Clinicopathological Analysis of Ultrasound-guided Vacuum-assisted Breast Biopsy for the Diagnosis and Treatment of Breast Disease

Discussion

The incidence of breast cancer has steadily increased in South Korea as well as in many Western countries. Young premenopausal women in Asian countries, including Korea, have a higher incidence of breast cancer than their Western counterparts; this has increased the level of interest in screening for early-stage breast cancer.

Recent advances in ultrasonography have allowed early detection of non-palpable malignant lesions smaller than 1 cm, contributing to the increase in breast cancer survival rates. Although most surgeons still opt for traditional excision biopsy of palpable lesions, this method is expected to be used increasingly less frequently. Worldwide, core-needle biopsy is the most commonly used method for both non-palpable and palpable lesions, given its high accuracy, cost effectiveness, low complication rate (hemorrhage or infection), and convenience. Despite these advantages, it is not a fail-proof method due to its inherent drawbacks, namely, high rates of false-negative findings and misclassification (underestimation of disease), which are inevitable because the procedure samples only parts of the target lesion, not the entire lesion. Morris et al. reported that 29% of breast lesions are heterogeneous, yielding different histologic results from the center and periphery; thus, targeting one part of a heterogeneous mass could lead to misdiagnosis (2). Moreover, Joshi et al. reported that core biopsy findings of ADH underestimate the diagnosis of malignancy by 18-88% (3).

Carder et al. performed additional excision biopsies on 26 patients who were diagnosed with B3 lesions (uncertain malignant potential) on core needle biopsy specimens. Although 22 had benign pathology at surgical excision, four had malignant lesions. According to the authors of that study, the difficulty in establishing a histologic diagnosis from specimens obtained via core needle biopsy stems largely from the limited amount of tissue obtained and uncertainties over whether the tissue is representative. They maintained that newer vacuum-assisted techniques, such as the mammotome, provide larger tissue samples and may help to reduce the rate of misclassification of tumors as being benign (4). VABB, developed to overcome the shortcomings of percutaneous biopsy, is now widely used because it allows complete removal of the target lesion and guarantees high diagnostic accuracy and therapeutic results (5).

The main indications for VABB at our institution are palpable or non-palpable BI-RADS ultrasound category 3 or 4a lesions, for which complete excisional VABB is usually performed. Category 4b-5 lesions are generally subjected to core biopsy, but incisional biopsy with VABB may be performed to acquire a larger sample for diagnostic purposes in patients on neoadjuvant chemotherapy who require a more accurate histologic and molecular biological diagnosis (5, 6). No consensus has yet been reached on how to deal with USG BI-RADS category 3 lesions, which are commonly found on ultrasonography. The conventional management of category 3 lesions, which have a malignancy rate of 0.5-2%, is 3-6-monthly follow-up with ultrasonography; this is less invasive and more cost-effective than percutaneous core biopsy. Some reports suggest that the triple test can differentiate between benign and malignant lesions with an accuracy of approximately 95%, and routine follow-up for 1-3 years to monitor changes in the features of the lesion is recommended (6-8). Proponents of this conservative approach argue that most fibroadenomas, which make up the vast majority of breast benign lesions, have already stopped growing at the time of their discovery, and even if they grow further, such growth generally stops at 2-3 cm in diameter. In fact, the lesions decrease in size in 15% of cases, and continued growth occurs in only 5-10% of all cases. Moreover, progression of a fibroadenoma to breast cancer is extremely rare (6-9).

Even though follow-up at 6-monthly intervals is recommended in most patients with category 3 lesions, the authors recommend VABB as an option for the following patients: i) those with a low probability of attending regular follow-up, ii) those planning a pregnancy or breast augmentation surgery, iii) those with extreme anxiety related to the lesion, iv) those with a lesion that increases in size during the course of follow-up, v) those with subjective symptoms or pain with BI-RADS ultrasound category 3-4 lesions, vi) those with a palpable mass, vii) those with a personal or family history of breast cancer, viii) those with BI-RADS ultrasound category 3 lesions planning conservative surgery in the ipsilateral breast, and ix) those with suspected multifocal or multicentric breast cancer (10, 11).

If the option of biopsy is chosen for category 3 or 4 lesions, one should decide whether to use VABB or core needle biopsy using an automated biopsy gun. Parker recommends excision via VABB for lesions <1.5 cm and conventional core needle biopsy for larger lesions, because accurate targeting is crucial for small lesions, especially those <1 cm; this is a great challenge when using a biopsy gun (11). Even if the surgeon succeeds in obtaining five tissue samples or more, it is often difficult to confirm whether tissue from the area of interest has been sampled. Such uncertainty increases if the biopsy results point to nonspecific benign disease and if the biopsy results contradict the ultrasonographic findings; this may prompt the surgeon to perform additional surgical excision. The problem of uncertainty can be resolved by completely removing the lesion via ultrasound-guided VABB (12).

In this study, the majority of patients (58.2%) were women aged <39 years, a considerably higher proportion than those reported in studies by Kim D et al. (13) and Kim SH et al. (14), of 37.2% and 35.2%, respectively. Various factors may be behind the different results found in studies from Korea, such as lower childbirth rates associated with delayed marriage, higher rates of smoking and alcohol consumption, and higher workplace stress. More importantly, however, it may be attributable to the increased interest and more active participation in breast self-examination and ultrasound screening by Korean women in their 20s and 30s, spurred on by the increased incidence of breast cancer among premenopausal women in Korea.

In this study, fibroadenoma was the most common histopathological type of benign breast tumor, accounting for 46.6% (n=5,226) of cases. It is still debatable whether fibroadenomas are likely to develop into breast cancer. Based on long-term follow-up results, many studies have reported that patients with biopsy-confirmed fibroadenomas have a high relative risk of developing breast cancer (15-18). In a study by Carter et al., among 685 patients with fibroadenoma followed up for 8.3 years, the relative risk of breast cancer was 1.7 (16). Thus, breast specialists should recommend active follow-up of patients with histologically confirmed fibroadenomas. Moreover, if a breast tumor is discovered in a patient with a growing fibroadenoma or with a family history of breast cancer, a more aggressive follow-up approach should be adopted (such as removing the tumor or performing a biopsy), rather than opting for routine follow-up.

In the current study, it was observed that patients commonly presented with multiple lesions (78.2%), and that most lesions (n=4,443 [39.6%]) were 0.6-1.0 cm in size. The largest lesion was a fibroadenoma measuring 7.9 cm. Surgical excision biopsy was recommended for this patient, but she refused this option as she had a tendency to develop hypertrophic scarring. She thus underwent successful excisional VABB with no complications or residual lesion. The procedure time was approximately 15 min, and 97 core pieces were removed.

In this study, 1,276 (11.4%) lesions were defined as being relatively large (>2.1 cm in long-axis diameter on ultrasonography). Approximately two-third of the lesions occurred in women younger than 40 years, and 10.2% were found to be malignant. The rate of malignancy increased with age. In other words, whereas the rate of having relatively large tumors was higher among women aged <40 years than ≥40 years, the probability of relatively large tumors being malignant tended to increase with increasing age. There is no clear explanation for the high prevalence of breast tumors among young women aged 20-39 years, but it is suspected that this may be attributable to physical growth and increasing obesity rates due to Westernized diet and lifestyle, as well as excessive alcohol consumption (19-21). A further finding was that approximately 43.3% of relatively large tumors were non-palpable, emphasizing the need for greater awareness of the necessity to perform regular ultrasonographic breast examinations among women younger than 40 years. Although no comparative analysis can be made due to lack of research reports on positive correlations between tumor size (measured by ultrasound) and malignancy rate, our findings suggest the need to biopsy relatively large tumors found on ultrasound examination in patients aged ≥40 years to determine whether such lesions are malignant.

According to the first edition of BI-RADS, ultrasound category 3 lesions typically have a malignancy rate of 0.5-2%, whereas the overall positive predictive values for categories 4 and 5 are 3-94% and >95%, respectively (22). In this study, the PPVs for BI-RADS ultrasound categories 3, 4a, 4b, 4c, and 5 were 0.6%, 3.4%, 34.8%, 66.2%, and 93.8%, respectively. These findings are fairly consistent with results from previous studies. Wiratkapun et al. reported that 94% of 497 patients with category 5 lesions were diagnosed with breast cancer and that categories 4a, 4b, and 4c showed PPVs of 9%, 21%, and 57%, respectively (23). Yoon et al. reported PPVs of 7.6% (149/1963), 37.8% (68/180), and 81.9% (235/287) for category 4a, 4b, and 4c lesions, respectively (24). These results, obtained from different research institutions, are all very similar. Furthermore, the ratio of invasive to noninvasive carcinomas were 31.0%, 42.8%, 68.8%, 96.1%, and 86.8% for BI-RADS ultrasound categories 3, 4a, 4b, 4c, and 5, respectively. Linear regression analysis demonstrated an increasing trend in the ratio of invasive to non-invasive carcinoma with an increasing level of BI-RADS ultrasound category (p<0.05).

These results may be attributable to the BI-RADS ultrasound classification system used. Various studies, such as those conducted by Kim EK et al. (25) and Jales et al. (26), have reported that suspicious ultrasound findings can be divided into major and minor suspicious findings: Major suspicious findings include an irregular shape, spiculated margin, and microcalcifications, whereas minor suspicious findings include a microlobulated or angular margin, nonparallel orientation, duct extension, complex echogenicity, and posterior shadowing. Accordingly, the categories were re-categorized as category 4a (one or more minor suspicious findings), category 4b (more than three minor suspicious findings), category 4c (one major suspicious feature with or without minor suspicious features), and category 5 (two or more major suspicious findings). Since higher category levels include major findings such as spiculated margin and/or irregular shape, the proportion of invasive carcinomas rather than in situ lesions would increase.

In terms of the number of core specimens required and VABB procedure time, from our findings it is inferred that to successfully achieve complete excision of lesions <2.0 cm in size using an 8-gauge probe, at least 12 core samples are needed, while an average of 18 samples are needed for lesions <3 cm in size. Although no comparative analysis can be made on the number of cores removed or the procedure time due to the lack of related research data, the results of this study demonstrate that the mean procedure time is close to three minutes, which is considerably shorter than the time required for conventional surgical excision biopsy. Even if an additional 5-10 min is required to apply pressure to stop bleeding and it may take longer than 3 minutes for less experienced surgeons, the time-saving benefit of VABB remains clear.

To date, several complications have been reported with the use of VABB. However, in this study, there were very few complications, such as damage to breast implants; severe bleeding requiring conversion to open surgery; and misclassification (histological underestimation), except in cases of ADH. Regarding the success rate of VABB, complete removal was confirmed in 94.4% of all cases followed-up; residual lesions were identified in the remaining 5.6% of cases. For patients with malignant tumors, residual lesions were not problematic because additional breast cancer surgery was performed. In patients with benign tumors, the residual lesion typically did not require excision surgery. However, because the residual lesion can regrow, follow-up ultrasound examination is required. In this study, four patients required a repeat VABB procedure because regrowth was detected during the follow-up period – the lesions in all four cases were confirmed to be benign.

Many studies have reported on histologic underestimation of cases of ADH diagnosed by ultrasound-guided VABB. Grady et al. reported that the rate of ADH underestimation was 33% (6 of 18) when incisional VABB was performed and 0% (0 of 29) when complete ultrasound-guided, percutaneous excision was performed, a result equivalent to open surgical biopsy. They stated that complete ultrasound-guided, vacuum-assisted, percutaneous excision is more accurate than non-excisional ultrasound-guided biopsy, and that patients so diagnosed have very low underestimation rates and may not require open surgical re-excision (27). In this study, eight (14.0%) of the 57 cases of ADH that were diagnosed after complete excision biopsy using ultrasound-guided VABB had histological underestimation. Hence, patients diagnosed with ADH by VABB should undergo additional surgical biopsy if deemed necessary. Moreover, further research is required to determine the indications for additional surgical biopsy in patients with ADH diagnosed on VABB specimens.

This was a retrospective study of the effectiveness ultrasound-guided VABB in Korean patients, the clinical profile of such patients, and the histopathological profile of the biopsied lesions. The study demonstrated a high incidence of tumors in young Korean women, unlike in the West. However, the PPV of VABB, based on BI-RADS ultrasound category, was very similar to those reported in previous Western studies. The authors also examined how many core specimens need to be removed for lesions of different sizes to achieve complete excision of the target lesion using 8-gauge probe VABB; these results are considered to be especially meaningful. However, this study also had some limitations. Because the sample size was too large, it was not possible to conduct an analysis by size and age in patients with residual lesions. The authors were also unable to investigate the incidence of hematoma and minor bleeding, which were the most common complications.

In conclusion, VABB excision of breast lesions is safe and useful for both diagnostic and therapeutic purposes. Its use is associated with minimal serious adverse events such as severe bleeding, histological underestimation, and incomplete removal of target lesions. It overcomes the drawbacks of ultrasound-guided core biopsy and surgical excision biopsy methods, that is expected to completely replace. The participants of this study are still being followed-up with ultrasound examinations at 6-monthly intervals, and the authors plan to report the long-term follow-up findings.