| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Departments of Surgery (SP, DH, YS), Pathology (DJ), and Radiology (KS, JT), York Hospital, York, Pennsylvania.
Correspondence: Address correspondence and reprint requests to: Steven Pandelidis, MD, Apple Hill Surgical Associates, 25 Monument Rd., Suite 220,York, PA 17403; Fax: 717-741-3604; E-mail: sjasj{at}aol.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Methods: Between October 1996 and March 2000, 1341 patients underwent stereotactic 11-gauge vacuum-assisted biopsy. Patients with ADH or DCIS were encouraged to undergo excisional biopsy.
Results: Surgical excision of 37 ADH lesions revealed 5 missed DCIS lesions and 1 missed invasive cancer. Twelve of 91 DCIS lesions were upstaged to invasive cancer upon excision. The underestimation rate was highest in patients with DCIS when the target lesion for biopsy was a zone of calcifications >1.5 cm. No correlation existed between the histologic features of DCIS lesions diagnosed by stereotactic biopsy and the presence of infiltrating disease on excision.
Conclusions: Vacuum-assisted 11-gauge stereotactic core biopsy understages 13.2% and 13.5% of DCIS and ADH lesions, respectively. In patients with DCIS found by stereotactic biopsy, a target zone of calcifications >1.5 cm is associated with a higher underestimation rate of infiltrating disease.
Key Words: 11-gauge vacuum-assisted core biopsy • Mammographic • Stereotactic • Breast lesions
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Veteran surgeons practicing today began their careers with a straightforward approach to breast cancer. A patient, under general anesthesia, would undergo excisional or incisional biopsy of a breast mass. If cancer was found on frozen section, then the surgeon would proceed with a mastectomy and axillary dissection. The surgeon was primarily responsible for the diagnosis and hopefully the cure for the disease. There was little role for radiologists, medical oncologists, and radiation oncologists in the disease’s treatment.
Increased public awareness together with improvements in mammography has led to earlier diagnosis. In recent years, cancer is often diagnosed when it is in situ. The majority of women diagnosed with breast cancer are candidates for wide excision and radiation rather than mastectomy. Routine axillary dissection is being replaced by sentinel lymph node biopsy and selective axillary dissection.
The development of stereotactic biopsy enables radiologists or surgeons to make a diagnosis on an awake patient. If that diagnosis is accurate, then the patient can have her definitive surgical treatment as an outpatient with one trip to the operating room. If sentinel node surgical techniques are used in combination with selective axillary dissection, the surgery can often be done with local anesthesia and intravenous sedation.
Early stereotactic techniques were problematic. Smaller gauge needles would often under sample the lesions in question and lead to false-negative results. Also, multiple passes were required to accurately sample a mammographic abnormality. An early report by Jackman et al. revealed that 9 of 18 lesions showing atypical ductal hyperplasia on a14-gauge core needle biopsy ultimately proved to be carcinomas on excisional biopsy.2 A more recent report by the same author suggests that sampling of such lesions can be improved with the use of an 11-gauge vacuum-assisted device.3
The current study seeks to define the accuracy of stereotactic biopsy when an 11-gauge vacuum-assisted device is used in diagnosing atypical ductal hyperplasia (ADH) and ductal carcinoma in situ (DCIS). A secondary aim of the study was to try to define mammographic and histologic features of DCIS lesions, which would increase the likelihood of diagnosis of infiltrating cancer upon excision.
The overall accuracy, sensitivity, and specificity of stereotactic biopsy in diagnosing malignant breast disease in women with suspicious mammograms were not evaluated in this study.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A dedicated Lorad stereotactic table (Danbury, CT) was used in combination with the Ethicon Endo-surgery Biopsys system (Cincinnati, OH). Multiple 11-gauge vacuum-assisted biopsies were acquired from each suspicious lesion. When calcifications were the target in question, specimen radiographs confirmed adequate sampling of the microcalcifications.
Core biopsy specimens were sampled by standard pathologic techniques. Any patient who was found to have ADH, atypical lobular hyperplasia (ALH), lobular carcinoma in situ (LCIS), or DCIS went on to needle localization and excisional biopsy. Patients with palpable breast masses or enlarged lymph nodes were not included in the current study.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Five of 37 ADH lesions (13.5%) on 11-gauge vacuum-assisted core biopsy were found to harbor carcinoma on surgical excision. Four patients were found to have DCIS, and one patient was found to have infiltrating ductal cancer (Table 1).
|
On histologic analysis of DCIS lesions, there were 35 lesions with comedonecrosis (CN). Fifty-six lesions were moderately or poorly differentiated. All 35 lesions with CN were either moderately or poorly differentiated.
On further histologic analysis, we found that 6 of 35 (17.1%) DCIS lesions with CN were upstaged to invasive cancer on excisional biopsy compared with 7 of 56 (12.5%) DCIS lesions without CN (P = .09). Eight of 63 (12.7%) moderate or high-grade DCIS lesions were upstaged to invasive cancer compared with 5 of 28 (17.9%) low-grade DCIS lesions (P = .75). Also, 5 of 35 (14.3%) DCIS lesions with both CN and moderate or poor histologic grade compared with 8 of 56 (14.3%) low-grade DCIS lesions with or without CN were upstaged to invasive cancer on excisional biopsy (P < .05). Therefore, neither the histologic grade of DCIS nor the presence of CN was predictive of the discovery of infiltrating ductal carcinoma (IDC) on excisional biopsy.
Calcifications were the target of biopsy in 86 of the 91 DCIS lesions diagnosed by stereotactic sampling. In 10 of these 86 lesions, the target zone of calcifications on mammography was > 1.5 cm. Sixty percent of these DCIS lesions were upgraded to infiltrating cancer upon excision. Only 5 of 76 target lesions (15%) with a breadth of calcifications 
1.4 cm were upgraded to infiltrating cancer. Therefore, zones of calcification > 1.5 cm were more likely to ultimately harbor infiltrating cancer than smaller zones of calcification (P < .05).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Less mutilating definitive surgery is used to treat the disease in the breast, and presently, obligatory axillary dissection is losing its prominent role. Radiation oncologists have made it possible for surgeons to preserve women’s breasts, and nuclear medicine physicians have been helpful in the development and practice of sentinel lymph node surgery.
With the expansion of mammographic screening and the widespread use of image-guided core biopsy techniques, a new area of controversy has evolved. Biopsy techniques that only sample a portion of a mammographic lesion will inevitably have less specificity than excisional biopsies.
To avoid the danger of false-negative biopsy results, any lesions that reveal ADH are subjected to excisional biopsy. Evaluation of the entire lesion can lead to the identification of DCIS or IDC. Similarly, lesions found to contain DCIS on core biopsy will sometimes demonstrate infiltrating cancer upon excision.
The key to accurately assessing a mammographic lesion is evaluation of an adequate volume of tissues. Early stereotactic techniques relied on fine-needle aspiration for cytologic analysis of a lesion. Subsequently, automated core biopsy devices were developed to produce a histologic sampling of tissue. Fourteen-gauge needles were used to sample the lesion. To make up for the small volume of tissue extracted per pass, multiple passes were made through the lesions. The procedure was therefore time-consuming for the physician and uncomfortable for the patient.
Subsequently, vacuum-assisted 14-gauge needle devices were developed. This technology eliminated the need for multiple percutaneous "sticks" but still was limited by the volume of tissue harvested. The next evolution in technology was the 11-gauge vacuum-assisted biopsy device. The larger needle combined with vacuum-assisted harvesting of tissue allows for extraction of a significant volume of tissue, thus allowing the pathologist to better characterize the mammographic lesion.
As previously stated, automated 14-gauge core biopsy devices can only procure a limited amount of tissue. On average, a 14-gauge automated core biopsy device will harvest 17 mg of tissue, whereas a vacuum-assisted 14-gauge device will harvest 44 mg of tissue, and lastly, an 11-gauge vacuum-assisted device can harvest 105 mg of tissue.4,5
In a study by Jackman et al., stereotactic core biopsy with a 14-gauge core device found 24 ADH lesions. On needle localization and excisional biopsy, 14 (58%) of these lesions revealed a neoplasm (9 lesions DCIS and 5 lesions IDC).6 Similarly, Liberman et al. demonstrated that ADH discovered by 14-gauge core biopsy often under represents the severity of a mammographic lesion. In eight such patients who went on to surgical excision, four were found to have DCIS, and two were found to have infiltrating cancer.7
In another study by Jackman et al., the reliability of 14-gauge vacuum-assisted biopsy was evaluated. In this study, 13 (18%) of 74 lesions showing ADH on vacuum-assisted core biopsy proved to be DCIS or IDC on needle localization and excisional biopsy.8 In a similar study, Burbank assessed the ability of the same technique to accurately diagnose DCIS. Nine of 55 lesions (16%) demonstrating DCIS on core biopsy were ultimately found to represent IDC upon excision. In 18% of biopsies showing DCIS or ADH on core sampling, no residual lesional tissue was evident upon excision.9
To overcome undersampling with stereotactic techniques, the 11-gauge vacuum-assisted core biopsy device was developed. Meyer et al. demonstrated improved diagnostic accuracy with utilization of the larger device. Only 1 of 9 ADH lesions found on vacuum-assisted core biopsy was upgraded to DCIS on excision. Similarly, only 1 of 28 lesions was upgraded from DCIS to infiltrating ductal cancer.10
In our study the underestimation rate of 13% by 11-gauge stereotactic vacuum-assisted core biopsy is concurrent with data published by other investigators in the field. Burak et al. found that 13% (6 of 46) and 11% (10 of 89) of ADH and DCIS lesions on 11-gauge stereotactic vacuum-assisted core biopsy were upstaged on surgical excision.11 Similarly, Maganini et al. found that 13% (4 of 32) of ADH lesions on 11-gauge stereotactic vacuum-assisted core biopsy were upgraded on surgical excision.12 More recent data by Brem et al. demonstrated a similar underestimation rate of 11% (4 of 38) on surgical excision of DCIS lesions sampled by 11-gauge stereotactic vacuum-assisted core biopsy.13
Underestimation of malignant disease with 11-gauge stereotactic vacuum-assisted core biopsy is increased when 10 or fewer cores are obtained per lesion.14 In our study the number of cores obtained per lesion ranged from 3 to 20 cores with a median of 6 and a mean of 9 cores. A low number of cores per lesion were obtained in the early course of the study. Also, if nodules were the targets for biopsy, fewer cores were taken. With the emergence of new data, we increased the number of cores per lesion, and currently our practice is to perform 10 or more cores per lesion.
Any biopsy technique that only samples a portion of a lesion will be a challenge to the pathologist. The greatest difficulty arises when calcifications are the target of stereotactic biopsy. A continuous spectrum of proliferative change exists between ADH, low-grade DCIS, high-grade DCIS with CN, and microinvasive ductal carcinoma. Although the differentiation of the above pathologic entities can be subjective, certain objective criteria exist to aid pathologists in the categorization of proliferative and neoplastic lesions. ADH had been subjectively defined as a lesion containing some but not all features of DCIS but involving only a single duct.12 More recently, DCIS has been differentiated from ADH on the basis of volume. Lesions fulfilling the definition of DCIS but having an aggregate diameter < 2 mm are defined as ADH.13 Clearly, if a pathologist is given only limited tissue to analyze, he or she may be unable to find an adequate volume of neoplastic change to make a diagnosis of DCIS.14 The threshold between pathologic entities can be subjective and poorly reproducible, especially between ADH and low-grade non-CN DCIS.15
Our data demonstrated a statistically significant increase in the underestimation rate of invasive cancer in DCIS when the target zone of calcification was > 1.5 cm. Hence, one needs to be vigilant for the possibility of IDC on excision of DCIS lesions when the target zone of calcification on mammography is > 1.5 cm.
There were no histologic factors analyzed that proved to be predictive of the underestimation of malignant disease. In summary, 17.1% (6 of 35) of DCIS lesions with CN compared with 12.5% (7 of 56) of DCIS lesions without CN were upstaged to invasive cancer on excisional biopsy. Also, 12.5% (8 of 63) of moderate or high-grade DCIS lesions and 17.9% (5 of 28) of low-grade DCIS lesions upgraded to IDC on excisional biopsy (P = .75).
Eleven-gauge vacuum-assisted core biopsy is an acceptable means to sample mammographic abnormalities with a 13.5% and 13.2% underestimation rate of malignant disease in ADH and DCIS, respectively. Hence, even with 11-gauge vacuum-assisted biopsy, patients found to have ADH or DCIS require needle localization and excisional biopsy to accurately characterize their lesions. Moreover, the operating surgeon performing an excisional biopsy on a patient with DCIS on whom the target of stereotactic biopsy was a zone of calcification > 1.5 cm, must be vigilant of the possibility of finding IDC. Larger expanses of microcalcifications are probably best initially sampled by needle localization and excisional biopsy rather than stereotactic core biopsy.16–18
|
Footnotes |
|---|
Received for publication May 2, 2002. Accepted for publication September 11, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  P. R. Eby, J. E. Ochsner, W. B. DeMartini, K. H. Allison, S. Peacock, and C. D. Lehman Is Surgical Excision Necessary for Focal Atypical Ductal Hyperplasia Found at Stereotactic Vacuum-Assisted Breast Biopsy? Ann. Surg. Oncol., November 1, 2008; 15(11): 3232 - 3238. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Sohn, Z. Arthurs, G. Herbert, J. Keylock, J. Perry, M. Eckert, D. Fellabaum, D. Smith, and T. Brown Atypical Ductal Hyperplasia: Improved Accuracy with the 11-Gauge Vacuum-Assisted versus the 14-Gauge Core Biopsy Needle Ann. Surg. Oncol., September 1, 2007; 14(9): 2497 - 2501. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Liberman, A. E. Holland, D. Marjan, M. P. Murray, L. Bartella, E. A. Morris, D. D. Dershaw, and R. T. Wynn Underestimation of Atypical Ductal Hyperplasia at MRI-Guided 9-Gauge Vacuum-Assisted Breast Biopsy Am. J. Roentgenol., March 1, 2007; 188(3): 684 - 690. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M F Dillon, C M Quinn, E W McDermott, A O'Doherty, N O'Higgins, and A D K Hill Diagnostic accuracy of core biopsy for ductal carcinoma in situ and its implications for surgical practice. J. Clin. Pathol., July 1, 2006; 59(7): 740 - 743. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
