Abstract
Objective:
To compare the incremental cancer detection rate (ICDR) using bilateral whole-breast ultrasonography (BWBUS) vs dynamic contrast-enhanced MRI in patients with primary breast cancer.
Methods:
A retrospective database search in a single institution identified 259 patients with breast cancer diagnosed from January 2011 to August 2014 who underwent mammography, BWBUS and MRI before surgery. Patient characteristics, tumour characteristics and lesions seen on each imaging modality were recorded. The sensitivity, specificity and accuracy for each modality were calculated. ICDRs according to index tumour histology and receptor status were also evaluated. The effect of additional cancer detection on surgical planning was obtained from the medical records.
Results:
A total of 266 additional lesions beyond 273 index malignancies were seen on at least 1 modality, of which 121 (45%) lesions were malignant and 145 (55%) lesions were benign. MRI was significantly more sensitive than BWBUS (p = 0.01), while BWBUS was significantly more accurate and specific than MRI (p < 0.0001). Compared with mammography, the ICDRs using BWBUS and MRI were significantly higher for oestrogen receptor-positive and triple-negative cancers, but not for human epidermal growth factor receptor 2-positive cancers. 22 additional malignant lesions in 18 patients were seen on MRI only. Surgical planning remained unchanged in 8 (44%) of those 18 patients.
Conclusion:
MRI was more sensitive than BWBUS, while BWBUS was more accurate and specific than MRI. MRI-detected additional malignant lesions did not change surgical planning in almost half of these patients.
Advances in knowledge:
BWBUS may be a cost-effective and practical tool in breast cancer staging.
INTRODUCTION
The accurate evaluation of the extent of locoregional disease in patients with newly diagnosed breast cancer is crucial for treatment planning, as it determines whether the patients are candidates for breast-conserving therapy. Diagnostic mammography remains the standard of care for the initial diagnosis and staging of breast cancer. However, the sensitivity of mammography in dense breasts has been reported to be as low as 30–48%.1,2 Two adjunct imaging modalities commonly used to stage breast cancer are ultrasonography and dynamic contrast-enhanced (DCE) MRI. Pre-operative bilateral whole-breast ultrasonography (BWBUS) has been shown to reveal mammographically occult malignant foci in 14% of ipsilateral breasts and 4% of contralateral breasts.3 Numerous studies have also shown that MRI improves the detection of occult tumour foci in the ipsilateral and contralateral breasts in patients with newly diagnosed breast cancer, leading to changes in surgical planning.4–15 Systematic reviews of the literature found that pre-operative MRI detected additional ipsilateral disease in 16–20% of patients and occult synchronous contralateral cancer in 4–5% of patients.16–18
Despite the widespread use of breast MRI for pre-operative staging in recent years, this use remains controversial. There is a lack of evidence of long-term benefits from routine breast MRI for pre-operative staging.19–23 Two randomized controlled trials [comparative effectiveness in MRI in breast cancer (COMICE) and MR mammography of nonpalpable breast tumours (MONET)] have shown that adding breast MRI to conventional assessment (clinical examination, diagnostic mammography and breast ultrasonography) did not significantly reduce the reoperation rate and, rather, increased the re-excision rate in patients with non-palpable breast cancer.19,20 In addition, the routine use of breast MRI for pre-operative staging has been associated with increased ipsilateral and bilateral mastectomy rates, leading to concerns of overdiagnosis and overtreatment. Furthermore, adding breast MRI to the staging work-up may increase cost and patient anxiety and cause undue delays in definitive treatment.24–27 While the standard policy is to complete a breast MRI examination and/or MRI-guided breast biopsy within 5 working days in pre-treatment evaluation at our institution, there have been reports in the literature that pre-operative breast MRI led to 22–30-day delays to surgery.24,27
The objective of this study was to compare DCE MRI with BWBUS for the incremental detection of breast cancer beyond mammography in patients with biopsy-proven primary breast cancer.
METHODS AND MATERIALS
Patient selection
Approval was obtained from our institutional review board and informed consent was waived. We retrospectively searched the radiology patient database at our institution to identify patients who were diagnosed with primary breast carcinoma during the period from January 2011 to August 2014.
The inclusion criteria were: (1) biopsy-proven primary breast carcinoma and (2) having undergone bilateral diagnostic mammography, staging BWBUS including regional nodal basins and staging bilateral DCE breast MRI within a 30-day period before neoadjuvant chemotherapy or surgery at our institution. Patients who received unilateral breast ultrasonography for staging or who did not receive bilateral DCE breast MRI before treatment were excluded. Patients who had breast MRI after neoadjuvant chemotherapy or surgery were also excluded.
Imaging techniques
At our institution, the standard imaging protocol for newly diagnosed breast cancer in current use is the same as that in use during the study period. This protocol includes bilateral full-field digital diagnostic mammography and staging whole-breast ultrasonography including regional nodal basins, performed by our dedicated breast radiologists and sonographers. The staging whole-breast ultrasonography is typically unilateral. BWBUS is performed for staging if the patient has heterogeneously dense or extremely dense breast parenchyma or if the patient has mammographic findings in the contralateral breast requiring sonographic evaluation. Bilateral DCE breast MRI is performed for staging for any of the following indications, per our imaging protocol: invasive lobular carcinoma, dense breast parenchyma posing difficulty in the evaluation of disease extent on mammography and ultrasonography high familial risk or genetic predisposition and suspected chest wall involvement.
During the study period, bilateral digital mammography was performed using dedicated mammography units (Hologic, Marlborough, MA). Routine craniocaudal and mediolateral oblique views of the breasts and spot views or spot magnification views of the areas of interest were obtained. Three-dimensional tomosynthesis was not routinely used in the diagnostic setting during the study period.
BWBUS was performed by one of the dedicated breast sonographers under the supervision of a breast radiologist. There were 10 dedicated breast sonographers and 17 breast radiologists at our institution during the study period. Real-time ultrasonography was performed with either Siemens ultrasound units (Munich, Germany) using 13.5-MHz or Philips Medical Systems ultrasound units (Cleveland, OH) using 12–18-MHz high-frequency linear-array probes. The patients were scanned in the supine position for medial breast and in the contralateral posterior oblique position with the ipsilateral arm raised above the head for lateral breast. The regional nodal basins, including the axillary, infraclavicular and internal mammary regions, were evaluated per the ultrasound imaging protocol at our institution. In selected cases, supraclavicular nodal basins were also assessed if suspicious lymph nodes were seen in the ipsilateral infraclavicular region. All significant discrete lesions (e.g. solid masses and complex or complicated cysts) on ultrasonography were measured in transverse and longitudinal orientations and the distance from each discrete lesion to the nipple was documented. Other significant findings such as shadowing, dilated ducts, oedema and skin thickening were also recorded.
Breast MRI was performed using a 3.0-T or 1.5-T magnet (General Electric Medical Systems, Milwaukee, WI) with a dedicated 16-channel phased-array breast coil with the patients in the prone position. Our MRI protocol includes DCE images (one pre-contrast run and five post-contrast runs in the sagittal plane using a fat-suppressed T1 weighted three-dimensional gradient-echo sequence), fat-suppressed T2 weighted fast spin-echo images in the sagittal plane and delayed post-contrast images in the axial plane. Gadopentetate dimeglumine (Magnevist®; Schering, Berlin, Germany) at 0.1 mmol kg−1 of body weight was given using a power injector followed by a saline flush. In a few patients with renal impairment, a half dose of gadobenate dimeglumine (MultiHance®; Bracco Diagnostics, Monroe Township, NJ) was used.
The images from diagnostic mammography, BWBUS and DCE MRI were interpreted by one of our specialized breast radiologists. BWBUS was interpreted with knowledge of the diagnostic mammographic findings and any outside imaging studies. DCE MRI was interpreted with knowledge of the diagnostic mammographic findings, BWBUS findings and any outside imaging studies.
Biopsy techniques
All index cancers were sampled via core-needle biopsy techniques. Ultrasonography guidance was used whenever possible. The majority of patients were biopsied and a cancer diagnosis was rendered at an outside facility before referral to our tertiary cancer centre. In these patients, the outside pathology slides were reviewed by dedicated breast pathologists and the cancer diagnoses were confirmed.
An additional lesion considered suspicious for malignancy on any of the three imaging modalities was biopsied if a malignant result would change the surgical planning. Additional lesions without histopathologic correlation were excluded from the analysis, including those lesions not biopsied before the initiation of neoadjuvant therapy.
132 additional lesions were biopsied via core biopsy technique using ultrasonography, stereotactic or MRI guidance. 95 of the additional lesions were biopsied via fine-needle aspiration technique using ultrasonography guidance. 39 additional lesions were not biopsied prior to mastectomy. The diagnoses for these 39 lesions were confirmed by examining the mastectomy specimen and correlating with imaging findings by size, o'clock position and distance to the nipple.
An additional lesion that was seen only on mammography or only on MRI was sampled via core biopsy using stereotactic or MRI guidance, respectively. When an additional suspicious lesion was seen on BWBUS or MRI-directed ultrasonography, it was typically sampled via ultrasonography-guided fine-needle aspiration biopsy (FNA) unless a highly suspicious additional lesion was found in the contralateral breast or the on-site cytopathologist deemed that core biopsy was necessary for histopathological diagnosis after FNA. Furthermore, when a suspicious lymph node was seen on BWBUS, it was sampled via ultrasonography-guided FNA. FNA of additional lesions and suspicious lymph nodes is coupled with on-site cytopathology reporting to facilitate treatment planning for many patients who travel from great distances or from out of state for same-day imaging and tertiary referral clinic appointments.
Ultrasonography-guided core biopsy was performed using a 14- to 18-gauge automated biopsy device (Bard Urological Division, Covington, GA). A minimum of four samples were obtained from each lesion. Ultrasonography-guided FNA was performed using a 2-inch-long 21-gauge needle. Typically, 1–2 passes were made. The samples from FNAs were evaluated by dedicated on-site cytopathologists. Marker clips (Bard Urological Division, Covington, GA) were placed in the biopsy cavities after core-needle biopsies or when FNA of the additional lesions revealed malignancy.
Stereotactic biopsy was performed for lesions seen only on mammography, such as calcifications. The patient was placed prone on a digital stereotactic table (Hologic, Marlborough, MA) and the lesion was targeted with mammographic guidance using stereotactic technique. A 9- to 11-gauge needle (Suros ATEC; Hologic, Marlborough, MA) was used to obtain multiple cores. Specimen radiography was performed to confirm retrieval of calcifications. A marker clip (Hologic, Marlborough, MA) was then placed in the biopsy cavity.
When a lesion was initially seen on breast MRI only, MRI-directed ultrasonography was performed unless the interpreting breast radiologist felt that the second-look ultrasonography would likely not yield additional information, typically for small non-mass enhancement (NME) lesions. If a correlating lesion was found on second-look ultrasonography, it was sampled under ultrasonography guidance as described above. If second-look ultrasonography did not reveal a correlating lesion, the lesion was sampled under MRI guidance using a dedicated breast biopsy coil. A marker clip (Hologic, Marlborough, MA) was then placed in the biopsy cavity.
Marker clip location was routinely confirmed on post-biopsy mammograms in craniocaudal and lateral projections. Any migration of the marker clip from the biopsy cavity was recorded.
Data collection
The medical records of each patient including imaging, imaging reports, clinical notes and pathology reports were reviewed. Patient demographics, including age at diagnosis, sex and race, were recorded. Index tumour characteristics, including histologic type (e.g. ductal carcinoma in situ, invasive ductal carcinoma, invasive lobular carcinoma or mucinous carcinoma), histologic grade and receptor status [i.e. oestrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (HER2)], were also recorded.
All imaging studies including mammography, ultrasonography and MRI were reviewed. Breast composition seen on mammography was reclassified according to the breast imaging-reporting and data system (BI-RADS) 5 lexicon28 (almost entirely fatty, scattered areas of fibroglandular density, heterogeneously dense or extremely dense).
The size and imaging features of each index cancer and each additional lesion considered suspicious (BI-RADS category 4 or 5) on mammography, BWBUS and DCE MRI were recorded according to the BI-RADS 5 lexicon28 and correlated among modalities. The histopathology of each additional lesion was correlated with imaging findings to determine the performance of each modality and combination of modalities. An additional lesion was considered true-positive result if the biopsy or surgery revealed malignancy at that site and was considered false-positive result if the biopsy or surgery revealed a benign lesion. There were seven lesions with high-risk histopathology (including four lesions with atypical ductal hyperplasia and three lesions with atypical lobular hyperplasia) results after stereotactic or MRI-guided biopsies were performed for additional suspicious lesions. These high-risk histopathology results were considered benign. These seven high-risk lesions in ipsilateral or contralateral breasts were considered benign because no malignancy was identified at the sites of these lesions in the mastectomy specimen or at follow-up ranging from 1 to 4 years.
Multifocal, multicentric and contralateral disease seen on each imaging modality was recorded. Multifocal disease was defined as two or more separate sites of breast cancer in the same breast at least 0.5 cm but <4 cm apart typically in the same duct system. Multicentric disease was defined as two or more separate sites of breast cancer where the greatest distance between two sites was at least 4 cm, typically in different breast quadrants.29,30 There were 36 patients with multifocal/multicentric disease present on mammography before BWBUS or MRI. Surgical planning at the first surgeon–patient interaction and the final surgical procedure performed were compared to determine whether malignant lesions detected by DCE MRI only altered the surgical option (segmentectomy vs mastectomy).
Statistical analysis
All statistical analyses were performed by our dedicated statistician using SAS® v. 9 software (SAS Institute Inc., Cary, NC). Descriptive statistics were calculated for patient age; race; breast composition; index tumour characteristics including histologic type, grade and receptor status; the presence of multifocal, multicentric or bilateral disease; and the index tumour histology and receptor status of additional malignant lesions. The sensitivity, specificity, positive-predictive value, negative-predictive value and accuracy of mammography, BWBUS, DCE MRI, mammography plus BWBUS and mammography plus DCE MRI were calculated. Accuracy indicates true-positive and true-negative lesions depicted by an imaging modality divided by the entire number of lesions. Sensitivity, specificity and accuracy in detecting additional lesions were compared between the modalities and combinations of modalities using the McNemar test. The incremental cancer detection rate (ICDR) was compared between modalities in ER-positive, HER2-positive and triple-negative index tumours using the McNemar test or signed-rank test. The ICDR based on magnet strength on MRI was compared using Fisher's exact test.
RESULTS
A total of 259 patients with 273 primary breast carcinomas were included in the study. 14 patients presented with bilateral breast cancer. All patients were female. Patient demographic characteristics and index tumour characteristics are summarized in Table 1.
Table 1.
Demographic characteristics of 259 patients and tumour characteristics in 273 breasts in those patients
| Variable | Number (%) |
|---|---|
| Age, mean ± standard deviation | 52.1 years ± 12.2 yearsa |
| Race | |
| African-American | 25 (9.7)a |
| Hispanic | 47 (18.1)a |
| White | 160 (61.8)a |
| Other | 27 (10.4)a |
| Breast composition | |
| Fatty | 2 (0.8)a |
| Scattered density | 32 (12.4)a |
| Heterogeneously dense | 209 (80.7)a |
| Extremely dense | 16 (6.2)a |
| Histologic type | |
| IDC | 171 (62.6)b |
| DCIS | 23 (8.4)b |
| ILC/IMC | 71 (26.0)b |
| Other | 8 (2.89)b |
| Histology grade | |
| 1 | 45 (16.5)b |
| 2 | 124 (45.4)b |
| 3 | 101 (37.0)b |
| Unknown | 3 (1.1)b |
| ER status | |
| Negative | 62 (22.7)b |
| Positive | 205 (75.1)b |
| Unknown | 6 (2.2)b |
| PR status | |
| Negative | 98 (35.9)b |
| Positive | 168 (61.5)b |
| Unknown | 7 (2.6)b |
| HER2 status | |
| Negative | 209 (76.6)b |
| Positive | 43 (15.8)b |
| Unknown | 21 (7.6)b |
| Triple-negative status | |
| No | 210 (76.9)b |
| Yes | 41 (15.0)b |
| Unknown | 22 (8.1)b |
DCIS, ductal carcinoma in situ; ER, oestrogen receptor; HER2, human epidermal growth factor receptor 2; IDC, invasive ductal carcinoma; ILC/IMC, invasive lobular carcinoma/invasive mammary carcinoma with mixed ductal and lobular features; PR, progesterone receptor;.
Data are in numbers (%) except where otherwise indicated.
n = 259 patients.
n = 273 breasts.
A total of 266 additional lesions in the ipsilateral or contralateral breasts in addition to the 273 index malignancies were seen on at least one modality (mammography, BWBUS or DCE MRI) pre-operatively. Of these, 121 (45%) lesions were malignant and 145 (55%) lesions were benign at histopathologic analysis. Table 2 summarizes the performance of each modality and combination of modalities in detecting additional malignant lesions. The sensitivities of mammography, BWBUS and DCE MRI were 36, 71 and 85%, respectively, and their specificities were 76, 68 and 39%, respectively. The accuracies of mammography, BWBUS and DCE MRI were 58, 69 and 60%, respectively. Two malignant lesions not shown by any imaging modality were incidentally found in the surgical specimens. No imaging correlate was identified for these two imaging occult lesions at the time of imaging interpretation or in retrospect. BWBUS alone had the highest accuracy, at 69%, of all the modalities and modality combinations in incremental cancer detection. The accuracy of BWBUS was 83% if counting both index tumour and additional lesions.
Table 2.
Diagnostic performance of imaging modalities in the 266 additional lesions
| Modality | Sensitivity | Specificity | PPV | NPV | Accuracy |
|---|---|---|---|---|---|
| Mammography | 43/121 (35.5) | 110/145 (75.9) | 43/78 (55.1) | 110/188 (58.5) | 153/266 (57.5) |
| BWBUS | 86/121 (71.1) | 98/145 (67.6) | 86/133 (64.5) | 98/133 (73.7) | 184/266 (69.2) |
| DCE MRI | 103/121 (85.1) | 57/145 (39.3) | 103/191 (53.4) | 57/75 (76.0) | 160/266 (60.2) |
| Mammography and BWBUS | 95/121 (78.5) | 72/145 (49.7) | 95/168 (56.5) | 72/98 (73.5) | 167/266 (62.8) |
| Mammography and DCE MRI | 112/121 (92.6) | 30/145 (20.7) | 112/227 (49.3) | 30/39 (76.9) | 142/266 (53.4) |
BWBUS, bilateral whole-breast ultrasonography; DCE, dynamic contrast-enhanced; NPV, negative-predictive value; PPV, positive-predictive value.
The number in parenthesis after the number of lesions in each cell indicates percentage.
Both BWBUS and DCE MRI had higher sensitivity for the detection of additional malignant lesions than mammography alone (p < 0.0001). DCE MRI had higher sensitivity than BWBUS (p = 0.01); and mammography plus DCE MRI was more sensitive than mammography plus BWBUS (p = 0.003). On the other hand, BWBUS was more accurate and specific than DCE MRI (p < 0.0001); and mammography plus BWBUS was also more accurate and specific than mammography plus DCE MRI (p < 0.0001). There was no significant difference in ICDR on DCE MRI based on magnet strength (p = 0.21). 10 additional malignant lesions in 8 patients were identified on the 155 MRIs performed on 3.0-T magnet. 12 additional malignant lesions in 10 patients were identified on the 104 MRIs performed on 1.5-T magnet.
Characteristics of the 121 additional malignant lesions are summarized in Table 3 according to the index tumour histologic subtype, receptor status and the imaging modalities by which the additional lesions were detected. Most additional cancer lesions were found in the breasts harbouring primary invasive ductal carcinoma [82/121 (68%) lesions] and primary invasive lobular carcinoma/invasive mammary carcinoma [33/121 (27%) lesions], a histologic distribution similar to that of the index tumours (Table 1). Of the 121 additional malignant lesions, 95 (79%) lesions were associated with ER-positive index cancer, 23 (19%) lesions were associated with HER2-positive index cancer and 15 (12%) were associated with triple-negative index cancer. The ICDRs of BWBUS and DCE MRI were higher than that of mammography for ER-positive (p < 0.0001 for both BWBUS and MRI) and triple-negative index cancers (p = 0.02 for BWBUS and p = 0.003 for MRI). However, the ICDRs did not significantly differ between the three imaging modalities for HER2-positive index cancers.
Table 3.
Histologic types and receptor statuses of index tumours with additional malignant lesions according to imaging modality
| Index tumour characteristic | Number of additional malignant lesions (n = 121) | Detected on mammography (number) (%) | Detected on BWBUS (number) (%) | Detected on DCE MRI (number) (%) |
|---|---|---|---|---|
| Histologic type | ||||
| IDC | 82 | 34/82 (41.5) | 58/82 (70.7) | 72/82 (87.8) |
| DCIS | 4 | 0/4 (0.0) | 2/4 (50.0) | 4/4 (100.0) |
| ILC/IMC | 33 | 8/33 (24.2) | 24/33 (72.7) | 25/33 (75.8) |
| Other | 2 | 1/2 (50.0) | 2/2 (100.0) | 2/2 (100.0) |
| ER status | ||||
| Negative | 23 | 9/23 (39.1) | 16/23 (69.6) | 23/23 (100.0) |
| Positive | 95 | 34/95 (35.8) | 70/95 (73.7) | 77/95 (81.1) |
| Unknown | 3 | 0/3 (0.0) | 0/3 (0.0) | 3/3 (100.0) |
| PR status | ||||
| Negative | 32 | 16/32 (50.0) | 26/32 (81.3) | 29/32 (90.6) |
| Positive | 85 | 27/85 (31.8) | 59/85 (69.4) | 70/85 (82.4) |
| Unknown | 4 | 0/4 (0.0) | 1/4 (25.0) | 4/4 (100.0) |
| HER2 status | ||||
| Negative | 91 | 29/91 (31.9) | 67/91 (73.6) | 78/91 (85.7) |
| Positive | 23 | 13/23 (56.5) | 16/23 (69.6) | 18/23 (78.3) |
| Unknown | 7 | 1/7 (14.3) | 3/7 (42.9) | 7/7 (100.0) |
| Triple-negative status | ||||
| Yes | 15 | 6/15 (40.0) | 12/15 (80.0) | 15/15 (100.0) |
| No | 98 | 36/98 (36.7) | 70/98 (71.4) | 80/98 (81.6) |
| Unknown | 8 | 1/8 (12.5) | 4/8 (50.0) | 8/8 (100.0) |
| Total | 121 | 43 | 86 | 103 |
BWBUS, bilateral whole-breast ultrasonography; DCE, dynamic contrast-enhanced; DCIS, ductal carcinoma in situ; ER, oestrogen receptor; HER2, human epidermal growth factor receptor 2; IDC, invasive ductal carcinoma; ILC/IMC, invasive lobular carcinoma/invasive mammary carcinoma with mixed ductal and lobular features; PR, progesterone receptor.
There were 22/121 (18%) additional malignant lesions identified by DCE MRI only in 18 patients. The additional malignant lesions were ipsilateral in 15 patients and contralateral in 3 patients. At the same time, 72/145 (50%) additional biopsies were performed on benign lesions seen on DCE MRI only. The size, imaging features, histologic type and grade of the DCE MRI-detected additional malignant lesions are summarized in Table 4. Of these 22 lesions, 13 (59%) lesions were masses (size range 0.6–1.2 cm, mean 0.8 cm) and 9 (41%) lesions were NMEs (size range 0.9–8.3 cm, mean 3.3 cm). Of the 13 masses, 10 (77%) masses were invasive carcinoma (ductal, lobular or mixed ductal and lobular; grade 1–2) and 3 (23%) masses were ductal carcinoma in situ (grade 1–3). In contrast, 6 (67%) of the nine NMEs were ductal carcinoma in situ (grade 1–3); the other 3 (33%) NMEs were invasive carcinoma (ductal or mixed ductal and lobular; grade 1–2).
Table 4.
Characteristics of the 22 lesions detected by dynamic contrast-enhanced MRI only in 18 patients
| Lesion | Size (cm) | Imaging feature | Biopsy method/imaging guidance | Histologic type/grade | Disease extent | Change in surgery | Final surgery |
|---|---|---|---|---|---|---|---|
| 1 | 0.8 | Mass | CN/ultrasonography | IDC/1 | Contralateral | No | BM |
| 2 | 0.7 | Mass | FNA/ultrasonography | IDC/1 | Contralateral | No | BM |
| 3 | 0.6 | Mass | Surgery | IDC/1 | Contralateral | No | BM |
| 4 | NA | NME | CN/MRI | DCIS/3 | MF | Yes | UM |
| 5 | 0.9 | Mass | Surgery | ILC/1 | MC | No | UM |
| 6 | 0.7 | Mass | Surgery | ILC/1 | MC | No | UM |
| 7 | 1.2 | NME | CN/MRI | IMC/1 | MC | Yes | DL |
| 8 | 8 | NME | Surgery | DCIS/3 | MC | No | UM |
| 9 | 1.2 | NME | CN/MRI | DCIS/3 | MF | Yes | SL |
| 10 | 1.2 | Mass | Surgery | IDC/1 | MC | No | BM |
| 11 | 1 | Mass | Surgery | IDC/1 | MC | No | BM |
| 12 | 0.6 | Mass | FNA/ultrasonography, surgery | DCIS/3 | MF | Yes | DL |
| 13 | 0.6 | Mass | CN/MRI | DCIS/1 | MC | Yes | BM |
| 14 | 2 | NME | CN/ultrasonography | DCIS/1 | Contralateral | Yes | BM |
| 15 | 0.9 | NME | CN/MRI | DCIS/3 | MF | No | BM |
| 16 | 8.3 | NME | CN/MRI | DCIS/2 | MC | Yes | BM |
| 17 | 1 | Mass | CN/MRI | IDC/2 | Contralateral | Yes | UM + SL |
| 18 | 0.7 | Mass | CN/MRI | DCIS/2 | MF | No | BM |
| 19 | 0.8 | Mass | CN/MRI | IMC/2 | MF | Yes | UM |
| 20 | 0.6 | Mass | Surgery | IMC/2 | MC | No | BM |
| 21 | 3.1 | NME | CN/MRI | IDC/1 | MC | No | UM |
| 22 | 1.8 | NME | CN/ultrasonography | IMC/2 | MC | Yes | SL |
BM, bilateral mastectomy; CN, core-needle biopsy; DCIS, ductal carcinoma in situ; DL, double lumpectomy; FNA, fine-needle aspiration biopsy; IDC, invasive ductal carcinoma; ILC, invasive lobular carcinoma; IMC, invasive mammary carcinoma with mixed ductal and lobular features; MC, multicentric; MF, multifocal; NME, non-mass enhancement; SL, single lumpectomy; UM, unilateral mastectomy.
Lesions 1–3 were from the same breast, lesions 5–6 were from the same breast and lesions 10–11 were from the same breast.
Disease extent (multifocal, multicentric and contralateral), change in surgical plans based on imaging and final surgical procedures performed for the 22 DCE MRI-detected additional malignant lesions are summarized in Table 4. Of the 6 patients in this group with multifocal disease, 4 patients had surgical plan change as a result of the detection of additional malignant lesions. Of the 9 patients in this group with 11 multicentric lesions, 4 patients had surgical plan change because of the additional lesions. Five malignant lesions in three contralateral breasts led to surgical plan change in two patients. One patient had three DCE MRI-detected malignant lesions in the contralateral breast; however, the patient had already expressed a preference for bilateral mastectomy at the initial surgical consultation before biopsies of the contralateral breast; so, her surgical plan did not change (Figure 1). Overall, 8/18 (44%) patients did not have changes in their surgical plans despite the diagnosis of multifocal, multicentric or contralateral disease detected by DCE MRI. These eight patients included the patients who preferred bilateral mastectomy and three patients with multicentric disease who had large index lesions and small breasts (Figure 2).
Figure 1.
A 61-year-old female with newly diagnosed low-grade invasive lobular carcinoma of right breast. The patient did not receive neoadjuvant therapy before surgery. (a) Sagittal dynamic contrast-enhanced (DCE) fat-suppressed T1 weighted three-dimensional gradient-echo sequence showing a biopsy-proven 10-cm index malignancy (arrows) in right central breast causing nipple retraction. (b, c) Sagittal post-subtraction DCE fat-suppressed T1 weighted three-dimensional gradient-echo sequence showing three additional suspicious enhancing masses (arrows) in left breast, which were confirmed to be low-grade invasive ductal carcinoma via biopsies and at surgery.
Figure 2.
A 51-year-old female with newly diagnosed low-to-intermediate-grade invasive mammary carcinoma with mixed ductal and lobular features in the left breast. Patient did not receive neoadjuvant therapy before surgery. (a) Panoramic sonographic image of the left breast from 10–12 o'clock in transverse plane showing a 5-cm irregular mass with posterior shadowing (arrows), correlating with a palpable finding on clinical examination. Ultrasound-guided core-needle biopsy confirmed malignancy. (b) Axial post-processing maximum intensity projection MR image again showing 5-cm index cancer (arrows) in upper inner quadrant of left breast. (c) Axial post-contrast MR image showing a 0.6-cm enhancing mass (arrow) at 5 o'clock of the left breast with delayed washout kinetics, suspicious of another site of cancer. This was confirmed to be low-to-intermediate-grade invasive mammary carcinoma with mixed ductal and lobular features at surgery. L, left breast; LT, left breast; R, right breast; Trans, transverse plane.
93 ipsilateral axillary, 34 infraclavicular, 7 internal mammary, 10 supraclavicular and 7 contralateral axillary nodal metastases were identified on BWBUS and confirmed via ultrasonography-guided fine-needle aspiration immediately following BWBUS. Among the biopsy-proven nodal metastases, 90 ipsilateral axillary, 20 infraclavicular, 7 internal mammary and 7 contralateral axillary nodal metastases were also seen on subsequent MRI examinations. The field of view in a standard breast MRI protocol does not include the supraclavicular and often only includes part of the infraclavicular nodal basin.
DISCUSSION
Our results in this non-consecutive series of patients with newly diagnosed breast cancer support the findings of multiple prior studies in confirming the higher sensitivity of breast MRI in incremental breast cancer detection.4–15 Of the 22 malignant lesions identified only on DCE MRI, 13 (59%) lesions were invasive carcinoma and 5 (56%) lesions of the 9 DCE MRI-detected ductal carcinomas in situ were grade 3, with 1 lesion being 8 cm in size. These clinically significant lesions would have been missed if pre-operative DCE MRI had not been performed, lending some support for breast MRI in staging newly diagnosed breast cancer. A single additional malignant lesion was found on ultrasonography only. In this patient, BWBUS showed an irregular 1.3-cm malignant mass 0.5 cm from the index tumour. MRI failed to show this mass secondary to artefact from the adjacent biopsy clips. While an additional malignant lesion 0.5 cm away from the index tumour was considered multifocal at our institution, this may be considered part of the index tumour at other institutions. Also, MRI studies were interpreted with the knowledge of findings on diagnostic mammography and BWBUS in this study.
Interestingly, in nearly half of the patients with malignant lesions found only on DCE MRI in this study, surgical planning was not affected by the additional malignant lesions detected, largely because some patients opted for bilateral mastectomies and others had large index tumours but small breasts, making mastectomy the surgical option of choice regardless of the additional malignant lesions. These findings reflect the difficulties encountered in the decision-making process for patients along the cancer treatment pathway; these patients, not infrequently, switch from segmentectomy to mastectomy or vice versa when the choice is available.
A recent Canadian population-based study showed that the use of pre-operative breast MRI had increased substantially in routine clinical practice and was associated with increases in additional tests, wait time to surgery, ipsilateral mastectomy rates and contralateral prophylactic mastectomy rates.27 These results are similar to those from earlier retrospective studies from the USA.24–26 Furthermore, a meta-analysis of the effect of MRI on the surgical management of ductal carcinoma in situ did not demonstrate any improvement in surgical outcomes.31 These findings raised concerns about “overdiagnosis and overtreatment” in the absence of long-term benefits of the routine use of pre-operative breast MRI.19–23 In our series, nearly half of the patients with additional malignancy found on MRI did not have surgical plan changes, suggesting potential “overdiagnosis” but not “overtreatment”.
Notably, despite the high sensitivity of DCE MRI, BWBUS was the most accurate imaging modality in incremental cancer detection among all modalities and modality combinations. DCE MRI had a high false-positive rate, with 72/145 (50%) of additional biopsies performed for lesions detected only with DCE MRI yielding benign results. This contrasts with results published by Berg et al,4 which reported a combination of mammography and clinical examination to be the most accurate, considering the high false-positive rates of both BWBUS and MRI. The accuracy of BWBUS in our series including both index tumours and additional malignancies was 83% compared with 68% by Berg et al. The difference in numbers may be related to improved resolution of ultrasonography equipment, higher frequency transducers and the evolution of technical skill of breast ultrasound over the past two decades.
Several systematic literature reviews have reported the detection of additional ipsilateral disease in 16–20% of patients and occult synchronous contralateral cancer in 4–5% of patients with newly diagnosed breast cancer upon pre-operative MRI.16–18 In our series, only 15/259 (5.8%) of patients had additional ipsilateral malignant lesions found by DCE MRI and 3/259 (1.2%) of patients had contralateral cancers that were revealed by DCE MRI. Recently, Kim et al32 also reported a low ICDR (2.0% for the ipsilateral breast and 0.9% for the contralateral breast) in a group of 1038 patients. The authors attributed these low ICDRs to the use of good-quality BWBUS in their entire patient population. There are differences between the study by Kim et al and our study. All patients without contraindications in the study by Kim et al underwent pre-operative staging MRI, while only selected patients at our institution underwent pre-operative staging MRI per our imaging protocol. We had a disproportionately high percentage of patients with dense breasts (87%) and invasive lobular carcinoma/invasive mammary carcinoma (26%) in this study population. We suspect that if DCE breast MRI were performed on all patients for staging at our institution, the ICDR of DCE MRI would have been even lower, probably similar to that reported in the series by Kim et al.
We found that the ICDRs using BWBUS and DCE MRI significantly increased for ER-positive and triple-negative index cancers, but not for HER2-positive index cancers, compared with the ICDRs using mammography. This finding contrasts with a recent study reporting that pre-operative MRI may benefit patients with HER2-positive tumours most, as there is an increased likelihood of the presence of additional disease.33 Both that study and our study were limited by small numbers of patients with HER2-positive cancer. In a large cohort of 398 patients with triple-negative breast cancer, Bae et al34 reported that the absence of pre-operative MRI was associated with an increased risk of recurrence. Further research is needed to define the role of MRI in evaluating breast cancer molecular subtypes, as these subtypes carry important treatment and prognostic implications.35–39
One advantage of BWBUS over MRI in the pre-operative evaluation of patients with breast cancer is the regional staging of nodal basins. Although BWBUS and MRI were equally effective in depicting suspicious lymph nodes in the ipsilateral or contralateral axillary and internal mammary nodal basins, the standard field of view on a breast MRI does not include the supraclavicular and often only includes part of the infraclavicular nodal basin, which is readily accessible by BWBUS. Furthermore, suspicious lymph nodes can be sampled under ultrasonography guidance at the time of BWBUS.
The limitations of our study include its retrospective nature. Only patients who underwent both BWBUS and DCE MRI were included. As a result, we had a disproportionately high percentage of patients with dense breasts and with invasive lobular carcinoma in our cohort as a result of our imaging protocol. This contrasts with patients in the COMICE trial,19 the MONET trial20 and the series reported by Berg et al,4 in which breast MRI was performed in consecutive patients with both fatty and dense breast parenchyma. The fact that the ICDR of DCE MRI in our patient cohort was still much lower than that reported in the literature further supports the value of high-quality BWBUS in the pre-operative staging of breast cancer. This fact also supports the notion that breast MRI may be necessary only in select patients. As another limitation, no long-term outcome data including local recurrence, development of distant metastasis and overall survival were available in this study. Since we included only patients diagnosed during 2011–2014, such long-term outcome data would require additional follow-up. A third potential limitation is whether our results can be reproduced in other centres and institutions since BWBUS is operator dependent. As demonstrated by Kim et al,32 we believe that high-quality BWBUS can be achieved in other centres with the advancement of ultrasonography equipment and the improvement of ultrasonography techniques over the past two decades, coupled with the training of dedicated breast sonographers through certification and continuing education modules.
As the debate over pre-operative breast MRI continues, a cost-effective and practical approach may be to use BWBUS for staging.40,41 Breast MRI can be performed in select cases if it is available. A European Society of Breast Cancer Specialists working group published a position paper in 2010 that included indications for pre-operative MRI (invasive lobular cancer, a tumour size discrepancy between mammography and ultrasonography of >1 cm, high familial risk and candidacy for partial breast irradiation).42 A single-centre analysis of 1102 consecutive patients with primary breast cancer found that pre-menopausal females with dense breasts and patients with lobular histology seemed to benefit from pre-operative MRI.43
The controversy over pre-operative staging breast MRI is not likely to be resolved any time soon. Future research including prospective randomized multi-institutional trials involving large numbers of patients is needed to further define the role of pre-operative breast MRI in light of the low recurrence rate of breast cancer with the current standard treatment regimen.44,45
Conflict of interest
Wei T Yang is a consultant for GE Healthcare.
Contributor Information
Hongying He, Email: hongying.he@uth.tmc.edu.
Jeri S Plaxco, Email: JSPlaxco@mdanderson.org.
Wei Wei, Email: WWei@mdanderson.org.
Lei Huo, Email: leihuo@mdanderson.org.
Rosalind P Candelaria, Email: RCandelaria@mdanderson.org.
Henry M Kuerer, Email: hkuerer@mdanderson.org.
Wei T Yang, Email: wyang@mdanderson.org.
REFERENCES
- 1.Mandelson MT, Oestreicher N, Porter PI, White D, Finder CA, Taplin SH, et al. Breast density as a predictor of mammographic detection: comparison of interval- and screen-detected cancers. J Natl Cancer Inst 2000; 92: 1081–7. doi: 10.1093/jnci/92.13.1081 [DOI] [PubMed] [Google Scholar]
- 2.Kolb TM, Lichy J, Newhouse JH. Comparison of the performance of screening mammography, physical examination, and breast US and evaluation of factors that influence them: an analysis of 27,825 patient evaluations. Radiology 2002; 225: 165–75. doi: 10.1148/radiol.2251011667 [DOI] [PubMed] [Google Scholar]
- 3.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: 569–76. doi: 10.1148/radiol.2242011215 [DOI] [PubMed] [Google Scholar]
- 4.Berg WA, Gutierrez L, NessAiver MS, Carter WB, Bhargavan M, Lewis RS, et al. Diagnostic accuracy of mammography, clinical examination, US, and MR imaging in preoperative assessment of breast cancer. Radiology 2004; 233: 830–49. doi: 10.1148/radiol.2333031484 [DOI] [PubMed] [Google Scholar]
- 5.Fisher U, Kopka L, Grabbe E. Breast carcinoma: effect of preoperative contrast-enhanced MR imaging on the therapeutic approach. Radiology 1999; 213: 881–8. [DOI] [PubMed] [Google Scholar]
- 6.Hollingsworth A, Stough R, O'Dell C. Breast magnetic resonance imaging for preoperative locoregional staging. Am J Surg 2008; 196: 389–97. doi: 10.1016/j.amjsurg.2007.10.009 [DOI] [PubMed] [Google Scholar]
- 7.Liberman L, Morris E, Dershaw D. MR imaging of the ipsilateral breast in women with percutaneously proven breast cancer. Am J Roentgenol 2003; 180: 901–10. doi: 10.2214/ajr.180.4.1800901 [DOI] [PubMed] [Google Scholar]
- 8.Orel SG, Schnall MD, Powell CM. Staging of suspected breast cancer: effect of MR imaging and MR-guided biopsy. Radiology 1995; 196: 115–22. doi: 10.1148/radiology.196.1.7784554 [DOI] [PubMed] [Google Scholar]
- 9.Bilimoria KY, Cambic A, Hansen NM, Bethke KP. Evaluating the impact of preoperative breast magnetic resonance imaging on the surgical management of newly diagnosed breast cancers. Arch Surg 2007; 142: 441–7. doi: 10.1001/archsurg.142.5.441 [DOI] [PubMed] [Google Scholar]
- 10.Deurloo EE, Peterse JL, Rutgers EJ. Additional breast lesions in patients eligible for breast-conserving therapy by MRI: impact on preoperative management and potential benefit of computerized analysis. Eur J Cancer 2005; 41: 1393–401. doi: 10.1016/j.ejca.2005.03.017 [DOI] [PubMed] [Google Scholar]
- 11.Schnall M, Blume J, Bluemke DA. MRI detection of distinct incidental cancer in woman with primary breast cancer studied in IBMC 6883. J Surg Oncol 2005; 92: 32–8. doi: 10.1002/jso.20381 [DOI] [PubMed] [Google Scholar]
- 12.Lehman CD, Gatsonis C, Kuhl CK. MRI evaluation of the contralateral breast in women with recently diagnosed breast cancer. N Engl J Med 2007; 356: 1295–303. doi: 10.1056/NEJMoa065447 [DOI] [PubMed] [Google Scholar]
- 13.Sardenelli F, Giuseppetti GM, Panizza P, Bazzocchi M, Fausto A, Simonetti G, et al. Sensitivity of MRI versus mammography for detecting foci of multifocal, multicentric breast cancer in fatty and dense breasts using the whole-breast pathologic examination as a gold standard. Am J Roentgenol 2004; 183: 1149–57. [DOI] [PubMed] [Google Scholar]
- 14.Gutierrez RL, DeMartini WB, Silbergeld JJ. High cancer yield and positive predictive value: outcomes at a center routinely using preoperative breast MRI for staging. Am J Roentgenol 2011; 196: W93–9. doi: 10.2214/AJR.10.4804 [DOI] [PubMed] [Google Scholar]
- 15.Bedrosian I, Mick R, Orel SG, Schnall M, Reynolds C. Changes in the surgical management of patients with breast carcinoma based on preoperative magnetic resonance imaging. Cancer 2003; 98: 468–73. doi: 10.1002/cncr.11490 [DOI] [PubMed] [Google Scholar]
- 16.Houssami N, Ciatto S, Macaskill P. Accuracy and surgical impact of magnetic resonance imaging in breast cancer staging: systematic review and meta-analysis in detection of multifocal and multicentric cancer. J Clin Oncol 2008; 26: 3248–58. doi: 10.1200/JCO.2007.15.2108 [DOI] [PubMed] [Google Scholar]
- 17.Plana MN, Carreira C, Muriel A. Magnetic resonance imaging in the preoperative assessment of patients with primary breast cancer: systematic review of diagnostic accuracy and meta-analysis. Eur Radiol 2012; 22: 26–38. doi: 10.1007/s00330-011-2238-8 [DOI] [PubMed] [Google Scholar]
- 18.Brennan ME, Houssami N, Lord S, Macaskill P, Irwig L, Dixon JM, et al. Magnetic resonance imaging screening of the contralateral breast in women with newly diagnosed breast cancer: systemic review and meta-analysis of incremental cancer detection and impact on surgical management. J Clin Oncol 2009; 27: 5640–9. doi: 10.1200/JCO.2008.21.5756 [DOI] [PubMed] [Google Scholar]
- 19.Turnbull L, Brown S, Harvey I. Comparative effectiveness of MRI in breast cancer (COMICE) trial: a randomized controlled trial. Lancet 2010; 375: 563–71. doi: 10.1016/S0140-6736(09)62070-5 [DOI] [PubMed] [Google Scholar]
- 20.Peters N, van Esser S, van den Bosch M. Preoperative MRI and surgical management in patients with nonpalpable breast cancer: the MONET—randomized controlled trial. Eur J Cancer 2011; 47: 879–86. doi: 10.1016/j.ejca.2010.11.035 [DOI] [PubMed] [Google Scholar]
- 21.Hwang N, Schiller D, Crystal P, Maki E, McCready DR. Magnetic resonance imaging in the planning of initial lumpectomy of invasive breast carcinoma: its effect on ipsilateral breast tumor recurrence after breast-conservation therapy. Ann Surg Oncol 2009; 16: 3000–9. doi: 10.1245/s10434-009-0607-1 [DOI] [PubMed] [Google Scholar]
- 22.Vos EL, Voogd AC, Verhoef C. Benefits of preoperative MRI in breast cancer surgery studied in a large population-based cancer registry. Br J Surg 2015; 102: 1649–57. doi: 10.1002/bjs.9947 [DOI] [PubMed] [Google Scholar]
- 23.Solin LJ, Orel SG, Hwang WT. Relationship of breast magnetic resonance imaging to outcome after breast-conservation treatment with radiation for women with early-stage invasive breast carcinoma or ductal carcinoma in situ. J Clin Oncol 2008; 26: 386–91. doi: 10.1200/JCO.2006.09.5448 [DOI] [PubMed] [Google Scholar]
- 24.Bleicher RJ, Ciocca RM, Egleston BL. Association of routine pretreatment magnetic resonance imaging with time to surgery, mastectomy rate, and margin status. J Am Coll Surg 2009; 209: 180–7. doi: 10.1016/j.jamcollsurg.2009.04.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Sorbero ME, Dick AW, Beckjord EB, Ahrendt G. Diagnostic breast magnetic resonance imaging and contralateral prophylactic mastectomy. Ann Surg Oncol 2009; 16: 1597–605. doi: 10.1245/s10434-009-0362-3 [DOI] [PubMed] [Google Scholar]
- 26.Heil J, Rauch G, Szabo AZ. Breast cancer mastectomy trends between 2006 and 2010: association with magnetic resonance imaging, immediate breast reconstruction, and hospital volume. Ann Surg Oncol 2013; 20: 3839–46. doi: 10.1245/s10434-013-3097-0 [DOI] [PubMed] [Google Scholar]
- 27.Arnaout A, Catley C, Booth CM. Use of preoperative magnetic resonance imaging for breast cancer: a Canadian population-based study. JAMA Oncol 2015; 1: 1238–50. doi: 10.1001/jamaoncol.2015.3018 [DOI] [PubMed] [Google Scholar]
- 28.American College of Radiology. Breast imaging reporting and data system. 5th edn. Reston, VA: American College of Radiology; 2013. [Google Scholar]
- 29.Berg WA, Yang WT. Diagnostic imaging: breast. 2nd edn. Salt Lake, UT: Amirsys; 2013. [Google Scholar]
- 30.Berg WA, Gilbreath PL. Multicentric and multifocal cancer: whole-breast US in preoperative evaluation. Radiology 2000; 214: 59–66. doi: 10.1148/radiology.214.1.r00ja2559 [DOI] [PubMed] [Google Scholar]
- 31.Fancellu A, Turner RM, Dixon JM. Meta-analysis of the effect of preoperative breast MRI on the surgical management of ductal carcinoma in situ. Br J Surg 2015; 102: 883–93. doi: 10.1002/bjs.9797 [DOI] [PubMed] [Google Scholar]
- 32.Kim J, Han W, Moon HG. Low rates of additional cancer detection by magnetic resonance imaging in newly diagnosed breast cancer patients who undergo preoperative mammography and ultrasonography. J Breast Cancer 2014; 17: 167–73. doi: 10.4048/jbc.2014.17.2.167 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Ha R, Jin B, Mango V, Friedlander L, Miloshev V, Malak S. Breast cancer molecular subtype as a predictor of the utility of preoperative MRI. Am J Roentgenol 2015; 204: 1354–60. doi: 10.2214/AJR.14.13666 [DOI] [PubMed] [Google Scholar]
- 34.Bae MS, Moon HG, Han W. Early stage triple-negative breast cancer: imaging and clinical-pathologic factors associated with recurrence. Radiology 2016; 278: 356–64. doi: 10.1148/radiol.2015150089 [DOI] [PubMed] [Google Scholar]
- 35.Reis-Filho JS, Rusztai L. Gene expression profiling in breast cancer: classification, prognostication, and prediction. Lancet 2011; 378: 1812–23. doi: 10.1016/S0140-6736(11)61539-0 [DOI] [PubMed] [Google Scholar]
- 36.Smid M, Wang Y, Zhang Y. Subtypes of breast cancer show preferential site of relapse. Cancer Res 2008; 68: 3108–14. doi: 10.1158/0008-5472.CAN-07-5644 [DOI] [PubMed] [Google Scholar]
- 37.Carey LA, Perou CM, Livasy CA. Race, breast cancer subtypes, survival in the Carolina Breast Cancer Study. JAMA 2006; 295: 2492–502. doi: 10.1001/jama.295.21.2492 [DOI] [PubMed] [Google Scholar]
- 38.Sorlie T. Molecular portraits of breast cancer: tumour subtypes as distinct disease entities. Eur J Cancer 2004; 40: 2667–75. [DOI] [PubMed] [Google Scholar]
- 39.Voduc KD, Cheang MC, Tyldesley S. Breast cancer subtypes and the risk of local and regional relapse. J Clin Oncol 2010; 28: 1684–91. doi: 10.1200/JCO.2009.24.9284 [DOI] [PubMed] [Google Scholar]
- 40.Fornage BD. Local and regional staging of invasive breast cancer with sonography: 25 years of practice at MD Anderson Cancer Center. Oncologist 2014; 19: 5–15. doi: 10.1634/theoncologist.2013-0323 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Yang WT. Staging of breast cancer with ultrasound. Semin Ultrasound CT MR 2011; 32: 331–41. doi: 10.1053/j.sult.2011.02.008 [DOI] [PubMed] [Google Scholar]
- 42.Sardanelli F, Boetes C, Borisch B. Magnetic resonance imaging of the breast: recommendations from the EUSOMA working group. Eur J Cancer 2010; 46: 1296–316. doi: 10.1016/j.ejca.2010.02.015 [DOI] [PubMed] [Google Scholar]
- 43.Debald M, Abramian A, Nemes L, Döbler M, Kaiser C, Keyver-Paik MD, et al. Who may benefit from preoperative breast MRI? A single-center analysis of 1102 consecutive patients with primary breast cancer. Breast Cancer Res Treat 2015; 153: 531–7. doi: 10.1007/s10549-015-3556-3 [DOI] [PubMed] [Google Scholar]
- 44.Recht A, Solin LJ. Breast-conserving surgery and radiotherapy in early-stage breast cancer: the importance of local control. Semin Radiat Oncol 2011; 21: 3–9. doi: 10.1016/j.semradonc.2010.08.001 [DOI] [PubMed] [Google Scholar]
- 45.Clarke M, Collins R, Darby S, Davies C, Elphinstone P, Evans V, et al. Effect of radiotherapy and of difference in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomized trials. Lancet 2005; 366: 2087–106. doi: 10.1016/S0140-6736(05)67887-7 [DOI] [PubMed] [Google Scholar]