Special report. Immunocytochemistry in diagnostic cytology: a 12-year perspective

Immunocytochemistry for predictive biomarker testing in lung cancer cytology

With an escalating number of predictive biomarkers emerging in non-small cell lung carcinoma (NSCLC), immunohistochemistry (IHC) is being used as a rapid and cost-effective tool for the screening and detection of many of these markers. In particular, robust IHC assays performed on formalin-fixed, paraffin-embedded (FFPE) tumor tissue are widely used as surrogate markers for ALK and ROS1 rearrangements and for detecting programmed death ligand 1 (PD-L1) expression in patients with advanced NSCLC; in addition, they have become essential for treatment decisions. Cytology samples represent the only source of tumor in a significant proportion of patients with inoperable NSCLC, and there is increasing demand for predictive biomarker testing on them. However, the wide variation in the types of cytology samples and their preparatory methods, the use of alcohol-based fixatives that interfere with immunochemistry results, the difficulty in procurement of cytology-specific controls, and the uncertainty regarding test validity have resulted in underutilization of cytology material for predictive immunocytochemistry (ICC), and most cytopathologists limit such testing to FFPE cell blocks (CBs). The purpose of this review is to: 1) analyze various preanalytical, analytical, and postanalytical factors influencing ICC results; 2) discuss measures for validation of ICC protocols; and 3) summarize published data on predictive ICC for ALK, ROS1, EGFR gene alterations and PD-L1 expression on lung cancer cytology. Based on our experience and from a review of the literature, we conclude that cytology specimens are in principal suitable for predictive ICC, but proper optimization and rigorous quality control for high-quality staining are essential, particularly for non-CB preparations.

Endoscopic ultrasound-guided fine-needle aspiration with liquid-based cytologic preparation in the diagnosis of primary pancreatic lymphoma

OBJECTIVES

The diagnosis subtyping of lymphoma on specimens collected by endoscopic ultrasound fine-needle aspiration (EUS-FNA) can be extremely difficult. When a cytopathologist is available for the on-site evaluation, the diagnosis may be achieved by applying flow cytometric techniques. We describe our experience with immunocytochemistry (ICC) and molecular biology studies applied on EUS-FNA specimens processed with a liquid-based cytologic (LBC) preparation for the diagnosis of primary pancreatic lymphoma (PPL).

METHODS

Three patients with a pancreatic mass underwent EUS-FNA. The collected specimens were processed with the ThinPrep method for the cytologic diagnosis and eventual additional investigations.

RESULTS

A morphologic picture consistent with PPL was found on the LBC specimens of the 3 patients. Subsequent ICC and molecular biology studies for immunoglobulin heavy chain gene rearrangement established the diagnosis of pancreatic large B-cell non-Hodgkin lymphoma in 2 patients and a non-Hodgkin lymphoma with plasmoblastic/immunoblastic differentiation in the remaining one.

CONCLUSIONS

An LBC preparation can be used to diagnose and subtype PPL by applying ICC and molecular biology techniques to specimens collected with EUS-FNA. This method can be an additional processing method for EUS-FNA specimens in centers where on-site cytopathologist expertise is not available.

Relevance of immunocytochemistry on thin-layer cytology in thyroid lesions suspicious for medullary carcinoma: a case-control study

BACKGROUND

Fine needle aspiration cytology represents the most important tool in the diagnosis of thyroid nodules, mostly in discriminating malignant from benign lesions. The diagnosis of medullary thyroid carcinoma (MTC) may present some problems related to its deceptive morphologic picture. This diagnosis may be supported by immunocytochemistry (ICC), which may be difficult to carry out on the conventional smears.

DESIGN

The diagnostic efficacy of ICC for the diagnosis of MTC with respect to other thyroid neoplasms on slides processed by thin-layer cytology (TLC) is evaluated.

PATIENTS

In the period between January 2002 and December 2005, 8,200 FNAB were processed. ICC on TLC slides was required in 33 cases. Conventional smears were fixed in ethanol, whereas TLC slides were processed with the Thin Prep 2000 method. All slides were then stained with Papanicolaou. In all cases where MTC was morphologically suspected, ICC for calcitonin, monoclonal carcinoembryonic antigen, and thyroglobulin was carried out only on TLC slides.

RESULTS

Thirty-three thyroid cytologic cases had ICC on the TLC slides, including 22 follicular proliferations and 11 malignant lesions. The application of ICC on TLC was conclusive in 32 cases and inconclusive in 1 case. Twenty cases underwent surgery. No false-positive and false-negative cases were found. Sensitivity and specificity were 100%, and the overall diagnostic accuracy was 100%.

CONCLUSIONS

ICC can be successfully applied on TLC slides. The combined results of morphology and a small immunopanel including thyroglobulin, calcitonin, and carcinoembryonic antigen yields a 100% diagnostic efficacy for MTC.

CONDENSED ABSTRACT

Fine needle aspiration cytology is an excellent technique for diagnosing malignant neoplasms of the thyroid, especially those derived from the follicular cells. A correct preoperative diagnosis of C-cell-derived tumors (MTC), which is essential for both the surgical approach to the primary tumor and the management of the patient, should rely not only on the morphologic picture but also on the immunocytochemical yielding using an immunopanel, which is particularly satisfactory on the TLC slides.

Lymphatic neoangiogenesis in human renal allografts: results from sequential protocol biopsies

Neoangiogenesis of lymphatic vessels may be important for the cellular immune response in renal transplants. To determine the prevalence and chronology of lymph vessel proliferation and its relation to cellular infiltrates and allograft function, we analyzed sequential protocol biopsies (n = 162), taken at 6, 12 and 26 weeks after transplantation. Biopsies were stained with an antibody against podoplanin and lymphatic vessel density was quantified per square millimeter. The prevalence of lymph vessel-positive biopsies and the lymph vessel density were similar at 6, 12 and 26 weeks after transplantation. Biopsies with acute cellular rejection showed no significantly different lymph vessel density compared to those below the threshold for acute rejection or chronic allograft nephropathy. While lymphatic neoangiogenesis was equally prevalent in biopsies with and without infiltrates, the lymph vessel density was significantly higher in areas with cellular infiltrates than in areas without. Graft function at 1 year after transplantation was better in cases with lymph vessels in their infiltrates compared to cases with lymph vessel-free infiltrates. In conclusion, lymphangiogenesis not only shows a clear association with cellular infiltrates but might also have an impact on the pathogenicity of these cellular infiltrates.

Use of p63 for distinction of glandular versus squamous lesions in cervicovaginal specimens

BACKGROUND

Differentiating primary glandular from high-grade squamous intraepithelial lesions (HSIL) that involve endocervical glands is not an uncommon diagnostic problem in liquid-based gynecological cytology. Squamous and atypical glandular cell lesions may show similar cytomorphologic features. The aim of this study was to evaluate the use of p63 as a marker of basal and/or squamous cell derivation in this differential diagnosis.

METHODS

Of 59,257 liquid-based cervicovaginal specimens collected over a 3-year period, 149 were diagnosed as atypical glandular cells of uncertain significance (AGUS) or adenocarcinoma and had histological follow-up. Ten cases (8AGUS and 2 adenocarcinomas) were proven to be high-grade dysplasia on cervical biopsies and the remaining cases represented glandular pathology. Slides from discrepant cases were stained with p63 antibody. In addition, the authors stained 25 control cases (10 adenocarcinomas, 10 HSIL, and 5 negative cervicovaginal specimens).

RESULTS

In all 10 discrepant cases, the abnormal groups originally interpreted as glandular in origin showed a homogeneous strong nuclear staining for p63 that indicated their squamous origin. Nuclei of isolated HSIL cells and basal cells from atrophic smears were also positive for p63. Benign and malignant glandular cells were uniformly negative. Isolated metaplastic, intermediate, and superficial squamous cells were likewise negative for this antibody.

CONCLUSIONS

p63 is a useful immunocytochemical marker for differentiating primary glandular pathology from HSIL in cervicovaginal specimens. It also detects isolated HSIL cells ("litigation cells"). This antibody is not expressed in AGUS, adenocarcinoma, or normal glandular cells. p63 stains basal cells and may be a diagnostic pitfall in atrophic cervicovaginal specimens.

Immunohistochemical Staining as a Potential Method for the Identification of Vaginal Epithelial Cells in Forensic Casework

There is currently no accurate method to identify vaginal epithelial cells uniquely. This study aimed to use a cell extraction procedure compatible with routine forensic sampling methods, and to investigate the expression of cytokeratin (CK), estrogen receptor-alpha (ERalpha), and phosphodiesterase 5 (PDE5) in order to distinguish between skin, buccal, vaginal, and external penile epithelial cells. Seminal fluid samples were also examined. Epithelial cell samples were fixed in formalin, embedded in agarose, and processed using histological methods. Antigen-antibody reactions were detected using the DAKO Envision+ detection system. CK was present in all cells from all five sources confirming the origin of cells as epithelial. Both ERalpha and PDE5 positively labeled vaginal, buccal, and skin epithelial cells. Although an antigen unique to vaginal epithelial cells was not identified, we have described a cell extraction procedure for use in the immunohistochemical detection of a wide range of antigens, an approach compatible with forensic diagnostics.

Immunophenotyping lymphomas in dogs: a comparison of results from fine needle aspirate and needle biopsy samples

The reliability of utilizing cytologic samples for immunophenotyping canine lymphomas was evaluated by a systematic comparison of results from fine needle aspirate (FNA) to needle biopsy specimens stained by immunochemical methods. The specific reactivity of a selected panel of 12 antibodies to cell surface markers and intermediate filaments was assessed in 11 dogs by comparing cytologic to histologic samples. There was excellent correlation of results between immunostained cytologic and histologic samples. FNA sampling is a simple, noninvasive method for determining the lymphoid phenotype in canine lymphoma. In addition, this technique may be useful as a cytodiagnostic aid in differentiating lymphoid from non-lymphoid tumors and in assessing neoplastic vs. reactive or hyperplastic processes.

Thyroid transcription factor 1: a marker for lung adenoarinoma in body cavity fluids

BACKGROUND

Adenocarcinomas are the most common epithelial malignancies in body cavity fluids. Subclassification of adenocarcinomas according to primary site can be a challenging task. Thyroid transcription factor 1 (TTF-1) is a nuclear transcription factor that is expressed in normal lung, in thyroid, and in their neoplasms. Because thyroid carcinomas rarely metastasize to the serosal surfaces, the authors used TTF-1 as a marker to distinguish adenocarcinomas of the lung from carcinomas of other organs.

METHODS

The authors studied 113 body cavity fluids (92 pleural fluid samples and 21 ascitic samples) from 113 patients with a diagnosis of adenocarcinoma on the basis of routine cytology. The primary sites of origin were confirmed clinically or histologically for all patients. There were 39 adenocarcinomas of the lung, 24 adenocarcinomas of the breast, 34 adenocarcinomas of the genitourinary tract, and 16 adenocarcinomas of the gastrointestinal tract. Archival Papanicolaou-stained, cytocentrifuged slides were used without destaining for immunocytochemistry with the monoclonal TTF-1 antibody using a commercial available method.

RESULTS

TTF-1 was expressed in 21 of 39 lung adenocarcinomas (54%). Intense nuclear staining was present in tumor cells that occurred in groups or in isolated form. In contrast, none of the other types of adenocarcinomas expressed TTF-1.

CONCLUSIONS

TTF-1 is a highly specific marker for adenocarcinomas of the lung in body cavity fluids. Immunocytochemistry using this antibody can be performed easily on archival Papanicolaou-stained, cytocentrifuged slides of fluid specimens.

Fine-needle aspiration biopsy of the retroperitoneum: a series of 111 cases not including specific organs

We report on our experience in fine-needle aspiration (FNA) biopsy of the retroperitoneum: 111 FNA biopsies performed on 99 patients. Cytologic diagnoses were divided into four groups: nondiagnostic (unsatisfactory samples because of a low cellularity and/or improperly prepared smears) aspirates (20%), benign (16%), suspicious for malignancy (13%), and malignant (50%). There were no known false-positive samples. We had two false-negative diagnoses due to sampling errors. Among diagnostic smears, the procedure showed a sensitivity of 97% and a specificity of 100%. The predictive value of a positive result was 100% and the predictive value of a negative result was 90%. The overall accuracy was 98%. Metastatic carcinomas accounted for the largest number of lesions in the group of malignant tumors. A primary tumor site was known for the majority of the cases before the aspiration was performed. In the remaining cases we were unable to suggest an origin. It is therefore important to emphasize the role of ancillary studies in patients that are at the first assessment of the disease or when a second intercurrent malignancy is suspected. In our limited experience, a suggestion of the correct subtype of retroperitoneal sarcoma was not possible. As in the rest of cytopathology, a multidisciplinary approach is mandatory in this setting to improve patient management.

Fine-needle aspiration biopsy of the adrenal glands: A ten-year experience

We report on our experience in FNA biopsy of the adrenal gland: 177 biopsies performed in the last 10 years. Cytologic diagnoses were divided into four groups: nondiagnostic aspirates (28%), primary adrenal lesions (13%), metastatic neoplasms (33%), and negative cases with known extra-adrenal malignancies (25%). Among diagnostic smears and excluding the latter group, the procedure was 100% specific for malignancy, and 98% of the lesions were correctly diagnosed. There were no known false-positive or false-negative samples. Quality of diagnosis improves with careful smearing (avoids artifacts) and immediate evaluation (raises adequacy rates) by the pathologist. Although the primary or secondary nature of most adrenal masses is readily apparent, it is essential to correlate the clinical, laboratory, and cytologic findings to reach the correct diagnosis. Furthermore, we believe that the primary site of many adrenal metastases must be defined on the basis of clinical data. Diagn. Cytopathol. 1999;21:92-97.

Cytology: screening or diagnostic tool?

Previously used only as a screening tool, cytology now emerges as a powerful diagnostic technique, especially since the advent of the fine needle aspiration (FNA) biopsy. This article highlights the use of ancillary techniques, primarily electron microscopy (EM), and immunohistochemistry (IHC). When coupled with routine cytological examination such as FNA and body cavity fluid cytology, EM and IHC can refine the diagnosis and make it more precise. The authors discuss how to solve common diagnostic dilemmas by the use of cytology along with IHC and EM. The following common diagnostic problems are addressed: mesothelioma versus adenocarcinoma, neuroendocrine neoplasms and their mimickers, melanoma versus carcinoma versus sarcoma, hepatocellular carcinoma versus adenocarcinoma and adenocarcinomas of unknown primary.

Immunocytochemistry controls using cell culture

A continuing problem in cytology laboratories is the lack of adequate control material for immunocytochemical testing. Usual control procedures involve testing paraffin-embedded control materials along with the patient specimens. These control materials are fundamentally unlike cytologic preparations. We have developed a method to make control preparations for immunocytochemical analysis using cultured anaplastic cells with known antigenic features from commercial sources. Cell lines included melanoma, rhabdomyosarcoma, T-cell leukemia, and squamous-cell carcinoma. Modified Saccomano and acetone fixation coupled with the cytospin technique enabled good-quality preparations. Cell lines were tested with antibodies for HMB-45, actin, leukocyte common antigen (LCA) and cytokeratin, which avidin-biotin immunoperoxidase and diaminobenzidine (DAB) as chromogens. Our final preparations were easily interpretable with excellent morphologic preservation of cellular detail. Cultured cells provide a superior method for preparing almost unlimited numbers of control slides for immunocytochemistry for laboratories with access to a tissue culture facility.

Growth fractions in canine non-Hodgkin's lymphomas as determined in situ by the expression of the Ki-67 antigen

The proportion of proliferating cells in non-Hodgkin's lymphomas (NHLs) as determined in situ by the expression of the Ki-67 antigen, has prognostic value in human oncology, and is strongly related to the different grades of malignancy. The evaluation of the Ki-67 index in canine NHLs may be useful in assessing the individual variability of the growth fraction in the different sub-types of lymphoma, and also the validity of the classification in terms of grade of malignancy. The growth fraction was evaluated in 92 canine NHLs, previously classified according to the Kiel classification (as adapted to the canine species), by determining the expression of the Ki-67 antigen with the MIB1 antibody on (1) paraffin-wax tissue sections in all 92 cases, and (2) fine-needle aspirates or tumour imprints in 30 cases. The labelling appeared satisfactory in 88% of the cases, with good concordance between the histological and cytological data. A highly significant correlation (P < 0.001) was established between the proportion of Ki-67+ cells and the classification into low-grade (Ki-67 index < 21%) and high-grade malignancy (Ki-67 index > 21% and usually > 29%). In the low-grade lymphoma group, a macronucleolated medium-sized-cell lymphoma not found in man had the lowest proliferation index. In the high-grade malignancy group, the number of Ki-67+ cells seemed to be proportional to cell size, whatever the phenotype, with the rare exceptions of some unclassifiable small-cell Burkitt-type or plasmacytoid lymphomas, which were highly proliferating. The classification of lymphomas into low-grade and high-grade appears to correlate well with their proliferative index. The existence of individual variations, within given categories of canine NHL, suggests that, as in human medicine, prognosis may be assisted by determining the growth fraction at initial diagnosis, and by fine-needle aspiration at relapses.

Immunocytochemical staining of smears and corresponding cell blocks from serous effusions: a follow-up and comparative investigation

Recently, we have recommended immunocytochemistry on serous effusions with the monoclonal antibodies Ber-EP4 and EMA to be used as a routine procedure. In this study, our earlier defined immunocytochemical profiles were tested in daily diagnostic work for a period and the profiles were applicated on the corresponding cell blocks from the effusions, too. It is concluded that routine use of the benign, malignant epithelial, and malignant mesothelial immunocytochemical profiles is valuable and superior to cytomorphology alone. Additionally, immunocytochemical staining of smears proved slightly more sensitive than immunohistochemistry performed on sections from the cell blocks.

DNA and keratin analysis of oral exfoliative cytology in the detection of oral cancer

Refinements in oral exfoliative cytology may make it a suitable screening technique for the early diagnosis of oral cancer. In this study DNA range profiles were combined with keratin expression in an attempt to improve the diagnostic accuracy of oral exfoliative cytology. Smears were taken from 33 biopsy-proven oral cancers and the contralateral normal site. For DNA range profiles the smears underwent Feulgen hydrolysis, with DNA distribution being assessed using the Vickers M85 microdensitometer. For keratin expression a panel of antikeratin antibodies were applied. The smears for keratin expression were then graded on a three-point scale. Abnormal DNA range profiles were observed in 23 of 33 smears taken from oral cancers and in two smears from normal oral mucosa (sensitivity 70%, specificity 90%, positive predictive value 90%). The simple epithelial keratins 8 and 19 were identified in the majority of oral cancer smears. The sensitivity of keratin 19 was greater (90%). However, keratin 8 was the most useful keratin marker associated with malignancy (sensitivity 62%, specificity 100%, positive predictive value 100%). The combination of simple keratin expression and DNA content improved the cancer detection rate beyond that obtainable with DNA range profile alone.

Retrospective c-erbB-2 immunostaining in aspiration cytology of breast cancer

Fifty fine-needle aspiration cytologies of breast that were diagnosed as carcinomas were retrieved from the files and retrospectively evaluated for the expression of c-erbB-2 oncoprotein using standard immunocytochemical methods. Corresponding histologic sections of all tumors were similarly studied. Seventeen fine-needle aspirates (34%) reacted positively for the presence of c-erbB-2 oncoprotein. All but one (32%) of the corresponding tissue sections were also positive for c-erbB-2 by immunohistochemistry. All positive cases were infiltrative ductal carcinomas with a preponderance of the comedo type. Positive reactions were localized in the cytoplasmic membrane of tumor cells. The staining was either present in all cells throughout a tumor, or it was completely absent. We conclude that immunocytochemistry for c-erbB-2 oncoprotein can be performed on fine-needle aspiration cytology samples that are previously fixed and stained with the Papanicolaou technique. Furthermore, the sensitivity of immunostaining results are comparable to that obtained in histologic sections.

Preoperative diagnosis of renal angiomyolipoma: fine needle aspiration cytology and immunocytochemical characterization

A preoperative diagnosis of renal angiomyolipoma (AML) is of great importance for a correct management of these patients with this tumor. In fact when the lesion is small and asymptomatic a conservative approach may be considered. We have evaluated the radiographic and fine needle aspiration cytology (FNAB) findings in 8 cases of AML. In 3 cases both radiology and cytology were suggestive of carcinoma and thus the patients underwent surgery. In one case both techniques suggested AML but surgery was performed because the lesion was large and symptomatic. In 4 cases where both radiology and cytology suggested AML no surgery was performed. Follow-up data are consistent with the benign nature of the lesions. The immunocytochemical analysis of the FNAB with a panel of antibodies including keratin, vimentin, actin and HMB-45 was indicative of AML in 7 of 8 cases, including 2 of the 3 cases misdiagnosed as carcinomas. The presence of HMB-45-positive perivascular epithelioid cells in the FNABs was the most significant finding. It is concluded that immunocytochemical analysis of FNAB with this monoclonal antibody panel can increase the accuracy of preoperative diagnosis of AML, and allow consideration of a conservative approach in selected cases.

Immunocytochemical staining of serous effusions: an additional method in the routine cytology practice?

In a consecutive and prospective cytomorphologic and immunocytochemical study we have examined 100 serous fluids with a panel of antibodies. Three different immunocytochemical patterns of staining were recognized: (i) a benign profile showing no Ber-EP4 or CEA-positive cells; (ii) a malignant profile with Ber-EP4 and strongly EMA-positive epithelial cells; and (iii) a malignant profile in which mesothelial cells were strongly positive for EMA. By applying these profiles the number of malignant cases recognized increased from 19 to 38. All cytomorphologic malignant fluids showed a malignant profile, but in two cases a malignant epithelial profile was found in patients without otherwise proven malignant disease (false positive staining). Immunocytochemistry with anti-Ber-EP4 and anti-EMA can be recommended as a routine procedure, but the marker result should always be correlated with cytomorphology, eventual histologic data and clinical records.

Cytokeratin expression in oral exfoliative cytology: effect of temperature and fixation

The identification of keratin expression within oral cytology may be useful in the diagnosis of clinically suspicious oral mucosal lesions. There may be wide variation in the temperature at which such smears are stored, prior to processing. Conventionally, rapid fixation or storage at low temperatures is recommended to preserve kertin expression within tissue biopsies. No previous study has assessed whether this is true for oral cytology. Smears were taken from clinically normal buccal mucosa. For each temperature assessed (-70, -40, -22, +5, +20 and +26 degrees C), one smear was spray-fixed (Vale Smear Fix) and one air-dried, prior to storage for 4 days, and then staining with the pan-epithelial antikeratin antibody, LP34. Preservation of keratin expression was assessed as either weak (zero to few positive cells) or strong (most cells positive). The results were analysed using logistic regression with the statistical modelling package, GLIM. Over the range of temperatures studied, spray fixation did not appear to improve the identification of keratin expression. Although the best preservation was obtained at lower temperatures, keratin expression was still adequate after 4 days at 20 degrees C. Hence, a delay in processing of 4 days would still allow detectable expression in oral exfoliative cytology even at room temperature.