| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Departments of Diagnostic Radiology (C.H.L., L.E.P., L.J.H., R.C.L., I.T.) and Pathology (D.C., M.E.C.), Yale University School of Medicine, 333 Cedar St, New Haven, CT 06520. From the 1996 RSNA scientific assembly. Received November 15, 1999; revision requested January 7, 2000; revision received March 7; accepted March 30. Address correspondence to C.H.L. (e-mail: Leec@biomed.med.Yale.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: From July 1992 to March 1999, DCIS without invasion was diagnosed by means of SCNB in 59 patients. Seventeen (29%) were found to have invasive disease after surgery. The underestimation rate for SCNB was compared with that obtained by means of open surgical biopsy. Mammographic and histologic features of cases with and those without invasion were compared.
RESULTS: All patients had calcifications on mammograms. There was no significant difference (P = .26) between the underestimation rate for SCNB with the 11-gauge vacuum-assisted device and that for open surgical biopsy. No statistically significant differences between cases with and those without invasion were seen in patient age, mean number of core specimens, level of suspicion, size of lesion, distribution and morphology of the calcifications, presence of an associated mass or density, subtype of DCIS, nuclear grade, or presence of necrosis or desmoplasia.
CONCLUSION: Mammographic and histologic features cannot be used reliably to predict cases that are underestimated with SCNB. However, SCNB with the 11-gauge vacuum-assisted device was as reliable as open surgical biopsy for diagnosing DCIS without invasion.
Index terms: Breast, biopsy, 00.1261 • Breast, calcification, 00.813 • Breast, surgery • Breast neoplasms, diagnosis, 00.324
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
At our institution, SCNB is offered to all women with nonpalpable mammographic abnormalities for which biopsy is recommended, regardless of the size of the lesion or level of suspicion. Cases are excluded from consideration for SCNB only if they are thought to be technically inaccessible, if the patient cannot tolerate a prone position, or if patient weight is more than 300 pounds. SCNB was performed on a dedicated prone stereotactic unit (Stereoguide, LoRad, Danbury, Conn or Universal, US Surgical, Norwalk, Conn). The first 25 cases were performed with an automated biopsy gun (Biopty; Bard Urological Division, Covington, Ga) and a 14-gauge needle with a 23-mm excursion (Biopty-Cut; Bard Urological Division). The remaining 34 cases were performed with a vacuum-assisted directional biopsy device (Mammotome; Biopsys/Ethicon Endo-Surgery, Cincinnati, Ohio) with an 11-gauge needle (Biopsys/Ethicon Endo-Surgery) (Table 1).
|
Each mammogram was reviewed retrospectively by a radiologist (L.E.P. or I.T.) who was experienced in breast imaging and blinded as to whether invasive tumor was eventually found. By using the terminology of the BI-RADS lexicon where appropriate, the following mammographic features were determined: size of the mammographic abnormality, distribution of calcifications, morphology of calcifications, and presence of associated mass or density. Size was measured as the longest length encompassed by the lesion in three dimensions on the standard craniocaudal and mediolateral oblique views. Magnification images were available for review in 52 of the 59 cases.
Three to six levels were obtained from the tissue blocks, and more levels were not routinely obtained for specimens obtained with the 11-gauge device as compared with those obtained with the 14-gauge needle. For all cases, an experienced breast pathologist (D.C.) retrospectively reviewed histologic slides of the specimens obtained with SCNB and the final surgical specimen obtained with either lumpectomy or mastectomy. Histologic slides of core specimens were reviewed before review of the slides from subsequent surgery, without knowledge of whether invasive tumor was ultimately found.
The following histologic features were noted: the presence or absence of invasive tumor, the subtype of DCIS (comedo or noncomedo), nuclear grade (1, 2, or 3, with 1 being the most well differentiated), presence of necrosis, presence of desmoplasia, presence of inflammatory infiltrate, and the location of the calcifications within the specimen (within areas of invasive tumor or DCIS, in adjacent benign tissue, or in both benign and malignant tissue). Invasion was defined as clear evidence of carcinoma in the stroma. Nuclear grade was determined according to the Nottingham modification of Bloom, Scarff, and Richardson (8). Necrosis was defined as death of numerous cells with residual necrotic or apoptotic nuclei; desmoplasia was scar associated with the periphery of the duct in the basement membrane region; and inflammation was defined as the presence of inflammatory cells, predominantly lymphocytes, associated with the basement membrane area of DCIS.
For each case, surgical management and outcome were reviewed. This included the number of times definitive surgical treatment was achieved as a one-step procedure, the type of surgical treatment (lumpectomy or mastectomy), the number and types of additional surgical procedures, and whether lymph node dissection was performed.
All needle-localized excisional biopsies performed for calcifications and not preceded by SCNB at our facility during the same time period as this study were reviewed to identify cases of DCIS without invasion. The number of times unsuspected invasive disease was found at definitive surgery in these cases was noted to compare this underestimation rate with that occurring after DCIS was diagnosed with SCNB.
The data were analyzed by using a statistics program (SYSTAT version 5.2; SYSTAT, Evanston, Ill). Contingency tables were analyzed by using the 
2 test, with Yates correction. Statistical significance was accepted if the P value was equal to or less than .05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Comparative features of the DCIS and invasive groups are presented in Tables 2 and 3. There was no statistically significant difference in patient age, mean number of core specimens obtained, level of suspicion, lesion size, distribution of calcifications, morphology of calcifications, or presence of an associated mass or density between the DCIS and the invasive groups. There was also no significant difference between the two groups in the subtype of DCIS, the nuclear grade of tumor, and the presence or absence of necrosis or desmoplasia. More cases that ultimately had invasion showed the presence of inflammation at histologic examination of the core specimen (Figure). This difference was statistically significant (P < .001).
|
|
|
Twenty-six of the 42 patients in the DCIS group were treated with lumpectomy followed by radiation, and 16 were treated with mastectomy. Thirty-nine (93%) of the 42 cases in the DCIS group were treated successfully with a single surgical procedure after SCNB. Two patients required reexcision after lumpectomy to achieve negative margins, and one patient underwent mastectomy after attempted lumpectomy yielded positive margins. In the invasive group, six patients were treated with lumpectomy and radiation and 11 underwent mastectomy. Nine (53%) of the 17 cases in the invasive group were treated successfully with a single surgical procedure after SCNB. Of the eight women who required more than one surgical procedure, four underwent attempted lumpectomy but had positive margins and underwent mastectomy, two underwent excisional biopsy prior to definitive surgery because of the surgeons’ reluctance to rely on the results of SCNB, and two underwent axillary lymph node dissection as a separate procedure after initial simple mastectomy revealed the presence of invasive tumor.
Eighteen of the 59 women in this series underwent axillary lymph node dissection. In 10 of these women, axillary lymph node dissection was performed at the time of mastectomy (n = 9) or lumpectomy (n = 1), despite the diagnosis of DCIS without invasion with SCNB, because of the suspicion that invasive tumor could be present based on mammographic and/or histologic findings such as extensive calcifications or comedonecrosis. In seven of these 10 patients, no invasive tumor was found after definitive surgery, and the axillary lymph nodes were negative for tumor. In the other three patients, invasive tumor was present in the mastectomy specimen, and one of the three had positive axillary nodes. Of the remaining eight patients who underwent axillary lymph node dissection, all had invasive disease documented at surgery after SCNB but prior to axillary lymph node dissection.
During the period of this study, 558 needle localization procedures were performed for calcifications but were not preceded by SCNB. DCIS without invasion was found in 74 cases. In 18 (24%) of these 74 cases, no further surgery was performed because it was thought, based on histologic examination of the excised tissue and mammographic extent of the tumor, that the malignancy had been adequately excised. The remaining 56 of the 74 patients with DCIS without invasion underwent further surgery after biopsy, either lumpectomy (33 cases) or mastectomy (23 cases), and unsuspected invasive tumor was found in six (11% of 56 in the reexcised group, 8% of 74 [total] cases of DCIS). This degree of underestimation was significantly less than that obtained with the 14-gauge automated device (P < .001) but was not significantly different from the underestimation rate obtained with the 11-gauge vacuum-assisted device (P = .26).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Given these reports correlating mammographic and histologic findings to subtype of DCIS and the risk of occult invasion, we questioned if factors could be identified that would predict those cases of invasive tumor likely to be underestimated as DCIS without invasion at SCNB. If this were true, the decision as to whether to perform axillary node dissection or sentinel node biopsy could be made prior to definitive surgery. Our results showed, however, that mammographic and most histologic factors, including mammographic lesion size, distribution of calcifications, presence of an associated mass or density, histologic subtype of DCIS, and nuclear grade as determined with core specimens obtained at biopsy, could not be used to predict cases that were associated with the presence of invasion.
We found, however, that the presence of inflammatory infiltrate around areas of DCIS at histologic examination of core specimens was significantly associated with the presence of invasive tumor. The reason for this can only be conjectured. It may be that the patient’s immune system mounts an inflammatory response to the tumor once invasion occurs, and even if the invasion is not detected histologically, the inflammatory infiltrate is evident. Despite the fact that significantly more cases with invasion had inflammatory infiltrate, this was also present in 21% (nine of 42) of cases without invasive tumor and was absent in 24% (four of 17) of cases with invasion. Therefore, the clinical utility of this finding is limited.
Although larger size of tumor and the presence of invasion was not statistically significant (P = .06), there was a trend in this direction. Perhaps with a greater number of cases, lesion size would have been significantly associated with invasive disease. The recognition that mammographic and histologic features of DCIS usually associated with an increased likelihood of invasion cannot be used to reliably predict its presence after SCNB is important so that the decision to perform lymph node dissection or sentinel node biopsy is not made solely on the basis of these factors after DCIS without invasion is diagnosed with SCNB. Although this is not standard practice, axillary lymph node dissection was performed in 10 patients in our series, seven of whom did not show invasion after further surgery, largely because of the surgeons’ reluctance to rely on the diagnosis of DCIS without invasion made on the basis of SCNB findings.
Recent reports of experience with the vacuum-assisted directional probe for performing SCNB have shown a higher degree of concordance between the results of the core biopsy specimens and the final histopathologic diagnosis in cases of DCIS (14,15). Burbank (14) reported that invasive tumor was found at subsequent excision in nine (16%) of 55 cases of DCIS diagnosed with SCNB performed with the automated biopsy device. However, none of the 32 cases of DCIS diagnosed with the directional, vacuum-assisted device had invasion after definitive surgery. This was thought to be related in part to the larger number of core specimens obtained with the vacuum device (mean 26 vs 19 for the automated device) and greater mean weight of the core specimens.
Our series also showed substantially fewer cases of histologic underestimation with the use of the 11-gauge vacuum-assisted device (18%) compared with the 14-gauge automated gun (44%). Although this difference did not reach statistical significance (P = .06), it is likely that with a greater number of cases, this difference would be statistically significant. Our results are similar to those of Won et al (16) who found 15% (three of 20) underestimation of invasive tumor as DCIS with the 11-gauge vacuum-assisted device and 35% (seven of 20) with the 14-gauge automated gun. Again, this may be related to the greater number of core specimens obtained with the vacuum-assisted device in the two series.
In addition, the vacuum-assisted device obtains tissue samples in a contiguous manner around the targeted area. As we have shown, calcifications tend to occur in areas of DCIS and/or adjacent benign tissue rather than in invasive tumor, and contiguous sampling of tissue surrounding the microcalcifications may be as important as increased volume of tissue removed in correctly diagnosing invasive carcinoma when it is present. However, discordance between SCNB results and those of subsequent excision persisted in our series with the use of the 11-gauge vacuum-assisted device.
Because calcifications are specifically targeted during the SCNB procedure, areas of invasive tumor that do not commonly contain calcifications may not be included within the core specimens despite contiguous sampling. Although removing a greater volume of tissue may improve concordance between core specimen and final histologic results, it is doubtful that this will entirely eliminate underestimation. This is further substantiated by our finding of invasive tumor in 11% of cases in which DCIS without invasion was diagnosed by means of needle localization followed by open surgical biopsy of calcifications. Similarly, Rosen et al (17) found unsuspected invasion in the mastectomy specimens in three (6%) of 50 cases after DCIS was diagnosed with excision, and Carter and Smith (18) showed that invasive tumor was found at mastectomy in seven (18%) of 38 cases of DCIS removed by means of excisional biopsy.
In conclusion, we found that calcifications usually occur in areas of DCIS and/or adjacent benign tissue and rarely in the invasive component of the tumor. Therefore, because SCNB specifically targets calcifications, histologic underestimation of cases initially diagnosed as DCIS without invasion will continue to occur regardless of the number of core specimens obtained. We also found that the underestimation rate for cases undergoing SCNB with the 11-gauge vacuum-assisted device was not significantly different from the underestimation rate of open surgical biopsy preceded by needle localization. Therefore, in our series, SCNB performed with the 11-gauge vacuum-assisted device was as reliable for diagnosing DCIS as was open surgical biopsy. Finally, we found that the factors usually thought to be predictive of occult invasion such as size, nuclear grade, and subtype of DCIS do not correlate with the presence of invasive tumor. It is important that physicians who perform SCNB and who perform subsequent definitive surgical therapy in these patients are aware of these points in order to choose the appropriate biopsy technique, optimize treatment strategies, and avoid potentially unnecessary axillary lymph node dissections or sentinel node biopsy.
|
FOOTNOTES |
|---|
Abbreviations: BI-RADS = Breast Imaging Reporting and Data System, DCIS = ductal carcinoma in situ, SCNB = stereotactic core needle biopsy
Author contributions: Guarantor of integrity of entire study, C.H.L.; study concepts and design, C.H.L., D.C., M.E.C., I.T.; definition of intellectual content, C.H.L.; literature research, C.H.L.; clinical studies, C.H.L., D.C., M.E.C., L.E.P., L.J.H., I.T.; data acquisition, D.C., L.E.P., M.E.C., L.J.H., I.T.; data analysis, C.H.L., R.C.L.; statistical analysis, R.C.L.; manuscript preparation, C.H.L., D.C., R.C.L.; manuscript editing, L.E.P., L.J.H., I.T., D.C.; manuscript review, D.C., L.E.P., M.E.C., L.J.H., R.C.L., I.T.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. J. Shin, H. H. Kim, S. M. Kim, H. R. Yang, J.-H. Sohn, G. Y. Kwon, and G. Gong Papillary Lesions of the Breast Diagnosed at Percutaneous Sonographically Guided Biopsy: Comparison of Sonographic Features and Biopsy Methods Am. J. Roentgenol., March 1, 2008; 190(3): 630 - 636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E A Rakha and I O Ellis An overview of assessment of prognostic and predictive factors in breast cancer needle core biopsy specimens J. Clin. Pathol., December 1, 2007; 60(12): 1300 - 1306. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J.-m. Lee, J. B. Kaplan, M. P. Murray, M. Mazur-Grbec, T. Tadic, D. Stimac, and L. Liberman Underestimation of DCIS at MRI-Guided Vacuum-Assisted Breast Biopsy Am. J. Roentgenol., August 1, 2007; 189(2): 468 - 474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. W. M. Lam, W. C. W. Chu, A. P. Y. Tang, G. Tse, and T. K. F. Ma Role of Radiologic Features in the Management of Papillary Lesions of the Breast. Am. J. Roentgenol., May 1, 2006; 186(5): 1322 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. M. Lomoschitz, T. H. Helbich, M. Rudas, G. Pfarl, K. F. Linnau, A. Stadler, and R. J. Jackman Stereotactic 11-gauge Vacuum-assisted Breast Biopsy: Influence of Number of Specimens on Diagnostic Accuracy Radiology, September 1, 2004; 232(3): 897 - 903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Leonard and S. M. Swain Ductal Carcinoma In Situ, Complexities and Challenges J Natl Cancer Inst, June 16, 2004; 96(12): 906 - 920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. E. Hoorntje, M. E. I. Schipper, P. H. M. Peeters, F. Bellot, R. K. Storm, and I. H. M. Borel Rinkes The Finding of Invasive Cancer After a Preoperative Diagnosis of Ductal Carcinoma-In-Situ: Causes of Ductal Carcinoma-In-Situ Underestimates With Stereotactic 14-Gauge Needle Biopsy Ann. Surg. Oncol., August 1, 2003; 10(7): 748 - 753. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Jakub, S. Pendas, and D. S. Reintgen Current Status of Sentinel Lymph Node Mapping and Biopsy: Facts and Controversies Oncologist, February 1, 2003; 8(1): 59 - 68. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. L. Rosen, R. C. Bentley, J. A. Baker, and M. S. Soo Imaging-Guided Core Needle Biopsy of Papillary Lesions of the Breast Am. J. Roentgenol., November 1, 2002; 179(5): 1185 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Liberman, J. B. Kaplan, E. A. Morris, A. F. Abramson, J. H. Menell, and D. D. Dershaw To Excise or To Sample the Mammographic Target: What Is the Goal of Stereotactic II-Gauge Vacuum-Assisted Breast Biopsy? Am. J. Roentgenol., September 1, 2002; 179(3): 679 - 683. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Jackman, R. L. Birdwell, and D. M. Ikeda Atypical Ductal Hyperplasia: Can Some Lesions Be Defined as Probably Benign after Stereotactic 11-gauge Vacuum-assisted Biopsy, Eliminating the Recommendation for Surgical Excision? Radiology, August 1, 2002; 224(2): 548 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Matsumura, J. S. T. Yao, A. A. Nemcek Jr., and M. Remy-Jardin Helical CT Angiography of Thoracic Outlet Syndrome Am. J. Roentgenol., September 1, 2001; 177(3): 714 - 715. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
