

C.W.

and Ellis

I.O.

, Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: Experience from a large study with long-term follow-up, Histopathology 41(3A) (2002), 154–161.

14.

Evans

, Whelehan

, Thomson

et al., Invasive breast cancer: Relationship between shear-wave elastographic findings and histologic prognostic factors, Radiology 263(3) (2012), 673–677.

15.

Foulkes

W.D.

, Smith

I.E.

and Reis-Filho

J.S.

, Triple-negative breast cancer, N Engl J Med 363(20) (2010), 1938–1948.

16.

Grimm

L.J.

, Johnson

K.S.

, Marcom

P.K.

, Baker

J.A.

and Soo

M.S.

, Can breast cancer molecular subtype help to select patients for preoperative MR imaging? Radiology 274(2) (2015), 352–358.

17.

Kojima

and Tsunoda

, Mammography and ultrasound features of triple-negative breast cancer, Breast Cancer 18(3) (2011), 146–151.

18.

Lee

S.H.

, Chang

J.M.

, Kim

W.H.

, Bae

M.S.

et al., added Value of shear-Wave elastography for evaluation of Breast Masses Detected with screening Us imaging, Radiology 273(1) (2014), 61–69.

19.

Mendelson

E.B.

, Böhm-Vélez

, Berg

W.A.

et al., ACR BIRADS^® ultrasound, In: ACR BI-RADS^® Atlas, Breast Imaging Reorting and Data System. American College of Radiology 2013, Reston (2013), 121–132.

20.

Meng

, Zhang

, Wu

et al., Preliminary results of acoustic radiation force impulse (ARFI) ultrasound imaging of breast lesions, Ultrasound Med Biol 37(9) (2011), 1436–1443.

21.

Nightingale

, Soo

M.S.

, Nightingale

and Trahey

, Acoustic radiation force impulse imaging:} demonstration of clinical feasibility, Ultrasound Med Biol 28(2) (2002), 227–235.

22.

Perou

C.M.

, Sorlie

, Eisen

M.B.

, van de Rijn

, et al., Molecular portraits of human breast tumours, Nature 406(6797) (2000), 747–752.

23.

Sarvazyan

, Hall

T.J.

, Urban

M.W.

, Fatemi

, Aglyamov

S.R.

and Garra

B.S.

, An overview of elastography – an emerging branch of medical imaging, Curr Med Imaging Rev 7(4) (2011), 255–282.

24.

Soliman

A.A.

, Wojcinski

and Degenhardt

, Ultrasonographic examination of the endometrium and myometrium using acoustic radiation force impulse (ARFI) imaging technology: An initial experience with a new method, Clin Hemorheol Microcirc 59(3) (2015), 235–243.

25.

Tozaki

, Isobe

and Sakamoto

, Combination of elastography and tissue quantification using the acoustic radiation force impulse (ARFI) technology for differential diagnosis of breast masses, Jpn J Radiol 30(8) (2012), 659–670.

26.

Uematsu

, Kasami

and Yuen

, Triple-negative breast cancer: Correlation between MR imaging and pathologic findings, Radiology 250(3) (2009), 638–647.

27.

Weigelt

, Horlings

H.M.

, Kreike

, Hayes

M.M.

et al., Refinement of breast cancer classification by molecular characterization of histological special types, J Pathol 216(2) (2008), 141–150.

28.

Whitman

G.J.

, Albarracin

C.T.

and Gonzalez-Angulo

A.M.

, Triple-negative breast cancer: What the radiologist needs to know, Semin Roentgenol 46(1) (2011), 26–39.

29.

Wojcinski

, Soliman

A.A.

, Schmidt

, Makowski

, Degenhardt

and Hillemanns

, Sonographic features of triple-negative and non-triple-negative breast cancer, J Ultrasound Med 31(10) (2012), 1531–1541.

30.

Wojcinski

, Stefanidou

, Hillemanns

and Degenhardt

, The biology of malignant breast tumors has an impact on the presentation in ultrasound: An analysis of 315 cases, BMC Women’s Health 13 (2013), 47. Published online 2013 November 19.

31.

J.M.

, Xu

X.H.

, Xu

H.X.

et al., Conventional US, Us elasticity imaging, and acoustic radiation Force impulse imaging for Prediction of Malignancy in Thyroid nodules, Radiology 272(2) (2014), 557–586.

32.

Yao

, Wu

, Zou

, Xu

, Xie

, Wu

and Xu

, Diagnostic Value of Virtual Touch Tissue Quantification for Breast Lesions with Different Size, BioMed Research International 2014 2014 (2014), 1–7.

33.

Youk

J.H.

, Gweon

H.M.

, Son

E.J.

, Kim

J.A.

and Jeong

, Shearwave elastography of invasive breast cancer: Correlation between quantitative mean elasticity value and immunohistochemical profile, Breast Cancer Res Treat 138(1) (2013), 119–126.

34.

Zhai

, Palmeri

M.L.

, Bouchard

R.R.

et al., An integrated indenter-ARFI imaging system for tissue stiffness quantification, Ultrason Imaging 30(2) (2008), 95–111.

	Triple-Negative Breast	ER-positive/PR-positive/HER2-negative
Basic Characteristic of Lesions	Cancer (n = 60)	Breast Cancer (n = 118)	P value
Age(y)^# #	45.57±11.1	56.42±11.8	<0.001^*
Distribution of lesion			0.216
No. of patients with single lesion	44 (73%)	84 (71%)
No. patients with multiple lesions	16 (27%)	34 (29%)
Histologic tumor grade			<0.001^*
Grade 1	1 (2%)	6 (5%)
Grade 2	29 (48%)	86 (73%)
Grade 3	30 (50%)	26 (22%)
Lymph node metastasis			<0.001^*
Yes	26 (43%)	29 (25%)
No	34 (57%)	89 (75%)
Involvement numbers^# #	4.46±0.783	2.37±0.298	0.001^*

	Triple-Negative Breast Cancer	ER-positive/PR-positive/HER2-negative
Lesion type on US imaging	(n = 60)	Breast Cancer(n = 118)	P value
Tumor size			0.008^*
5 mm ≤ d <10 mm	3 (5%)	12 (10%)
10 mm ≤ d <20 mm	17 (28%)	54 (46%)
20 mm ≤ d	40 (67%)	52 (44%)
Shape			0.032^*
Irregular	25 (42%)	99 (84%)
Regular	35 (58%)	19 (16%)
Boundary			0.090
Poorly defined	45 (75%)	92 (78%)
Well defined	15 (25%)	26 (22%)
Margin			0.090
Smooth	8 (13%)	18 (15%)
Rough	52 (87%)	100 (85%)
Ratio between tall and wide			0.170
≥1	22 (36%)	51 (43%)
<1	38 (64%)	67 (57%)
Internal echo			0.069
Evenly	5 (8%)	13 (11%)
Unevenly	55 (92%)	105 (89%)
Calcification			0.118
No Calcification	21 (35%)	36 (31%)
Microcalcification(d≥1 mm)	19 (32%)	55 (47%)
Macrocalcification(<1 mm)	20 (33%)	27 (23%)
Blood flew signals			0.382
Detected	41 (68%)	86 (73%)
Undetected	19 (32%)	32 (27%)

	Triple-Negative	ER-positive/PR-positive/
Lesions Characteristic		Breast Cancer	HER2-negative Breast	P value
on ARFI		(n = 60)	Cancer(n = 118)
VTI grade				0.006^*
Grade 2		2 (3%)	1 (1%)
Grade 3		2 (3%)	30 (25%)
Grade 4	4a	40 (67%)	70 (59%)
	4b	16 (27%)	17 (14%)
VTI boundary				0.033^*
Poorly defined		44 (73%)	67 (57%)
Well defined		16 (27%)	51 (43%)
SWV(m/s)				<0.01^*
≥3.51		38 (63%)	51 (43%)
<3.51		22 (37%)	67 (57%)

		Lymphatic metastasis
ARFI Findings		Yes (26)	No (34)	P value
VTI score				0.004^*
Score 2		0 (0)	2 (6%)
Score3		0 (0)	2 (6%)
Score4	4a	14 (54%)	26 (76%)
	4b	12 (46%)	4 (12%)
Boundary on VTI				0.227
Clear		9 (35%)	7 (21%)
Uclear		17 (65%)	27 (79%)
SWV(m/s)				0.002^*
≥4.15		18 (69%)	11 (32%)
<4.15		8 (31%)	23 (68%)