Decisive role of intermediate filament typing of tumor cells in the differential diagnosis of difficult fine needle aspirates

Thirty-six diagnostically difficult fine needle aspirates from enlarged lymph nodes and malignant soft tissue tumors, containing tumor cells with scanty or no obvious light microscopic features indicative of their differentiation, were assessed by a panel of six cytopathologists. Their diagnoses were recorded and then compared with the definitive diagnosis established by combining the cytologic findings with the results of intermediate filament typing of tumor cells in the smears using monoclonal antibodies specific for each filament type. The results show that use of these antibodies can markedly improve the accuracy of the cytologic diagnosis of tumor type as well as revise or prevent erroneous cytologic diagnoses in difficult cases. This pertains especially to the differential diagnoses of carcinoma versus malignant lymphoma, carcinoma versus malignant melanoma, carcinoma versus sarcoma and squamous carcinoma versus carcinoma of simple epithelia. Intermediate filament typing of tumor cells in aspirates as an objective histogenetic criterium makes the differential diagnosis of the difficult aspirates much more reliable and reproducible, provided that appropriate questions are asked, monoclonal antibodies with well-defined specificities are used and the antigenicity of the intermediate filaments in smears is preserved.     

Diagnosis of metastatic renal cell and thyroid carcinomas by intermediate filament typing and cytology of tumor cells in fine needle aspirates

Coexpression of keratin and vimentin was found in carcinoma cells of 13 fine needle aspirates of metastatic lesions that showed some cytologic features considered to be consistent with a renal or thyroid origin, but in which a large number of other possible primary sites would have to be taken into account on the basis of the morphologic evidence alone. Immunochemistry thus narrowed the cytologic differential diagnosis to thyroid, renal, endometrial and ovarian carcinomas, which are known to show true coexpression of keratin and vimentin. In most cases, clinical data available at the time of the fine needle aspiration supported the thyroid or renal origin of the carcinoma cells found in the aspirates. In two cases, which lacked significant clinical information, the diagnosis of metastatic renal cell carcinoma was provided on the basis of the combined morphologic and immunocytochemical evidence. In these two cases, computed tomography or ultrasonography revealed kidney tumors, which were removed and confirmed histologically to be clear cell carcinomas.     

Neuroendocrine (Merkel-cell) carcinoma of the skin. Cytology, intermediate filament typing and ultrastructure of tumor cells in fine needle aspirates

The light and transmission electron microscopic findings and the intermediate filament typing of tumor cells from fine needle aspirates of primary, recurrent and metastatic neuroendocrine (Merkel-cell) carcinoma of the skin are described. The tumor cells in the smears coexpressed keratins and neurofilaments and were characterized by "intermediate filament buttons," i.e., buttonlike fragments of cytoplasm of tumor cells, which could be observed either by staining with intermediate filament-specific antibodies using immunofluorescence or by their appearance in hematoxylin-and-eosin-stained smears. These features may help in the differential diagnosis of fine needle aspirates of neuroendocrine carcinomas of the skin.     

Differential diagnosis of lymph node aspirates by intermediate filament typing of tumor cells

Twenty-one lymph node aspirates for which a differential diagnosis was difficult or not possible by light microscopy alone were selected for further study by intermediate filament typing. The use of four monoclonal antibodies specific for different keratin subsets or for vimentin resulted in a definitive diagnosis in all instances. The results show that use of these antibodies can improve the accuracy of cytologic diagnosis and provide further evidence that intermediate filament typing can help differentiate (1) malignant melanoma from adenocarcinoma, (2) malignant lymphoma from small-cell anaplastic carcinoma and (3) squamous-cell carcinoma from adenocarcinoma. Intermediate filament typing can also be used to identify very small numbers of carcinoma cells in specimens that are apparently negative for tumor cells by light microscopy. The method is quick, relatively simple, reliable and unambiguous, provided that appropriate questions are asked and antibodies with well-defined specificities are used.     

Malignant melanoma in fine needle aspirates and effusions. An immunocytochemical study using monoclonal antibodies

The value of monoclonal antibodies (MAbs) for the immunodetection of keratin, vimentin and two melanoma-associated antigens recognized by NKI/C3 and NKI/Bteb for the diagnosis of malignant melanoma has been previously established on histologic preparations. In the present study, cytologic preparations from 20 fine needle aspirates and effusions from patients with malignant melanoma were evaluated using these antibodies. Twenty of 20 smears were negative for keratin, and 20 of 20 smears were positive for vimentin. Positivity for NKI/C3 was seen in 12 of 12 cases studied, and for NKI/Bteb in 12 of 13 cases. These results indicate that a panel of MAbs consisting of anti-keratin, anti-vimentin, NKI/C3 and NKI/Bteb is useful for a more accurate diagnosis of malignant melanomas on cytologic preparations. The expression of these antigens in melanoma cells in cytologic smears can be a valuable aid in the detection of primary (noncutaneous) and metastatic melanomas by fine needle aspiration.     

Association of mast cells with Warthin's tumor in fine needle aspirates of the salivary gland

OBJECTIVE: To determine the significance of the presence of mast cells in Warthin's tumor by evaluating the occurrence of these cells in cellular and immunohistochemical preparations. STUDY DESIGN: Specimens derived from five cases of FNAC were examined. A total of four slides from five cases were prepared from each: two air-dried smears were stained with May-Grünwald-Giemsa (MGG) stain and two with Hansel's stain. The other two were alcohol fixed and stained using the Papanicolaou method. The smears were evaluated for the presence of mast cells, especially associated with oxyphilic cells. In order to investigate the location of mast cells, we also counted those cells by means of immunohistochemistry using anti-mast cell monoclonal antibody AA1. RESULTS: The Hanselstained cellular sample from Warthin's tumor contained numerous mast cells, associated mainly with large, oxyphilic cell sheets. The number of AA1-positive cells (mast cells) stained with immunohistochemistry was greater in epithelial component than in lymphoid stroma. CONCLUSION: Mast cells in a salivary gland aspirate might be indicative of Warthin's tumor; therefore, MGG-stained slides offer the advantage of ease of preparation, particularly when the typical cytologic features are not present.     

Chondroblastoma in fine needle aspirates.

The cytologic features of smears of fine needle aspirates from four chondroblastomas were studied. In an ideal fine needle aspiration biopsy sample of chondroblastoma, all the diagnostic details can be found: chondroid matrix surrounding individual round to oval mononuclear cells, calcifications among cells and multinucleate osteoclasts. Chondroid matrix, which we believe to be the decisive feature, and calcifications are seen better in Giemsa-than in Papanicolaou-stained smears.     

Telomerase activity in thyroid fine needle aspirates

OBJECTIVE: To study the utility of telomerase activity (TA) detection in thyroid fine needle aspirates (FNAs). STUDY DESIGN: One hundred two thyroid nodules were studied: 70 in FNA samples and 32 in frozen tumors. From among FNA samples, there were 57 nodules of the goiter, 1 adenoma, 7 papillary carcinomas, 1 medullary carcinoma and 1 lymphoma. Three cases of thyroiditis were excluded. The 32 frozen tissue tumors consisted of 5 follicular adenomas, 4 follicular carcinomas and 23 papillary carcinomas. TA was analyzed by Telomerase PCR ELISA (Roche Diagnostics, Indianapolis, Indiana, U.S.A.). RESULTS: TA was negative in the 57 nonneoplastic nodules and 6 follicular adenomas and positive in 3 of 4 follicular carcinomas and 10 of 30 papillary carcinomas. TA sensitivity was 41.4% and specificity 100%. Sensitivity for malignancy was higher (85.7%) in FNAs than in TA. CONCLUSION: TA seems highly specific for neoplasms of the thyroid. Further studies are needed to confirm whether TA detection could contribute to identifying neoplasms when FNAs are inconclusive for malignancy and in cases of scanty material.     

Microfilariae in fine needle aspirates from epididymal lesions

Microfilariae of Wuchereria bancrofti were observed in fine needle aspiration smears from three epididymal nodules, and degenerating microfilariae suggestive of Brugia malayi were found in the smears from a fourth case. The smears in all four cases showed a polymorphonuclear inflammatory cell component as well as epithelioid cell granulomata. While blood eosinophilia was present in all four cases, eosinophilia was present in the aspiration smears in only one case. Microfilariae could be demonstrated in the peripheral blood in only one case.     

Immunocytochemistry on fine needle aspirates in paraffin miniblocks

A simplified method of processing of fine needle aspirates for paraffin miniblocks suitable for both morphologic and immunocytochemical evaluation is described. Aspirates were fixed in ethanol at 4 degrees C, dehydrated in acetone and xylene and embedded in paraffin (58 degrees C). All steps were carried out in a single Eppendorf centrifuge tube; the total process took less than four hours. Deparaffinized sections were stained using the alkaline phosphatase-antialkaline phosphatase technique with monoclonal and conventional antibodies helpful in the differential cytologic diagnosis of alcohol-fixed aspiration biopsy specimens. Antibodies to keratin, vimentin, desmin, neurofilaments, glial fibrillary acidic protein, leukocyte-common antigen, synaptophysin and immunoglobulin kappa and lambda light chains reacted positively on the miniblock material. Since the paraffin miniblocks combine the histologic pattern of the tumor with the differentiation-specific information provided by immunocytochemistry, their use can improve the accuracy of tumor typing in aspirates.     

p53 mutations as tumor markers in fine needle aspirates of palpable breast masses

OBJECTIVE: Mutations in p53 exons 5-8 are found in 40-50% of breast carcinomas. We performed a retrospective analysis of p53 mutations in fluid-based, archival fine needle aspirates (FNAs) of breast masses to determine their potential diagnostic utility as breast tumor cell markers. STUDY DESIGN: Residual, fluid-based, archival FNAs of 27 breast masses were retrospectively evaluated by polymerase chain reaction (PCR), single-strand conformational polymorphism analysis (SSCP) and sequencing for p53 exons 5-8. Results were compared with the morphologic diagnoses and genotyping of available excisional biopsy tissue. RESULTS: Six of the twenty-seven cases were found to have a clonal mutation in p53; all six mutated cases showed carcinoma on subsequent excisional biopsy. Definitive cytologic diagnosis of cancer had been possible in only four of the six cases. Identical mutations were found in the excised carcinomas in the five cases with available tissue. None of the 14 aspirates with benign cytology had detectable mutations in p53. CONCLUSION: p53 Mutational analysis by PCR/SSCP/sequencing deserves to be critically studied as a diagnostic criterion in patients with indeterminate or suspicious cytology. Validation studies should be performed to test p53 mutations as molecular diagnostic markers in breast cytology specimens.     

Comparison of fine needle aspirates of breast cancers to imprint smears by means of digital cell image analysis

Morphonuclear assessments were performed using the SAMBA 2005 cell image processor on cell nuclei in fine needle aspirates and corresponding imprint smears from 17 not-otherwise-specified (NOS) breast carcinomas to study the influence of cell sampling on the morphonuclear measurements. Fourteen parameters related to densitometric (nuclear DNA content), morphometric (nuclear area) and textural (chromatin organization and distribution) characteristics were computed for each nucleus. The results demonstrated that such morphonuclear features evolved significantly and positively with respect to conventional histopathologic grading. The method of cell sampling significantly influenced the results, but without altering the general conclusions regarding evolution of the morphonuclear features.     

Ringlike structures in fine needle aspirates from the breast

Spherical ringlike structures of various sizes resembling Liesegang rings (LRs) are described in four fine needle aspirates (FNAs) from breast lesions over a period of 13 years. A characteristic finding in these structures was a distinct double layer outer wall with striations and an amorphous central nidus. Under polarized light they were non-refractile and no birefringence was noted in Congo red. Immunohistochemical stains for calcium, iron, mucus, glycogen, amyloid, cytokeratin and epithelial membrane antigen in all the cases were found to be negative. Since LRs can be mistaken for ova or parasites, their presence in aspirates of breast should be kept in mind to avoid misdiagnosis.     

Immunocytochemistry of fine needle aspirates. A tactical approach

To maximize the potential of immunocytodiagnosis for fine needle aspiration (FNA) samples, it is necessary to be aware of the pitfalls and limitations of these techniques and to formulate a strategy to deal with the many variables involved. Five cases are presented to illustrate some of these variables, which include determining the adequacy of the FNA specimen, selecting tactics for cytologic and immunocytochemical studies, selecting methods for processing the FNA sample, preparing smears to enrich and preserve cells of interest, selecting enzyme labeling methods to optimize sensitivity and specificity, selecting monoclonal antibodies to make the study efficient and pertinent and interpreting the study results. The adequacy of the FNA specimen could be determined by an immediate cytologic assessment of the aspirate as it was obtained. Alcohol-fixed smears and formalin-fixed tissue sections prepared from the aspirate were used for diagnosis; the immunocytochemical studies were used as a diagnostic adjunct for accurate cell identification. Immunocytochemical studies were done on air-dried cytocentrifuge smears of pre-washed cells. While both immunoperoxidase and immunoalkaline phosphatase methods were suitable, we recommend the immunoperoxidase method for the study of aspirates from nonhemopoietic tissues and the immunoalkaline phosphatase method for the study of aspirates with many blood cells present. The proper selection of monoclonal antibodies and the interpretation of the results are best made in the context of the cytologic characteristics of the FNA sample and the clinical features of the patient.     

Diagnostic schemes for fine needle aspirates of breast masses

A comparison was made of four statistically based schemes for classifying epithelial cells from 243 fine needle aspirates of breast masses as benign or malignant. Two schemes were computer-generated decision trees and two were user generated. Eleven cytologic characteristics described in the literature as being useful in distinguishing benign from malignant breast aspirates were assessed on a scale of 1 to 10, with 1 being closest to that described as benign and 10 to that described as malignant. The original computer-generated dichotomous decision tree gave 6 false negatives and 12 false positives on the data set; another tree generated from the current data improved performance slightly, with 5 false negatives and 10 false positives. Maximum diagnostic overlap occurred at the cut-point of the original dichotomous tree. The insertion of a third node evaluating additional parameters resulted in one false negative and seven false positives. This performance was matched by summing the scores of the eight characteristics that individually were most effective in separating benign from malignant. We conclude that, while statistically designed, computer-generated dichotomous decision trees identify a starting sequence for applying cytologic characteristics to distinguish between benign and malignant breast aspirates, modifications based on human expert knowledge may result in schemes that improve diagnostic performance.