Introduction: B3-lesions of the breast are a heterogeneous group of neoplasms, associated with a higher risk of breast cancer. Recent studies show a low upgrade rate into malignancy after subsequent open surgical excision (OE) of most B3-lesions when proven by vacuum-assisted biopsy (VAB). However, there is a lack of long-term follow-up data after VAB of high-risk lesions. The primary aim of this study was to demonstrate whether follow-up of B3 lesions is a beneficial and reliable alternative to OE in terms of long-term outcome. The secondary aim was to identify patient and lesion characteristics of B3 lesions for which OE is still necessary. Methods: This retrospective multicenter study was conducted at 8 Swiss breast centers between 2010 and 2019. A total of 278 women (mean age: 53.5 ± 10.7 years) with 286 B3-lesions who had observation only and who had at least 24 months of follow-up were included. Any event during follow-up (ductal carcinoma in situ [DCIS], invasive cancer, new B3-lesion) was systematically recorded. Data from women who had an event during follow-up were compared with those who did not. The results for the different B3 lesions were analyzed using the t test and Fisher’s exact test. A p value of <0.05 was considered statistically significant. Results: The median follow-up interval was 59 months (range: 24–143 months) with 52% (148/286) having a follow-up of more than 5 years. During follow-up, in 42 women, 44 suspicious lesions occurred, with 36.4% (16/44) being invasive cancer and 6.8% (3/44) being DCIS. Thus, 6.6% (19/286) of all women developed malignancy during follow-up after a median follow-up interval of 6.5 years (range: 31–119 months). The initial histology of the B3 lesion influenced the subsequent occurrence of a malignant lesion during follow-up (p < 0.038). The highest malignancy-developing rate was observed in atypical ductal hyperplasia (ADH) (24%, 19/79), while all other B3-lesions had malignant findings ipsi- and contralateral between 0% and 6%. The results were not influenced by the VAB method (Mx-, US-, magnetic resonance imaging-guided), the radiological characteristics of the lesion, or the age or menopausal status of the patient (p > 0.12). Conclusion: With a low risk of <6% of developing malignancy, VAB followed by long-term follow-up is a safe alternative to OE for most B3-lesions. A higher malignancy rate only occurred in ADH (24%). Based on our results, radiological follow-up should be bilateral, preferable using the technique of initial diagnosis. As we observed a late peak (6–7 years) of breast malignancies after B3-lesions, follow-up should be continued for a longer period (>10 years). Knowledge of these long-term outcome results will be helpful in making treatment decisions and determining the optimal radiological follow-up interval.

Breast lesions of uncertain malignant potential or high-risk (B3) are a heterogeneous group of abnormalities with a low but variable risk of associated malignancy [1‒3]. B3 lesions are typically diagnosed by core needle biopsy or by vacuum-assisted biopsy (VAB) with a 7G–11G device under either ultrasound, stereotactic, or magnetic resonance imaging (MRI) guidance [4]. The management of B3 lesions remains a significant challenge in clinical practice. For many years, the standard approach to B3 lesions has been open surgical excision (OE), especially for those with atypia detected by diagnostic VAB techniques. The rationale for OE is to detect any coincident significant malignant disease that was not detected by VAB, as in B3 lesions, the upgrade rate to ductal carcinoma in situ (DCIS) or invasive cancer has been shown to be 20–30% [5‒7]. However, malignant lesions occurring among B3 lesions are usually associated with a favorable biological profile and prognosis, and the rate of progression to malignancy occurs over a long time [8, 9]. This must be taken into account when discussing optimal treatment options for high-risk lesions [10, 11].

Recent studies have demonstrated that VAB is a reliable and effective alternative to OE of B3 lesions without atypia, such as papillomas and radial scars (RSs) [6, 12‒17]. This has led to an increasing use of VAB and subsequent observation in nonsurgically managed B3 lesions in some countries [18‒21].

This management recommendation has been supported by the first and second International Consensus Conference on lesions of uncertain malignant potential [18, 19], the US guidelines [20], and the UK guidelines [6]. Although there are some differences in these recommendations between the continental, US and UK guidelines, performing VAB with consecutive follow-up (without OE) can be considered the new gold standard for managing of the majority of B3 lesions without atypia in several countries [16‒22]. However, long-term results from patients with B3 lesions who did not undergo OE are still lacking.

The aim of this study was to analyze the long-term outcome of patients with B3 lesions after imaging-guided VAB who underwent follow-up instead of OE. The primary objective was to determine the upgrade rates of B3 lesions and the occurrence of new malignancies or B3 lesions during the follow-up period. The study aimed to demonstrate whether follow-up of high-risk lesions is a safe and beneficial management with an acceptable low upgrade rate during follow-up and an alternative to OE.

The secondary objective was to define patients at risk and histopathological features of B3 lesions at higher risk of upgrading into malignancy leading to subsequent OE recommendation. Overall, the results of this study will provide valuable information for clinical decision-making and optimal management of B3 lesions.

Study Design and Patient Cohort

This retrospective observational study was conducted between 2010 and 2019 at nine breast centers in Switzerland (Breast-Center Zurich, Cantonal Hospital Lucerne, Cantonal Hospital Schaffhausen, Cantonal Hospital Frauenfeld, Cantonal Hospital Baden, Cantonal Hospital Aarau, Radiological Center Lugano, Radiological Center Lausanne, Breast Center Lindenhofgruppe Bern). The breast centers involved were certified as VAB-performing centers by the Minimal Invasive Breast Biopsy (MIBB) working group, which guarantees a minimal number of VAB procedures performed by trained and certified operators [18]. All certified institutions entered relevant information about all VAB they performed into the database, including the method of detection, type of imaging found, histological result, management, and follow-up data. The MIBB database was initiated in 2002 with the aim of quality control and improvement, and scientific research of breast VAB. The study enrolled patients who gave informed consent, who had a confirmed B3 lesion in the VAB specimen, and who had a minimum follow-up period of 24 months. Women who were found to have a coincident malignancy in the breast biopsy specimen, whose data were incomplete, or who had a shorter follow-up time were excluded. Follow-up included a radiological examination and documentation of patient’s medical history.

All women who underwent VAB and were eventually diagnosed with a B3 lesion were included in this study. Women with additional invasive cancers and DCIS in the breast biopsy specimen were excluded.

For analysis of long-term outcomes of women with biopsy-proven B3 lesions, only women, who had follow-up of at least 24 months after the initial VAB to confirm the lesion stability and thus exclude malignancy during follow-up were included. Follow-up results were obtained either by review of imaging results or, in few patients, by a patient interview. Women with incomplete datasets and for whom follow-up was not available were excluded from the study cohort. The study was approved by the Cantonal Ethics Committee.

Imaging and Biopsy Methods

All study participants underwent digital mammography with or without tomosynthesis. Additional imaging procedures such as ultrasound, MRI, or additional mammographic images were ordered if necessary. All physicians were specialized in multimodality breast imaging and certified by the MIBB to perform VAB [23, 24]. All imaging examinations and biopsies were performed using current quality standards in accordance with national and international practice guidelines [22, 23]. All examinations were read and classified according to the American College of Radiology (ACR) Breast Imaging-Reporting and Data System (BI-RADS) lexicon [25, 26]. For indeterminate or suspicious breast lesions (BI-RADS IV and V) on ultrasound, tomosynthesis/mammography, or MRI-VAB was performed using the imaging modality that best depicted the suspicious lesion [23, 24, 27, 28]. All biopsies were performed with modern, fully automated biopsy devices with a needle size of at least 12 (7–12) gauge.

Histologic Processing and Classification of Biopsy Specimens

Biopsy specimens from the study patients were processed according to a standard protocol based on current national and European guidelines and diagnosed by specialized breast pathologists within the breast core team [6, 18, 19, 29‒32]. When necessary for classification, further pathological assessment was performed at the pathologist’s discretion, including additional stratification or immunohistochemistry. All diagnosed B3 lesions were categorized with regard to current guidelines [6, 18, 19, 29, 30, 32]. The B3 lesions analyzed in this study were RS, atypical ductal hyperplasia (ADH), flat epithelial atypia (FEA), classical type of lobular neoplasia (LN), and papilloma without atypia. Papilloma with atypia were not included, as these lesions have a significantly higher risk of upgrade and should always undergo a subsequent OE [19, 21, 32].

Interdisciplinary radiological-pathological correlation was performed within the scope of a weekly preoperative interdisciplinary discussion for each target lesion to ensure that the histopathological diagnosis made in the VAB correlated with the imaging lesion. If a clear radiological-pathological correlation was given, a multidisciplinary breast center conference decided under consideration of timely current clinical management guidelines whether the given B3 lesion should be managed with follow-up or subsequent OE.

Women who were found to have new suspicious lesions during follow-up underwent standard diagnostic and treatment, including VAB and OE. The management of the suspicious lesions was identical to that of the initial breast lesion, including VAB. The final postoperative surgical pathology report was included in addition to the VAB diagnosis in this study.

Data Collection

Based on the entries in the MIBB database, all cases that met the inclusion criteria were enrolled in the study. For each study participant, the following data were prospectively recorded in the MIBB database: demographic data, detection modality of the target lesion, localization within the breast, characteristics of the target lesion on mammography, presence or absence of lesion-associated calcifications, ultrasound or MRI, final histological result of the VAB specimen, management recommendation of the B3 lesion (observation vs. OE), reason for not performing OE, imaging follow-up results or patient-reported events, and histology of any subsequent surgical breast resection or biopsy.

Data Analysis and Statistics

Histology and imaging size of initial high-risk lesions, the imaging method used to detect the lesion, imaging characteristics of the lesion, final histological result of all suspicious lesion during follow-up or at OE, duration of follow-up intervals, time interval between initial B3 diagnosis and the occurrence of a suspicious lesion during follow-up were analyzed for all study participants. The selection criteria were the presence of a relevant event during follow-up, defined as any B3 lesion, in situ lesion, and malignant lesion that occurred.

Data from women who had normal follow-up were compared with data from women who developed a B3 lesion or malignant breast lesion during follow-up. Differences between the two groups in the method of lesion detection, type of the lesion, size of lesion, and histology of the initial lesion were analyzed.

The t test for independent variables was used to test whether the differences between patient groups were statistically significant. The distribution of the different B3 categories was examined with the Mann-Whitney’s U test. For categorical and continuous variables, respectively, Fisher’s exact tests and χ2 test were used, and a two-sided p value of <0.05 considered statistically significant. All analyses were performed using SPSS 28.0.1 [33].

Study Cohort and Follow-Up Interval

In total, 15,321 VAB were performed in the participating centers between 2010 and 2019. Out of those, 3,163 women were diagnosed with B3 lesions at VAB, of whom 378 (12.0%) underwent documented observational follow-up instead of OE. Due to several reasons, such as shorter follow-up interval or coincidence of malignant lesions, 26.4% (100/378) of the women were excluded from data analysis (Fig. 1). A total number of 286 B3-lesions among 278 study participants were finally analyzed. The median patient’s age was 52 years (range: 18–85 years). The main reason for not performing OE of the B3 lesion was the radiological complete removal of the lesion (calcification or mass) in 66.1% (189/286). In one-third of all patients (33.9%; 97/286), the multidisciplinary team recommended observation because of the patient’s advanced age, comorbidities, or an estimated very low risk of upgrade into malignancy (Table 1). The majority of the B3 lesions were diagnosed by mammographic-guided VAB (79.4%, 177/286), with microcalcifications as the most frequent target lesion (92.7%, 164/177). Ultrasound-guided VAB was performed in 15.0% (43/286), and MRI-guided VAB was performed in 23.1% (66/286) (Table 1). The study cohort, as well as the characteristics of the initial target lesion, showed characteristics that are commonly seen and representative of most women who are seen in specialized breast centers in Switzerland and most European countries.

Fig. 1.

Flowchart study cohort.

Fig. 1.

Flowchart study cohort.

Close modal
Table 1.

Patient characteristics, biopsy method used, and follow-up intervals

AllPatients/lesions with normal follow-upPatients/lesions with event during follow-up (B3/DCIS/invasive ca)p value
Age, years n = 278 n = 236 n = 42 >0.728 
 Mean/SD 53.5/±10.7 53.6/±11.0 53.2/±10.2  
 Median/range 52/18–85 52/18–85 51/31–79  
Menopausal status n = 278 n = 236 n = 42 >0.278 
 Premenopausal 179 64.4% 155 65.7% 24 57.1%  
 postmenopausal 99 35.6% 81 34.3% 18 42.9%  
Reason for omitting surgery n = 286 n = 242 n = 44 >0.321 
 Lesion completely removed by VAB 189 66.1% 155 64.0% 34 77.3%  
 Surgical and esthetical risk 74 25.9% 69 28.5% 11.4%  
 Comorbidity 1.4% 0.8% 4.5%  
 Refusal of the patient 10 3.5% 3.3% 4.5%  
 Other 3.1% 3.3% 2.3%  
Method of VAB n = 286 n = 242 n = 44 >0.978 
 Mammography/tomosynthesis 177 62.2% 150 62.0% 27 61.4%  
 Ultrasound 43 15.0% 29 12.0% 14 31.8%  
 MRI 66 23.1% 63 26.0% 6.8%  
Time interval follow-up, months n = 278 n = 236 n = 42 >0.350 
 Mean/SD 58.9 24.3 56.4 23.2 75.3 23.6  
 Median/range 61 24–143 59 24–143 80 25–119  
 24–35 months 26 9.4% 25 10.6% 2.4%  
 36–47 months 31 11.2% 28 11.9% 7.1%  
 48–59 months 36 12.9% 33 14.0% 7.1%  
 60–71 months 39 14.0% 36 15.3% 7.1%  
 72–83 months 49 17.6% 45 19.1% 9.5%  
 84–96 months 57 20.5% 46 19.5% 11 26.2%  
 ≥96 months 40 14.4% 23 9.7% 17 40.5%  
AllPatients/lesions with normal follow-upPatients/lesions with event during follow-up (B3/DCIS/invasive ca)p value
Age, years n = 278 n = 236 n = 42 >0.728 
 Mean/SD 53.5/±10.7 53.6/±11.0 53.2/±10.2  
 Median/range 52/18–85 52/18–85 51/31–79  
Menopausal status n = 278 n = 236 n = 42 >0.278 
 Premenopausal 179 64.4% 155 65.7% 24 57.1%  
 postmenopausal 99 35.6% 81 34.3% 18 42.9%  
Reason for omitting surgery n = 286 n = 242 n = 44 >0.321 
 Lesion completely removed by VAB 189 66.1% 155 64.0% 34 77.3%  
 Surgical and esthetical risk 74 25.9% 69 28.5% 11.4%  
 Comorbidity 1.4% 0.8% 4.5%  
 Refusal of the patient 10 3.5% 3.3% 4.5%  
 Other 3.1% 3.3% 2.3%  
Method of VAB n = 286 n = 242 n = 44 >0.978 
 Mammography/tomosynthesis 177 62.2% 150 62.0% 27 61.4%  
 Ultrasound 43 15.0% 29 12.0% 14 31.8%  
 MRI 66 23.1% 63 26.0% 6.8%  
Time interval follow-up, months n = 278 n = 236 n = 42 >0.350 
 Mean/SD 58.9 24.3 56.4 23.2 75.3 23.6  
 Median/range 61 24–143 59 24–143 80 25–119  
 24–35 months 26 9.4% 25 10.6% 2.4%  
 36–47 months 31 11.2% 28 11.9% 7.1%  
 48–59 months 36 12.9% 33 14.0% 7.1%  
 60–71 months 39 14.0% 36 15.3% 7.1%  
 72–83 months 49 17.6% 45 19.1% 9.5%  
 84–96 months 57 20.5% 46 19.5% 11 26.2%  
 ≥96 months 40 14.4% 23 9.7% 17 40.5%  

The mean follow-up period in all women was about 5 years (mean 59 months, median 61 months, range 24–143 months; SD 24.3). The majority of all women underwent a follow-up period of more than 3 years (177/278), with 52.5% (146/278) having a follow-up of more than 5 years. A very long follow-up of more than 8 years had 14.4% (49/278) of study participants with a maximum follow-up period of almost 12 years (Table 1; Fig. 2). No significant difference regarding patient characteristics was present between women who developed a high-risk or malignant lesion during follow-up and those who did not (p > 0.05).

Fig. 2.

Follow-up interval in women with B3-lesion with and without events during follow-up.

Fig. 2.

Follow-up interval in women with B3-lesion with and without events during follow-up.

Close modal

Histology of the Initial B3 Lesions at VAB

No difference could be observed regarding the histology of the initial B3 lesion in the two groups with and without follow-up events (Table 2). The majority of the initially detected B3-lesions belonged to the histological groups of classical type of LN (31.1%, 89/286), FEA (15.7%, 45/286), and in particular ADH (27.6%, 79/286). In 16.4% (47/286), a complex B3-lesion was present with a combination of more than one B3 histology, of which ADH + FEA, ADH + LN, and LN + FEA were the most common combinations (Table 2).

Table 2.

Characteristics of B3-lesion at initial VAB

B3 lesion at initial VABAll B3 lesions (n = 286)Ipsilateral event at biopsy site (n = 22)Ipsilateral event not at biopsy (n = 8)Contralateral event (n = 14)
B3 histology p = 0.038 
ADH 79 27.6% 36.4% 62.5% 42.9% 
Classical type of LN 89 31.1% 27.3% 25.0% 28.6% 
FEA 45 15.7% 18.2% 12.5% 14.3% 
Phyllodes tumor 1.7%   
Radical scar 21 7.3% 9.1% 
Complex lesions 47 16,4% 9.1%   14.3% 
 ADH + FEA 2.1% 
 ADH + LN 17 5.9% 4.6% 7.1% 
 FEA + LN 20 7.0% 4.6% 7.1% 
 Others 1.4% 
Lesion size, mma n =251a n =21 n = 7 n = 12 
 Mean/SD 13.1 ±11.8 12.6 ±6.8 15.4 ±5.9 11.7 ±17.6 
 Median/range 10 2–85 10 4–30 17 7–22 5–70 
   p > 0.393 
B3 lesion at initial VABAll B3 lesions (n = 286)Ipsilateral event at biopsy site (n = 22)Ipsilateral event not at biopsy (n = 8)Contralateral event (n = 14)
B3 histology p = 0.038 
ADH 79 27.6% 36.4% 62.5% 42.9% 
Classical type of LN 89 31.1% 27.3% 25.0% 28.6% 
FEA 45 15.7% 18.2% 12.5% 14.3% 
Phyllodes tumor 1.7%   
Radical scar 21 7.3% 9.1% 
Complex lesions 47 16,4% 9.1%   14.3% 
 ADH + FEA 2.1% 
 ADH + LN 17 5.9% 4.6% 7.1% 
 FEA + LN 20 7.0% 4.6% 7.1% 
 Others 1.4% 
Lesion size, mma n =251a n =21 n = 7 n = 12 
 Mean/SD 13.1 ±11.8 12.6 ±6.8 15.4 ±5.9 11.7 ±17.6 
 Median/range 10 2–85 10 4–30 17 7–22 5–70 
   p > 0.393 

*n = 35 n.a. (no exact lesions size available)

Events during Follow-Up

Most patients (84.6%, n = 236) had an uneventful follow-up, and only 42 women developed 44 (15.4%) radiologically suspicious lesion during follow-up. One in two lesions (22/44) occurred at the exact site of the initial B3-target lesion, and 18.2% (8/44) were detected in the same breast but in a different area from the initial B3-lesion. One in three suspicious lesions (31.8%, 14/44) was described in the contralateral breast (Table 3). The follow-up interval, demographic data, and lesion characteristics of these women did not differ significantly between the different groups (p > 0.20) and were comparable with the group of women with uneventful follow-up (Tables 1-3).

Table 3.

Characteristics of all lesions that occurred during follow-up (n = 44)

Ipsilateral event at biopsy site during follow-upIpsilateral event not at biopsy site during follow-upContralateral event during follow-up
All (n = 44) 22 50% 18.2% 14 31.8% 
B3 lesion (n = 25) 11 50.0% 50% 10 71.4% 
 ADHa 13.6% 12.5% 21.4% 
 Classical type of LN (LCIS/AHL)a 13.6% 25.0% 28.5% 
 FEAa 9.1% 12.5% 7.1% 
 Phyllodes tumor 
 Radical scar 4.5% 14.2% 
 Others 9.1% 
DCIS (n = 3) 4.5% 12.5% 
 Size of lesion at VAB with malignancy during follow-up in mm 10 mm and 4 mm 10 mm  
Invasive cancer (n = 16) 40.9% 37.5% 28.6% 
 Size of lesion at VAB with malignancy during follow-up in mm       
 Mean/SD 11.1 ±5.7 19.7 ±2.1 ±0.7 
 Median/range 10 5–26 20 17–22 7–8 
 p > 0.067 
Ipsilateral event at biopsy site during follow-upIpsilateral event not at biopsy site during follow-upContralateral event during follow-up
All (n = 44) 22 50% 18.2% 14 31.8% 
B3 lesion (n = 25) 11 50.0% 50% 10 71.4% 
 ADHa 13.6% 12.5% 21.4% 
 Classical type of LN (LCIS/AHL)a 13.6% 25.0% 28.5% 
 FEAa 9.1% 12.5% 7.1% 
 Phyllodes tumor 
 Radical scar 4.5% 14.2% 
 Others 9.1% 
DCIS (n = 3) 4.5% 12.5% 
 Size of lesion at VAB with malignancy during follow-up in mm 10 mm and 4 mm 10 mm  
Invasive cancer (n = 16) 40.9% 37.5% 28.6% 
 Size of lesion at VAB with malignancy during follow-up in mm       
 Mean/SD 11.1 ±5.7 19.7 ±2.1 ±0.7 
 Median/range 10 5–26 20 17–22 7–8 
 p > 0.067 

aBecause of the small numbers, complex lesions were put in together with the pure lesions regarding the most prominent histology within the lesion.

Characteristics of Lesions Occurred during Follow-Up

Of all lesions detected radiologically during follow-up, 43.2% (19/44) of lesions were malignant; with 36.4% (16/44) being invasive breast cancers and 6.8% (3/44) being DCIS. Accordingly, in 6.6% (19/286) of all initial B3-lesions, a malignant lesion occurred during follow-up, with 5.3% (15/286) in the ipsilateral breast, mainly at the initial biopsy site (3.8% 11/286), and 1.4% (4/286) in the contralateral breast.

More than half (25/44, 56.8%) of all relevant lesions that were detected during follow-up were B3-lesions, which in the majority occurred at the former VAB biopsy site (3.8% (11/286) and in the contralateral breast and (3.5% (10/286). All details of during follow-up detected relevant lesions are given in Tables 2 and 3.

With regard to the type of the initial B3 lesions, the percentage of invasive cancers that occurred during follow-up was quite low and uniform after the diagnosis of LN, FEA, and even after complex B3 lesions, ranging from 0.0% to 5.7%. By far the highest rate of an invasive cancer was after the initial diagnosis of ADH (24.1%; 19/79) (Fig. 3).

Fig. 3.

Rate of DCIS and invasive cancer during follow-up based on the initial B3-lesion proven by VAB.

Fig. 3.

Rate of DCIS and invasive cancer during follow-up based on the initial B3-lesion proven by VAB.

Close modal

Time Interval of Follow-Up Lesion Detection

During follow-up, the time interval between the initial VAB diagnosis and the newly detected relevant lesions (B3 and malignant lesions) ranged from 31 months to 119 months (median 79 months). The mean time interval for newly detected invasive carcinoma was longest at 77 months, compared with 62 months for DCIS and 74 months for new B3-lesions (p > 0.723). Only a minority (4.5%, 2/44) of all new lesions were detected within 3 years after the initial B3 diagnosis (Table 4).

Table 4.

Time interval between initial VAB and event during follow-up

B3 lesion during follow-upDCIS during follow-upInvasive cancer during follow-upp value
Time interval VAB – malignancy, months    >0.723 
Mean 73.8 62.3 76.8  
Median 72 41 82  
Range 33–92 39–107 31–119  
Standard deviation 16.1 38.7 25.8  
B3 lesion during follow-upDCIS during follow-upInvasive cancer during follow-upp value
Time interval VAB – malignancy, months    >0.723 
Mean 73.8 62.3 76.8  
Median 72 41 82  
Range 33–92 39–107 31–119  
Standard deviation 16.1 38.7 25.8  
Ipsilateral malignancy at biopsy site during follow-upIpsilateral malignancy not at biopsy site during follow-upContralateral malignancy during follow-upp value
Time interval VAB – malignancy, months    <0.024 
Mean 88.0 45.5 67.4  
Median 90 37 72  
Range 39–110 31–87 32–92  
Standard deviation 20.1 28.9 22.3  
Ipsilateral malignancy at biopsy site during follow-upIpsilateral malignancy not at biopsy site during follow-upContralateral malignancy during follow-upp value
Time interval VAB – malignancy, months    <0.024 
Mean 88.0 45.5 67.4  
Median 90 37 72  
Range 39–110 31–87 32–92  
Standard deviation 20.1 28.9 22.3  

Malignancy during follow-up occurred earlier in the ipsilateral breast than in the contralateral breast (mean 56 months vs. 67 months). New lesions in the ipsilateral breast occurred significantly earlier not at the area of the initially diagnosed B3-lesion, compared to new lesions that were diagnosed at the biopsy side (mean 46 months vs. 88 months) (p < 0.024) (Table 4).

Influence of the Detection Type

Invasive breast cancer occurred in 7% (16/177) of all B3-lesions initially biopsied by mammographic-guided VAB, and DCIS in only 1 patient (0.44% [1/177]). Almost all cases of malignancy during follow-up after mammographically diagnosed B3 lesions were detected by calcifications on mammogram (94.2% [16/17]). Among the initially ultrasound-guided VAB lesions, one invasive cancer (2.3%, 1/43) and one DCIS occurred at the biopsy site (2.3%, 1/43). No malignant lesion developed in the patients whose initially B3-lesion was proven by MRI-VAB.

B3-lesions are frequent diagnoses among radiologically suspicious breast lesions and can be found in 20–30% of imaging-guided VAB [2, 3, 9]. As B3-lesions are highly heterogeneous, the different types of B3-lesions are associated with a variable long-term risk of developing malignancy, although this risk is relatively low. The good overall prognosis of B3-lesions on the one hand and their tremendous variability on the other hand have an impact on the clinical management that is to avoid overdiagnosis as well as underdiagnosis. Until recently, the OE of B3-lesions was still recommended, as earlier studies described upgrade rates up to malignancy in up to 30% [10, 30]. However, as recent studies consider established method of VAB and improvement of diagnostic methods, they relativize the upgrade risk to a much lower number, and therefore, international guidelines aim to reduce an overtreatment of B3-lesions [31]. Within this approach, imaging-based follow-up instead of OE has been discussed to become an alternative recommendation for the management of B3-lesions. Nevertheless, most guidelines are reluctant to commit to clear therapy recommendations [29, 31]. This is also true for the recommendation of follow-up interval, frequency, and method of therapy.

The International Consensus Conferences on B3 lesions from 2016 to 2019 recommended regular senological surveillance following a diagnosis of B3-lesions, including non-atypical lesions (papilloma, RS) and a more frequent surveillance for classical type of LN [18, 19]. The current UK guidelines recommend annual mammographic follow-up for 5 years followed by return to the biannual routine breast cancer screening [6].

Malignancy during Long-Term Follow-Up

In order to review the current trend of treatment de-escalation in patients with B3-lesions, our retrospective observational multicenter study analyzed a noticeable long-term follow-up of affected women without subsequent OE. The follow-up data of up to 12 years with a mean of 6.1 years may thereby provide data of high clinical relevance [14]. Furthermore, the study aims to identify potential risk factors for the emergence of malignancy during the follow-up. The study cohort consisted of 286 B3-lesions in 278 women. The majority of B3-lesions were diagnosed by mammography-guided VAB, with microcalcification as the most frequent target lesion. During follow-up, 15.1% (42/278) of women developed a radiologically suspicious lesion. The malignancy rate among these women during follow-up was remarkably low (6.6% [19/286]), with 5.6% (16/286) for invasive cancers. However, the general bilateral risk is three to four times higher than the age-related risk of the normal female population [34]. Our findings are consistent with previous studies reporting that the risk of malignancy after a biopsy-driven diagnosis of a B3-lesion is relatively low (1–16%) and depends on the specific B3 histology [1‒7]. In our study, the most frequent B3-lesions were classical types of LN, FEA, and ADH. The highest risk of developing malignancy was observed in women with ADH or complex B3-lesions, which is in line with previous reports [2‒6].

Implications for Follow-Up Regimen

ADH has been long regarded as a biologically distinct type of B3-lesion. It shares cytological and morphological similarities with DCIS, only being a small lesion not exceeding 2 mm in maximum extension. Therefore, it is comprehensible that ADH has the highest rate of developing malignancy (24%) during follow-up in our study group. Despite some authors with lower patient numbers and shorter follow-up time describe a rate of development of malignancy lower than 8%, our long-term data emphasize the treatment of ADH with OE [18, 19, 35]. In contrast, in all other B3 lesions except ADH, radiological follow-up and avoidance of OE seems to be appropriate.

Almost all cases of malignancy during follow-up after mammographically diagnosed B3-lesions were detected by calcifications in the mammogram. Based on these results, a suggested imaging protocol for women with a B3 diagnosis (except for ADH), made by VAB of calcifications, could include bilateral digital full-field mammography or tomosynthesis biannually with supplemental ultrasound in women with dense breasts.

Another important finding is that the time interval between the initial diagnosis of a B3-lesion and the detection of lesions during follow-up ranged from 21 to 119 months. This underlines the importance of long-term follow-up in women with a B3-lesion, which is especially true for ADH and LN. In addition, our findings suggest that the follow-up interval should be individualized based on the initial B3 histology, with more frequent follow-up recommended for women with ADH or complex B3-lesions [6, 31, 36]. These results imply that there might be no need for annual follow-up or even shorter intervals in the first years after the VAB diagnosis of a noncomplex, non-ADH B3-lesion after omitting a subsequent OE [18, 19]. However, in case of B3 lesion-associated microcalcifications leading to the indication for VAB, a mammographic confirmation of the complete resection of microcalcifications should be performed within 6 months [37]. There might also be an indication to control the clip localization, if a marker was placed [38]. The median time interval between the initial VAB diagnosis of a B3-lesion and the occurrence of a new lesion during follow-up was 6.5 years, with only one case with small DCIS occurring 24 months after the initial VAB. As malignant lesions occurred after 5 years and later, a long-term follow-up for many years seems to be reasonable. As the risk of developing malignancy concerns both breasts even after an unilateral B3-lesion, radiological follow-up should always consider both sides [39]. It is important to remain sensitive to the fact that close monitoring can trigger unnecessary fear of cancer, and patients should be informed that early detected breast malignancy has an excellent prognosis in most cases.

Strengths and Limitations

To the best of our knowledge, this is the first study with a noticeable long-term follow-up after omitting subsequent surgery of B3-lesions. All physicians who performed the VAB interventions were certified and routine operators, working in certified breast centers, guaranteeing a qualified level of performance and providing comparability.

Our study has some limitations. First, it was a single-country study, and the results may not be generalizable to other populations. Second, the study was based on data from the MIBB database, which may not have captured all relevant information, such as the patients’ family history of breast cancer, or their lifestyle factors. Third, the study had a relatively small sample size, which may limit the statistical power of the analysis. This is especially true for the low number of B3 lesions diagnosed by MRI-guided and US-guided VAB. Furthermore, our retrospective study did not compare the study group with women who received subsequent OE after the diagnosis of B3-lesion. Ideally, a prospective randomized trial comparing OE and radiological observation with a high number of patients and long-term follow-up should be initiated to verify our results.

Follow-up seems to be an acceptable and safe management for most B3-lesions diagnosed by VAB, with a low malignancy rate of <3% during follow-up in the ipsilateral breast. However, structured long-term follow-up protocols for patients’ surveillance should be followed individually by performing the most sensitive imaging method for the initially detected B3-lesion. ADH differs significantly from other B3-lesions regarding its long-term outcome and should be treated with OE.

The study was approved by the Ethics Committee Lucerne (BASEC ID 2020-00075) in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Written informed consent was obtained from the participants.

All authors declare that they have no conflict of interest.

The study was self-founded.

Constanze Elfgen and Simone Schrading: writing; Zsuzsanna Varga and Rahel A. Kubik-Huch: analysis; Katrin Breitling, Daniela Schwegler-Guggemos, and Gert Kampmann: review; and Domenico Lepori, Martin Sonnenschein, and Christoph Tausch: research.

The datasets used and analyzed during this study are available from the corresponding author on reasonable request.

1.
Santen
RJ
,
Mansel
R
.
Benign breast disorders
.
N Engl J Med
.
2005
;
353
(
3
):
275
85
.
2.
Manfrin
E
,
Mariotto
R
,
Remo
A
,
Reghellin
D
,
Falsirollo
F
,
Dalfior
D
et al
.
Benign breast lesions at risk of developing cancer--a challenging problem in breast cancer screening programs: five years’ experience of the Breast Cancer Screening Program in Verona (1999-2004)
.
Cancer
.
2009
;
115
(
3
):
499
507
.
3.
Orsaria
P
,
Grasso
A
,
Carino
R
,
Caredda
E
,
Sammarra
M
,
Altomare
C
et al
.
Heterogeneous risk profiles among B3 breast lesions of uncertain malignant potential
.
Tumori
.
2020
;
106
(
2
):
115
25
.
4.
Catanzariti
F
,
Avendano
D
,
Cicero
G
,
Garza-Montemayor
M
,
Sofia
C
,
Venanzi Rullo
E
et al
.
High-risk lesions of the breast: concurrent diagnostic tools and management recommendations
.
Insights Imaging
.
2021
;
12
(
1
):
63
.
5.
Gagnon
N
,
Martel
E
,
Cadrin-Chenevert
A
,
Ledoux
E
,
Racicot
C
,
Villiard
R
.
Upgrade rate of atypical ductal hyperplasia: ten years experience and predictive factors
.
J Surg Res
.
2021
;
266
:
311
8
.
6.
Pinder
SE
,
Shaaban
A
,
Deb
R
,
Desai
A
,
Gandhi
A
,
Lee
AHS
et al
.
NHS Breast Screening multidisciplinary working group guidelines for the diagnosis and management of breast lesions of uncertain malignant potential on core biopsy (B3 lesions)
.
Clin Radiol
.
2018
;
73
(
8
):
682
92
.
7.
Mohrmann
S
,
Maier-Bode
A
,
Dietzel
F
,
Reinecke
P
,
Krawczyk
N
,
Kaleta
T
et al
.
Malignancy rate and malignancy risk assessment in different lesions of uncertain malignant potential in the breast (B3 lesions): an analysis of 192 cases from a single institution
.
Breast Care
.
2022
;
17
(
2
):
159
65
.
8.
Eusebi
V
,
Feudale
E
,
Foschini
MP
,
Micheli
A
,
Conti
A
,
Riva
C
et al
.
Long-term follow-up of in situ carcinoma of the breast
.
Semin Diagn Pathol
.
1994
;
11
(
3
):
223
35
.
9.
Renshaw
AA
,
Gould
EW
.
Long term clinical follow-up of atypical ductal hyperplasia and lobular carcinoma in situ in breast core needle biopsies
.
Pathology
.
2016
;
48
(
1
):
25
9
.
10.
Richter-Ehrenstein
C
,
Maak
K
,
Roger
S
,
Ehrenstein
T
.
Lesions of “uncertain malignant potential” in the breast (B3) identified with mammography screening
.
BMC Cancer
.
2018
;
18
(
1
):
829
.
11.
Strachan
C
,
Horgan
K
,
Millican-Slater
RA
,
Shaaban
AM
,
Sharma
N
.
Outcome of a new patient pathway for managing B3 breast lesions by vacuum-assisted biopsy: time to change current UK practice
.
J Clin Pathol
.
2016
;
69
(
3
):
248
54
.
12.
Bennett
IC
.
The changing role of vacuum-assisted biopsy of the breast: a new prototype of minimally invasive breast surgery
.
Clin Breast Cancer
.
2017
;
17
(
5
):
323
5
.
13.
Rakha
EA
,
Ho
BC
,
Naik
V
,
Sen
S
,
Hamilton
LJ
,
Hodi
Z
et al
.
Outcome of breast lesions diagnosed as lesion of uncertain malignant potential (B3) or suspicious of malignancy (B4) on needle core biopsy, including detailed review of epithelial atypia
.
Histopathology
.
2011
;
58
(
4
):
626
32
.
14.
Nguyen
CV
,
Albarracin
CT
,
Whitman
GJ
,
Lopez
A
,
Sneige
N
.
Atypical ductal hyperplasia in directional vacuum-assisted biopsy of breast microcalcifications: considerations for surgical excision
.
Ann Surg Oncol
.
2011
;
18
(
3
):
752
61
.
15.
Nicosia
L
,
Latronico
A
,
Addante
F
,
De Santis
R
,
Bozzini
AC
,
Montesano
M
et al
.
Atypical ductal hyperplasia after vacuum-assisted breast biopsy: can we reduce the upgrade to breast cancer to an acceptable rate
.
Diagnostics
.
2021
;
11
(
6
):
1120
.
16.
Schrading
S
,
Strobel
K
,
Keulers
A
,
Dirrichs
T
,
Kuhl
CK
.
Safety and efficacy of magnetic resonance-guided vacuum-assisted large-volume breast biopsy (MR-Guided VALB)
.
Invest Radiol
.
2017
;
52
(
3
):
186
93
.
17.
Krischer
B
,
Forte
S
,
Singer
G
,
Kubik-Huch
RA
,
Leo
C
.
Stereotactic vacuum-assisted breast biopsy in ductal carcinoma in situ: residual microcalcifications and intraoperative findings
.
Breast Care
.
2020
;
15
(
4
):
386
91
.
18.
Rageth
CJ
,
O'Flynn
EA
,
Comstock
C
,
Kurtz
C
,
Kubik
R
,
Madjar
H
et al
.
First International Consensus Conference on lesions of uncertain malignant potential in the breast (B3 lesions)
.
Breast Cancer Res Treat
.
2016
;
159
(
2
):
203
13
.
19.
Rageth
CJ
,
O’Flynn
EAM
,
Pinker
K
,
Kubik-Huch
RA
,
Mundinger
A
,
Decker
T
et al
.
Second International Consensus Conference on lesions of uncertain malignant potential in the breast (B3 lesions)
.
Breast Cancer Res Treat
.
2019
;
174
(
2
):
279
96
.
20.
NBS. P. Clinical guidance for breast cancer screening assessment. NHSBSP publication number 49. 2016 [4th]. Available from: https://www.gov.uk/government/publications/breast-screening-clinical-guidelines-for-screening-management.
21.
Schiaffino
S
,
Calabrese
M
,
Melani
EF
,
Trimboli
RM
,
Cozzi
A
,
Carbonaro
LA
et al
.
Upgrade rate of percutaneously diagnosed pure atypical ductal hyperplasia: systematic review and meta-analysis of 6458 lesions
.
Radiology
.
2020
;
294
(
1
):
76
86
.
22.
Perry
N
,
Broeders
M
,
de Wolf
C
,
Tornberg
S
,
Holland
R
,
von Karsa
L
.
European guidelines for quality assurance in breast cancer screening and diagnosis. Fourth edition--summary document
.
Ann Oncol
.
2008
;
19
(
4
):
614
22
.
23.
Wallis
M
,
Tardivon
A
,
Helbich
T
,
Schreer
I
European Society of Breast Imaging
.
Guidelines from the European society of breast imaging for diagnostic interventional breast procedures
.
Eur Radiol
.
2007
;
17
(
2
):
581
8
.
24.
Dhamija
E
,
Singh
R
,
Mishra
S
,
Hari
S
.
Image-guided breast interventions: biopsy and beyond
.
Indian J Radiol Imaging
.
2021
;
31
(
2
):
391
9
.
25.
D’Orsi
CJSE
,
Mendelson
EB
,
Morris
EA
ACR BI-RADS® atlas: breast imaging reporting and data system
Reston, VA
American College of Radiology
.
2013
.
26.
ACR BI-RADS Atlas: follow-up and outcome monitoring
. Available from: https://www.acr.org/-/media/ACR/Files/RADS/BI-RADS/FUOM-Basic-Audit.pdf.
27.
Schrading
S
,
Distelmaier
M
,
Dirrichs
T
,
Detering
S
,
Brolund
L
,
Strobel
K
et al
.
Digital breast tomosynthesis-guided vacuum-assisted breast biopsy: initial experiences and comparison with prone stereotactic vacuum-assisted biopsy
.
Radiology
.
2015
;
274
(
3
):
654
62
.
28.
Lambert
J
,
Steelandt
T
,
Heywang-Kobrunner
SH
,
Gieraerts
K
,
Van Den Berghe
I
,
Van Ongeval
C
et al
.
Long-term MRI-guided vacuum-assisted breast biopsy results of 600 single-center procedures
.
Eur Radiol
.
2021
;
31
(
7
):
4886
97
.
29.
Cancer NGB
NCCN guidelines breast cancer
.
2022
. Available from: https://www.nccn.org/guidelines/guidelines-detail?category=1&id=1419.
30.
El-Sayed
ME
,
Rakha
EA
,
Reed
J
,
Lee
AH
,
Evans
AJ
,
Ellis
IO
.
Predictive value of needle core biopsy diagnoses of lesions of uncertain malignant potential (B3) in abnormalities detected by mammographic screening
.
Histopathology
.
2008
;
53
(
6
):
650
7
.
31.
S3-Richtlinien. Deutsche Krebsgesellschaft DK, AWMF. S3-Leitlinie Früherkennung, Diagnose, Therapie und Nachsorge des Mammakarzinoms, 2021 [Version 4.4]. Available from: https://www.awmf.org/uploads/tx_szleitlinien/032-045OLl_S3_Mammakarzinom_2021-07.pdf.
32.
WHO
Breast tumours WHO classification of tumours
5th ed
2019
.
33.
IBM SPSS Chicago IL
.
2022
. Available from: https://www.jmp.com/en_gb/offers/statistical-analysis-software.html?msclkid=ea4c9ca569fe1a68616cf8ba4505c4fb&utm_source=bing&utm_medium=cpc&utm_campaign=(JMP)%20EMEA%20-%20ENG%20-%20Search%20-%20NB%20-%20Competitors&utm_term=spss%20software&utm_content=SPSS%20Software.
34.
Services. NCIaIM
DevCan: Probability of developing or dying of cancer software [6.7.6]
.
2019
. Available from: https://surveillance.cancer.gov/devcan/ (accessed on 14 April 2021).
35.
Schiaffino
S
,
Massone
E
,
Gristina
L
,
Fregatti
P
,
Rescinito
G
,
Villa
A
et al
.
Vacuum assisted breast biopsy (VAB) excision of subcentimeter microcalcifications as an alternative to open biopsy for atypical ductal hyperplasia
.
Br J Radiol
.
2018
;
91
(
1085
):
20180003
.
36.
Huang
YX
,
Chen
YL
,
Li
SP
,
Shen
JP
,
Zuo
K
,
Zhou
SC
et al
.
Development and validation of a simple-to-use nomogram for predicting the upgrade of atypical ductal hyperplasia on core needle biopsy in ultrasound-detected breast lesions
.
Front Oncol
.
2020
;
10
:
609841
.
37.
Hann
LE
,
Liberman
L
,
Dershaw
DD
,
Cohen
MA
,
Abramson
AF
.
Mammography immediately after stereotaxic breast biopsy: is it necessary
.
AJR Am J Roentgenol
.
1995
;
165
(
1
):
59
62
.
38.
Lee
IT
,
Ma
KS
,
Luan
YZ
,
Chen
JL
.
Immediate clip migration after breast biopsy: a meta-analysis for potential risk factors
.
Br J Radiol
.
2022
;
95
(
1140
):
20220195
.
39.
Hartmann
LC
,
Radisky
DC
,
Frost
MH
,
Santen
RJ
,
Vierkant
RA
,
Benetti
LL
et al
.
Understanding the premalignant potential of atypical hyperplasia through its natural history: a longitudinal cohort study
.
Cancer Prev Res
.
2014
;
7
(
2
):
211
7
.