Morphologic and Immunocytochemical Performances of Effusion Cell Blocks Prepared Using 3 Different Methods

Size: px

Start display at page:

Download "Morphologic and Immunocytochemical Performances of Effusion Cell Blocks Prepared Using 3 Different Methods"

Harry Wilkins
6 years ago
Views:

1 Anatomic Pathology / Performance of Different Cell Block Preparation Techniques Morphologic and Immunocytochemical Performances of Effusion Cell Blocks Prepared Using 3 Different Methods Xin Jing, MD, 1 Qing Kay Li, MD, PhD, 2 Ursula Bedrossian, PhD, 3 and Claire W. Michael, MD 4 Key Words: Cell block preparations; Morphologic performance; Immunocytochemical performance; Serous effusions Abstract With increased use of the ThinPrep method for nongynecologic specimens, cell blocks are more commonly prepared by harvesting cells that are fixed in CytoLyt solution. The current study compared morphologic and immunocytochemical performance of effusion cell blocks prepared using CytoLyt-prefixed thrombin clot (CTC) with plasma thrombin clot (PT) and HistoGel (HG) preparation. The study included a total of 25 malignant or benign serous fluids. Three individual cell block materials were simultaneously prepared from each of the 25 effusion specimens using the CTC, PT, or HG method. H&E staining and immunostaining for pancytokeratin (pan-ck), carcinoembryonic antigen (CEA), epithelial membrane antigen (EMA), B72.3, HBME-1, estrogen receptor (ER), progesterone receptor (PR), CD45, CD20, and CD3 were then performed. The CTC preparation revealed compatible cellularity and good cellular details. In addition, CTC cell blocks revealed a similar percentage of cells with positive immunostaining along with the strongest intensity and the least background staining. The CTC method can be used reliably as an adjunct to other preparation techniques. The cell block preparation has been commonly used as a complementary adjunct to cytologic evaluation of nongynecologic specimens. 1-4 Morphologic and immunocytochemical evaluation of cell block material commonly provides additional information that is crucial to resolve diagnostic dilemmas. Cell blocks may be prepared through various techniques. Traditionally, cells are harvested by centrifuging the collection tube, and cell blocks are then prepared from the cell palette using either the plasma thrombin clot (PT; Jones Pharma, St Louis, MO) or HistoGel (HG; commercially available agar, Thermo Scientific, Waltham, MA) methods. With increasing use of the ThinPrep method for nongynecologic specimens, cell blocks are more commonly prepared by harvesting cells that were previously collected and fixed in CytoLyt solution, a methanol-based medium (Hologic, Marlborough, MA). This is particularly true for fine-needle aspiration (FNA) specimens in which CytoLyt solution becomes the collecting medium of choice in many laboratories. Many laboratories use the HG method for preparing cell blocks from FNA specimens because the PT method becomes technically difficult with the introduction of methanol, the principal alcohol contained in CytoLyt solution, and the hemolysis of blood. In the current study, we explored whether the morphologic and immunocytochemical performance of cell blocks prepared from effusion specimens may vary depending on the collecting medium and other variables during cell block preparation using the PT, HG, and CTC methods. Materials and Methods The study consisted of a total of 25 specimens of serous fluids, including 16 adenocarcinomas, 2 mesotheliomas, 4 Downloaded from Am J Clin Pathol 2013;139:

2 Jing et al / Performance of Different Cell Block Preparation Techniques lymphomas, and 3 negative effusions. Although effusions are not routinely collected in CytoLyt solution, effusions were selected as the sample of choice for the following reasons: (1) the specimens represented a wide range of diseases and (2) the specimens provided enough cellularity for simultaneous preparation of the 3 cell blocks needed in this study. Preparation of Cell Pellets and Cell Blocks Each fluid specimen was vigorously agitated, collected in three 50-mL aliquots, and centrifuged at 2,320 rotations per minute for 5 minutes. The supernatant was discarded and 3 individual cell pellets were randomly chosen for the 3 different cell block preparation techniques. For the PT method, 2 to 3 drops each of plasma and thrombin (1,000 U/mL) were added to and mixed with one of the pellets. The mixture was set aside at room temperature for 30 seconds and allowed to clot. For the HG method, HistoGel was heated at 60 C and converted to a liquid state which is maintained at 50 C. Four to 5 drops of melted HistoGel were then added to and mixed with 1 pellet. The mixture was placed at room temperature and allowed to solidify. For the CTC method, the remaining cell pellet was resuspended in 30 ml of CytoLyt solution and allowed to sit for at least 30 minutes to simulate the process used for CytoLyt collected specimens. The samples were recentrifuged and the supernatant was discarded. A cell block was then prepared using the aforementioned PT preparation. The PT method could still be performed in effusion specimens despite the CytoLyt fixation, and was therefore selected to ensure separate evaluation of the methanol prefixation compared with the HG method as 2 different variables. Morphologic and Immunocytochemical Examination Each cell block was placed in a tissue cassette and fixed in 10% formalin for 24 hours and processed using the traditional method. Briefly, a total of eleven 4-μm sections were cut from each paraffin-embedded cell block. One section was stained with H&E stain. The remaining 10 sections were mounted on Baxter M6146 PLUS slides (Baxter Scientific, Deerfield, IL) and heated at 60 C for 30 to 60 minutes. Using the standard avidin-biotin-peroxidase complex technique, immunostaining was performed on Dako Autostainer (Dako, Carpinteria, CA). Antibodies against pancytokeratin (pan- CK), carcinoembryonic antigen (CEA), B72.3, epithelial membrane antigen (EMA), HBME-1, estrogen receptor (ER), progesterone receptor (PR), CD45, CD20, and CD3 were used. Manufacturers, dilutions, and methods of pretreatment for the individual antibodies are shown in Table 1. Appropriate positive and negative controls were obtained. H&E-stained cell block slides were assessed for morphologic appearance including cellularity; cell distribution within the section; nuclear-cytoplasmic preservation; and presence of artifacts such as exaggerated cytoplasmic vacuoles, denser cytoplasm, more frayed cytoplasmic borders, cellular shrinkage, and increased cytoplasmic and nuclear holes. Morphologic preservation was graded as good and fair when cytoplasmic and nuclear details were well-preserved without the aforementioned artifacts in more than 50% and less than 50% of the cells of interest, respectively. The immunostaining was assessed based on the approximate percentage of cells of interest showing positive cytoplasmic membrane staining for pan-ck, CEA, EMA, B72.3, HBME-1, CD45, CD20, and CD3, as well as nuclear staining for ER and PR. Semiquantitative scores were given as follows: 0 (complete lack of or blush staining), 1 (5%-25%), 2 (25%- 50%), 3 (50%-75%), and 4 (75%-100%). The immunostaining was considered to be positive when the score was 1 or more. Results Technical Issues Preparation of the PT block involved the least technical details. It was the simplest and the least time-consuming process and could be performed at room temperature. The HG method was a more tedious process as HistoGel had to be initially converted and maintained in the liquid state. CTC was the most difficult to prepare as it took a longer time to clot. In cell blocks prepared with the PT and CTC methods, cells tended to aggregate in the center of the clot and the cut sections whereas in the HG preparation, the cells gravitated to the bottom of the hardened pellet. This was addressed by dividing the HG pellet in half and embedding the resulting semispheres at the cut side. 5 Morphologic Appearance The cellularity of the CTC sections was comparable to that of the PT while the HG sections appeared to have the least cellularity, with the cellular groups interspersed among large Table 1 Description of the Antibodies Antibody Manufacturer Dilution Pretreatment pan-ck: AE1/3 Chemocon 1:200 Protease Cam5.2 Becton Dickinson 1:40 None CEA Dako 1:1000 ph = 8.5 EMA Ventana Prediluted ph = 8.5 HBME-1 Dako 1:400 None ER Ventana Prediluted ph = 8.5 PR Ventana Prediluted ph = 8.5 CD45 Ventana Prediluted ph = 8.5 CD20 Ventana Prediluted ph = 8.5 CD3 Ventana Prediluted ph = 8.5 CEA, carcinoembryonic antigen; EMA, epithelial membrane antigen; ER, estrogen receptor; pan-ck, pancytokeratin; PR, progesterone receptor. 178 Am J Clin Pathol 2013;139: Downloaded 178 from

Anatomic Pathology / Original Article agar lakes. Among all 3 preparations, PT blocks revealed the best overall morphologic preservation with minimal artifacts.

increased nuclear-cytoplasmic holes Image 1.

3 Anatomic Pathology / Original Article agar lakes. Among all 3 preparations, PT blocks revealed the best overall morphologic preservation with minimal artifacts. Cells in the HG sections exhibited exaggerated cytoplasmic vacuoles, denser cytoplasm, and more frayed cytoplasmic borders, whereas cells in the CTC sections exhibited more cellular shrinkage and increased nuclear-cytoplasmic holes Image 1. However, proportions of cell blocks with good and fair morphologic preservation (defined as artifacts seen in more and less than 50% of the cells of interests, respectively) were similar among all 3 methods Table 2. Immunohistochemical Performance Immunoreactivity, in terms of approximate percentage of cells with positive staining, was detected to be similar, regardless of the preparations used Table 3. In comparison with cell blocks prepared with the PT or HG method, CTC cell blocks showed the most intense staining and the least background staining Image 2. False-positive staining was not seen in any of the 3 preparations. Discussion Ideally, cellular components revealed by cell blocks are expected to exhibit the following features: resemblance to corresponding cells in alcohol-fixed Papanicolaou (Pap) stained smears, adequate preservation of nuclear-cytoplasmic details, easy recoverability, and suitability for ancillary studies such as immunostaining. An ideal method for preparation would be simple, reproducible, and readily adaptable in a routine setting. 4 For nongynecologic specimens such as fluids and FNA specimens, diagnostic usefulness of cell blocks for providing additional cytomorphologic and/or immunocytochemical information is well documented. 3,6-8 Those studies used A B C Image 1 Metastatic adenocarcinoma of esophagus. H&E-stained cell blocks were prepared using the plasma thrombin (PT) (A), HistoGel (HG) (B) and CytoLyt-prefixed thrombin (CTC) (C) clot methods. CTC and PT sections reveal similar cellularity whereas the HG section has the least cellularity. PT blocks showed the best cellular detail and the least artifacts among all 3 methods (H&E, 400). Downloaded from Am J Clin Pathol 2013;139:

4 Jing et al / Performance of Different Cell Block Preparation Techniques Table 2 Morphologic Appearance of H&E-Stained Cell Blocks Prepared With Different Methods a Cellularity Cellular Preservation Preparation Methods Moderate/High Low Good Fair Presence of Artifacts CTC (n = 25) 21 (84) 4 (16) 21 (84) 4 (16) Yes PT (n = 25) 22 (88) 3 (12) 22 (88) 3 (12) Minimal HG (n = 25) 20 (80) 5 (20) 21 (84) 4 (16) Yes CTC, CytoLyt-prefixed thrombin clot; HG, HistoGel; PT, plasma thrombin clot. a Data are given as number (percentage) unless otherwise indicated. Table 3 Immunostaining Performance in Cell Blocks Prepared With Different Methods a CTC HG PT Antibodies Positive Negative Positive Negative Positive Negative Epithelial markers (n = 18) pan-ck 17 (94) 1 (6) 16 (89) 2 (11) 17 (94) 1 (6) CEA 5 (28) 13 (72) 6 (33) 12 (67) 5 (28) 13 (72) EMA 18 (100) 0 (0) 18 (100) 0 (0) 18 (100) 0 (0) B (33) 12 (67) 5 (28) 13 (72) 4 (22) 14 (78) HBME 3 (17) 15 (83) 3 (17) 15 (83) 3 (17) 15 (83) ER 9 (50) 9 (50) 9 (50) 9 (50) 9 (50) 9 (50) PR 3 (17) 15 (83) 3 (17) 15 (83) 3 (17) 15 (83) Lymphoid markers (n = 7) CD45 7 (100) 0 (0) 6 (86) 1 (14) 7 (100) 0 (0) CD20 4 (57) 3 (43) 3 (43) 4 (57) 3 (43) 4 (57) CD3 0 (0) 7 (100) 0 (0) 7 (100) 0 (0) 7 (100) CEA, carcinoembryonic antigen; CTC, CytoLyt-prefixed thrombin clot method; EMA, epithelial membrane antigen; ER, estrogen receptor; HG, HistoGel; pan-ck, pancytokeratin; PR, progesterone receptor; PT, plasma thrombin clot. a Data are given as number (percentage) unless otherwise indicated. traditional preparation techniques including the PT or agar methods. Cell blocks prepared from ThinPrep specimens with the PT method have been used mainly for gynecologic Pap specimens preserved in PreservCyt (Hologic). 9,10 To prevent alcohol from interfering with coagulation, the sediment is washed by resuspending in normal saline and recentrifuging. Data on the performance of cell blocks prepared from nongynecologic specimens fixed in CytoLyt are sparse. It is widely assumed by pathologists that the methanol contained in CytoLyt may interfere with the immunocytochemical stains, and potentially introduce cellular artifacts such as considerable shrinkage. We are aware of only 1 relevant study that consisted of 3 serous fluids and 10 fine-needle aspirates. 11 The authors of that study indicated that the cell blocks prepared with the PT method for nongynecologic specimens provided superior cellularity, cell distribution, and background quality compared with other methods, including inverted filter sedimentation, albumin method, and simple sedimentation. In the current study, effusions were selected as the sample of choice because the specimens represented a wide range of diseases. Furthermore, each sample provided enough cellularity to simultaneously prepare the 3 cell blocks needed for the study. Our data generated from 25 serous fluid specimens demonstrated that, compared with the 2 traditionally used methods (PT and HG), CTC preparation provided compatible cellularity and good cellular details. In addition, CTC cell blocks revealed compatible immunostaining performance manifested by similar percentage of cells of interest with positive staining and absence of false-positive staining. It is noteworthy that CTC cell blocks revealed the most intense staining and the least background staining. In a previous study we also demonstrated that CytoLyt preservation did not interfere with S-100 protein stain, which is known to be inhibited by alcohol fixation. This performance can probably be explained by the low concentration of methanol in CytoLyt. 12 The presence of finely granular background has been reported with CytoLyt preparations 13 but it was absent in our specimens. Although CytoLyt solution was the collecting medium in both the studies, the former study 13 used agar and ethanol gel to prepare the cell blocks rather than PT. Among all 3 preparations, HG cell blocks presented more pronounced artifacts, mainly heat-related, such as exaggerated vacuoles, dense cytoplasm, and shrunken cells with frayed cytoplasmic borders. CTC cell blocks showed shrinkage of cells and increased nuclear-cytoplasmic holes. The latter may 180 Am J Clin Pathol 2013;139: Downloaded 180 from

Anatomic Pathology / Original Article Image 2 Representative example of immunostaining in cell blocks prepared using the plasma thrombin (PT), HistoGel (HG) and CytoLyt-prefixed thrombin (CTC) clot

5 Anatomic Pathology / Original Article Image 2 Representative example of immunostaining in cell blocks prepared using the plasma thrombin (PT), HistoGel (HG) and CytoLyt-prefixed thrombin (CTC) clot methods: pancytokeratin (pan-ck) in esophageal adenocarcinoma; epithelial membrane antigen (EMA) in mesothelioma; HBME-1 in reactive mesothelial cells; estrogen receptor (ER) in ovarian adenocarcinoma; and CD20 in B-cell lymphoma. CTC cell blocks appeared to show the most intense staining and the least background staining (immunostain, 400). Downloaded from Am J Clin Pathol 2013;139:

6 Jing et al / Performance of Different Cell Block Preparation Techniques be related to hemolytic and mucolytic agents in CytoLyt solution. Preparation of cell blocks with PT method was relatively easier, whereas HG was more technically involved. CTC cell blocks took a longer time to clot, most likely as a result of the hemolytic agents in CytoLyt solution. In conclusion, effusion cell blocks prepared with the CTC method provided compatible cellularity and good cellular details. The preservation in methanol-containing CytoLyt did not interfere with the immunostaining performance. It can be used reliably as an adjunct to the preparation techniques commonly used in the cytologic laboratory. From the 1 Department of Pathology, University of Michigan, Ann Arbor, MI; 2 Department of Pathology, The Johns Hopkins University, Baltimore, MD; 3 College of Science and Math, Madonna University, Livonia, MI; and 4 Case Western University/ University Hospitals Case Medical Center, Cleveland, OH. Address reprint requests to Dr Jing: Dept of Pathology, University of Michigan, 1500 E Medical Center Dr, Ann Arbor, MI ; xinjing@med.umich.edu. References 1. Zito FA, Gadaleta CD, Salvatore C, et al. A modified cell block technique for fine needle aspiration cytology. Acta Cytol. 1995;39: Schieven LW, Smedts F, Hopman AH, et al. Fine needle aspiration using improved agar microbiopsy is highly concordant with renal mass final diagnosis and subclassification. J Urol. 2009;182: Kulkarni MB, Desai SB, Ajit D, et al. Utility of the thromboplastin-plasma cell-block technique for fine-needle aspiration and serous effusions. Diagn Cytopathol. 2009;37: Nathan NA, Narayan E, Smith MM, et al. Cell block cytology: improved preparation and its efficacy in diagnostic cytology. Am J Clin Pathol. 2000;114: Bales C. Laboratory techniques. In: Koss L, ed. Koss Diagnostic Cytopathology And Its Histopathologic Bases. Philadelphia, PA: Lippincott Williams & Wilkins, 2006: Fetsch PA, Simsir A, Brosky K, et al. Comparison of three commonly used cytologic preparations in effusion immunocytochemistry. Diagn Cytopathol. 2002;26: Yang GC, Wan LS, Papellas J, et al. Compact cell blocks: use for body fluids, fine needle aspirations and endometrial brush biopsies. Acta Cytol. 1998;42: Smedts F, Schrik M, Horn T, et al. Diagnostic value of processing cytologic aspirates of renal tumors in agar cell (tissue) blocks. Acta Cytol. 2010;54: Keyhani-Rofagha S, Vesey-Shecket M. Diagnostic value, feasibility, and validity of preparing cell blocks from fluidbased gynecologic cytology specimens. Cancer. 2002;96: Rowe LR, Marshall CJ, Bentz JS. Cell block preparation as an adjunctive diagnostic technique in ThinPrep monolayer preparations: a case report. Diagn Cytopathol. 2001;24: Nigro K, Tynski Z, Wasman J, et al. Comparison of cell block preparation methods for nongynecologic ThinPrep specimens. Diagn Cytopathol. 2007;35: Jing X, Michael CW, Theoharis CG. The use of immunocytochemical study in the cytologic diagnosis of melanoma: evaluation of three antibodies [published online ahead of print October 24]. Diagn Cytopathol Maksem JA, Weidmann J. Specialized preparative devices are not needed for liquid-based, thin-layer cytology: an alternate manual method using a metastable alcoholic gel. Diagn Cytopathol. 2001;25: Am J Clin Pathol 2013;139: Downloaded 182 from

What kind of material should we use for ICC in our daily routine. Torill Sauer Department of Pathology, Akershus University Hospital

What kind of material should we use for ICC in our daily routine Torill Sauer Department of Pathology, Akershus University Hospital Diversity of preparing cytological material Cell block Direct smears