432
Original Article
Additional Malignant Breast Lesions Detected on Second-Look US After Breast MRI vs. Additional Malignant Lesions Detected on Initial US in Breast Cancer Patients: Comparison of US Characteristics
Authors
V. Y. Park, M. J. Kim, H. J. Moon, E. K. Kim
Affiliation
Department of Radiology, Research Institute of Radiological Science, Yonsei University, College of Medicine, Seoul
Key words
Zusammenfassung
Abstract
"
!
!
Ziel: Ziel der Studie war der Vergleich von US-Befunden synchroner maligner Mammaläsionen, die nach der Diagnose des Indexkarzinoms im Second-Look US zusätzlich diagnostiziert wurden im Vergleich zu solchen, die unmittelbar im initialen US dargestellt werden konnten. Hiermit sollten unterschiedliche Charakteristika bestimmt werden, die bei der Diagnosestellung unterstützen und das Ergebnis der initialen US Untersuchung essentiell verbessern können. Material und Methoden: Bei 38 Mammakarzinom Patienten erfolgte eine retrospektive Durchsicht 39 mammografisch okkulter Läsionen, die synchron mit dem Indexkarzinom assoziiert waren (21 Läsionen: Detektion mittels Second-Look US, 18 Läsionen: Detektion im initialen US). Alle Patienten erhielten initial eine Mammografie, eine bilaterale Mamma Sonografie und eine MammaMRT, und alle Läsionen wurden pathologisch durch eine Biopsie oder präoperative Lokalisierung bestätigt. Ergebnisse: Zusätzliche maligne Mammaläsionen, die mittels initialem US oder im Second-Look US detektiert wurden, waren häufig dezenter und zeigten nicht die klassischen Malignitätskriterien. Die mittels Second-Look US diagnostizierten Läsionen (Median, 7.0 mm; Range, 3 – 22 mm) waren kleiner, als solche, die im initialen US dargestellt werden konnten (Median, 9.0 mm; Range 3 – 45 mm), obschon der Unterschied nicht signifikant war (p = 0,134). Second-Look Läsionen zeigten keine Schallfortleitung (p = 0,037) und einen signifikant größerer Anteil an Läsionen hatte einen begrenzten oder unscharfen Rand im Vergleich zu Läsionen, die sich im initialen US darstellten (p = 0,042). Second-Look-Läsionen waren signifikant häufiger subareolär oder relativ weit distant (> 5 cm) von der Mamille entfernt gelegen im Vergleich zu den initial detektierten Läsionen.
Purpose: The purpose of our study was to review and compare the US findings of synchronous malignant breast lesions other than the index cancer additionally detected on second-look US with those detected on initial US, and therefore to determine differing characteristics that may aid in diagnosis and essentially improve the performance of the initial US examination. Materials and Methods: A retrospective review of 39 mammographically occult synchronous malignant lesions other than the index cancer from 38 patients was performed (21 lesions: detected on second-look US, 18 lesions: detected on initial US). All patients underwent initial mammography, bilateral whole breast US (BWBU) and breast MRI, and all lesions were confirmed pathologically by biopsy or preoperative localization. Results: Additional malignant breast lesions detected on both initial US and second-look US tended to be subtle and often did not show classic malignant findings. Second-look US lesions (median, 7.0 mm; range, 3 – 22 mm) tended to be smaller than initial US lesions (median, 9.0 mm; range 3 – 45 mm), although the difference was not statistically significant (p = 0.134). Secondlook US lesions also showed no posterior acoustic features (p = 0.037) and a significantly higher proportion of lesions with circumscribed or indistinct margins compared to initial US lesions (p = 0.042). Second-look US lesions were significantly subareolar or relatively far (> 5 cm) from the nipple than initial US lesions (p = 0.048). Conclusion: Second-look US lesions showed more subtle findings of posterior acoustic features and margins, and tended to be subareolar or relatively far from the nipple compared to initial US lesions. However, both groups showed subtle US findings and there was no significant difference in other features.
● breast ● ultrasound ● second-look " "
received accepted
23.11.2012 30.4.2013
Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1335663 Published online: February 7, 2014 Ultraschall in Med 2014; 35: 432–439 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0172-4614
Correspondence Prof. Min Jung Kim Department of Radiology, Research Institute of Radiological Science, Yonsei University, College of Medicine 50 Yonsei-ro Seodaemun-gu 120–752 Seoul Korea, Republic of Tel.: 82-2-22 28-74 00 Fax: 82-2-3 93-30 35 [email protected]
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
Detektion zusätzlicher maligner Mammaläsionen mittels Second-Look US nach Mamma-MRT vs. Detektion zusätzlicher Läsionen im initialen US bei Mammakarzinom Patienten: Vergleich von US Charakteristika
Original Article
433
Schlussfolgerung: Mit Second-Look US diagnostizierte Läsionen zeigten dezentere Befunde bezüglich Schallfortleitung und Berandung und lagen eher subareolär oder relativ weit von der Mamille entfernt im Vergleich zu Läsionen im initialen US. Insgesamt zeigten beide Gruppen dezente US Befunde und es ergaben sich keine signifikanten Unterschiede anderer US Charakteristika.
!
It is essential to accurately evaluate the local extent of malignancy and to detect any contralateral lesions preoperatively for appropriate treatment plans to be established in breast cancer patients, especially when breast-conserving surgery is under consideration. For these purposes, both bilateral whole-breast US (BWBU) and breast MR imaging are used as adjuncts to mammography in the preoperative evaluation of breast cancer patients, with its detection rate of additional lesions reported to be between 14 – 27 % [1 – 7]. Either or both modalities can be used according to the situation of each institution and the accessibility of imaging methods. Some studies reported that additional lesions have been depicted with MR imaging even in cases in which whole-breast US was performed [3]. In a more recent study, MRI mammographically and sonographically identified occult synchronous malignant lesions in 9 % of patients with newly diagnosed breast cancer [8]. Interestingly, several reports showed that such MR-detected synchronous lesions could be identified on sonogram with the application of second-look US [5, 9 – 17]. We may define this as ‘re-identification with second-look US’, which means that synchronous lesions identified on second-look US were not previously detected because they were not visualized during the initial US examination [17]. This led us to pose the question why these lesions failed to be identified in the initial US examination. We hypothesized that if a large amount of information on MR-detected synchronous malignant lesions re-identified on secondlook US was analyzed, such lesions may be detected on initial US of other breast cancer patients in the future. Although MRI provides the highest accuracy in targeting by visualizing enhancing lesions, it is inferior to US-guided examination in terms of accessibility, cost, and patient discomfort and therefore second-look US is commonly prescribed in clinical settings for further characterization and for biopsy if possible [11, 14, 17]. Moreover, it would be of great clinical benefit if we could find MR-detected synchronous lesions in the initial US examination. Yet in a few previous studies, synchronous malignant breast lesions diagnosed with MRI and second-look US other than the index cancer were reported with a tendency to have subtle or no classic malignant US findings [11, 15, 18]. However, those studies have focused on imaging features on second-look US that may aid in the assessment of malignancy and thus have included both benign and malignant unexpected breast MRI lesions. Comparison between synchronous benign and malignant lesions have shown that malignant lesions more often have an ultrasound correlate (43 % in carcinoma vs. 10 % in benignity) [9]. However, to our knowledge, there have been no reports comparing imaging features between synchronous malignant lesions detected on initial US and those detected on second-look US. The purpose of our study was to review and compare US findings of synchronous malignant breast lesions other than the index cancer detected on second-look US with those detected on initial US, and therefore to determine differing characteristics
that may aid in diagnosis and essentially improve the performance of the initial US examination.
Materials and Methods !
This retrospective study was approved by our Institutional Review Board and informed consent was waived.
Study Population From October 2009 to October 2010, 652 patients, including 560 patients newly diagnosed with breast cancer, underwent breast MR examinations at our institution. MR imaging was performed for either preoperative or pre-chemotherapy staging, or to monitor response to chemotherapy. Among the initial 560 breast cancer patients, 143 patients underwent further image-guided biopsy or preoperative localization for additional suspicious breast lesions other than the index cancer. 96 patients were excluded because the histopathology results were benign (n = 60), because we could not find a separate pathologic diagnosis in the pathologic report at surgery (n = 15) or because breast MR images were obtained before the initial US examination (n = 21), making it difficult to determine whether the lesion was detected by initial US or MRI. Nine patients were further excluded because their additional malignant lesions were detectable on preoperative mammography. The remaining 38 patients with 39 additional malignant lesions constitute the population of our study. Because an index breast cancer lesion was present in each patient at the time of surgery, these lesions were regarded as synchronous additional breast lesions. 21 lesions in 21 patients were additional malignant lesions detected on second-look US (hereafter referred to as “second-look " Fig. 1), and 18 lesions in 17 patients were addiUS lesions”) (● tional malignant lesions detected on initial US (hereafter referred to as “initial US lesions”). All lesions were confirmed pathologically by biopsy (n = 26) or preoperative localization (n = 13). All patients underwent surgical excision with subsequent histological examination.
Diagnostic Strategy at Our Institution At our facility, when breast cancer is suspected or has been diagnosed at an outside hospital, mammography and breast ultrasound are initially performed and breast MR examinations are performed within 2 weeks of the initial examination, in accordance with our MRI room schedule. An additional biopsy is recommended when an additional suspicious breast lesion is detected on any of the aforementioned modalities. If suspicious findings are found on the initial breast US, US-guided biopsy is usually performed on the same day or after breast MRI is performed in patients with known breast cancer. If suspicious additional findings are seen only on mammography, patients undergo stereotactic biopsy or mammography-guided localization. If breast
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
Introduction
Original Article
Fig. 1 A 49-year-old women with known palpable breast cancer in the right breast and a synchronous malignant lesion detected on initial US. MLO a and CC b mammographic views demonstrate focal asymmetry corresponding to the palpable area in the right upper central breast. c The transverse US image shows the palpable index cancer (arrow) found in the right breast, 1 o'clock position. d In the initial US examination, an additional 13-mm suspicious lesion (arrow) was found in the ipsilateral breast, 9 o'clock position. This lesion is isoechoic, has an irregular shape, non-parallel orientation, angular margin and is without posterior acoustic shadowing. e The axial T1-weighted dynamic subtracted MR image shows the enhancing mass (arrow) in the right outer center breast. US-guided core needle biopsy revealed DCIS. Abb. 1 Eine 49jährige Frau mit bekanntem Mammakarzinom der rechten Brust und einer synchronen beim initialen US detektierten malignen Läsion. MLO a und CC b Mammografien zeigen eine fokale Asymmetrie korrespondierend zu einer palpablen Masse in der rechten oberen zentralen Brust. c Der US Transversalschnitt zeigt das palpable Indexkarzinom (Pfeil) in der rechten Brust bei ein Uhr. d Bei der initialen US Untersuchung zeigte sich eine zusätzliche suspekte 13 mm große Läsion (Pfeil) in der ipsilateralen Brust bei 9 Uhr. Die Läsion ist isoechogen, hat eine irreguläre Form, eine nicht-parallele Orientierung, einen angulierten Rand und zeigt keine Schallfortleitung. e Das axiale T1-gewichtete dynamische Subtraktions-MRT Bild zeigt eine Kontrast anreichernde Raumforderung (Pfeil) im rechten oberen zentralen Brustanteil. Die US-gestützte Feinnadel Biopsie ergab ein DCIS.
MRI reveals additional suspicious breast lesions, second-look US and US-guided biopsy are performed. MRI-guided biopsy is recommended if no correlate can be found on the second-look US.
Imaging Study with MMG, US, and Image-Guided Biopsy All patients underwent mammography before MRI, which primarily included craniocaudal (CC) and mediolateral-oblique (MLO) views. When needed, additional imaging, such as magnification or compression views, was obtained. All patients also underwent initial bilateral whole-breast US performed by one of six
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
breast imaging radiologists (including authors M. J. K., E. K. K.), each of whom had various levels of breast imaging experience (range: 1 – 11 years), using a 5 – 12 MHZ (HDI 5000;Philips-ATL, Bothell, WA) or another 5 – 12 MHZ (iU22;Philips Medical Systems) linear array transducer. US-guided core biopsy or localization was performed by the radiologist who performed breast US for any additionally detected lesions classified as Breast Imaging Reporting and Data System (BI-RADS®) category 4 or 5 or for any suspicious lesion seen on MRI which had a US correlate on second-look US [19, 20]. Irregular shape, non-parallel orienta-
Downloaded by: University of Queensland. Copyrighted material.
434
Original Article
MRI and Second-Look Ultrasound MRI was performed using a 3.0-T system (TIM Trio; Siemens Healthcare, Erlangen, Germany) with a dedicated, bilateral breast coil (Siemens Healthcare). All patients were examined in the prone position. The MRI protocol included 3 D coronal T1weighted turbo fast low-angle shot sequence (TR/TE, 326/2.6; field of view 360 × 360 mm; section thickness 3 mm), transverse T2-weighted turbo spin-echo imaging (TR/TE, 4360/82; field of view 360 × 360 mm; section thickness 3 mm), and dynamic contrast-enhanced MRI (TR/TE, 280/2.6; field of view 360 × 360 mm; section thickness 3 mm) every minute until a 7-minute period had been concluded. Second-look US examinations were recommended for lesions categorized as suspicious findings based on the BI-RADS MRI lexicon at the discretion of the interpreting radiologists, all of whom had breast MR imaging experience (range: 1 – 15 years) [13]. At the time of second-look US, MR images were made available for direct correlation. First, the area for US was determined by assessing the corresponding breast quadrant and the distance between the lesion and the nipple on MR images. Afterwards, the size and depth of the lesion and its relationship with the surrounding tissue, including the mammary fascia, Cooper’s ligament, and its location within each mammary zone, were evaluated for accurate correlation between MR images and US [21]. For US correlates, US-guided biopsy (n = 8) or preoperative localization (n = 13) was performed.
Image Review For the 39 lesions in 38 women, the medical records, histological findings and imaging studies were retrospectively reviewed by 2 radiologists(M. J.K, V. Y. P.). Imaging information, including location of breast lesions, distance from nipple and density of breast tissue, was obtained from imaging reports. The density of each breast tissue was categorized according to mammography criteria [15, 19] (0 – 25 % parenchyma = fatty, 25 – 50 % = mild, 50 – 75 % = moderate, and 75 – 100 % = dense). Each lesion was classified as a mass or non-mass lesion based on MR images. US features were interpreted and classified in consensus by two radiologists according to the BI-RADS US lexicon [15, 19], which classifies the shape, orientation, margin, boundary, echo, posterior echoes (shadow or enhancement) and presence of calcification on US. The location of each additional lesion was classified as either ipsilateral (same or different quadrant) or contralateral with respect to the index cancer, and the depth of each lesion was further classified according to the location of the majority of the lesion as superficial, middle or deep (superficial = within the subcutaneous fat, middle = within the glandular tissue, and deep = within the retromammary fat). The distance of each lesion from the nipple was obtained from the US report. The lesion size was measured on US according to the largest diameter. The number of suspicious US findings was also counted for each lesion.
Data Analysis The chi-square test or Fisher’s exact test was used for comparison of sonographic findings (i. e., shape, orientation, margin, boundary, echogenicity, posterior echoes, presence of calcification, distance from nipple, lesion depth, location) histology, density of breast tissue and MRI presentation (mass or non-mass) between initial US lesions and second-look US lesions. The student’s t-test was also used to compare patient age between the two groups. The Mann-Whitney U test was used to compare lesion size between the two groups because the data was not normally distributed. Statistical analysis was performed using the SPSS statistical analysis software (PASW Statistics, version 18.0.0; SPSS, Chicago, Ill), with the significance level set at a two-sided p-value of 0.05.
Results !
The patient and lesion characteristics of initial and second-look " Table 1. US lesions are summarized in ●
Patients and Lesions The patient ages ranged from 25 to 68 years (mean, 46.6 years). There was no significant difference in patient age between second-look US lesions and initial US lesions (p = 0.713). The median size of all lesions was 8.0 mm (range, 3 – 45 mm). Second-look US lesions (median, 7.0mm; range, 3 – 22 mm) tended to be smaller than initial US lesions (median, 9.0mm; range, 3 – 45 mm), although the difference was not statistically significant (p = 0.134).
Histological Characteristics of Additional Malignant Lesions On histological examination, all lesions were confirmed as malignant; 16 (41 %) invasive ductal carcinomas (IDC), 5 (13 %) invasive lobular carcinomas (ILC), and 18 (46 %) ductal carcinomas in situ (DCIS). Among the second-look US lesions, 11 (53 %) were DCIS, while 6 (33 %) of (the) initial US lesions were confirmed as DCIS (p = 0.385).
US Characteristics of Lesions Lesion characteristics were recorded according to the BI-RADS lexicon for breast US. There was no significant difference in the proportion of irregular shape lesions (p = 0.762) or with respect to lesion orientation (p = 0.414) between the two groups. Also, there was no significant difference in the composition of circumscribed and non-circumscribed lesions (p = 0.417), boundary (p = 0.889), or echo pattern (p = 0.167). Second-look US lesions more frequently showed indistinct margins, although this difference was not statistically significant [6 (29 %) second-look US lesions vs. 1 (5 %) initial US lesion]. However, when a broader combined grouping of circumscribed and indistinct margins was done, second-look US lesions showed a significantly higher proportion of lesions with circumscribed or indistinct margins compared with initial US lesions (p = 0.042). Second-look US lesions tended to show no posterior acoustic features compared to initial US lesions (p = 0.037). Calcification was not seen on US in all second-look lesions and in 94 % (17 of 18) of initial US lesions (p = 0.461). Overall, there were no suspicious US features in 19 % (4/21) of second-look US lesions, while 6 % (1/18) of initial US lesions showed no suspicious findings. Both groups showed a similar percentage of lesions with only one suspicious US finding – 33.3 % (7/21) in the second-look US group and 28 % Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
tion, non-circumscribed margin, echogenic halo, posterior acoustic shadowing and microcalcification were considered suspicious US findings. All US-guided biopsies were performed with a 14gauge dual-action semiautomatic core biopsy needle (Stericut with coaxial; TSK Laboratory, Tochigi, Japan). The throw of the biopsy needle was 2.2 cm. Four or five core samples per lesion were obtained according to standard protocol.
435
Original Article
lesion characteristics
total (n = 39)
initial US lesions
second-look
(n = 18)
US lesions
p
(n = 21) patient age (years) lesion size (mm)
46.6 ± 1.4
46.1 ± 2.0
47.2 ± 2.1
8.0 (3 – 45)
9.0 (3 – 45)
7.0 (3 – 22)
lesion location
0.713 0.134 > 0.05
same quadrant
16 (41)
10 (55)
6 (28)
other quadrant
15 (38)
5 (28)
10 (48)
8 (21)
3 (17)
5 (24)
DCIS
18 (46)
7 (39)
11 (52)
invasive
21 (54)
11 (61)
10 (48)
2 (25 – 50 % dense)
3 (8)
0 (0)
3 (14)
3 (50 – 75 % dense)
30 (77)
15 (83)
15 (72)
4 (75 – 100 % dense)
6 (15)
3 (17)
3 (14)
contralateral breast
Table 1 Patient and lesion characteristics of initial US lesions versus second-look US lesions.
histological type
> 0.05
breast density on MMG
> 0.05
US findings
> 0.05
shape round
7 (18)
3 (17)
4 (19)
oval
21 (54)
9 (50)
12 (57)
irregular
11 (28)
6 (33)
5 (24)
parallel
19 (49)
7 (39)
12 (57)
non-parallel
20 (51)
11 (61)
9 (43)
orientation
0.414
margin
> 0.05 0.042 1
circumscribed
7 (18)
2 (11)
angular
8 (20)
5 (28)
3 (14)
17 (44)
10 (56)
7 (33)
indistinct
7 (18)
1 (5)
6 (29)
spiculated
0 (0)
0 (0)
0 (0)
36 (92)
17 (94)
19 (90)
3 (8)
1 (6)
2 (10)
isoechoic
23 (59)
8 (44)
15 (71)
hypoechoic
15 (38)
9 (50)
6 (29)
1 (3)
1 (6)
0 (0)
35 (90)
14 (78)
21 (100)
microlobulated
5 (24)
boundary abrupt echogenic
0.889
echogenicity
mixed (hypo and hyper)
0.167
posterior features none
0.037
posterior shadowing
2 (5)
2 (11)
0 (0)
posterior enhancement
2 (5)
2 (11)
0 (0)
38 (97)
17 (94)
21 (100)
1 (3)
1 (6)
0 (0)
5 (13)
1 (6)
4 (19)
calcification none calcification in mass
0.461
distance from nipple subareolar
0.048
> 5 cm
3 (8)
0 (0)
3 (14)
≤ 5 cm
31 (79)
17 (94)
14 (67)
7 (18)
2 (11)
5 (24)
middle
19 (49)
10 (56)
9 (43)
deep
13 (33)
6 (33)
7 (33)
31 (79)
14 (78)
17 (81)
8 (21)
4 (22)
4 (19)
depth superficial
0.638
MRI presentation mass non-mass
0.878
Percentage of each category within each group is show in parentheses. Patient age is expressed as mean ± standard deviation and lesion size is expressed as median and range. The student’s t-test was used for comparison of patient age and the Mann-Whitney U test was used for comparison of lesion size. The chi-square test or Fisher’s exact test was used for comparison of lesion location, histology, breast density, US findings, distance from the nipple, lesion depth and MRI presentation between the two groups. 1 p-value when combined grouping of circumscribed and indistinct margins is performed.
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
436
Fig. 2 Second-look US lesions were more frequently subareolar or relatively far from the nipple. a, b A 62-year-old woman with known breast cancer in the right breast, 8 o'clock position. a The axial T1-weighted dynamic subtracted MR image shows an unexpected 4-mm enhancing lesion (arrow) in the right subareolar area. b Second-look US found a probable complicated cyst-looking correlate (arrow) without any suspicious findings, which was prospectively classified as BI-RADS category 3. However, in retrospective review, it seems to be a complex echoic lesion with adjacent ductal structures rather than a complicated cyst, which may have raised suspicion for this small lesion. US-guided localization was performed, and it was confirmed as DCIS.
Abb. 2 Second-Look US Läsion waren häufiger subareolär oder relative weit von der Mamille entfernt gelegen a, b. Eine 62jährige Frau mit bekanntem Mammakarzinom der rechten Brust bei 8 Uhr. a Das axiale T1-gewichtete dynamische Subtraktions-MRT Bild zeigt unerwartet eine 4 mm große Kontrast anreichernde Läsion (Pfeil) rechts subareolär. b Der SecondLook US zeigte das Korrelat einer wahrscheinlich komplizierten Zyste (Pfeil) ohne suspekte Zeichen, was prospektiv als BI-RADS Kategorie 3 klassifiziert wurde. In der retrospektiven Analyse handelte es sich aber eher um eine komplexe echogene Läsion mit benachbarten duktalen Strukturen denn um eine komplizierte Zyste, was diese kleine Läsion verdächtig hätte erscheinen lassen können. Nach US-gestützter Entnahme ergab sich ein DCIS.
(5/18) in the initial US group. Second-look US lesions also tended to be either subareolar or relatively far (> 5 cm) from the nipple " Fig. 2). There was no compared to initial US lesions (p = 0.048) (● significant difference in lesion depth between the two groups (p = 0.638). Among the 39 additional malignant lesions, 16 were located in the same quadrant as the index cancer, 15 were located in a different quadrant in the ipsilateral breast, and 8 were located in the contralateral breast. There was no significant difference between the two groups, second-look US lesions and initial US lesions, in the percentage of ipsilateral or contralateral lesions (p = 0.702), or in the percentage of same quadrant lesions (p = 0.167). Overall on MRI, 31(79 %) of the 39 lesions presented as mass lesions and 8 (21 %) presented as non-mass-like enhancements. There was no significant difference between the two groups on MRI presentation (p = 0.878).
which were later accurately identified by second-look US. Possible reasons for failed detection on initial US are: first, the lesions could have been too subtle; second, they could have been too small to be detected on initial US; and finally, they may have been neglected because of a lack of attention from the US operator. By comparing the US findings of synchronous malignant initial US lesions with second-look US lesions, we hoped to identify different US findings that would help increase the sensitivity of the initial US and preoperative evaluation of breast cancer patients, especially in those who did not undergo MRI. Mammographically occult malignancy detected on US is known as a small but invasive lesion with less typical US findings of malignancy. The probability of malignancy has been reported to be much higher in synchronous nodules found in patients with breast cancer, even when there are no suspicious US findings [16]. Kim et al. found that the rate of malignancy among BIRADS category 3 nodules found in breast cancer patients was 11.4 %, and that 90.7 % of malignant nodules were mammographically occult [22]. Furthermore, the risk of malignancy was related to the distance from the primary malignant tumor [22]. Our results also show that synchronous nodules detected on initial US tend to show less malignant features (no suspicious findings in 5.6 %, one of 18; only one suspicious finding in 33.3 %, 6 of 18) and are mostly located in the ipsilateral breast (83.3 %, 15 of 18), and especially in the same quadrant (44.4 %, 8 of 18), all of which is consistent with previous literature. There have also been several studies reporting the US features of MR-detected malignancy on second-look US [15, 18]. Abe et al. [15] found that malignant breast lesions initially detected on MRI tended to be subtle and that the classic malignant US findings were often absent, with 33 % showing no suspicious US features. However,
Discussion !
Today, whole-breast ultrasound is increasingly included in the preoperative evaluation of breast cancer patients, as it aids the detection of mammographically occult additional lesions [1, 2]. Although breast MRI has been shown to depict synchronous lesions even after whole-breast US was performed [3], there are limitations to its routine use in all breast cancer patients due to its high cost and low accessibility for biopsy. In our study, all included patients underwent initial bilateral whole-breast US examination before breast MRI. Therefore, we could assume that all second-look US lesions were not visualized during the initial US examination, i. e., initial US could not identify existing lesions
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
437
Downloaded by: University of Queensland. Copyrighted material.
Original Article
Original Article
only 76 of the 148 patients included in their study underwent US before breast MRI and therefore some lesions could have been initially detected on US. In a different study in which all patients underwent initial US, Laguna et al. [18] also reported that classic malignant US findings were not often present with only 4 – 8 % showing posterior acoustic shadowing. Our study also showed similar results – 19 % (4/21) of second-look US lesions showed no suspicious US findings, which was a higher percentage than that seen in initial US lesions. Also 33.3 % (7/21) of cases showed only one suspicious finding. A non-circumscribed margin was the only suspicious US feature in 28.6 % (6/21) of cases. Second-look US lesions showed a higher percentage of parallel-oriented lesions (57 % vs. 39 %) and isoechoic lesions (71 % vs. 44 %). These show the same trend as those of Abe in which among 33 malignant lesions on second-look US, 61 % were parallel-oriented and 36 % were isoechoic. Interestingly, all second-look US lesions showed no posterior acoustic features (100 % vs. 78 %). Posterior shadowing aids lesion identification, and its absence may be one of the reasons why the lesions were not detected in the initial US examination. In such cases, the use of tissue harmonic imaging may be helpful in lesion detection by its amplification of posterior acoustic features [23, 24]. In our study, 76 % of second-look US lesions showed a non-circumscribed margin which is similar to the 73 % reported by Abe, but there was no significant difference when compared with initial US lesions (82 %). We also found that a higher percentage of second-look US lesions showed no calcification on US (100 % vs. 94 %), but none of these differences were statistically significant. However, second-look US lesions showed a significantly higher proportion of lesions with circumscribed or indistinct margins: 11 (53 %) second-look US lesions vs. 3 (16 %) initial US lesions (p = 0.042). A possible explanation is that circumscribed lesions may have been considered benign on initial US, or were hard to distinguish from the background due to combined isoechogenicity. Additionally, lesions with indistinct margins may have been difficult to initially detect. Interestingly, second-look US lesions were either subareolar or relatively far (> 5 cm) from the nipple compared to initial US lesions (p = 0.048). Subareolar lesions may have been easy to miss on initial US due to obscuration by nipple shadows, which is well known as one of the pitfalls of breast US [25]. Also, lesions relatively far from the nipple may have been missed due to the complacency of the examiner as he or she reached the periphery of the breast. Furthermore, initial US lesions showed a higher percentage of same quadrant lesions compared with secondlook US lesions (p = 0.167). Second-look US lesions also tended to be smaller in size, with a mean size of 7.84 mm (p = 0.134). This study demonstrates that additional malignant breast lesions detected on both initial US and second-look US tend to be subtle and often do not show classic malignant findings. As both groups show subtle US features, the clinical significance of a considerable number of the obtained results in this study is questionable. Rather, our study will have clinical value with respect to reviewing which aspects need improvement in the initial US examination – emphasizing the need for meticulous examination on initial preoperative bilateral whole-breast US, including the subareolar area and the periphery of the breast. Our study has several limitations. First, the number of included lesions was relatively small, with an overall number of 39 lesions. In addition, there were a slightly larger number of second-look US lesions than initial US lesions, which does not accurately represent the actual clinical setting where a much larger number of additional malignant lesions are detected on initial US. This is
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
probably because we only included patients who underwent preoperative breast MRI. In cases where obvious suspicious synchronous lesions were detected on initial US, the treatment plan would have been changed from breast-conserving surgery to mastectomy and preoperative breast MRI would not have been performed. Also, our study reviewed US features of only pathologically confirmed malignant lesions and therefore a comparison with benign lesions was not performed. Thus, this study may be of less value in the differentiation between benign and malignant additional lesions. However, we believe that understanding the US features of additional malignant lesions will essentially help in the determination of malignancy. As both groups showed subtle US features, our study supports findings from previous studies that a lower threshold must be applied in the case of a known index cancer when suspecting malignancy [15, 18, 22]. As our study focused on identifying US findings of additional malignant lesions, all MR-detected lesions naturally showed a US correlate. Hence this study would be of little value in the management of MR-detected lesions without a US correlate. However, similar to previous studies [5, 11, 12, 14 – 16], we also found that 80 % of second-look lesions presented as mass-like lesions on MRI. This supports the current literature that mass-like lesions on MRI are more likely to be visualized on US. In summary, we found that synchronous malignant breast lesions detected on both initial US and second-look US tend to be subtle and often do not show classic malignant findings. Second-look US lesions show more subtle findings of posterior acoustic features and margins and tend to be subareolar or relatively far from the nipple. Therefore, meticulous examination on initial preoperative US with special attention to the subareolar area and to the periphery of the breast, with simultaneous awareness of the above findings, needs to be emphasized to ensure thorough evaluation for possible additional malignant lesions.
Reference 01 Berg WA, Gilbreath PL. Multicentric and multifocal cancer: wholebreast US in preoperative evaluation. Radiology 2000; 214: 59 – 66 02 Moon WK, Noh DY, Im JG. Multifocal, multicentric, and contralateral breast cancers: bilateral whole-breast US in the preoperative evaluation of patients. Radiology 2002; 224: 569 – 576 03 Berg WA, Gutierrez L, NessAiver MS et al. Diagnostic accuracy of mammography, clinical examination, US, and MR imaging in preoperative assessment of breast cancer. Radiology 2004; 233: 830 – 849 04 Pediconi F, Miglio E, Telesca M et al. Effect of preoperative breast magnetic resonance imaging on surgical decision making and cancer recurrence rates. Invest Radiol 2012; 47: 128 – 135 05 Luciani ML, Pediconi F, Telesca M et al. Incidental enhancing lesions found on preoperative breast MRI: management and role of secondlook ultrasound. Radiol Med 2011; 116 (6): 886 – 904 06 Gutierrez RL, DeMartini WB, Silbergeld JJ et al. High cancer yield and positive predictive value: outcomes at a center routinely using preoperative breast MRI for staging. Am J Roentgenol Am J Roentgenol 2011; 196: W93 – 99 07 Sohn YM, Kim MJ, Kim EK. Impact of Preoperative Bilateral Whole Breast Sonography in Patients with Invasive Lobular Carcinoma: Results from Two Medical Centers. Ultraschall in Med 2012 08 Girardi V, Carbognin G, Camera L et al. Multifocal, multicentric and contralateral breast cancers: breast MR imaging in the preoperative evaluation of patients with newly diagnosed breast cancer. Radiol Med 2011; 116: 1226 – 1238 09 LaTrenta LR, Menell JH, Morris EA et al. Breast lesions detected with MR imaging: utility and histopathologic importance of identification with US. Radiology 2003; 227: 856 – 861 10 Shin JH, Han BK, Choe YH et al. Targeted ultrasound for MR-detected lesions in breast cancer patients. Korean J Radiol 2007; 8: 475 – 483
Downloaded by: University of Queensland. Copyrighted material.
438
Original Article
18 Laguna AD, Arranz SJ, Checa VQ et al. Sonographic findings of additional malignant lesions in breast carcinoma seen by second look ultrasound. J Clin Imaging Sci 2011; 1: 34 19 American College of Radiology. Breast Imaging Reporting and Data System (BI -RADS). 4th ed. Reston, VA: American College of Radiology; 2003 20 Hille H, Vetter M, Hackeloer BJ. The accuracy of BI-RADS classification of breast ultrasound as a first-line imaging method. Ultraschall in Med 2012; 33: 160 – 163 21 Park VY, Kim MJ, Kim EK et al. How To Do Second-Look Ultrasound For Finding Unexpected Lesions On Breast MRI Radiographics. This article is supposed to be published in July 2013 22 Kim SJ, Ko EY, Shin JH et al. Application of sonographic BI-RADS to synchronous breast nodules detected in patients with breast cancer. Am J Roentgenol Am J Roentgenol 2008; 191: 653 – 658 23 Oktar SO, Yucel C, Ozdemir H et al. Comparison of conventional sonography, real-time compound sonography, tissue harmonic sonography, and tissue harmonic compound sonography of abdominal and pelvic lesions. Am J Roentgenol Am J Roentgenol 2003; 181: 1341 – 1347 24 Mesurolle B, Helou T, El-Khoury M et al. Tissue harmonic imaging, frequency compound imaging, and conventional imaging: use and benefit in breast sonography. J Ultrasound Med 2007; 26: 1041 – 1051 25 Stavros AT. Breast ultrasound equipment requirements. Breast ultrasound. Philadelphia: Lippincott Williams & Wilkins; 2003
Downloaded by: University of Queensland. Copyrighted material.
11 Wiratkapun C, Duke D, Nordmann AS et al. Indeterminate or suspicious breast lesions detected initially with MR imaging: value of MRI-directed breast ultrasound. Acad Radiol 2008; 15: 618 – 625 12 Demartini WB, Eby PR, Peacock S et al. Utility of targeted sonography for breast lesions that were suspicious on MRI. Am J Roentgenol Am J Roentgenol 2009; 192: 1128 – 1134 13 Destounis S, Arieno A, Somerville PA et al. Community-based practice experience of unsuspected breast magnetic resonance imaging abnormalities evaluated with second-look sonography. J Ultrasound Med 2009; 28: 1337 – 1346 14 Meissnitzer M, Dershaw DD, Lee CH et al. Targeted ultrasound of the breast in women with abnormal MRI findings for whom biopsy has been recommended. Am J Roentgenol Am J Roentgenol 2009; 193: 1025 – 1029 15 Abe H, Schmidt RA, Shah RN et al. MR-directed (“Second-Look”) ultrasound examination for breast lesions detected initially on MRI: MR and sonographic findings. Am J Roentgenol Am J Roentgenol 2010; 194: 370 – 377 16 Carbognin G, Girardi V, Calciolari C et al. Utility of second-look ultrasound in the management of incidental enhancing lesions detected by breast MR imaging. Radiol Med 2010; 115: 1234 – 1245 17 Candelaria R, Fornage BD. Second-look US examination of MR-detected breast lesions. J Clin Ultrasound 2011; 39: 115 – 121
439
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Original Article
Additional Malignant Breast Lesions Detected on Second-Look US After Breast MRI vs. Additional Malignant Lesions Detected on Initial US in Breast Cancer Patients: Comparison of US Characteristics
Authors
V. Y. Park, M. J. Kim, H. J. Moon, E. K. Kim
Affiliation
Department of Radiology, Research Institute of Radiological Science, Yonsei University, College of Medicine, Seoul
Key words
Zusammenfassung
Abstract
"
!
!
Ziel: Ziel der Studie war der Vergleich von US-Befunden synchroner maligner Mammaläsionen, die nach der Diagnose des Indexkarzinoms im Second-Look US zusätzlich diagnostiziert wurden im Vergleich zu solchen, die unmittelbar im initialen US dargestellt werden konnten. Hiermit sollten unterschiedliche Charakteristika bestimmt werden, die bei der Diagnosestellung unterstützen und das Ergebnis der initialen US Untersuchung essentiell verbessern können. Material und Methoden: Bei 38 Mammakarzinom Patienten erfolgte eine retrospektive Durchsicht 39 mammografisch okkulter Läsionen, die synchron mit dem Indexkarzinom assoziiert waren (21 Läsionen: Detektion mittels Second-Look US, 18 Läsionen: Detektion im initialen US). Alle Patienten erhielten initial eine Mammografie, eine bilaterale Mamma Sonografie und eine MammaMRT, und alle Läsionen wurden pathologisch durch eine Biopsie oder präoperative Lokalisierung bestätigt. Ergebnisse: Zusätzliche maligne Mammaläsionen, die mittels initialem US oder im Second-Look US detektiert wurden, waren häufig dezenter und zeigten nicht die klassischen Malignitätskriterien. Die mittels Second-Look US diagnostizierten Läsionen (Median, 7.0 mm; Range, 3 – 22 mm) waren kleiner, als solche, die im initialen US dargestellt werden konnten (Median, 9.0 mm; Range 3 – 45 mm), obschon der Unterschied nicht signifikant war (p = 0,134). Second-Look Läsionen zeigten keine Schallfortleitung (p = 0,037) und einen signifikant größerer Anteil an Läsionen hatte einen begrenzten oder unscharfen Rand im Vergleich zu Läsionen, die sich im initialen US darstellten (p = 0,042). Second-Look-Läsionen waren signifikant häufiger subareolär oder relativ weit distant (> 5 cm) von der Mamille entfernt gelegen im Vergleich zu den initial detektierten Läsionen.
Purpose: The purpose of our study was to review and compare the US findings of synchronous malignant breast lesions other than the index cancer additionally detected on second-look US with those detected on initial US, and therefore to determine differing characteristics that may aid in diagnosis and essentially improve the performance of the initial US examination. Materials and Methods: A retrospective review of 39 mammographically occult synchronous malignant lesions other than the index cancer from 38 patients was performed (21 lesions: detected on second-look US, 18 lesions: detected on initial US). All patients underwent initial mammography, bilateral whole breast US (BWBU) and breast MRI, and all lesions were confirmed pathologically by biopsy or preoperative localization. Results: Additional malignant breast lesions detected on both initial US and second-look US tended to be subtle and often did not show classic malignant findings. Second-look US lesions (median, 7.0 mm; range, 3 – 22 mm) tended to be smaller than initial US lesions (median, 9.0 mm; range 3 – 45 mm), although the difference was not statistically significant (p = 0.134). Secondlook US lesions also showed no posterior acoustic features (p = 0.037) and a significantly higher proportion of lesions with circumscribed or indistinct margins compared to initial US lesions (p = 0.042). Second-look US lesions were significantly subareolar or relatively far (> 5 cm) from the nipple than initial US lesions (p = 0.048). Conclusion: Second-look US lesions showed more subtle findings of posterior acoustic features and margins, and tended to be subareolar or relatively far from the nipple compared to initial US lesions. However, both groups showed subtle US findings and there was no significant difference in other features.
● breast ● ultrasound ● second-look " "
received accepted
23.11.2012 30.4.2013
Bibliography DOI http://dx.doi.org/ 10.1055/s-0033-1335663 Published online: February 7, 2014 Ultraschall in Med 2014; 35: 432–439 © Georg Thieme Verlag KG Stuttgart · New York · ISSN 0172-4614
Correspondence Prof. Min Jung Kim Department of Radiology, Research Institute of Radiological Science, Yonsei University, College of Medicine 50 Yonsei-ro Seodaemun-gu 120–752 Seoul Korea, Republic of Tel.: 82-2-22 28-74 00 Fax: 82-2-3 93-30 35 [email protected]
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
Detektion zusätzlicher maligner Mammaläsionen mittels Second-Look US nach Mamma-MRT vs. Detektion zusätzlicher Läsionen im initialen US bei Mammakarzinom Patienten: Vergleich von US Charakteristika
Original Article
433
Schlussfolgerung: Mit Second-Look US diagnostizierte Läsionen zeigten dezentere Befunde bezüglich Schallfortleitung und Berandung und lagen eher subareolär oder relativ weit von der Mamille entfernt im Vergleich zu Läsionen im initialen US. Insgesamt zeigten beide Gruppen dezente US Befunde und es ergaben sich keine signifikanten Unterschiede anderer US Charakteristika.
!
It is essential to accurately evaluate the local extent of malignancy and to detect any contralateral lesions preoperatively for appropriate treatment plans to be established in breast cancer patients, especially when breast-conserving surgery is under consideration. For these purposes, both bilateral whole-breast US (BWBU) and breast MR imaging are used as adjuncts to mammography in the preoperative evaluation of breast cancer patients, with its detection rate of additional lesions reported to be between 14 – 27 % [1 – 7]. Either or both modalities can be used according to the situation of each institution and the accessibility of imaging methods. Some studies reported that additional lesions have been depicted with MR imaging even in cases in which whole-breast US was performed [3]. In a more recent study, MRI mammographically and sonographically identified occult synchronous malignant lesions in 9 % of patients with newly diagnosed breast cancer [8]. Interestingly, several reports showed that such MR-detected synchronous lesions could be identified on sonogram with the application of second-look US [5, 9 – 17]. We may define this as ‘re-identification with second-look US’, which means that synchronous lesions identified on second-look US were not previously detected because they were not visualized during the initial US examination [17]. This led us to pose the question why these lesions failed to be identified in the initial US examination. We hypothesized that if a large amount of information on MR-detected synchronous malignant lesions re-identified on secondlook US was analyzed, such lesions may be detected on initial US of other breast cancer patients in the future. Although MRI provides the highest accuracy in targeting by visualizing enhancing lesions, it is inferior to US-guided examination in terms of accessibility, cost, and patient discomfort and therefore second-look US is commonly prescribed in clinical settings for further characterization and for biopsy if possible [11, 14, 17]. Moreover, it would be of great clinical benefit if we could find MR-detected synchronous lesions in the initial US examination. Yet in a few previous studies, synchronous malignant breast lesions diagnosed with MRI and second-look US other than the index cancer were reported with a tendency to have subtle or no classic malignant US findings [11, 15, 18]. However, those studies have focused on imaging features on second-look US that may aid in the assessment of malignancy and thus have included both benign and malignant unexpected breast MRI lesions. Comparison between synchronous benign and malignant lesions have shown that malignant lesions more often have an ultrasound correlate (43 % in carcinoma vs. 10 % in benignity) [9]. However, to our knowledge, there have been no reports comparing imaging features between synchronous malignant lesions detected on initial US and those detected on second-look US. The purpose of our study was to review and compare US findings of synchronous malignant breast lesions other than the index cancer detected on second-look US with those detected on initial US, and therefore to determine differing characteristics
that may aid in diagnosis and essentially improve the performance of the initial US examination.
Materials and Methods !
This retrospective study was approved by our Institutional Review Board and informed consent was waived.
Study Population From October 2009 to October 2010, 652 patients, including 560 patients newly diagnosed with breast cancer, underwent breast MR examinations at our institution. MR imaging was performed for either preoperative or pre-chemotherapy staging, or to monitor response to chemotherapy. Among the initial 560 breast cancer patients, 143 patients underwent further image-guided biopsy or preoperative localization for additional suspicious breast lesions other than the index cancer. 96 patients were excluded because the histopathology results were benign (n = 60), because we could not find a separate pathologic diagnosis in the pathologic report at surgery (n = 15) or because breast MR images were obtained before the initial US examination (n = 21), making it difficult to determine whether the lesion was detected by initial US or MRI. Nine patients were further excluded because their additional malignant lesions were detectable on preoperative mammography. The remaining 38 patients with 39 additional malignant lesions constitute the population of our study. Because an index breast cancer lesion was present in each patient at the time of surgery, these lesions were regarded as synchronous additional breast lesions. 21 lesions in 21 patients were additional malignant lesions detected on second-look US (hereafter referred to as “second-look " Fig. 1), and 18 lesions in 17 patients were addiUS lesions”) (● tional malignant lesions detected on initial US (hereafter referred to as “initial US lesions”). All lesions were confirmed pathologically by biopsy (n = 26) or preoperative localization (n = 13). All patients underwent surgical excision with subsequent histological examination.
Diagnostic Strategy at Our Institution At our facility, when breast cancer is suspected or has been diagnosed at an outside hospital, mammography and breast ultrasound are initially performed and breast MR examinations are performed within 2 weeks of the initial examination, in accordance with our MRI room schedule. An additional biopsy is recommended when an additional suspicious breast lesion is detected on any of the aforementioned modalities. If suspicious findings are found on the initial breast US, US-guided biopsy is usually performed on the same day or after breast MRI is performed in patients with known breast cancer. If suspicious additional findings are seen only on mammography, patients undergo stereotactic biopsy or mammography-guided localization. If breast
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
Introduction
Original Article
Fig. 1 A 49-year-old women with known palpable breast cancer in the right breast and a synchronous malignant lesion detected on initial US. MLO a and CC b mammographic views demonstrate focal asymmetry corresponding to the palpable area in the right upper central breast. c The transverse US image shows the palpable index cancer (arrow) found in the right breast, 1 o'clock position. d In the initial US examination, an additional 13-mm suspicious lesion (arrow) was found in the ipsilateral breast, 9 o'clock position. This lesion is isoechoic, has an irregular shape, non-parallel orientation, angular margin and is without posterior acoustic shadowing. e The axial T1-weighted dynamic subtracted MR image shows the enhancing mass (arrow) in the right outer center breast. US-guided core needle biopsy revealed DCIS. Abb. 1 Eine 49jährige Frau mit bekanntem Mammakarzinom der rechten Brust und einer synchronen beim initialen US detektierten malignen Läsion. MLO a und CC b Mammografien zeigen eine fokale Asymmetrie korrespondierend zu einer palpablen Masse in der rechten oberen zentralen Brust. c Der US Transversalschnitt zeigt das palpable Indexkarzinom (Pfeil) in der rechten Brust bei ein Uhr. d Bei der initialen US Untersuchung zeigte sich eine zusätzliche suspekte 13 mm große Läsion (Pfeil) in der ipsilateralen Brust bei 9 Uhr. Die Läsion ist isoechogen, hat eine irreguläre Form, eine nicht-parallele Orientierung, einen angulierten Rand und zeigt keine Schallfortleitung. e Das axiale T1-gewichtete dynamische Subtraktions-MRT Bild zeigt eine Kontrast anreichernde Raumforderung (Pfeil) im rechten oberen zentralen Brustanteil. Die US-gestützte Feinnadel Biopsie ergab ein DCIS.
MRI reveals additional suspicious breast lesions, second-look US and US-guided biopsy are performed. MRI-guided biopsy is recommended if no correlate can be found on the second-look US.
Imaging Study with MMG, US, and Image-Guided Biopsy All patients underwent mammography before MRI, which primarily included craniocaudal (CC) and mediolateral-oblique (MLO) views. When needed, additional imaging, such as magnification or compression views, was obtained. All patients also underwent initial bilateral whole-breast US performed by one of six
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
breast imaging radiologists (including authors M. J. K., E. K. K.), each of whom had various levels of breast imaging experience (range: 1 – 11 years), using a 5 – 12 MHZ (HDI 5000;Philips-ATL, Bothell, WA) or another 5 – 12 MHZ (iU22;Philips Medical Systems) linear array transducer. US-guided core biopsy or localization was performed by the radiologist who performed breast US for any additionally detected lesions classified as Breast Imaging Reporting and Data System (BI-RADS®) category 4 or 5 or for any suspicious lesion seen on MRI which had a US correlate on second-look US [19, 20]. Irregular shape, non-parallel orienta-
Downloaded by: University of Queensland. Copyrighted material.
434
Original Article
MRI and Second-Look Ultrasound MRI was performed using a 3.0-T system (TIM Trio; Siemens Healthcare, Erlangen, Germany) with a dedicated, bilateral breast coil (Siemens Healthcare). All patients were examined in the prone position. The MRI protocol included 3 D coronal T1weighted turbo fast low-angle shot sequence (TR/TE, 326/2.6; field of view 360 × 360 mm; section thickness 3 mm), transverse T2-weighted turbo spin-echo imaging (TR/TE, 4360/82; field of view 360 × 360 mm; section thickness 3 mm), and dynamic contrast-enhanced MRI (TR/TE, 280/2.6; field of view 360 × 360 mm; section thickness 3 mm) every minute until a 7-minute period had been concluded. Second-look US examinations were recommended for lesions categorized as suspicious findings based on the BI-RADS MRI lexicon at the discretion of the interpreting radiologists, all of whom had breast MR imaging experience (range: 1 – 15 years) [13]. At the time of second-look US, MR images were made available for direct correlation. First, the area for US was determined by assessing the corresponding breast quadrant and the distance between the lesion and the nipple on MR images. Afterwards, the size and depth of the lesion and its relationship with the surrounding tissue, including the mammary fascia, Cooper’s ligament, and its location within each mammary zone, were evaluated for accurate correlation between MR images and US [21]. For US correlates, US-guided biopsy (n = 8) or preoperative localization (n = 13) was performed.
Image Review For the 39 lesions in 38 women, the medical records, histological findings and imaging studies were retrospectively reviewed by 2 radiologists(M. J.K, V. Y. P.). Imaging information, including location of breast lesions, distance from nipple and density of breast tissue, was obtained from imaging reports. The density of each breast tissue was categorized according to mammography criteria [15, 19] (0 – 25 % parenchyma = fatty, 25 – 50 % = mild, 50 – 75 % = moderate, and 75 – 100 % = dense). Each lesion was classified as a mass or non-mass lesion based on MR images. US features were interpreted and classified in consensus by two radiologists according to the BI-RADS US lexicon [15, 19], which classifies the shape, orientation, margin, boundary, echo, posterior echoes (shadow or enhancement) and presence of calcification on US. The location of each additional lesion was classified as either ipsilateral (same or different quadrant) or contralateral with respect to the index cancer, and the depth of each lesion was further classified according to the location of the majority of the lesion as superficial, middle or deep (superficial = within the subcutaneous fat, middle = within the glandular tissue, and deep = within the retromammary fat). The distance of each lesion from the nipple was obtained from the US report. The lesion size was measured on US according to the largest diameter. The number of suspicious US findings was also counted for each lesion.
Data Analysis The chi-square test or Fisher’s exact test was used for comparison of sonographic findings (i. e., shape, orientation, margin, boundary, echogenicity, posterior echoes, presence of calcification, distance from nipple, lesion depth, location) histology, density of breast tissue and MRI presentation (mass or non-mass) between initial US lesions and second-look US lesions. The student’s t-test was also used to compare patient age between the two groups. The Mann-Whitney U test was used to compare lesion size between the two groups because the data was not normally distributed. Statistical analysis was performed using the SPSS statistical analysis software (PASW Statistics, version 18.0.0; SPSS, Chicago, Ill), with the significance level set at a two-sided p-value of 0.05.
Results !
The patient and lesion characteristics of initial and second-look " Table 1. US lesions are summarized in ●
Patients and Lesions The patient ages ranged from 25 to 68 years (mean, 46.6 years). There was no significant difference in patient age between second-look US lesions and initial US lesions (p = 0.713). The median size of all lesions was 8.0 mm (range, 3 – 45 mm). Second-look US lesions (median, 7.0mm; range, 3 – 22 mm) tended to be smaller than initial US lesions (median, 9.0mm; range, 3 – 45 mm), although the difference was not statistically significant (p = 0.134).
Histological Characteristics of Additional Malignant Lesions On histological examination, all lesions were confirmed as malignant; 16 (41 %) invasive ductal carcinomas (IDC), 5 (13 %) invasive lobular carcinomas (ILC), and 18 (46 %) ductal carcinomas in situ (DCIS). Among the second-look US lesions, 11 (53 %) were DCIS, while 6 (33 %) of (the) initial US lesions were confirmed as DCIS (p = 0.385).
US Characteristics of Lesions Lesion characteristics were recorded according to the BI-RADS lexicon for breast US. There was no significant difference in the proportion of irregular shape lesions (p = 0.762) or with respect to lesion orientation (p = 0.414) between the two groups. Also, there was no significant difference in the composition of circumscribed and non-circumscribed lesions (p = 0.417), boundary (p = 0.889), or echo pattern (p = 0.167). Second-look US lesions more frequently showed indistinct margins, although this difference was not statistically significant [6 (29 %) second-look US lesions vs. 1 (5 %) initial US lesion]. However, when a broader combined grouping of circumscribed and indistinct margins was done, second-look US lesions showed a significantly higher proportion of lesions with circumscribed or indistinct margins compared with initial US lesions (p = 0.042). Second-look US lesions tended to show no posterior acoustic features compared to initial US lesions (p = 0.037). Calcification was not seen on US in all second-look lesions and in 94 % (17 of 18) of initial US lesions (p = 0.461). Overall, there were no suspicious US features in 19 % (4/21) of second-look US lesions, while 6 % (1/18) of initial US lesions showed no suspicious findings. Both groups showed a similar percentage of lesions with only one suspicious US finding – 33.3 % (7/21) in the second-look US group and 28 % Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
tion, non-circumscribed margin, echogenic halo, posterior acoustic shadowing and microcalcification were considered suspicious US findings. All US-guided biopsies were performed with a 14gauge dual-action semiautomatic core biopsy needle (Stericut with coaxial; TSK Laboratory, Tochigi, Japan). The throw of the biopsy needle was 2.2 cm. Four or five core samples per lesion were obtained according to standard protocol.
435
Original Article
lesion characteristics
total (n = 39)
initial US lesions
second-look
(n = 18)
US lesions
p
(n = 21) patient age (years) lesion size (mm)
46.6 ± 1.4
46.1 ± 2.0
47.2 ± 2.1
8.0 (3 – 45)
9.0 (3 – 45)
7.0 (3 – 22)
lesion location
0.713 0.134 > 0.05
same quadrant
16 (41)
10 (55)
6 (28)
other quadrant
15 (38)
5 (28)
10 (48)
8 (21)
3 (17)
5 (24)
DCIS
18 (46)
7 (39)
11 (52)
invasive
21 (54)
11 (61)
10 (48)
2 (25 – 50 % dense)
3 (8)
0 (0)
3 (14)
3 (50 – 75 % dense)
30 (77)
15 (83)
15 (72)
4 (75 – 100 % dense)
6 (15)
3 (17)
3 (14)
contralateral breast
Table 1 Patient and lesion characteristics of initial US lesions versus second-look US lesions.
histological type
> 0.05
breast density on MMG
> 0.05
US findings
> 0.05
shape round
7 (18)
3 (17)
4 (19)
oval
21 (54)
9 (50)
12 (57)
irregular
11 (28)
6 (33)
5 (24)
parallel
19 (49)
7 (39)
12 (57)
non-parallel
20 (51)
11 (61)
9 (43)
orientation
0.414
margin
> 0.05 0.042 1
circumscribed
7 (18)
2 (11)
angular
8 (20)
5 (28)
3 (14)
17 (44)
10 (56)
7 (33)
indistinct
7 (18)
1 (5)
6 (29)
spiculated
0 (0)
0 (0)
0 (0)
36 (92)
17 (94)
19 (90)
3 (8)
1 (6)
2 (10)
isoechoic
23 (59)
8 (44)
15 (71)
hypoechoic
15 (38)
9 (50)
6 (29)
1 (3)
1 (6)
0 (0)
35 (90)
14 (78)
21 (100)
microlobulated
5 (24)
boundary abrupt echogenic
0.889
echogenicity
mixed (hypo and hyper)
0.167
posterior features none
0.037
posterior shadowing
2 (5)
2 (11)
0 (0)
posterior enhancement
2 (5)
2 (11)
0 (0)
38 (97)
17 (94)
21 (100)
1 (3)
1 (6)
0 (0)
5 (13)
1 (6)
4 (19)
calcification none calcification in mass
0.461
distance from nipple subareolar
0.048
> 5 cm
3 (8)
0 (0)
3 (14)
≤ 5 cm
31 (79)
17 (94)
14 (67)
7 (18)
2 (11)
5 (24)
middle
19 (49)
10 (56)
9 (43)
deep
13 (33)
6 (33)
7 (33)
31 (79)
14 (78)
17 (81)
8 (21)
4 (22)
4 (19)
depth superficial
0.638
MRI presentation mass non-mass
0.878
Percentage of each category within each group is show in parentheses. Patient age is expressed as mean ± standard deviation and lesion size is expressed as median and range. The student’s t-test was used for comparison of patient age and the Mann-Whitney U test was used for comparison of lesion size. The chi-square test or Fisher’s exact test was used for comparison of lesion location, histology, breast density, US findings, distance from the nipple, lesion depth and MRI presentation between the two groups. 1 p-value when combined grouping of circumscribed and indistinct margins is performed.
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
Downloaded by: University of Queensland. Copyrighted material.
436
Fig. 2 Second-look US lesions were more frequently subareolar or relatively far from the nipple. a, b A 62-year-old woman with known breast cancer in the right breast, 8 o'clock position. a The axial T1-weighted dynamic subtracted MR image shows an unexpected 4-mm enhancing lesion (arrow) in the right subareolar area. b Second-look US found a probable complicated cyst-looking correlate (arrow) without any suspicious findings, which was prospectively classified as BI-RADS category 3. However, in retrospective review, it seems to be a complex echoic lesion with adjacent ductal structures rather than a complicated cyst, which may have raised suspicion for this small lesion. US-guided localization was performed, and it was confirmed as DCIS.
Abb. 2 Second-Look US Läsion waren häufiger subareolär oder relative weit von der Mamille entfernt gelegen a, b. Eine 62jährige Frau mit bekanntem Mammakarzinom der rechten Brust bei 8 Uhr. a Das axiale T1-gewichtete dynamische Subtraktions-MRT Bild zeigt unerwartet eine 4 mm große Kontrast anreichernde Läsion (Pfeil) rechts subareolär. b Der SecondLook US zeigte das Korrelat einer wahrscheinlich komplizierten Zyste (Pfeil) ohne suspekte Zeichen, was prospektiv als BI-RADS Kategorie 3 klassifiziert wurde. In der retrospektiven Analyse handelte es sich aber eher um eine komplexe echogene Läsion mit benachbarten duktalen Strukturen denn um eine komplizierte Zyste, was diese kleine Läsion verdächtig hätte erscheinen lassen können. Nach US-gestützter Entnahme ergab sich ein DCIS.
(5/18) in the initial US group. Second-look US lesions also tended to be either subareolar or relatively far (> 5 cm) from the nipple " Fig. 2). There was no compared to initial US lesions (p = 0.048) (● significant difference in lesion depth between the two groups (p = 0.638). Among the 39 additional malignant lesions, 16 were located in the same quadrant as the index cancer, 15 were located in a different quadrant in the ipsilateral breast, and 8 were located in the contralateral breast. There was no significant difference between the two groups, second-look US lesions and initial US lesions, in the percentage of ipsilateral or contralateral lesions (p = 0.702), or in the percentage of same quadrant lesions (p = 0.167). Overall on MRI, 31(79 %) of the 39 lesions presented as mass lesions and 8 (21 %) presented as non-mass-like enhancements. There was no significant difference between the two groups on MRI presentation (p = 0.878).
which were later accurately identified by second-look US. Possible reasons for failed detection on initial US are: first, the lesions could have been too subtle; second, they could have been too small to be detected on initial US; and finally, they may have been neglected because of a lack of attention from the US operator. By comparing the US findings of synchronous malignant initial US lesions with second-look US lesions, we hoped to identify different US findings that would help increase the sensitivity of the initial US and preoperative evaluation of breast cancer patients, especially in those who did not undergo MRI. Mammographically occult malignancy detected on US is known as a small but invasive lesion with less typical US findings of malignancy. The probability of malignancy has been reported to be much higher in synchronous nodules found in patients with breast cancer, even when there are no suspicious US findings [16]. Kim et al. found that the rate of malignancy among BIRADS category 3 nodules found in breast cancer patients was 11.4 %, and that 90.7 % of malignant nodules were mammographically occult [22]. Furthermore, the risk of malignancy was related to the distance from the primary malignant tumor [22]. Our results also show that synchronous nodules detected on initial US tend to show less malignant features (no suspicious findings in 5.6 %, one of 18; only one suspicious finding in 33.3 %, 6 of 18) and are mostly located in the ipsilateral breast (83.3 %, 15 of 18), and especially in the same quadrant (44.4 %, 8 of 18), all of which is consistent with previous literature. There have also been several studies reporting the US features of MR-detected malignancy on second-look US [15, 18]. Abe et al. [15] found that malignant breast lesions initially detected on MRI tended to be subtle and that the classic malignant US findings were often absent, with 33 % showing no suspicious US features. However,
Discussion !
Today, whole-breast ultrasound is increasingly included in the preoperative evaluation of breast cancer patients, as it aids the detection of mammographically occult additional lesions [1, 2]. Although breast MRI has been shown to depict synchronous lesions even after whole-breast US was performed [3], there are limitations to its routine use in all breast cancer patients due to its high cost and low accessibility for biopsy. In our study, all included patients underwent initial bilateral whole-breast US examination before breast MRI. Therefore, we could assume that all second-look US lesions were not visualized during the initial US examination, i. e., initial US could not identify existing lesions
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
437
Downloaded by: University of Queensland. Copyrighted material.
Original Article
Original Article
only 76 of the 148 patients included in their study underwent US before breast MRI and therefore some lesions could have been initially detected on US. In a different study in which all patients underwent initial US, Laguna et al. [18] also reported that classic malignant US findings were not often present with only 4 – 8 % showing posterior acoustic shadowing. Our study also showed similar results – 19 % (4/21) of second-look US lesions showed no suspicious US findings, which was a higher percentage than that seen in initial US lesions. Also 33.3 % (7/21) of cases showed only one suspicious finding. A non-circumscribed margin was the only suspicious US feature in 28.6 % (6/21) of cases. Second-look US lesions showed a higher percentage of parallel-oriented lesions (57 % vs. 39 %) and isoechoic lesions (71 % vs. 44 %). These show the same trend as those of Abe in which among 33 malignant lesions on second-look US, 61 % were parallel-oriented and 36 % were isoechoic. Interestingly, all second-look US lesions showed no posterior acoustic features (100 % vs. 78 %). Posterior shadowing aids lesion identification, and its absence may be one of the reasons why the lesions were not detected in the initial US examination. In such cases, the use of tissue harmonic imaging may be helpful in lesion detection by its amplification of posterior acoustic features [23, 24]. In our study, 76 % of second-look US lesions showed a non-circumscribed margin which is similar to the 73 % reported by Abe, but there was no significant difference when compared with initial US lesions (82 %). We also found that a higher percentage of second-look US lesions showed no calcification on US (100 % vs. 94 %), but none of these differences were statistically significant. However, second-look US lesions showed a significantly higher proportion of lesions with circumscribed or indistinct margins: 11 (53 %) second-look US lesions vs. 3 (16 %) initial US lesions (p = 0.042). A possible explanation is that circumscribed lesions may have been considered benign on initial US, or were hard to distinguish from the background due to combined isoechogenicity. Additionally, lesions with indistinct margins may have been difficult to initially detect. Interestingly, second-look US lesions were either subareolar or relatively far (> 5 cm) from the nipple compared to initial US lesions (p = 0.048). Subareolar lesions may have been easy to miss on initial US due to obscuration by nipple shadows, which is well known as one of the pitfalls of breast US [25]. Also, lesions relatively far from the nipple may have been missed due to the complacency of the examiner as he or she reached the periphery of the breast. Furthermore, initial US lesions showed a higher percentage of same quadrant lesions compared with secondlook US lesions (p = 0.167). Second-look US lesions also tended to be smaller in size, with a mean size of 7.84 mm (p = 0.134). This study demonstrates that additional malignant breast lesions detected on both initial US and second-look US tend to be subtle and often do not show classic malignant findings. As both groups show subtle US features, the clinical significance of a considerable number of the obtained results in this study is questionable. Rather, our study will have clinical value with respect to reviewing which aspects need improvement in the initial US examination – emphasizing the need for meticulous examination on initial preoperative bilateral whole-breast US, including the subareolar area and the periphery of the breast. Our study has several limitations. First, the number of included lesions was relatively small, with an overall number of 39 lesions. In addition, there were a slightly larger number of second-look US lesions than initial US lesions, which does not accurately represent the actual clinical setting where a much larger number of additional malignant lesions are detected on initial US. This is
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
probably because we only included patients who underwent preoperative breast MRI. In cases where obvious suspicious synchronous lesions were detected on initial US, the treatment plan would have been changed from breast-conserving surgery to mastectomy and preoperative breast MRI would not have been performed. Also, our study reviewed US features of only pathologically confirmed malignant lesions and therefore a comparison with benign lesions was not performed. Thus, this study may be of less value in the differentiation between benign and malignant additional lesions. However, we believe that understanding the US features of additional malignant lesions will essentially help in the determination of malignancy. As both groups showed subtle US features, our study supports findings from previous studies that a lower threshold must be applied in the case of a known index cancer when suspecting malignancy [15, 18, 22]. As our study focused on identifying US findings of additional malignant lesions, all MR-detected lesions naturally showed a US correlate. Hence this study would be of little value in the management of MR-detected lesions without a US correlate. However, similar to previous studies [5, 11, 12, 14 – 16], we also found that 80 % of second-look lesions presented as mass-like lesions on MRI. This supports the current literature that mass-like lesions on MRI are more likely to be visualized on US. In summary, we found that synchronous malignant breast lesions detected on both initial US and second-look US tend to be subtle and often do not show classic malignant findings. Second-look US lesions show more subtle findings of posterior acoustic features and margins and tend to be subareolar or relatively far from the nipple. Therefore, meticulous examination on initial preoperative US with special attention to the subareolar area and to the periphery of the breast, with simultaneous awareness of the above findings, needs to be emphasized to ensure thorough evaluation for possible additional malignant lesions.
Reference 01 Berg WA, Gilbreath PL. Multicentric and multifocal cancer: wholebreast US in preoperative evaluation. Radiology 2000; 214: 59 – 66 02 Moon WK, Noh DY, Im JG. Multifocal, multicentric, and contralateral breast cancers: bilateral whole-breast US in the preoperative evaluation of patients. Radiology 2002; 224: 569 – 576 03 Berg WA, Gutierrez L, NessAiver MS et al. Diagnostic accuracy of mammography, clinical examination, US, and MR imaging in preoperative assessment of breast cancer. Radiology 2004; 233: 830 – 849 04 Pediconi F, Miglio E, Telesca M et al. Effect of preoperative breast magnetic resonance imaging on surgical decision making and cancer recurrence rates. Invest Radiol 2012; 47: 128 – 135 05 Luciani ML, Pediconi F, Telesca M et al. Incidental enhancing lesions found on preoperative breast MRI: management and role of secondlook ultrasound. Radiol Med 2011; 116 (6): 886 – 904 06 Gutierrez RL, DeMartini WB, Silbergeld JJ et al. High cancer yield and positive predictive value: outcomes at a center routinely using preoperative breast MRI for staging. Am J Roentgenol Am J Roentgenol 2011; 196: W93 – 99 07 Sohn YM, Kim MJ, Kim EK. Impact of Preoperative Bilateral Whole Breast Sonography in Patients with Invasive Lobular Carcinoma: Results from Two Medical Centers. Ultraschall in Med 2012 08 Girardi V, Carbognin G, Camera L et al. Multifocal, multicentric and contralateral breast cancers: breast MR imaging in the preoperative evaluation of patients with newly diagnosed breast cancer. Radiol Med 2011; 116: 1226 – 1238 09 LaTrenta LR, Menell JH, Morris EA et al. Breast lesions detected with MR imaging: utility and histopathologic importance of identification with US. Radiology 2003; 227: 856 – 861 10 Shin JH, Han BK, Choe YH et al. Targeted ultrasound for MR-detected lesions in breast cancer patients. Korean J Radiol 2007; 8: 475 – 483
Downloaded by: University of Queensland. Copyrighted material.
438
Original Article
18 Laguna AD, Arranz SJ, Checa VQ et al. Sonographic findings of additional malignant lesions in breast carcinoma seen by second look ultrasound. J Clin Imaging Sci 2011; 1: 34 19 American College of Radiology. Breast Imaging Reporting and Data System (BI -RADS). 4th ed. Reston, VA: American College of Radiology; 2003 20 Hille H, Vetter M, Hackeloer BJ. The accuracy of BI-RADS classification of breast ultrasound as a first-line imaging method. Ultraschall in Med 2012; 33: 160 – 163 21 Park VY, Kim MJ, Kim EK et al. How To Do Second-Look Ultrasound For Finding Unexpected Lesions On Breast MRI Radiographics. This article is supposed to be published in July 2013 22 Kim SJ, Ko EY, Shin JH et al. Application of sonographic BI-RADS to synchronous breast nodules detected in patients with breast cancer. Am J Roentgenol Am J Roentgenol 2008; 191: 653 – 658 23 Oktar SO, Yucel C, Ozdemir H et al. Comparison of conventional sonography, real-time compound sonography, tissue harmonic sonography, and tissue harmonic compound sonography of abdominal and pelvic lesions. Am J Roentgenol Am J Roentgenol 2003; 181: 1341 – 1347 24 Mesurolle B, Helou T, El-Khoury M et al. Tissue harmonic imaging, frequency compound imaging, and conventional imaging: use and benefit in breast sonography. J Ultrasound Med 2007; 26: 1041 – 1051 25 Stavros AT. Breast ultrasound equipment requirements. Breast ultrasound. Philadelphia: Lippincott Williams & Wilkins; 2003
Downloaded by: University of Queensland. Copyrighted material.
11 Wiratkapun C, Duke D, Nordmann AS et al. Indeterminate or suspicious breast lesions detected initially with MR imaging: value of MRI-directed breast ultrasound. Acad Radiol 2008; 15: 618 – 625 12 Demartini WB, Eby PR, Peacock S et al. Utility of targeted sonography for breast lesions that were suspicious on MRI. Am J Roentgenol Am J Roentgenol 2009; 192: 1128 – 1134 13 Destounis S, Arieno A, Somerville PA et al. Community-based practice experience of unsuspected breast magnetic resonance imaging abnormalities evaluated with second-look sonography. J Ultrasound Med 2009; 28: 1337 – 1346 14 Meissnitzer M, Dershaw DD, Lee CH et al. Targeted ultrasound of the breast in women with abnormal MRI findings for whom biopsy has been recommended. Am J Roentgenol Am J Roentgenol 2009; 193: 1025 – 1029 15 Abe H, Schmidt RA, Shah RN et al. MR-directed (“Second-Look”) ultrasound examination for breast lesions detected initially on MRI: MR and sonographic findings. Am J Roentgenol Am J Roentgenol 2010; 194: 370 – 377 16 Carbognin G, Girardi V, Calciolari C et al. Utility of second-look ultrasound in the management of incidental enhancing lesions detected by breast MR imaging. Radiol Med 2010; 115: 1234 – 1245 17 Candelaria R, Fornage BD. Second-look US examination of MR-detected breast lesions. J Clin Ultrasound 2011; 39: 115 – 121
439
Park VY et al. Additional Malignant Breast … Ultraschall in Med 2014; 35: 432–439
