|Year : 2017 | Volume
| Issue : 3 | Page : 425-429
Utility of cell block to detect malignancy in fluid cytology: Adjunct or necessity?
Sumedha Dey1, Dipanwita Nag1, Ayandip Nandi1, Ranjana Bandyopadhyay2
1 Department of Pathology, Medical College, Kolkata, West Bengal, India
2 Department of Pathology, Calcutta National Medical College, Kolkata, West Bengal, India
|Date of Web Publication||31-Aug-2017|
48/7, Patuapara Lane, P.O. Serampore, Hooghly - 712 201, West Bengal
Source of Support: None, Conflict of Interest: None
Aim of the Study: Cell block (CB) technique when supplemented with conventional smear, provides increased cellularity, preservation of architectural pattern with excellent morphology, and a clear background. We compare the utility of CB technique compared to conventional smear in detection of malignancy in serous effusions.
Materials and Methods: An institution-based observational and analytical study was carried out over 1 year on 50 patients with effusions. The residual amount of centrifuged deposit after preparation of conventional smear was mixed with 10% alcohol-formalin solution, and CBs were prepared. Calretinin and cytokeratin 5 were used for reactive mesothelial cells and Wilms tumor 1, thyroid transcription factor 1, CDX2, and estrogen receptor were used to confirm the adenocarcinoma cells.
Results: Maximum patients belonged to the age group of 61–70 years. Male:female ratio 1:1.17. Most common cause of malignant peritoneal effusion was due to ovarian malignancies in females and adenocarcinoma of stomach in males while, in case of pleural effusion, it was breast carcinoma in females and lung carcinoma in males. Thirteen suspicious cases were subjected to immunohistochemistry (IHC). In 70% cases, CB findings were consistent with the findings of conventional smears. In 20% cases, the conventional smears were suspicious for malignancy, and malignancy was confirmed by CB technique, whereas in 10% cases, both smears and CB were suspicious for malignancy and the original nature of the lesion was confirmed by the IHC. Sensitivity and specificity of CB compared to conventional smear were 88.88% and 86.98%, respectively.
Conclusion: CB produced significantly better results (P = 0.0271) while detecting malignant lesions and reducing suspicious results (P = 0.0226).
Keywords: Cell block, immunohistochemistry, serous effusions
|How to cite this article:|
Dey S, Nag D, Nandi A, Bandyopadhyay R. Utility of cell block to detect malignancy in fluid cytology: Adjunct or necessity?. J Can Res Ther 2017;13:425-9
| > Introduction|| |
Examination of body fluids has been accepted as a routine laboratory procedure, not only to differentiate between transudative or exudative causes but also for the detection of unsuspected cancers and metastasis from cancer of unknown primary origin. The body fluid sent for cytological examination is subjected to total leukocyte count and differential count and presence of tumor cells. The residual material after cytologic examination which is usually discarded can be used in the preparation of cell block (CB). Richardson et al., Dekker and Bupp have shown that additional diagnosis of cancer can be obtained in fluid specimen if smear technique is supplemented by CB sections of residual material. It also provides material for special stains such as Periodic acid–Schiff (PAS), alcian blue and immunohistochemistry (IHC). By using a combination of the conventional smear and the CB methods for the reporting of effusions, malignancy, and the primary site could be determined with 81% accuracy.,
Aims and objective
Objective of our study was to compare the sensitivity and diagnostic specificity of CB preparation versus conventional smear method for detection of malignancy in fluid cytology, to assess the usefulness of IHC to distinguish reactive mesothelial cells from adenocarcinoma cells and to perform special stains when needed to reach a diagnosis of primary site of malignancy in cases of malignant effusions.
| > Materials and Methods|| |
An institution-based observational and analytical study was carried out with 50 patients with body cavity effusions over 1 year. The fluid aspirated was examined for physical characteristics such as appearance, color, and coagulum. Total leukocyte count was obtained using the Neubauers modified counting chamber. Conventional smears were prepared from the centrifuged deposit and stained with Leishman-Giemsa and Papanicolaou stain in combination for better categorization of atypical mononuclear cells. Now, the residual amount of centrifuged deposit is mixed with 10% alcohol-formalin solution and centrifuged again. The cell button obtained was kept for overnight fixation. The next day, the cell button obtained was taken in a filter paper and processed through three changes of alcohol, two changes of xylene, and two changes of paraffin and a CB was obtained. Microtomy was done, and sections were stained with hematoxylin and eosin. The suspicious specimens were subjected to special stains as PAS and IHC by peroxidase-antiperoxidase technique for further categorization. Calretinin and cytokeratin 5 were used for reactive mesothelial cells and Wilms tumor 1 (WT1), estrogen receptor (ER), CDX2, and thyroid transcription factor 1 (TTF-1) were used to confirm the adenocarcinoma cells from ovarian (serous), breast, intestinal, and lung primary, respectively. Statistical analysis was done in GraphPad InStat 3 (http://www.graphpad.com/scientific-software/instat/) software.
| > Results|| |
Fifty patients from age 24 to 82 years were included in our study. Among them, maximum number of patients belonged to the age group of 61–70 years. The female:male ratio was 1.17:1. Twenty-six cases presented with pleural effusion, whereas 24 cases were peritoneal. Among the 50 cases, 22 cases (44%) presented with effusion which was transudative in nature, whereas 28 cases (56%) presented with exudative effusion. 34% presented with hemorrhagic and straw colored fluid, and 16% presented with purulent effusion; 6% were turbid with deposit whereas 10% presented with serosanguinous fluid. While doing cell typing, lymphocytes were the most common cells found followed by polymorphs, mesothelial cells, and atypical mononuclear cells.
Most common cause of malignant peritoneal effusion was due to ovarian malignancies in females (14 cases, 51.8%) and adenocarcinoma of stomach in males (5 cases, 18.5%), whereas, in case of pleural effusion, it was breast carcinoma in females (3 cases, 11.2%) and lung carcinoma in males (3 cases, 11.2%).
The findings of conventional smears and that of CB prepared from pleural fluid are categorized into three classes: benign, suspicious, and malignant according to their cellularity, morphology, and architectural pattern. In conventional smears prepared from the pleural fluid, maximum outcome was of benign (50%) followed by malignancy (26.92%) followed by suspicious for malignancy (23.07%). However, in CB preparations, benign category reduced to 46.15%, malignancy increased to 42.30%, and suspicious category significantly reduced to 11.53%. In conventional smears prepared from the peritoneal fluid, six cases (25%) were malignant followed by nine cases (37.5%) were suspicious for malignancy and benign each. In CB prepared from the peritoneal fluid, malignant cases increased to 54.16% benign cases reduced to 33.33%, whereas only three cases (12.5%) were suspicious for malignancy [Table 1].
|Table 1: Analysis of discrepancies between conventional smears and cell block technique in pleural fluid|
Click here to view
Cellularity of the smears and block for detecting malignancy was graded according to the presence of tumor cells in the smears and CB into no cells, low (<10% tumor cells), Moderate (10–50% tumor cells) and High (>50% tumor cells). CB method yielded high cellularity in 74.07% cases compared to 29.62% in conventional smear, moderate yield in 14.81% compared to 33.3% in conventional smear, and low yield significantly reduced to 11.1% from 37.07% in conventional smear.
When conventional smears and CB findings are compared on the basis of preservation of morphological features (i.e., intactness of cell membrane, cytoplasmic characteristics, nuclear membrane, nuclear chromatin pattern, and presence or absence of nucleoli, presence of microvilli, and cellular window) [Figure 1]a; it was found that morphological features were far better preserved in CB technique (81.48%) as compared to conventional smears (51.85%) [Figure 2]b. When CB procedure was compared to conventional smear procedure in terms of preservation of architectural pattern (i.e., presence of three-dimensional cell clusters, cell balls, [Figure 2]a papillae [Figure 2]c, glandular or acinar formations [Figure 1]e, sheets, etc.), it was found that architectural pattern was far more well-preserved in CB specimens (88.88%) than in conventional smears (44.44%) [Figure 1]d and [Figure 1]e.
|Figure 1: (a) Reactive mesothelial cell cluster showing prominent microvilli (Pap, ×400). (b) Calretinin positivity in mesothelial cell (calretinin, ×400). (c) Cytokeratin 5/6 positivity in mesothelial cell (CK5/6, ×400). (d) Cell clock preparation showing prominent signet ring morphology in a case of gastric adenocarcinoma (H and E, ×400). (e) Cell block preparation showing acini formation in the same case (H and E, ×400)|
Click here to view
|Figure 2: (a) Conventional showing a cluster of adenocarcinoma cells with smooth three-dimensional cell ball configuration (Pap, ×400). (b) Conventional smear showing adenocarcinoma cells in papillae in a case of serous adenocarcinoma of ovary (Pap, ×200). (c) Cell block preparation showing neoplastic cells in papillary pattern (H and E, ×400). (d) Immunohistochemistry showing Wilms tumor 1 staining in papillary lining proving serous carcinoma of ovary (Wilms tumor 1, ×400)|
Click here to view
There were total 13 suspicious cases out of 50 patients whose CB preparations were made and subjected to IHC. One case (7.69%) had inadequate material, three cases (23.07%) which were suspicious for malignancy both on conventional smears as well as on CB technique was confirmed malignant by IHC. Three cases (23.07%) which appeared suspicious for malignancy both on CB and conventional smears were confirmed reactive by IHC, and in six cases (46.15%), IHC proved the CB findings [Table 2].
|Table 2: Distribution of patients according to their immunohistochemistry findings|
Click here to view
Among the 50 patients who were examined in 35 (70%) cases, CB findings were consistent with the findings of conventional smears. In 10 cases (20%), the conventional smears were suspicious for malignancy, and malignancy was confirmed by CB technique, whereas in five cases (10%), both smears and CB was suspicious for malignancy and the original nature of the lesion was confirmed by the IHC.
Sensitivity and specificity of CB compared to conventional smear in the detection of malignancy were 88.88% and 86.98%, respectively. Positive predictive value being 88.88% and negative predictive value being 86.95%.
While comparing the conventional technique and CB methods, it was found that CB produced significantly better results (P = 0.0271) while detecting malignant lesions and reducing suspicious results (P = 0.0226).
| > Discussion|| |
The CB technique is one of the oldest methods used for the evaluation of body fluids. Using 10% alcohol-formalin as a fixative increases the cellularity due to less destruction by the fixative, providing better morphological details and improving the sensitivity of the diagnosis. Multiple sections can be obtained by the CB method for the special stains and immunohistochemical studies.
We received 50 samples of body fluids of which pleural fluid was more compared to peritoneal fluid. Maximum samples were in the age group of 61–70 years. Least number of patients were in the age group of 11–20 years. Male:female ratio was 1:1.17. Similar findings were observed by Khan et al.
Most of the malignant peritoneal effusions were due to ovarian malignancy in females. The next common malignancy causing peritoneal effusion in males was due to primaries in the stomach followed by primaries in gallbladder and colon. The most common cause of malignant pleural effusion was lung carcinoma in males. In females, the most common cause of pleural effusion was breast carcinoma. Khan et al. showed carcinoma lung was the most common site of malignant effusion followed by carcinoma ovary and gastrointestinal tract. Similarly, Murphy and Ng described the most common primary lesions were in breast followed by lung and ovary. DiBonito et al. also reported a similar pattern of primaries in malignant pleural and peritoneal effusions.
We used 10% alcohol-formalin in 1:1 proportions as a fixative in formalin CB preparations. Similar fixatives were used in various other studies by Bodele et al., Vellios, and Griffin who used 50% ethanol with formalin in various ratio. Khan et al. used Bouin's fixative. Nathan et al. devised Nathan alcohol-formalin substitute consisting of absolute alcohol and 40% formaldehyde in 9:1 proportion.
The preservation of the morphological features in formalin CB was better and comparable with studies by Khan et al. and Dekker and Bupp. Moreover, using alcohol-formalin method, no additional material was required, and the process is very cheap, cost-effective yet highly diagnostic. The cellular yield was more with CB methods as compared to conventional smear method. The CB concentrated the cellular material into a small area which was useful in screening the material in lesser time. Similar findings were noted in studies by Yang et al., Dekker and Bupp, and Thapar et al.
In our study for denoting cellularity, we considered the recovery of tumor cells in malignant cases. Similar criteria were used in studies by Takagi Khan et al., Nathan et al., Dekker and Bupp., and Chapman and Whalen. The recovery rate of tumor cells by CB method in a study by Khan et al. was 20% greater than that obtained for specimen examined in conventional smears. Bodele et al. diagnosed additional 7% malignancies in CB methods when compared to conventional smear methods. Thapar et al. showed a diagnostic yield of 20% by CB preparations. In this study, additional 22% malignancies were diagnosed by CB preparations.
Reactive mesothelial cell mimicking malignancy in conventional smear could be differentiated in CB by architectural character. Similar findings were obtained by Shivakumarswamy et al. where CB diagnosed malignant lesions in 22% cases in comparison to 8% malignant lesions in conventional smears. We observed pericellular lacunae in more than 50% cases of adenocarcinoma cases considered to be helpful diagnostic feature in this cases. Price et al. also observed pericellular lacunae in the adenocarcinoma cases in CB method. Based on morphologic features alone, the cytologic differentiation of reactive mesothelial cells from adenocarcinoma can be difficult. Various cytologic features are characteristic of, but not specific for, mesothelial cells example, intercellular spaces (windows), commonly seen in cellular aggregates of mesothelial cells [Figure 1]a, also can be identified in 13% of cases of metastatic adenocarcinoma. Usefulness of cytochemical stains such as mucicarmine is limited because they are generally insensitive.
Several investigators have demonstrated that calretinin is a sensitive and specific marker for both benign and malignant mesothelial cells. In this study, the reactive mesothelial cells of all benign cases demonstrated nuclear enhancement with strong membrane positivity giving the cells a poached egg appearance [Figure 1]b, whereas all metastatic adenocarcinomas lacked positive nuclear staining to calretinin. Cytokeratin 5 is also good marker for identification of the reactive mesothelial cells [Figure 1]c. The mesothelial cells show a strong membrane positivity when stained with cytokeratin.
WT1 is a tumor suppressor gene implicated in the development of Wilms tumor and has been demonstrated in a high percentage of ovarian serous carcinomas (89–92%). As most of the cases of peritoneal effusions which we received were from patients whose presence of ovarian primary was diagnosed both clinically, radiographically, and histopathologically, we used WT1 as the marker for detecting malignant deposits in the CB sections. Most of the cases provided strong nuclear positivity when stained with WT1 antibody [Figure 2]d. Similarly, TTF-1 nuclear staining proved useful in lung carcinomas, ER/PR staining helped to distinguish breast primary, whereas CDX2 nuclear stain proved intestinal origin.
To conclude, CB techniques definitively increased detection of malignancy in body cavity effusion when used as an adjunct to conventional smears. Morphological and architectural features are better identified in CB technique thus improving sensitivity. Multiple sections can be taken from the same material for special stain, IHC thus improving specificity. 10% alcohol-formalin used as a fixative is a simple, inexpensive, and comparable alternative for CB preparation.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| > References|| |
Dekker A, Bupp PA. Cytology of serous effusions. An investigation into the usefulness of cell blocks versus smears. Am J Clin Pathol 1978;70:855-60.
Richardson HL, Koss LG, Simon TR. An evaluation of the concomitant use of cytological and histocytological techniques in the recognition of cancer in exfoliated material from various sources. Cancer 1955;8:948-50.
Khan N, Sherwani RK, Afroz N, Kapoor S. The cytodiagnosis of malignant effusions and determination of the primary site. J Cytol 2005;22:107-10.
Wong JW, Pitlik D, Abdul-Karim FW. Cytology of pleural, peritoneal and pericardial fluids in children. A 40-year summary. Acta Cytol 1997;41:467-73.
Grunze H. The comparative diagnostic accuracy, efficiency and specificity of cytologic technics used in the diagnosis of malignant neoplasm in serous effusions of the pleural and pericardial cavities. Acta Cytol 1964;8:150-63.
Murphy WM, Ng AB. Determination of primary site by examination of cancer cells in body fluids. Am J Clin Pathol 1972;58:479-88.
DiBonito L, Falconieri G, Colautti I, Bonifacio D, Dudine S. The positive pleural effusion. A retrospective study of cytopathologic diagnoses with autopsy confirmation. Acta Cytol 1992;36:329-32.
Bodele AK, Parate SN, Wadadekar AA, Bobhate SK, Munshi MM. Diagnostic utility of cell block preparation in reporting of fluid cytology. J Cytol 2003;20:133-5.
Vellios F, Griffin J. Examination of body fluids for tumors cells. Am J Clin Pathol 1954;24:676-81.
Nathan NA, Narayan E, Smith MM, Horn MJ. Cell block cytology. Improved preparation and its efficacy in diagnostic cytology. Am J Clin Pathol 2000;114:599-606.
Yang GC, Wan LS, Papellas J, Waisman J. Compact cell blocks. Use for body fluids, fine needle aspirations and endometrial brush biopsies. Acta Cytol 1998;42:703-6.
Thapar M, Mishra RK, Sharma A, Goyal V, Goyal V. Critical analysis of cell block versus smear examination in effusions. J Cytol 2009;26:60-4.
] [Full text]
Takagi F. Studies on tumor cells in serous effusion. Am J Clin Pathol 1954;24:663-75.
Chapman CB, Whalen EJ. The examination of serous fluids by the cell-block technic. N
Engl J Med 1947;237:215-20.
Shivakumarswamy U, Arakeri SU, Karigowdar MH, Yelikar B. Diagnostic utility of the cell block method versus the conventional smear study in pleural fluid cytology. J Cytol 2012;29:11-5.
] [Full text]
Price BA, Ehya H, Lee JH. Significance of pericellular lacunae in cell blocks of effusions. Acta Cytol 1992;36:333-7.
Yu GH, Sack MJ, Baloch ZW, DeFrias DV, Gupta PK. Occurrence of intercellular spaces (windows) in metastatic adenocarcinoma in serous fluids: A cytomorphologic, histochemical, and ultrastructural study. Diagn Cytopathol 1999;20:115-9.
Cibas ES, Corson JM, Pinkus GS. The distinction of adenocarcinoma from malignant mesothelioma in cell blocks of effusions: The role of routine mucin histochemistry and immunohistochemical assessment of carcinoembryonic antigen, keratin proteins, epithelial membrane antigen, and milk fat globule-derived antigen. Hum Pathol 1987;18:67-74.
Chhieng DC, Yee H, Schaefer D, Cangiarella JF, Jagirdar J, Chiriboga LA, et al
. Calretinin staining pattern aids in the differentiation of mesothelioma from adenocarcinoma in serous effusions. Cancer 2000;90:194-200.
Hammar SP, Bockus DE, Remington FL, Rohrbach KA. Mucin-positive epithelial mesotheliomas: A histochemical, immunohistochemical, and ultrastructural comparison with mucin-producing pulmonary adenocarcinomas. Ultrastruct Pathol 1996;20:293-325.
[Figure 1], [Figure 2]
[Table 1], [Table 2]