Development of a novel approach for breast cancer prediction and early detection using minimally invasive procedures and molecular analysis: how cytomorphology became a breast cancer risk predictor

With enhanced public awareness, advances in breast imaging, and emphasis on early breast cancer detection and prevention, more women are seeking consultation to assess the status of their breast health. Risk assessment has become an integral part of established multi-disciplinary breast care, and breast cancer risk reduction interventions have received a great deal of attention. Similarly, interest in identification of high-risk individuals has increased significantly. Atypical proliferative changes in breast epithelial cells are ranked high among various known breast cancer risk factors and, in recent years, have been the subject of several investigations. Breast tissue and fluid in the ductal system provide a rich source of cells and biomarkers that have the potential to aid in the assessment of short-term risk of breast cancer development, and assess responses to interventional prevention efforts. There are three minimally invasive procedures currently being utilized to sample breast tissue in asymptomatic high-risk individuals. These procedures are: fine-needle aspiration biopsy, nipple aspiration fluid, and ductal lavage. In this review article, the merits and limitations of each procedure are presented, and the contribution of cytomorphology and molecular analysis in breast cancer prediction is highlighted. In addition, the role of Masood Cytology Index as a surrogate endpoint biomarker in chemopreventative trials is discussed.

Cytomorphology as a risk predictor: experience with fine needle aspiration biopsy, nipple fluid aspiration, and ductal lavage

Primary prevention of breast cancer requires identification and elimination of cancer-causing agents, which is an incredibly difficult task to follow. Secondary prevention involves screening individuals who are at increased risk for breast cancer in hopes that early intervention will affect survival. In the 1980s, chemoprevention received serious attention. This approach was aimed at reducing cancer risk by administration of natural or synthetic clinical compounds that prevent, reverse, or suppress carcinogenesis in individuals at increased risk for cancer. It was not until 1998, however, when the first report from the National Surgical Adjunct Breast and Bowel Project (Breast Cancer Prevention Trial BCPT; P-1) randomized clinical trial appeared in the literature supporting the hypothesis that breast can-cer can be prevented. This study showed that administration of tamoxifen reduced the risk for invasive and noninvasive breast cancer by almost 50% in all age groups. With the current availability of tamoxifen as a chemopreventive agent and with the increasing emphasis on early breast cancer detection and prevention, more women seek consultation to determine their risk for breast cancer. However, in the absence of any detectable breast lesion, clinically and mammographically, only a few women may volunteer to have their breasts sampled by surgical biopsy for risk assessment. Other non-surgical procedures include fine needle aspiration biopsy (FNAB), nipple aspirate fluid (NAF), and the recently introduced procedure, ductal lavage. These techniques may provide better alternatives. These minimally invasive procedures are capable of recruiting cellular material for cytomorphologic interpretation and biomarker studies.

Cytomorphology of Common Malignant Tumors of the Breast

A definitive cytologic diagnosis of breast cancer is usually possible when using the six major criteria of malignancy (cellularity, dyshesion, monomorphism, anisonucleosis, irregular nuclear membranes, prominent nucleoli) as part of the triple test. Carcinomas of special type have unique clinical and cytologic features that pathologists need to consider, because these may confuse interpretation. Complete subtyping of carcinomas may not always be possible by fine needle aspiration. Diagnostic accuracy for breast carcinoma is excellent. False-negative diagnoses are infrequent and chiefly due to sampling issues. False-positive diagnoses are extremely rare. Uniform report terminology should be used to ensure that diagnostic information is conveyed appropriately and consistently to guide the next diagnostic or treatment step.

'Atypical' and 'suspicious' diagnoses in breast aspiration cytology

BACKGROUND

In 1996, the National Cancer Institute (NCI) recommended five categories for the diagnosis of breast aspiration cytology: benign, atypical, suspicious, malignant, and unsatisfactory. The authors evaluated the usefulness of separating inconclusive diagnoses into the aforementioned atypical and suspicious categories. The current study examined the correlation between cytologic and histologic findings made in breast aspiration cytology specimens that were categorized in accordance with these NCI guidelines.

METHODS

From 1992 to 2000, 7727 breast aspirations were performed at the University of Miami/Jackson Memorial Medical Center (Miami, FL). Aspirates were classified using criteria similar to the NCI recommendations. Four hundred eighty-nine aspirates (6%) were classified as being cytologically 'atypical' or 'suspicious'. Of those, 225 'atypical' aspirates and 162 'suspicious' aspirates had available histologic follow-up data and were included in the study.

RESULTS

Among the 'atypical' aspirates, 118 (52%) yielded malignant findings on histologic evaluation. Infiltrating ductal carcinoma (n = 87; 74%) was the most common malignant diagnosis, followed by infiltrating lobular carcinoma (n = 12; 10%). Among 'suspicious' aspirates, 135 (83%) yielded malignant findings on histologic analysis. The most common benign diagnosis in both 'atypical' and 'suspicious' aspirates was proliferative fibrocystic changes with or without atypia, followed by fibroadenoma.

CONCLUSIONS

Most aspirates that yielded suspicious findings on cytologic examination proved to be malignant, as did > 50% of aspirates that yielded atypical findings. Infiltrating lobular carcinoma commonly was underdiagnosed as being atypical on cytologic examination. Benign lesions with atypical or suspicious cytologic diagnoses included certain cases of atypical ductal hyperplasia. The authors concluded that the distinction between the atypical and suspicious categories, as recommended by the NCI, is not warranted. Therefore, they suggest the use of a single term, such as 'equivocal', to describe inconclusive diagnoses on breast fine-needle aspiration cytology.

The significance of the diagnosis of atypia in breast fine-needle aspiration

The diagnosis of atypia in breast fine-needle aspiration (FNA) continues to be an area of debate in cytology practice. The aim of this study was to assess the clinical significance of this term and to evaluate potential morphological criteria, which would determine the patient's outcome. A computer-based search was carried out to retrieve breast FNAs performed between 1990 and 2000 that were diagnosed as atypical. Cases followed by surgical resection were reexamined for the presence of morphological features potentially differentiating benign and malignant lesions. Out of 1,568 breast FNAs, there were 64 cases (4%) with a diagnosis of atypia. Thirty-eight cases had surgical follow-up material that revealed malignancy in 14 cases (37%) and benign lesions in 24 cases (63%). The benign diagnostic categories included fibrocystic change (12/24), fibroadenoma (3/24), tubular adenoma (2/24), and nonspecific findings (7/24). The malignant diagnoses included ductal carcinoma (9/14), lobular carcinoma (3/14), ductal carcinoma in situ (DCIS; 1/14), and tubular carcinoma (1/14). The evaluation of cytological criteria used to differentiate benign from malignant lesions (i.e., cellularity, loss of cohesion, myoepithelial cells, nuclear enlargement, nuclear overlap, prominent nucleoli) revealed significant overlap between benign and malignant cases, particularly in cases of fibroadenoma, tubular adenoma, and proliferative breast disease. The surgical follow-up of four hypocellular cases revealed lobular carcinoma in two cases and ductal carcinoma in the remaining two cases. Our study confirmed that the diagnosis of atypia is clinically significant because it is associated with a high probability of malignancy. No morphological criterion is able to reliably differentiate benign and malignant lesions in cases diagnosed with atypia. Diagnosis of atypia is particularly significant in hypocellular cases. We recommended that breast FNAs with a diagnosis of atypia be evaluated further histologically.

Critical clinicopathologic analysis of 23 cases of fine-needle breast sampling initially recorded as false-positive. The 44-year experience of the Institut Curie

BACKGROUND

Because false-positive cytologic diagnoses in breast tumors are rare, few cases have been reported, although their consequences may be highly detrimental to the patient. The authors report the Institut Curie's experience, by using a multidisciplinary approach.

METHODS

Of 9334 benign breast tumors examined preoperatively for cytologic diagnosis by fine-needle sampling (FNS), the 23 (0.25%) FNS cases considered to be false-positive were retrospectively reviewed and analyzed.

RESULTS

Tumors were situated close to the nipple in 7 cases and away from the nipple in 16 cases. Tumor stage was T0 for 1 case, T1 for 18 cases, and T2 for 4 cases. Radiologically, six tumors were classified as malignant, seven as indeterminate or suspicious, and nine as benign. Three of six tumors studied by flow cytometry were DNA aneuploid. Based on a multidisciplinary clinicopathologic review, 20 FNS cases were finally classified as false-positive, and the remaining 3 tumors with malignant FNS and subsequent benign histology were classified as true-positive, because local and/or metastatic progression was observed in the short term.

CONCLUSIONS

The authors' review suggests two categories of false-positive cases: the first in which cytologic benign patterns are overdiagnosed, and the second in which atypical morphologic criteria were present. Nevertheless, as shown by the malignant course in three cases, patients with malignant preoperative FNS and corresponding benign histology always require close clinical follow-up. Finally, surgical overtreatment rate could be decreased if all radiologically benign tumors with positive/suspicious FNS were subject to intraoperative frozen section examination.