Submit Your Article CMED MEACR meeting
Home Print this page Email this page Users Online: 626
Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2015  |  Volume : 4  |  Issue : 6  |  Page : 707-712

Comparing flow cytometry immunophenotypic and immunohistochemical analyses in diagnosis and prognosis of chronic lymphoproliferative disorders: Experience from a Tertiary Care Center


Department of Pathology and Molecular Medicine, Army Hospital Research and Referral, Delhi Cantonment, New Delhi, India

Date of Web Publication9-Nov-2015

Correspondence Address:
Tathagata Chatterjee
Department of Pathology and Molecular Medicine, Army Hospital Research and Referral, Delhi Cantonment, New Delhi - 110 010
India
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/2278-0513.165524

Rights and Permissions
  Abstract 

Background: The latest World Health Organization classification incorporates extensive description of immunophenotype of the neoplastic cells while describing chronic lymphoproliferative disorders (CLPDs). The present study was undertaken with an aim to identify and compare the roles of flow cytometry (FCM) and immunohistochemistry (IHC) as modalities of immunophenotyping in the diagnosis of CLPDs. Materials and Methods: Thirty untreated cases of CLPDs were enrolled in the study. Twenty eight cases of B-CLPD were divided into two groups - chronic lymphocytic leukemia (CLL) (21 patients) and non-CLL (7 patients). Peripheral blood/bone marrow aspirate samples were analysed by FCM using various panels of monoclonal antibodies. Immunohistochemical analysis of bone marrow biopsies obtained from these patients was also performed. Results: Panel A of monoclonal antibodies comprising CD5, CD23, CD22, surface membrane immunoglobulin (SmIg), FMC7 and Panel B comprising CD5, CD23, CD22, SmIg, FMC7, CD79b were useful (P < 0.01 and <0.001 respectively) while Panel C comprising CD5, CD23, SmIg, FMC7 and CD79b was not found to be useful in distinguishing CLL from non-CLL (P > 0.05) The concordance rate between FCM and IHC ranged from 80% to 100% for all comparable immunological markers. In all cases of CLPDs, we propose a screening panel comprising 9 markers including CD19, CD5, CD23, FMC7, CD10, CD20, CD3, kappa and lambda, which are important for specifying the lineage (B or T), to differentiate CLL from non-CLL group and for deciding the secondary panel. Conclusion: Scoring system using CD5, CD23, CD22, FMC7, CD79b, and SmIg is useful in differentiating CLL from non-CLL cases. Concordance rate of FCM and IHC in CLPDs is 93.3%. Using a panel comprising CD19, CD5, CD23, FMC7, CD10, CD20, CD3, kappa and lambda, a diagnosis of CLL, mantle cell, and follicular lymphoma, the three most common CLPDs can be made. Secondary panels for diagnosis of hairy cell leukemia and T-cell CLPD should be utilized.

Keywords: Chronic lymphocytic leukemia, chronic lymphoproliferative disorders, flow cytometry, immunohistochemistry


How to cite this article:
Dewan K, Mann N, Chatterjee T. Comparing flow cytometry immunophenotypic and immunohistochemical analyses in diagnosis and prognosis of chronic lymphoproliferative disorders: Experience from a Tertiary Care Center. Clin Cancer Investig J 2015;4:707-12

How to cite this URL:
Dewan K, Mann N, Chatterjee T. Comparing flow cytometry immunophenotypic and immunohistochemical analyses in diagnosis and prognosis of chronic lymphoproliferative disorders: Experience from a Tertiary Care Center. Clin Cancer Investig J [serial online] 2015 [cited 2019 Oct 21];4:707-12. Available from: http://www.ccij-online.org/text.asp?2015/4/6/707/165524


  Introduction Top


Chronic lymphoproliferative disorders (CLPDs) are a heterogeneous group of leukemias/lymphomas, which are defined by the proliferation of mature B and rarely T/NK lymphoid cells in the peripheral blood, bone marrow and/or lymph nodes/spleen and other lymphoid tissue. The present World Health Organization (WHO) classification incorporates extensive descriptions of the immunophenotype of the neoplastic cells while describing CLPDs.[1] Immunophenotyping can be done on fresh peripheral blood/bone marrow aspirate/lymph node aspirate samples using flow cytometry (FCM) or on formalin-fixed-paraffin-embedded tissue using immunohistochemistry (IHC). Each of the methods available comes with its advantages and disadvantages and contribution towards making the final hematological diagnosis. FCM provides a rapid, selective identification of normal and abnormal cell population excluding the dead cells and allows the detection of weakly expressed antigens. IHC on bone marrow biopsy sections, on the other hand, offers a visual impression of architectural relationships between different population of cells and to the stromal component.[2]

The present study was undertaken with an aim to identify and compare the roles of FCM and IHC in the diagnosis of CLPD.


  Materials and Methods Top


Thirty eight consecutive fresh untreated cases of CLPDs presenting in a tertiary care hospital were enrolled in this study conducted over a 2 years period from 2011 to 2013. Previously treated cases of CLPDs, cases with total nucleated cell count in peripheral blood/bone marrow that was insufficient for FCM or cases with inadequate length of bone marrow biopsies (<1.5 cm) and cases that did not have infiltration on bone marrow biopsy were excluded from the study. After excluding 8 such cases, 30 CLPD cases were finally included in the study.

The patient age ranged from 32 to 81 years (mean 62.1 years) and a male-female ratio of 2.7:1 was obtained. Of the 30 cases included in the study, 28 cases were B-CLPD and 2 cases were T-CLPD. The 28 B-CLPD cases were divided into two groups - chronic lymphocytic leukemia (CLL) (21 patients) and non-CLL (7 patients). The non-CLL group included three cases of hairy cell leukemia (HCL), two cases of follicular lymphoma (FL) and one case of mantle cell lymphoma (MCL) and splenic marginal zone lymphoma (SMZL). In the T-CLPD, one case of T-large granular lymphoma (T-LGL) and T-prolymphocytic lymphoma (T-PLL) were included.

Two milliliters of peripheral blood and bone marrow aspirate was collected separately in ethylene diamine tetraacetic acid tubes and was processed for FCM within 24 h. Fresh peripheral blood and bone marrow aspirate smears were prepared at the same time of collection and stained with Leishman stain for evaluation.

The FC-500 flow cytometer with CXP Software (Beckman Coulter) was used for FCM. Before acquisition os samples, calibration and fluorochrome compensation of FCM were performed according to manufacturer's instructions. Based on the cell count of the sample, it was diluted with isotone to achieve a total nucleated cell count between 5 and 10 lakh/μl. Monoclonal antibodies CD19, CD5, CD23, CD22, FMC7, kappa, lambda, CD11c, CD38, CD10, CD79b, CD25, CD3, CD4, and CD8 were used. Expression of a particular antigen by at least 20% of cells was required for considering it positive for a specific marker. Intensity of staining was measured as weak, moderate or bright depending on mean fluorescence intensity scale.

The nature of reactivity of CLL and non-CLL patients to Panel A of monoclonal antibodies which was comprising CD5, CD23, CD22, surface membrane immunoglobulin (SmIg), and FMC7, Panel B which was comprising CD5, CD23, CD22, SmIg, FMC7, and CD79b, and Panel C which was including CD5, CD23, SmIg, FMC7, and CD79b was observed.

Freshly obtained bone marrow biopsy specimens were left for fixation overnight (20–24 h) and then washed in distilled water for 30 min. Specimens were then decalcified in Gooding and Stewart's decalcification fluid (10% formic acid and 5% formaldehyde) for about 6 h before being processed and embedded in paraffin wax. Sections were stained routinely with hematoxylin and eosin, Perl's stain (for iron) and silver stain (for reticulin fibres). The infiltration pattern of biopsy was noted in each case.

For IHC, antigen retrieval was done using heat induced epitope retrieval in a pressure cooker and Streptavidin-biotin method with horse-radish peroxidase enzyme was used as detection kit (Dako, Denmark). The panel of antibodies used in bone marrow biopsy sections included CD19, CD5, CD23, CD22, CD10, kappa, lambda, CD11c, CD79b, CD25, CD3, CD4, CD8, and cyclin D1.

A Microsoft excel sheet was generated and SPSS Software version 11.5 (SPSS, Chicago, IL, USA) used to analyze the data collected. P < 0.05 was regarded as statistically significant.


  Results Top


Flow cytometric analysis

The mean total leukocyte count in peripheral blood was 72,800 cells/μl (range: 5100 cells/μl to 292,000 cells/μl). Percentage of gated lymphocytes in the bone marrow aspirate samples ranged from 10% to 90% (mean 71%) in all patients, 40–98% (mean 75%) in CLL patients, and 10–87% (mean 62%) in non-CLL patients.

Of the 21 cases of CLL, CD5 positivity was seen in 90.5% (19 of 21 cases), CD23 positivity was seen in 95.2% (20 of 21 cases), CD22 positivity was seen in 38% (8 of 21 cases), weak SmIg positivity in 100% cases, FMC7 positivity in 33.3% (7 of 21 cases), and CD79b positivity seen in 14.3% (3 of 21 cases) [Figure 1]. Of the 7 cases of non-CLL, CD5 positivity was seen in 28.5% (2 of 7 cases), CD23 positivity in 57.1% (4 of 7 cases), CD22 positivity in 42.8% (3 of 7 cases), weak SmIg positivity in 71.4% (5 of 7 cases), FMC7 positivity in 85.7% (6 of 7 cases), and CD79b positivity was seen in 71.4% (5 of 7 cases) [Table 1].
Figure 1: Bone marrow histopathology, immunohistochemistry and flow cytometric analysis in a case of chronic lymphocytic leukemia (a) nodular infiltration of bone marrow biopsy by chronic lymphocytic leukemia/small lymphocytic lymphoma, (b) CD5 and CD23 was positive in cells infiltrating the marrow, (c) 84.4% of cells in the case showed CD19 and CD23 coexpression, (d) 82.5% of cells showed CD5 and CD23 coexpression, (e) kappa and lambda plot showing kappa restriction, (f) CD3 and CD4 negativity by these cells, (g) expression of CD19 and CD23 by these cells, (h) the cells were negative for FMC7

Click here to view
Table 1: Reactivity of various monoclonal antibodies on FCM in CLL and non-CLL patients

Click here to view


After assigning a score of 1 for positive reactivity to CD5, CD23, weak SmIg reactivity and nonreactivity of FMC7, CD79b, and CD22, total score was calculated as proposed by Matutes et al. and Moreau et al.[3],[4] Using Panel A, a score of ≥4 was obtained in 18 of 21 CLL cases (85.7%) and a score of <4 was obtained in 85.7% (6 of 7) non-CLL cases and the results were statistically significant (P < 0.01). Using Panel B, a score of ≥4 was obtained in 95.2% (20 of 21) CLL cases and a score of <4 was obtained in 85.7% (6 of 7) non-CLL cases and the results were statistically significant (P < 0.001). Using Panel C, a score of ≥3 was obtained in 85.7% (18 of 21) CLL cases but a score of <3 in non-CLL cases was obtained in only 57.1% (4 of 7 cases) and the results were statistically insignificant (P > 0.05) [Table 2]. The results indicate that both Panels A and B are useful in distinguishing CLL from non-CLL, but a lower P value for Panel B suggests its superiority over Panel A.
Table 2: Comparison of Panels A, B and C in CLL and non-CLL

Click here to view


Histopathological and immunohistochemical analysis

In CLL, varied patterns of bone marrow infiltrations were observed including a diffuse pattern of infiltration 4/21 cases, nodular pattern in 4/21 cases, interstitial in 5/21 and mixed pattern in 8/21 cases. In HCL, 2/3 cases had diffuse and 1/3 cases had interstitial pattern of infiltration; both cases of FL had nodular pattern; single case of SMZL had interstitial pattern; MCL case had nodular pattern of infiltration. Among the T-CLPDs, single case of T-PLL and T-LGL showed diffuse and interstitial pattern, respectively.

One of the cases of bone marrow biopsy showed neoplastic cells with typical fried egg appearance on hematoxylin and eosin, chicken wire pattern of vasculature on reticulin staining, TRAP positivity of neoplastic cells on cytochemistry and positivity of neoplastic cells for CD11c, CD103, and DBA44 but negative staining for CD25 on both IHC and FCM. The patient was diagnosed as variant HCL (HCLv). A single case of SMZL was also included in the present study and it was not possible to differentiate SMZL from HCLv by morphology and routinely available markers in FCM and IHC. The 77-year-old lady diagnosed as SMZL showed a good response to therapy after being undiagnosed for a long time her repeated bone marrow aspiration showed a complete morphological remission and her spleen size returned to normal on follow-up.

Corelation between flow cytometry and immunohistochemistry

CD23 expression was noted in 20/21 CLL cases on FCM and 18/21 cases on IHC. Concordance between IHC and FCM was seen in 90% of the CLL cases for CD23 marker. Concordance between FCM and IHC for CD5 marker was noted in 100% of CLL cases and 96.6% (29/30 cases) of CLPD cases. CD20 disconcordance was found in only one case of CLPD by IHC and FCM. CD10 positivity was noted in one case of FL but was negative in the other case. Although CD10 negative, this case on morphology and complete immunophenotyping (BCL-2, CD20, and CD79b positivity) had a marrow infiltrate of FL. Lymph node biopsy in this case showed characteristic morphology of FL and IHC further confirmed the diagnosis of FL. Due to limited number of FL cases, significance of results about CD10 expression cannot be evaluated.

CD11c which is considered as a marker of HCL was found to be expressed in 4 out of 21 (19%) CLL but the intensity of CD11c expression was not as bright as was seen in HCL. Similarly, CD25 marker of HCL was found to be expressed in other types of B-CLPDs including nearly 20% of CLL. However, the expression of CD25 was brighter in HCL than in other CLPD cases.

CD19 was positive in all B-CLPDs and negative in both T-CLPDs by both FCM and IHC. CD20 was positive in 27 of 28 B-CLPD cases and negative in both T-CLPD cases by FCM but all B-CLPD cases were positive and both T-CLPD cases were negative by IHC. The concordance rates of various immunological markers on FCM and IHC is listed in [Table 3] and the concordance rate was ≥ 80% for all markers. The overall concordance between FCM and IHC was 93.3%.
Table 3: Comparison of FCM and IHC in chronic lymphoproliferative disorders (n=30)

Click here to view



  Discussion Top


The 2008 WHO classification has emphasized that CLPD's cannot be classified on the basis of morphology alone and supplementation with immunophenotyping is indispensible, especially in the diagnosis of difficult cases.[1] In the present study, we compared the role of FCM and IHC in the diagnosis and classification of CLPD. Thirty newly diagnosed cases of CLPD from a tertiary care hospital were evaluated by FCM on peripheral blood and/or bone marrow aspirate samples. We also assessed the morphological patterns in bone marrow biopsy and IHC on bone marrow biopsy.

Twenty eight of 30 cases were of B-type CLPD of which CLL was the most common entity (75%). We observed expression of CD5, CD23, weak SmIg positivity and CD79b negativity were the most significant findings in the CLL group. In the non-CLL group, negativity for CD5 and positivity for FMC7 and CD79b were the most significant findings [Table 1]. These findings are comparable to observations made by Moreau et al. except that we did not find SmIg to be a highly expressed in CLL cases.[3] In our study, 5 of 7 non-CLL cases also showed SmIg positivity. A comparison of the three panels of monoclonal antibodies proposed by Matutes et al. and Moreau et al. for the differentiation of CLL from non-CLL was made.[3],[4] We found that both Panels A and B were useful in differentiating CLL from non-CLL. The sensitivity of Panels A, B and C for CLL was 85.7%, 95.2% and 85.7%, respectively. The specificity of Panels A, B and C was 85.7%, 85.7% and 57.1%, respectively [Table 2]. The results were statistically more significant for Panel B, indicating its superiority over the other two panels. Moreau et al. had also found a higher accuracy of Panels B and C in comparison to Panel A.

The concordance between FCM and IHC calculated in our study was 93.3%. In a previous study by El-Sayed et al. in 2008 done on Egyptian patients, 88% concordance between FCM and histopathology/IHC in the diagnosis of lymphoma was reported.[5] Similar results were independently obtained by Mand'áková et al. and Martínez et al. who found a concordance between FCM and histopathology/IHC in 89% and 87.2% respectively.[6],[7] Sah et al. compared the results between the two methods of immunophenotyping in 110 patients of B-cell lymphomas and reported a concordance rate of 88%.[8] Carulli et al. reported that 89.5% of 114 cases studied showed concordance between FCM and histopathological examination.[9] Naughton et al. in 1998 reported a lower concordance between FCM and bone marrow biopsy BMB in 273 bone marrow samples from NHL patients, where FCM detected disease in only 60% cases with bone marrow infiltration.[10] An improvement in the technique and antibodies used presently could account for the difference in the rate of concordance. Concordance rate for various immunological markers was reported between 76% and 100% by Biesemier et al. with 100% concordance between FCM and frozen-section immunophenotyping in CD3, CD4 and CD8 markers, which is same as that observed in this study.[11] The concordance rate for CD5, kappa and lambda was reported as 86%, 70% and 60% respectively by Biesemier et al. which are lower than those obtained in this study.[11]

In this study, considerable discordance was noted between IHC and FCM results in detecting kappa and lambda chain expression. Sometimes, it was difficult to interpret the Ig light chain detection by IHC because of diffuse staining in paraffin embedded blocks. Whereas in contrast, clonality assignment of the same?? By FCM was easier. Discordance was seen in 6/30 cases for kappa and 3/30 cases in lambda. Leers et al. had pointed out that lack of contrast between surface-immunoglobulin staining and extracellular immunoglobulin staining was a major drawback of immunohistochemical detection of monoclonality in B-cell lymphoproliferative disorders.[12] They established monoclonality in 9 out of 10 NHL cases by FCM while only 6 of 9 cases were conclusive by IHC.[12] Difficulty in determination of monoclonality by immunoglobulin light chain restriction using IHC was also noted by Abdel-Ghafar et al. because of the destruction of some of the antigenic epitopes by the fixation and decalcification process used for BMB specimens.[13] A similar observation had been made by Dunphy who had highlighted the difficulty arising due to weak expression of antigen in paraffin tissue because of variations in tumor preservation and fixation.[2]

We propose a two tier approach in immunophenotypic analysis of all new CLPD cases in order to derive maximum benefit from a machine (FCM) that uses costly reagents since cost can be an important limiting factor in developing countries. The screening panel comprises 9 markers including CD19, CD5, CD23, FMC7, CD10, CD20, CD3, kappa and lambda, which are important for specifying the lineage (B or T), to differentiate CLL from non-CLL group and in deciding the secondary panel. The rationale for selecting a B-cell-centric screening panel for CLPD is that majority of the cases that we found in our study were of B-CLPDs. We propose that CLPD panel should be initially meant for characterizing CLL and to differentiate it from non-CLL. Scoring system using CD5, CD23, CD22, FMC7, CD79b and SmIg is very useful in differentiating CLL from non-CLL cases. Using the 9- marker panel, a diagnosis of CLL, MCL and FL, the three most common CLPDs could be made. In a case suspected to be HCL, a secondary panel comprising of CD19, CD25, CD103, CD11c, kappa and lambda light chains, annexin 1 and DBA 44 would be required. If the T-cell markers in the screening panel are positive, T-cell CLPDs should be further evaluated using CD4, CD8, CD7.


  Conclusion Top


The concordance rate between FCM on peripheral blood/bone marrow aspirate samples and IHC on bone marrow biopsy sections in our study was 93.3%. Scoring system using CD5, CD23, CD22, FMC7, CD79b and SmIg was found to be useful in differentiating CLL from other B-CLPDs. In a case of CLPD, a screening panel comprising CD19, CD5, CD23, FMC7, CD10, CD20, CD3, kappa and lambda would be successful in the diagnosis of most of CLPDs. This can be followed by a secondary panel as required.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

 
  References Top

1.
Swerdlow S, Harris NL, Stein H, Jaffe ES, Theile J, Vardiman JW. World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2008. p. 180-82.  Back to cited text no. 1
    
2.
Dunphy CH. Applications of flow cytometry and immunohistochemistry to diagnostic hematopathology. Arch Pathol Lab Med 2004;128:1004-22.  Back to cited text no. 2
    
3.
Moreau EJ, Matutes E, A'Hern RP, Morilla AM, Morilla RM, Owusu-Ankomah KA, et al. Improvement of the chronic lymphocytic leukemia scoring system with the monoclonal antibody SN8 (CD79b). Am J Clin Pathol 1997;108:378-82.  Back to cited text no. 3
    
4.
Matutes E, Owusu-Ankomah K, Morilla R, Garcia Marco J, Houlihan A, Que TH, et al. The immunological profile of B-cell disorders and proposal of a scoring system for the diagnosis of CLL. Leukemia 1994;8:1640-5.  Back to cited text no. 4
    
5.
El-Sayed AM, El-Borai MH, Bahnassy AA, El-Gerzawi SM. Flow cytometric immunophenotyping (FCI) of lymphoma: Correlation with histopathology and immunohistochemistry. Diagn Pathol 2008;3:43.  Back to cited text no. 5
    
6.
Mand'áková P, Campr V, Kodet R. Correlation of results of flow cytometry and morphologic findings in the diagnosis of malignant B-cell lymphoma. Cas Lek Cesk 2003;142:651-5.  Back to cited text no. 6
    
7.
Martínez A, Aymerich M, Castillo M, Colomer D, Bellosillo B, Campo E, et al. Routine use of immunophenotype by flow cytometry in tissues with suspected hematological malignancies. Cytometry B Clin Cytom 2003;56:8-15.  Back to cited text no. 7
    
8.
Sah SP, Matutes E, Wotherspoon AC, Morilla R, Catovsky D. A comparison of flow cytometry, bone marrow biopsy, and bone marrow aspirates in the detection of lymphoid infiltration in B cell disorders. J Clin Pathol 2003;56:129-32.  Back to cited text no. 8
    
9.
Carulli G, Canigiani S, Volpini M, Ciancia EM, Galimberti S, Zucca A, et al. Bone marrow infiltration in B-cell non-Hodgkin's lymphomas: Comparison between flow cytometry and bone marrow biopsy. Recenti Prog Med 2005;96:284-90.  Back to cited text no. 9
    
10.
Naughton MJ, Hess JL, Zutter MM, Bartlett NL. Bone marrow staging in patients with non-Hodgkin's lymphoma: Is flow cytometry a useful test? Cancer 1998;82:1154-9.  Back to cited text no. 10
    
11.
Biesemier KW, Dent GA, Pryzwansky KB, Folds JD. A comparative study of frozen-section immunoperoxidase and flow cytometry for immunophenotypic analysis of lymph node biopsies. Clin Diagn Lab Immunol 1994;1:299-303.  Back to cited text no. 11
    
12.
Leers MP, Theunissen PH, Ramaekers FC, Schutte B, Nap M. Clonality assessment of lymphoproliferative disorders by multiparameter flow cytometry of paraffin-embedded tissue: An additional diagnostic tool in surgical pathology. Hum Pathol 2000;31:422-7.  Back to cited text no. 12
    
13.
Abdel-Ghafar AA, El Din El Telbany MA, Mahmoud HM, El-Sakhawy YN. Immunophenotyping of chronic B-cell neoplasms: Flow cytometry versus immunohistochemistry. Hematol Rep 2012;4:e3.  Back to cited text no. 13
    


    Figures

  [Figure 1]
 
 
    Tables

  [Table 1], [Table 2], [Table 3]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction
Materials and Me...
Results
Discussion
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed3093    
    Printed54    
    Emailed1    
    PDF Downloaded20    
    Comments [Add]    

Recommend this journal