More than a quarter million cases of breast malignancy were diagnosed in 2005, including about 25% as in situ disease. The sine qua non of breast cancer is a tissue diagnosis. Clinical detection of cancer, usually secondary to a lump, has often prompted diagnosis by percutaneous techniques or excision. Until the 1990s, the former relied primarily on fine needle aspiration biopsy (FNAB) which invited insufficient material in 15% of 25% of cases and relied on expertise in cytologic evaluation that was not easily applied in a uniformly standardized manner. More recently, histologic diagnosis has been available for percutaneous biopsy with larger gauge needles, following the advancement of such technology prompted by its increasing use for image-guided biopsy attendant to the detection of clinically occult lesions through screening mammography. More than half of women over age 40 have had a mammogram within the past year, and more than 32 million mammograms are performed annually in the United States.
Although a number of noninvasive techniques have been and are being developed to assess the likelihood of a detected abnormality as representing cancer, the diagnosis, as mentioned, virtually always requires tissue (an exception is illustrated in the clinical diagnosis of inflammatory breast cancer). Again, prior to the 1990s, most clinically occult lesions were excised following the preoperative mammographically guided localization of such lesions by needles, dye, wires, or a combination thereof, first described in 1965. 1 Using freehand placement or employing grids (alphanumeric or fenestrated), the location of the needle tip could be radiographically confirmed. Reported rates of surgery failing to recover the suspect lesion ranged from 2% to 37%, but generally approximated 2% to 3%. 2 Similar logistics can be applied to localizing lesions by stereotactic, sonographic, or MRI guidance.
Because a majority of screening-detected lesions—by imaging or clinical examination—do not represent cancer, growing interest has prompted the development of new technologic methods for percutaneous tissue sampling often referred to as core needle biopsy (CNB). In addition, based on several single institutional or retrospective studies as well as the only published multi-institutional prospective study, certain lesions recovered at CNB require reassessment to determine if more tissue is required (by means of excision) because of an association with malignancy in the immediate anatomic focus. 3-5
Initial use of CNB was reported with the utilization of stereotactic imaging and extended to ultrasound-guided biopsies, generally with 14G needles; the use of smaller gauge needles has been reported as less accurate. Brenner et al reported on the results of a prospective study showing that with five samples obtained with a 14G needle, a 91% sensitivity could be achieved; a 92% sensitivity achieved if excision was prompted by the diagnosis of so-called “high risk” lesions; and a sensitivity of 98% achieved if a subsequent assessment of CNB histologic results and mammographic findings indicated concordance. 3 Note that such a false-negative rate of 2% is comparable to that achieved by a direct surgical approach, but involves individual variation in radiologic assessment of concordance (note also that there is variation in histologic interpretation as well). 2 Success varies for five CNB samples depending on the type of lesion subject to biopsy, highest accuracy for masses, and least accuracy for calcifications, the inference being that evaluation of calcifications required more tissue sampling.
|Needle localization of mass in upper outer quadrant by freehand technique. Tip abuts suspicious mass. (Click the image for a larger version.)|
In addition to obtaining more samples, operators could avail themselves of new technology. Larger gauge needles became available, as did single insertion devices that could be rotated within the breast to obtain tissue from different orientations. Often called vacuum-assisted devices, the use of vacuum technology helped secure the core sample within the device, but did not compensate for inadequate placement of the needle, a misunderstanding that continues even today. In the later 1990s, 11G needles were found to “undersample” lesions with less frequency and provided a more accurate assessment of the etiology of an imaged lesion. Using 12 samples of 11G CNB material, however, Philpotts reported that in lesions characterized as calcification lesions, 11G needles underestimated the severity of the disease in about 11% of cases. 6
CHALLENGES AND PITFALLS
Undersampling may refer to two types of situations. The first is identifying malignant disease but showing only intraductal disease (eg, DCIS) where invasive disease is present. The second is in identifying nonmalignant disease where in fact malignant disease is present; certain types of so-called high-risk lesions have been subject to closer evaluation. In an attempt to avoid this potential problem, even larger needles have been developed. Initial optimism about the ABBI (Advanced Breast Biopsy Instrument, US Surgical) device, which sought to obtain very large samples and required postprocedure suturing, was frustrated by frequent malfunctions. More recently, a device using radiofrequency transmission instead of a cutting needle has been introduced such that a 1 cm or even 2 cm intact sample can be removed percutaneously. Issues regarding RF artifact and outcomes are currently being explored with respect to its impact on clinical practice. A variety of devices and needle sizes are available in the marketplace.
The underestimation of malignant disease, although likely to prompt further surgery, is not trivial. In general, the diagnosis of DCIS that does not extend over a large area of the breast does not incur the morbidity of subsequent nodal sampling—sentinel node or full axillary sampling—for purposes of staging. Current techniques have not obviated nodal sampling when a malignant diagnosis is made.
More important, certain lesions are associated with malignant disease and, though benign on initial histologic evaluation, may prompt further excision. The most studied of these lesions is atypical ductal hyperplasia (ADH), where association with frank malignancy, though reported as high as 50%, is found in about 20% of cases. 3,7 Indeed, this is not surprising because the diagnosis of DCIS is often made on the basis of both quantitative aspects (eg, how many duct-lobular units are involved), as well as qualitative parameters (cytologic findings, architectural findings). Indeed, ADH and noncomedo DCIS share certain genetic markers.
There is sufficiently more controversy regarding lobular neoplasia, which encompasses both atypical lobular hyperplasia (ALH) and lobular carcinoma in situ (LCIS). Because so few lesions have been studied, simple conclusions regarding the need for excision cannot be drawn. Even one of the larger studies to date (with less than 40 cases) combined two institutional results where methods of biopsy were considerably different. 8 ALH has been shown to be associated with fewer instances of cancer development, but when cancer occurs, it tends to be more often associated with the ipsilateral breast in which ALH was identified. 9 The overall risk of cancer when lobular neoplasia is found, based on 20-year follow-up data, is estimated at about 19%, including both invasive ductal and lobular carcinoma in either breast at any location. Reports often suggest a 10% to 14% rate of cancer found at excision following the CNB diagnosis of ALH or LCIS. 10 However, similar rates of malignancy are also reported as incidental malignancy found during the removal of otherwise benign calcifications. 11
Radial scar is another designated “high risk” lesion. Historically confused with tubular carcinoma due to a similar architecture, immunohistochemical staining now distinguishes the two entities in a majority of cases by the demonstration of myoepithelial cells intimately associated with the epithelium of radial scar lesions. Prior literature indicating the need to excise radial scars found at CNB has been lacking sufficient data. A large multi-institutional study found that, when 12 or more samples were obtained at CNB, and no atypia was present, there were no cases of malignancy. 12
Papillary lesions are also considered “high risk” in that they may often contain both malignant and benign components. One of the larger series, which reported on only 46 cases, suggested a negative predictive value of 93% in the absence of atypia. While the authors concluded this might obviate the need for excision, others interpret the data as requiring excision of the anatomic focus subject to CNB, unless the entire papillary lesion has been removed. 13
DIFFICULT DEFINITIVE DIAGNOSES
Other entities may be difficult to diagnose definitively as benign, even though they have no clear association with malignancy. Some benign phyllodes tumors may have areas so indistinguishable from fibroadenomata that a misdiagnosis may be made. The presence of hypercellularity of stromal components usually alerts the pathologist to this possibility. Because the imaging findings of the two entities are similar, the continued growth of the lesion should prompt excision on this basis. Focal fibrosis is a somewhat nonspecific diagnosis (as is fibrocystic change) but can present as a nonspecific masslike appearance. Adequate sampling, corroborated by image documentation of the biopsy device transgressing the lesion, should help to reinforce the validity of the CNB diagnosis. Pseudoangiomatous stromal hyperplasia (PASH) is an unusual cause of new or developing focal asymmetry or mass on a mammogram and is a diagnosis that can again be made by CNB, provided there is adequate sampling and documentation as described above, because this is not a widely prevalent entity that is likely to be serendipitously found. In like manner, sclerosing adenosis can sometimes be difficult for the pathologist to establish from CNB but, when diagnosed correctly, requires no additional excision. Finally, columnar cell alteration is a recently described observation where epithelial cells have a more columnar appearance, often with secretions called apical snouts, than is normally found in ducts and lobules. Although insufficient data has been reported, in the absence of atypia, there appears to be no clear indication for further excision.
Future developments may include continued introduction of new technologies and better assessment of the indications for excision when certain lesions are found at CNB, whether applied in a stereotactic, ultrasound, or MR environment. Already, some of these devices are being investigated not only for diagnostic but also for therapeutic applications. For example, the same device that employs the freezing of a probe that stabilizes a lesion while percutaneous biopsy is performed has been modified to permit cryoablation of the entire lesion. 14 Removal of an entire lesion by percutaneous means, ablation by focused ultrasound, or radiofrequency methods are similar extensions of the techniques developed for breast biopsy. 15 Whether lesions need to be removed or destroyed may depend on future success with accuracy of percutaneous diagnosis. With so many lesions coming to clinical attention through screening efforts and long-term trials indicating similar mortality outcomes for limited treatment, 16 the impetus for better indications for intervention as well as less-invasive means is likely to continue.
R. James Brenner, MD, JD, FACR, FCLM, is chief of breast imaging and professor of radiology, University of California, San Francisco-Mt Zion Hospital Cancer Center.
- Dodd GD, Fry K, Delany W. Pre-op localization of occult carcinoma of the breast. In: Nealon TF, ed. Management of the Patient with Breast Cancer. Philadelphia: WB Saunders; 1965:88-113.
- Jackman RJ. Marzoni FA. Needle-localized breast biopsy: why do we fail. Radiology. 1997;204:677-684.
- Brenner RJ, Bassett LW, Fajardo L, et al. Stereotactic core breast biopsy: a multiinstitutional prospective trial. Radiology. 2001;218:866-872.
- Parker SH, Burbank F, Jackman RJ, et al. Percutaneous large-core breast biopsy: a multiinstitutional study. Radiology. 1994;193:359-364.
- Liberman L, Feng TL, Dershaw DD, Morris EA, Abramson AF. US-guided core breast biopsy: use and cost-effectiveness. Radiology. 1998;208:717-723.
- Philpotts LE, Shaheen NA, Carter D, Lange RC, Lee CH. Comparison of rebiopsy rates after stereotactic core needle biopsy of the breast with 11-gauge vacuum suction probe versus 14-gauge needle and automatic gun. AJR Am J Roentgenol. 1999;172:683-687.
- Reynolds HE. Core needle biopsy of challenging benign breast conditions: a comprehensive literature review. AJR Am J Roentgenol. 2000;174:1245-1250.
- Foster MC, Helvie MA, Gregory NE, Rebner M, Nees AV, Paramagul C. Lobular carcinoma in situ or atypical lobular hyperplasia at core-needle biopsy: is excisional biopsy necessary? Radiology. 2004;231:813-819.
- Page DL, Schuyler PA, Dupont WD, Jensen RA, Plummer WD Jr, Simpson JF. Atypical lobular hyperplasia as a unilateral predictor of breast cancer risk: a retrospective cohort study. Lancet. 2003; 361:125-29.
- Cohen MA. Cancer upgrades at excision biopsy after the diagnosis of atypical lobular hyperplasia or lobular carcinoma in situ at core-needle biopsy: some reasons why [editorial]. Radiology. 2004; 231:617-621.
- Rosenfeld I, Tartter PI, Gajdos C, Hermann G, Bleiweiss I. The significance of malignancies incidental to microcalcifications in breast spot localization biopsy specimens. Am J Surg. 2001;182:1-5.
- Brenner RJ, Jackman RJ, Parker SH, et al. Percutaneous core needle biopsy of radial scars of the breast: when is excision necessary. AJR Am J Roentgenol. 2002;179:1179-1184.
- Rosen EL, Bentley RC, Baker JA, Soo MS. Imaging-guided core needle biopsy of papillary lesions of the breast. AJR Am J Roentgenol. 2002;179:1185-1192
- Littrup PJ, Freeman-Gibb L, Andea A, et al. Cryotherapy for breast fibroadenomas. Radiology. 2005;234:63-72.
- Fornage BD, Sneige N, Ross MI, et al. Small (< or = 2-cm) breast cancer treated with US-guided radiofrequency ablation: feasibility study. Radiology. 2004;231:215-224
- Veronesi U, Cascinelli N, Mariani L, et al. Twenty-year follow-up of a randomized study comparing breast-conserving surgery with radical mastectomy for early breast cancer. N Engl J Med. 2002;347:1227-32.