mgh:cyto-1-14

• Breast_Cytology_Lecture.pdf

Indications for cytology examination

Breast lesions can be detected on physical examination by the patient or a physician, or by screening mammography. By palpation, benign masses tend to be soft, rubbery, and mobile with well defined margins. Masses that are hard or fixed with ill-defined or irregular borders are worrisome for malignancy. Mammography is the standard diagnostic imaging modality, but in younger women with dense breast tissue or postmenopausal women on hormone replacement therapy, mammography has limited utility and ultrasound needle used instead. As women age, their breasts are progressively replaced with fat and become less dense, enhancing the diagnostic accuracy of mammography. During pregnancy, ultrasound is used to evaluate breast masses so that radiation exposure can be avoided and also because mammography is less useful in the hormonally altered breast tissue where the amount of adipose tissue is decreased. For men, a combination of mammography and ultrasound is commonly used.

Fine needle aspiration (FNA) is a minimally invasive, cost effective means of evaluating breast masses detected by palpation or by radiographic imaging. The procedure is very well tolerated and can be performed in the clinic setting without anesthesia. Multiple studies have demonstrated the utility of FNA biopsy in evaluating breast lesions. However, the sensitivity (78% to 99%) and specificity (79% to 99%) of breast FNA vary widely among institutions. The rates in some studies are comparable to those for core needle biopsy, whereas other rates are less than optimal and support the use of core needle biopsy instead of FNA. The best results are obtained when the FNA biopsy is performed by skilled aspirators with on site pathologists who are available for adequacy evaluation and experienced pathologists to interpret the specimens. It is important that physicians who perform FNA biopsies receive supervised training in the technique. Formally trained physicians who have performed more than 150 supervised FNA biopsies have significantly lower false-negative and non-diagnostic rates been due informally trained physicians who have performed fewer than 10 supervised FNA biopsies and learned the technique through reading a description, attending a lecture, or by observing the procedure.

Diagnostic accuracy is greatly improved when the FNA results are correlated with clinical and radiologic findings (triple test). The use of the triple test decreases the false negative rate from 10% to less than 1% and the false positive rate from 1% to less than 0.2%. The most common cause of a false negative result is sampling error, which can be due to causes such as small lesion size, fibrosis, and/or inexperience of the aspirator. In contrast, a false positive result is commonly due to interpretive error and can involve hyperplastic processes such as fibroadenoma, proliferative fibrocystic change, papillomas, and lactational change. If there are discordant results between the clinical, radiographic, and/or FNA findings, excisional biopsy is recommended for further evaluation, especially in cases where definitive therapy is planned for a malignant cytologic diagnosis.

FNA causes few complications, most commonly temporary pain or bruising at the biopsy site. Hematomas can occur but are preventable by applying adequate pressure after obtaining the sample. Infection is uncommon. Rare instances of needle track seeding by tumor cells have been reported when samples are taken was larger bore needles (14- to 18-gauge) than are typically used for FNA biopsy (23- to 25-gauge). Pneumothorax is another rare complication and tends to occur more frequently in women with slight builds, smaller breasts, and deeper lesions. It can be avoided by keeping the path of the needle angled to the rib cage and not inserting the needle perpendicular to the chest wall. In some instances, FNA biopsy has been reported to cause infarction of the lesion, most commonly in papillomas and fibroadenomas, which can lead to diagnostic difficulty when evaluating subsequent excision specimens.

Procuring the specimen

Once the mass is fixed with the non-aspirating hand, the skin is cleaned with an alcohol swab at the planned needle entry site. The loosened needle cap is removed and the aspirating hand stabilized by resting the syringe barrel against the thumb or forefinger of the non-aspirating hand. This guards against any physiologic hand tremor and ensures precise needle placement, but is not be needed after insertion of the needle. The needle is inserted into the lesion and the syringe plunger pulled back to generate several cubic centimeters of vacuum. The vacuum is maintained until the needle is removed from the patient. With a straight wrist, the needle is moved back and forth quickly and repeatedly in a sawing motion (“excursions”) for a dwell time no longer than 15-20 seconds (approximately 40-60 excursions) along the original needle trajectory, alternately advancing into the mass and withdrawing to a superficial location without exiting the patient. Slower needle action will yield less material. A shorter dwell time (2 to 5 seconds) is recommended for highly vascular lesions. Each time the needle advances into the lesion, its cutting tip dislodges small tissue fragments; this cutting action is essential for a successful FNA. Negative pressure alone without needle movement will not procure enough tissue for diagnosis in solid lesions.1 The vacuum in the syringe helps conduct the tissue fragments into the needle shaft and hub. A slight acceleration of the needle as it advances into the mass enhances the cutting action of the needle tip. Keeping the needle tip within the mass avoids diluting the specimen with adjacent non-lesional tissue. Material can be seen accumulating in the needle hub, although absence of visible material does not signify an inadequate sample. If blood is rapidly entering the hub, withdraw the needle immediately.

Sampling with thinner needles (25g or 27g) is preferred for vascular lesions as well as for fibrous lesions like a fibroadenoma. When sampling a sclerotic lesion, the needle should be moved more vigorously. To sample more of the mass with one needle pass, withdraw the needle tip to a superficial location while maintaining vacuum, then redirect it to a different area by changing the angle of entry. “Fanning” allows for sampling of a larger area: after each excursion, when the needle tip is in a superficial location, change its angle of entry slightly until the entire region of interest is sampled. Avoid changing direction when the needle is deep in the lesion. This results in tissue tearing and hemorrhage, which compromises the diagnostic yield of subsequent passes.

Remove the needle from the patient after the last excursion is completed or when material or blood is visible in the needle hub, which can occur in less than 15 seconds. Withdraw the needle from the patient in a controlled manner and only after you have released the vacuum in the syringe. Failure to release suction before withdrawing the needle from the patient pulls the aspirated material into the barrel of the syringe, making it difficult to expel for smear preparation. Pressure to the site is applied immediately with gauze to minimize bleeding. The patient or an assistant should perform this step, because the aspirator needs to prepare the sample immediately.

An FNA procedure typically involves inserting the needle into the mass 2 or 3 times (“passes”) to obtain several samples. The center of the mass is often sampled on the first needle pass with the needle approximately perpendicular to the skin. Other areas of the mass are sampled on subsequent passes, especially if the initial material is necrotic, cystic, or otherwise non-diagnostic. Sampling the mass along its long axis tends to yield more cellular specimens compared with moving the needle along the short axis.

As the FNA needle penetrates the target mass, the aspirator can learn to assess the texture of the lesion. Various adjectives used to describe the texture include gritty, rubbery or leathery, and soft. Learning to "feel with the needle" in this way can yield very useful information about the lesion being sampled. Masses are often heterogeneous, and different regions not only feel different but may also yield different cytologic material.

Test platforms/specimen processing and triage

All breast cytology specimens (FNA, duct lavage, or nipple discharge) may be prepared as direct smears (air dried or alcohol fixed) or as liquid based thin layer slides (ThinPrep or SurePath). In general it is desirable to prepare both alcohol fixed and air dried smears; fixed smears are useful in the evaluation of nuclear detail, whereas air dried smears allow for easier identification of cytoplasmic quality and background features such as mucin and stroma. Liquid based techniques are best for specimens such as cyst fluid or duct lavage, which consist of a considerable volume of fluid. The aspirated material can be directly placed in to a vial of liquid based fixative.

If liquid based preparations are used, it should be noted that the processing required to make a thin layer slide can alter the specimen appearance from what is typically seen on a direct smear. It is important to be aware of the differences because the cytologic features used to support a diagnosis of malignancy differ in direct aspirate smears and liquid based preparations.

Cell blocks are useful for cases requiring immunohistochemistry for hormone receptor status or in situ hybridization for HER-2/neu. Because most immunohistochemistry laboratories have validated their stains for formalin fixed paraffin embedded tissue, it is best to place the FNA material directly into formalin. Alternatively, it can be transported in saline or Roswell Park Memorial Institute (RPMI) medium prior to fixing in formalin. It is less than optimal to prepare a cell block from material placed in liquid based fixative such as CytoLyt or CytoRich Red because typically the stains are not validated for these alternative alcohol based fixatives. The easiest method for making cell blocks is to allow bloody specimens to clot in the hub of the needle before flicking the material onto a glass slide and then rinsing the material into formalin. For nonbloody or non-clotted specimens, the material can be rinsed directly into formalin, but multiple visible particles are needed to prepare an adequate cell block. In these cases, addition of HistoGel or use of a collodion bag may be necessary to prevent loss of material during processing. To ensure that adequate material is obtained it is best to evaluate immediately a smear prepared from a drop of material from each aspirate sample.

Reporting and terminology

Standard reporting of breast FNA results uses 5 managerially relevant categories: No malignant cells identified, atypical, suspicious for malignancy, positive for malignant cells, and non-diagnostic/unsatisfactory. These were developed in 1996 as part of the National Institute of Health practice guidelines, referred to as the Uniform Approach to Breast Fine Needle Aspiration Biopsy. The benign category encompasses benign breast tissue, simple cysts, fibrocystic change, fibroadenoma, fat necrosis, lactational change, and inflammatory processes. The atypical/indeterminate category applies to cases where the lesion is more likely to be benign but a well differentiated carcinoma cannot be excluded. The suspicious for malignancy category is used when the lesion is more likely malignant but confirmatory biopsy is recommended before definitive treatment. The malignant category should be reserved for cases that are definitely malignant, the majority of which consist of primary breast adenocarcinomas (ductal and lobular). The atypical/indeterminate and suspicious for malignancy categories apply to cases where the FNA results are inconclusive. The distinction between these 2 categories can be somewhat subjective. A general statement regarding the likelihood of malignancy should be indicated in the report, but both atypical and suspicious diagnoses warrant a surgical biopsy. The unsatisfactory category is for specimens that are determined to be inadequate due to insufficient sampling or preparation artifact.

In general, determination of adequacy relies on assessment by the aspirator as to whether the sample is representative of the palpated or imaged lesion and by the pathologist as to whether the slides are well prepared and interpretable without significant artifacts of slide preparation such as air drying, obscuring blood, or overly thick smears. Therefore it is helpful if the aspirator and the interpreter are the same person and that person is able to review the sample for adequacy at the time the FNA procedure is completed. In this way, additional sampling may occur promptly, and the likelihood of obtaining an adequate specimen increases.

Basic cytomorphology

Benign Components – C93-17243

• Normal ductal cells form a single cell layer attached to a basement membrane. The 2-dimensionally arranged ductal cells are cohesive with each other
• Myoepithelial cells are another cell type that normally lies between the basement membrane and the ductal cells. They typically have slightly darker nuclei, and paler fragile cytoplasm that may rupture when the ductal cells strip from the basement membrane
• Apocrine cells have a prominent nucleoli, round nucleus with an abundant granular cytoplasm; exhibit cohesiveness while maintain their cell boarders.
• Adipocytes are large clear cells with well-defined cytoplasmic boarders, comparable to “soap suds”
• Foam cells or macrophages are inflammatory cells with hypervacuolated cytoplasm and a reniform shaped nuclei

Fibroadenoma – C01-43583

• Ductal cells are arranged in a "staghorn" or animal shapes or blunt branched duct configuration.
• Scattered stripped myoepithelial cells in the background
• Identification of stromal tissue fragments is important for a definitive diagnosis of fibroadenoma
• The fibromyxoid stromal fragment often has a smooth surface and bland morphology of the mesenchymal nuclei

Fat Necrosis – C95-1257

• Few to no epithelial cells- possible fragments of fibroadipose tissue
• No single epithelial cells
• Inflammatory cells, single histiocytes and giant multinucleated histocytes
• Relatively clean background

Papilloma –C94-17220

• Cellular aspirates, benign ductal cells in papillary groups that can mark atypia and necrosis in the presence of infarction
• Myoepithelial cells are present indicating the benign nature
• Apocrine metaplasia may be seen
• If enough material is available, immunostains for myoepithelial markers (e.g. p63) can help to recognize the myoepithelial cell population that underlies the ductal cells; excision may be necessary to definitively determine the nature of a papillary lesion on cytology

Gynecomastia – N11-11968

• Low to moderately cellular specimen which can contain group of ductal cells
• Ductal cells have small rounded nuclei with little pleomorphism
• Scattered myoepithelial cells
• Not much different from a female breast sample

Papillary Carcinoma – N13-7402

• Contain a monotonous neoplastic cell population of papillary groups and single cells; cells with randomized polarity forming smooth, rounded spaces
• Necrotic debris
• Areas of possible cribriforming and fibrovascular cores

Phyllodes Tumor – C98-11449

• Represents a biphasic breast tumor of uncertain malignant potential
• Both stromal components and ductal epithelial cells may be present
• Low grade phyllodes tumors often resemble fibroadenomas on cytology
• High grade malignant phyllodes tumors resemble sarcomas and may show bizarre spindle cells singly and in clusters on cytology
• In those cases, nuclear atypia is prominent, nuclei are plump and a small amount of wispy cytoplasm is present

Ductal Carcinoma – Low grade: C02-6087 High grade: C01-33524

• Discohesive monotonous ductal cell population with randomized polarity, 3D groups and single cells
• Nuclei containing prominent nucleoli, nuclear boarders are irregular atypical chromatin
• Apoptotic bodies, necrotic debris
• Absence of myoepithelial cells

Immunostaining for HER2/c-erbB-2 is done on cell blocks and can show overexpression of the HER2 protein. Fluorescence in situ hybridization (HER2-FISH) may show that the HER2 gene is amplified. If the test results indicate HER2 overexpression and/or HER2 amplification, then a patient may be eligible for treatment with the cancer drug Herceptin® (Trastuzumab)

Lobular Carcinoma – C02-22969

• Aspirates may be sparse but can also be cellular
• Tumor cells tend to be small with eccentric nuclei
• Nuclei are mostly uniform in size but may be irregular, with small nucleoli
• Lobular-type tumor cells may show targetoid intracytoplasmic mucin vacuoles
• Thin cords of lobular carcinoma cells can infiltrate the adipose tissue, "single file" pattern

A cytokeratin immunostain can be helpful to disclose the sparse and inconspicuous lobular carcinoma cells in aspirates, or in cavity fluids

Medullary Carcinoma – C92-2097

• Syncytial groups of anaplastic tumor cells with delicate cytoplasm
• Numerous malignant nuclei, often bare and devoid of cytoplasm
• Lymphocytes and necrosis may be present in the background
• The definitive diagnosis of medullary carcinoma of the breast depends on architectural features, which can only be made on surgical pathology specimens

Triple Test

“Triple Test” for Breast Lesions

• Has a high negative predictive value if all three modalities of a breast lesion are negative: clinical impression, imaging findings and tissue diagnosis, often by cytology
• Cytology findings have to be correlated with both clinical and imaging findings. If a breast sample appears benign on cytology but is clinically or radiologically suspicious – this is a discorrelation
• Do not mix up with “triple negative” breast cancer, which refers to breast cancers that are negative for estrogen and progesterone receptors, and HER2 proteins

Reference- McKee, Grace T. Cytopathology of the Breast, MGH 2002 Fisher, Andrew H. Breast FNA, Cytologystuff through Hologic, 2013