ORIGINAL ARTICLES. Materials and Methods

Transcription

1 ORIGINAL ARTICLES Role of Fine-Needle Aspirate Immunophenotyping by Flow Cytometry in Rapid Diagnosis of Lymphoproliferative Disorders Ritu Gupta, M.D., 1 * Shano Naseem, M.D., 1 Rajesh Kashyap, M.D., 1 and Lily Paul, M.D. 2 Immunophenotyping is an essential component in the diagnostic work-up of lymphoproliferative disorders (LPD). As compared to immunohistochemistry, flow cytometric immunophenotyping (FCMI) is rapid, quantitative and a more objective technique. This study was designed to evaluate the utility of FCMI on fine needle aspirates (FNA) in rapid diagnosis of LPD in routine clinical practice. FNA from 31 consecutive cases clinically suggestive of LPD were subjected to FCMI. Representative material for FCMI was obtained in 28 (90%) cases and a definite diagnosis established in 27 cases. Histopathogical correlation was available in 22 cases and concordance with FCMI results was observed in 19 (86.4%) cases. FCMI analysis was inconclusive in 4 cases. The results of FCMI were available the same day and were crucial for therapeutic purpose in 3 patients with superior vena cava syndrome. FCMI combined with cytological examination of aspirate smears permits rapid diagnosis with high level of accuracy resulting in efficient treatment planning for critically ill patients and those from far-off rural areas. Diagn. Cytopathol. 2007;35: ' 2007 Wiley-Liss, Inc. Key Words: FNA; immunophenotyping; flow cytometry; lymphoproliferative disorder 1 Department of Hematology, Sanjay Gandhi Postgraduate Institute of Medical Sciences (SGPGIMS), Lucknow, Uttar Pradesh Department of Pathology, SGPGIMS, Lucknow, Uttar Pradesh *Correspondence to: Dr. Ritu Gupta, M.D., B-26, Satyawati Colony, New Delhi , India. rituajay97@hotmail.com Received 27 January 2006; Accepted 6 April 2007 DOI /dc Published online in Wiley InterScience ( The fine needle aspiration (FNA) technique has been used in the diagnosis of lymph node disease since the early 20th century. The method has since been used primarily to differentiate between benign and malignant lymphadenopathy. If malignant, a lymph node biopsy is usually the investigation of choice, which allows study of the lymph node architecture and use of immunohistochemical stains as required. Excision biopsy, on the other hand, is traumatic, technically difficult if the lesion is deep seated and causes delay in diagnosis on account of histopathological processing. FNA is an economical and convenient alternative to open biopsy of lymph node, whose diagnostic accuracy can be further improved manifold when used in collaboration with other special techniques like flow cytometry, cytochemistry, bacterial culture, immunocytochemistry, electron microscopy, and in situ hybridization. We initiated this study to evaluate the diagnostic utility of flow cytometric immunophenotyping (FCMI) for rapid diagnosis of lymphoproliferative disorders (LPD) in routine clinical practice. Materials and Methods Fine needle aspirates (FNAs) from 31 consecutive cases with suspected LPD at our center from July 2003 to September 2005, were subjected to FCMI. Aspirations were performed with a 23-gauge needle. Syringing and suction was avoided to prevent contamination with peripheral blood. A minimum of two slides were prepared and airdried. One of these slides was stained with leishman and the second used for cytochemistry, if required. An average of three passes was made to provide adequate number of cells for flow cytometric evaluation. Material from these passes was suspended in phosphate buffered saline (PBS, ph 7.4) and immediately processed for FCMI. All the samples were processed within 30 min and acquired within 2 hr of collection to prevent degeneration of cells. The total leukocyte count of the aspirate cell suspension ' 2007 WILEY-LISS, INC. Diagnostic Cytopathology, Vol 35, No 7 381

2 GUPTA ET AL. Table I. Clinicopathological Features of Cases Subjected to Flow Cytometric Analysis on Fine-Needle Aspirate Specimens S No. FCMI utility Age/ Sex Site of aspirate PB findings BM findings Diagnosis Cytological FCM Histopathology 1 C 34/M Cervical 5% atypical cells >90% immature cells NHL NK lymphoma NHL-small 2 C 36/M Cervical WNL 50% immature cells NHL/AL B-lineage ALL Leukemia/ 3 C 5/M Cervical WNL Myclofibrosis NHL/AL BL NHL-high grade 4 C 64/M Thyroid nodule WNL WNL Plasmacytoma Plasmacytoma? Plasmacytoma 5 C 59/M Cervical WNL WNL NHL B-NHL High grade NHL 6 C 20/M Cervical WNL WNL NHL/AL B-NHL High grade NHL 7 C 66/M Axillary LE reaction Few atypical cells NHL B-NHL? Lymphoma 8 C 36/F Submandibular Pancytopenia 40% atypical cells NHL/AL T-ALL ALL 9 C 60/M RL preauricular WNL ND NHL MCL NHL-small 10 C 8/M Axillary LE reaction Few atypical cells NHL/AL T-NHL? Lymphoma 11 C 57/M Cervical LE reaction NHL infiltration NHL B-NHL NHL 12 C 8/M Mediastinal Few atypical cells NHL/AL NHL/AL T-ALL ND 13 C 47/M Cervical WNL WNL NHL B-NHL B-NHL 14 C 20/F Mediastinal WNL NHL/AL NHL/AL T-ALL ND 15 C 50/M Cervical WNL WNL NHL B-NHL NHL 16 C 47/M Cervical WNL LPD LPD B-NHL ND 17 C 5/M Cervical 4% atypical cells AL NHL BL ND 18 C 7/M Cervical WNL 7% blasts NHL T-LL ND 19 C 70/M Cervical WNL WNL NHL B-NHL NHL-DLBL 20 C 12/M Cervical WNL WNL NHL T-LL ND 21 C 25/M Chest swelling CML-CP CMPD Atypical cells CML-BC ND 22 C 43/M Cervical Few atypical cells CLPD NHL MCL NHL 23 C 63/M Inguinal WNL LPD NHL B-NHL (FL) NHL 24 C 54/M Cervical WNL WNL NHL MCL NHL 25 NC 45/F Cervical WNL WNL Metastasis Metastasis Metastasis 26 C 46/M Axillary WNL WNL Reactive LNitis Reactive LNitis ND 27 C 6/F Submandibular Thrombocytopenia WNL Reactive LNitis Reactive LNitis Reactive LNitis 28 NC 49/M Cervical WNL WNL Degenerated cells Degenerated Hodgkin s disease 29 NC 54/M Cervical LE reaction Myclofibrosis Degenerated cells Degenerated Hodgkin s disease 30 NC 34/M Cervical WNL WNL Degenerated cells Degenerated Castleman s 31 C 17/M Cervical LE reaction WNL Reactive LNitis Reactive LNitis ND PB, peripheral blood; BM, bone marrow; FCM, flow cytometry; WNL, within normal limits; LE, leukoerythroblastic; AL, acute leukemia; LPD, lymphoproliferative disorder; FL, follicular lymphoma; MCL, mantle cell lymphoma; LL, lymphoblastic lymphoma; BL, Burkitt s lymphoma; LNitis, Lymphadenitis; ND, not done; C, Contributory to diagnosis; NC, Non-contributory to diagnosis;?, probable diagnosis. was determined manually using modified neubauer chamber. The cytomorphological impression of the aspirate smears guided the panel of monoclonal antibodies (MoAbs) to be used in an individual case. Immunostaining was done with MoAbs preconjugated with fluorescein isothiocyanate (FITC) and phycoerythrin (PE). CD45 (Peridinin chlorophyll) or CD19 (Peridinin chlorophyll) was added to all the tubes, wherever required, to gate the desired cell population. A minimum of 50,000 cells were aliquoted in each tube and incubated with pretitrated volumes of MoAbs for 15 min, followed by incubation with 1 ml of ammonium chloride based lysis solution for 5 min. Cells were then washed with PBS and suspended in 1% paraformaldehyde. A minimum of 2,000 events were acquired in each tube and analyzed on Facs Calibur using Cell-Quest software (Becton Dickinson, BD Biosciences). The final diagnosis was based on clinical, cytomorphological, and FCMI findings and compared with tissue biopsy results and clinical follow up to evaluate the contribution of FCMI. The role of FCMI was considered contributory (C) when the results facilitated rapid institution of therapy essential to save patient s life or provided definitive diagnosis or disease subcategorization. Results In total, there were 31 patients with lymphadenopathy and clinical suspicion of LPD, who underwent FNA of lymph node followed by FCMI. Seven patients belonged to pediatric age group, remaining being adults, age ranging from 5 66 yr. All except 2 had superficial lymphadenopathy and the cervical region was the most common site involved (Table I). In 2 patients with mediastianal lymphadenopathy, FNA was performed under ultrasound guidance. Representative cytological material was obtained in 28 cases, remaining 3 showed only degenerated cells. Initial Cytological Evaluation Cytologically, a diagnosis of lymphoma/leukemia was suspected in 24 cases (77%), reactive lymphadenitis in 3 cases (9.6%), and metastatic carcinima in one case (1%). In 3 cases (9.6%), FNA revealed only degenerated cells (Table I). Of 24 cases with LPD, 17 were labeled as non-hodgkin s lymphoma (NHL) and 1 as plasmacytoma (extramedullary). There were 3 cases of leukemic lymphadenopathy where peripheral blood smear did not reveal immature 382 Diagnostic Cytopathology, Vol 35, No 7

3 ROLE OF FNA IMMUNOPHENOTYPING IN LPD Table II. Flow Cytometric Analysis of Lymph Node Aspirates S. No. Positive markers Negative markers Impression 1 CD 2, 7, 8, 38, 56* CD 5*, 10, 20, FMC-7, 19, 23, 20, 5, 4, NK cell leukemia/lymphoma TdT, 3*, 34, 13, 33, ab, gd, CD3 cytoplasmic, j, k 2 CD 10*, 2, 19*, 13, 34* CD 5, 20*, 22*, 23, 7, 33, 117, j, k B-Lineage ALL 3 CD 19, 20, 22, j* CD10, 2, 34, 3, 33, k* Burkitt s lymphoma 4 CD 38, j* CD 20*, k* Plasmacytoma 5 CD19, 20*, weak expression of 23 CD 5, 10, j, k B-NHL 6 CD 19, 20*, j* CD 3, 5, 10, 23, TdT, k* B-NHL 7 CD 19, 20*, 23, 38, CD 25, j* CD 5, 10, 103, k* B-NHL 8 CD 7, 13, 56* CD10, 19, 5*, 23, 2*, 33, 22, 34, 4, 8, 16, 117, 3* NK/T cell lymphoma 9 CD 5*, 19, 20*, 38 CD 10, FMC-7, 23* Mantle cell lymphoma 10 CD 2, 7, 3*, 45 CD10, 19, 13, 33, 22, 34, 117, 5*, 4*, 8*, TdT T-NHL 11 CD19, 20*, 38, j* CD5, 23, 10, 25, 103, FMC-7, k* B-NHL 12 CD 2, 7, 5*, 8*, 34* CD10, 19, 13, 33, 22, 3*, 20, 4*, 117, j, k T-ALL 13 CD 10*, 19*, 20 CD 5, 23, 4, 8, j, k* B-NHL 14 CD 3*, 5, 7, 10*, 34* CD 2* T-ALL 15 CD 19, 20*, j* CD 5, 10, 38, 23, 4, 8, 7, 3, 2, k* B-NHL 16 CD 19, 22, 38, 20*, 23, j* CD5, 10, 3, TdT, k B-NHL 17 CD10, 19, 13, 22, 20, k* CD 2, 7, 4, 8, 5, 3, 34, 33, 117, j Burkitt s lymphoma 18 CD2, 7, 4, 8, 5, 3cytoplasmic*, CD10, 19, 13, 33, 22, 20, 3surface* T lymphoblastic lymphoma 34*, TdT 19 CD19, 20*, 22, j* CD 10, 5, 3, 23, 34, TdT, k* B-NHL 20 CD5*, 4*, 8*, 10, TdT* CD 19, 22, 34, 20, 33, 3surface*, j, k T-lymphoblastic lymphoma 21 CD13, 33, 117, 41*, HLA-DR CD10, 19, 2, 7, 22, 34, 5, 3, 20, 4, 8 CML-BC 22 CD5*, 20*, 19, 22*, 25, k* CD23*, 3, 10, j Mantle cell lymphoma 23 CD 19, 20*, 22, 10*, j* CD5, 23, 3, 38, k* B-NHL (FCC) 24 CD5*, 10, 19, 20*, j* CD3, 22, 23*, 38, k* Mantle cell lymphoma j, kappa light chains; k, Lambda light chains; *, abnormal expression especially useful for diagnosis. cells and the bone marrow aspirate smears were nonrepresentative; 2 of these had deep seated, i.e. mediastianal lymphadenopathy. All 3 cases with benign lymphadenopathy were reactive in nature with smears showing polymorphous population of cells lymphoid cells in various stages of maturation, eosinophils, plasma cells, histiocytes, and tingible body macrophages. Flow Cytometric Immunophenotyping The cell count of FNAs ranged from cells/l to cells/l with a median of cells/l. On the basis of initial cytological assessment, a panel of markers for immunophenotyping was decided. Table II shows main flow cytometric findings for all the 24 cases with LPD. Viability of cells was assessed based on scattergrams on flow cytometer. All the samples contained viable cells when the entire process of sample preparation to acquisition on flow cytometer was carried out within 2 hr of sample collection. Following FCMI analysis, the diagnosis was established in 27/31 (87%) cases, further subcategorization and lineage assignment for therapeutic purpose could also be done in cases of lymphoma and leukemia. Three cases that showed degenerated cells on cytology were inconclusive by FCMI and a case, in which possibility of metastasis was considered, was positive for CD30 and negative for all other hemato-lymphoid markers. This case was confirmed as metastasis on histopathological examination (cytokeratin positive). All the FCMI results were reported the same day. Histopathological correlation Of the 31 cases, histopathological correlation was available in 22 cases. Histopathological diagnosis was concordant with FCMI results in 19 cases (86.4%), which included 15 cases of NHL, one case each of B-lineage ALL, plasmacytoma, metastasis, and reactive lymphadenitis. The 3 cases were FNA showed only degenerated cells, 2 were diagnosed as Hodgkin s disease, and 1 as Castleman s disease on histopathological examination. Histopathological correlation was not available for 4 cases of NHL, 2 cases of T-lineage ALL with mediastianal lymphadenopathy, 1 case of extramedullary blast crisis of chronic myeloid leukemia, and 2 cases with reactive lymphadenitis (Table I). The minimum time taken for reporting of histopathology results in all the cases was 4 days. Overall, FCMI was contributory to diagnosis in 27 (87%) cases and inconclusive or noncontributory in 4 (13%) cases. Discussion FNA offers a simple and safe technique for obtaining diagnostic material with excellent preservation of cytomorphology. When used in conjunction with FCMI, FNA has shown to have high diagnostic accuracy in LPD. 1 FCMI can demonstrate light chain restriction and aberrant phenotypic expression useful for B and T-lineage lymphoid neoplasm, respectively. In low grade B-NHL such as small lymphocytic lymphoma, follicular center cell Diagnostic Cytopathology, Vol 35, No 7 383

4 GUPTA ET AL. lymphoma, and mantle cell lymphoma, a combination of CD19, CD5, CD10, CD23, bcl-2, FMC-7, and surface immunoglobulins allows phenotypic definition of the neoplastic population into their respective category. 2,3 A correlation of cytological features of peripheral blood smear and bone marrow biopsy and, immunological studies is required to differentiate high grade NHL from acute leukemia as both cytomorphology and FCMI in isolation are inadequate. 4,5 In our study, differentiation of acute leukemia (4 cases) from high grade NHL (3 cases) was established by combining findings of peripheral blood smear, bone marrow biopsy, and FCMI of lymph node aspirates. Expression of markers of cell immaturity such as CD34, TdT, and monoclonal light chains were of particular help in differentiating these two entities (Table II). Definitive diagnosis of T-lymphoproliferative disorders by immunophenotyping using flow cytometry is often difficult, as there are no definite phenotypic markers of monoclonality for T cells, unlike B-NHL where light chain restriction is highly suggestive of a neoplastic process. The presence of aberrant phenotype, suggested by loss of 1 T-cell marker or asynchronous expression, is helpful in diagnosing T-cell lymphoproliferative disorder. 6 This feature of T-cell lymphoproliferative disorders is well characterized in our study (Table II). The excision biopsy is useful in assessing architectural patterns which cannot be evaluated on aspirate smears. Current classification of lymphoproliferative disorders using phenotypic and genotypic studies, as well as morphology, renders architecture less critical. 7 Nodularity in follicular lymphoma and grading of lymphoid neoplasm, assessed on histopathological examination, however, are of prognostic importance. 8 Of these, grading of lymphoid neoplasm can be done by flow cytometric analysis of cell cycle and S-phase fraction. 9 Immunohistochemistry (IHC) on tissue biopsy provides pertinent information for diagnosis and prognostication. As compared to FCMI, IHC is associated with altered immune reactivity because of tissue fixation, lower degree of sensitivity, and specificity of antibodies used for fixed tissue, inability to differentiate surface expression of antigen from cytoplasmic expression, unavailability of multiparameter evaluation of individual cell, and time delay due to processing and staining procedures involved. FCMI with FNA results were available the same day and thus considerably reduced the time required for diagnosis. This was found to be of particular importance in 3 patients with superior vena cava syndrome and respiratory distress where an early treatment was instituted after FCMI results became available. For patients referred to our center from far-off rural areas, adequate treatment planning could be achieved in single visit to the hospital. Sampling error is a potential problem in FNA of lymphoid proliferations. A reactive population may be all that is identified in aspiration smears from Hodgkin s disease, T-cell rich B-cell lymphoma, T-cell lymphoma associated with a large number of reactive cells, and cases of partial involvement by lymphoma/leukemia. 10,11 Reactive hyperplasia on FCMI was seen in 3 patients in our study. In 2 patients, the lymph node regressed and in one patient in whom the lymph node did not regress on follow-up, an excision biopsy was performed which revealed benign reactive proliferation. Contrary to the usual findings of reactive lymphadenitis reported in literature, the aspirate smears in both of our cases of Hodgkin s lymphoma showed degenerated cells. Similar findings were obtained in a case of Castleman disease. This phenomenon could be attributed to fibrosis associated with these conditions. The tissue biopsy should, therefore, be carried out, if the clinical features are suggestive of lymphoma, regardless of a reactive or inconclusive pattern on FCMI. Our study confirms the findings of previous studies, showing a high level of diagnostic accuracy (85%) when using FCMI in conjunction to aspirate cytology. 5,12 14 If specimen associated limitations are recognized and the antibody panel selection is based upon morphological assessment and the ability of the antibody to separate normal from abnormal cell population, the diagnostic accuracy is consistent with that of histopathology. To the best of our knowledge, this is the first series from India where FCMI on FNAs has been used for diagnosis and management of lymphoproliferative disorders. The rapid availability of test results with high level of diagnostic accuracy was instrumental in efficient treatment planning for critically ill patients and patients from far-off rural areas. References 1. Ravinksy E, Morales C. Cytodiagnosis of lymphoid proliferations by fine needle aspiration biopsy. Acta Cytol 1999;43: Young NA, Ehya H, Smith MR. Utilization of fine needle aspiration cytology and flow cytometry in the diagnosis and sub classification of primary and recurrent lymphoma. Cancer 1998;84: Verstovek G, Chakarborty S, Ramzy I, Jorgensen JL. Large B-cell lymphoma. Diagn Cytopathol 2002;27: Perikala VK, Karimi M, Monabati A, Sadheghipour AR, Tavangar SM, Moosavi A, Nourani H, Haghkshanas M, Bedayat GR. Cytology of leukemic lymphadenopathy. Acta Cytol 2002;46: Dey P, Amir T, Al Jassar A, Al Shemmari S, Jogai S, Bhat MG, Al Quallaf A, AlShammari Z. Combined applications of fine needle aspiration cytology and Flow cytometric immunphenotyping for diagnosis and classification of non Hodgkin Lymphoma. Cytojournal 2006;3: Zander DS, Iturraspe JA, Everett ET, Massey JK, Braylan RC. Flow cytometry: In vitro assessment of its potential application for the diagnosis and classification of lymphoid processes in cytological prepration from fine needle aspirate. Am J Clin Pathol 1994;101: Harris NL, Jaffe ES, Diebold J, Flandrin G, Muller-Hermelink HK, Vardiman J. The World Health Organization classification 384 Diagnostic Cytopathology, Vol 35, No 7

5 ROLE OF FNA IMMUNOPHENOTYPING IN LPD of neoplastic diseases of the hematopoietic and lymphoid tissues: Report of the clinical advisory committee meeting. Airlie house, Virginia, November Histopathology 2000;36: Ravinsky E, Morales C. Diagnosis of lymphoma by image- guided needle biopsies. Acta Cytol 2005;49: Nguyen DT, Diamond LW. Approach to flow cytometry. In: Nguyen DT, Diamond LW, editors. Diagnostic hematology: A pattern approach. Oxford: Butterworth-Heinemann; p Daskalopoulou D, Harhalakis N, Markidou SG. Fine needle aspiration cytology of NHL: A morphologic and immunophenotypic study. Acta Cytol 1995;39: Das DK, Gupta SK, Datta BN, Sharma SC. Fine needle aspiration cytodiagnosis of NHL and its subtyping under working formulation of 175 cases. Diagn Cytopathol 1991;1: Moriarty AT, Wiesema L, Snyder W, Kotylo PK, McCloskey DW. Immunophenotyping of cytologic specimen by flow cytometry. Diagn Cytopathol 1993;9: Simsir A, Fetsch P, Stetler-Stevenson M, Abati A. Immunophenotypic analysis of Non-Hodgkin s lymphoma in cytologic specimens. Diagn Cytopathol 1999;20: Zardari IM, Jain S, Bennett G. Flow cytometric algorithm on fine needle aspirates for the clinical workup of patients with lymphadenopathy. Diagn Cytopathol 1998;19: Diagnostic Cytopathology, Vol 35, No 7 385