5

  DETECTION OF LOW LEVELS OF B-CELL LYMPHOPROLIFERATIVE DISORDERS   RL MAIESE AND RC BRAYLAN   INTRODUCTION  
    Flow cytometry (FCM) is a useful adjunct to microscopy in the diagnosis and classification of lymphoproliferative disorders. It is objective and quantitative, and allows a higher level of sensitivity of detection and better characterization of B-cell lymphoproliferative disorders than conventional diagnostic techniques. (1,2) More recently, molecular genetic methods such as the polymerase chain reaction (PCR) have been applied to amplify rearranged immunoglobulin chains, detecting clonal B-cell expansions at a level of sensitivity similar to that of FCM. (3,4) We applied these techniques to morphologically unremarkable bone marrow and other clinical samples from patients suspected to have lymphoproliferative disorders and found that some of these specimens contain small numbers of monoclonal B-cells.
  FLOW CYTOMETRIC ANALYSIS OF B-CELL POPULATIONS FOR DETERMINATION OF CLONALITY   Rationale  
    In normal tissues, including peripheral blood, B-cells are polyclonal, composed of a mixture of cells individually expressing either kappa (k) or lambda (l) surface immunoglobulin (Ig) light chains (2,5) (Figure 1). The values of the kappa/lambda ratios in these tissues fall within a narrow range
(approximately 1.5:1). Conversely, in B-cell lymphomas/chronic leukemias, the neoplastic B-cells are clonal populations, which only express a single Ig light chain (Figure 1). In the past, studies using single-color fluorescence demonstrated disparities in the kappa and lambda distributions in lymphoproliferative processes. However, this type of analysis may be impaired by the presence of non-B-cells bearing passively adsorbed cytophilic Ig. The use of anti-kappa and anti-lambda antibodies in conjunction with anti-B-cell reagents, on the other hand, permits the analysis of light chain distributions exclusively on B-cells, eliminating the problem of non-specific binding of Ig to
Fc-receptor-bearing monocytes, NK cells or activated T cells. When B-cells in normal tissues or blood are stained for surface kappa or lambda Ig light chains, the resulting distributions are bimodal, reflecting the normal mixture of kappa- and lambda-expressing cells (2,5). In samples harboring B-cell lymphoma, staining of kappa and lambda surface Ig light chains on neoplastic B-cells results in unimodal, either positive or negative, distributions, except in cases of B-cell neoplasias lacking Ig expression, in which neither light chain is detected. The mutually exclusive expression of Ig light chains in B-cell neoplasia is particularly useful in the detection of a clonal expansion within populations of B-cells selected on the basis of their cell size or intensity of antigen expression. (2) This approach increases the sensitivity of detection of neoplastic B-cells to levels comparable to, or better than, that of molecular analysis and should be valuable in assessing minimal tumor involvement.
  Previous studies  
    In a study of normal peripheral blood samples (n=87), leukocytes were stained with a PE-labeled pan B-antibody (eg, CD19 or CD20) and, simultaneously, with either FITC-labeled anti-kappa or anti-lambda antibodies. The assessment of immunoglobulin light-chain expression was performed by analyzing immunoglobulin light-chain expression on CD19 or CD20-positive cells only. (2) The resulting average ratio of the percentage of kappa- to lambda-bearing B-cells was 1.51 (S.D.: ±0.31, range 0.75-2.46) (Figure 2). The sum of the percentages of kappa- and lambda-bearing cells, analyzed independently, was close to 100% (Figure 3). These results indicate that virtually all peripheral blood B-cells express surface immunoglobulin and that there is very little interference from non-specific Ig binding. The mean percentage of kappa- and lambda-bearing B lymphocytes was 60 (range: 43.3 to 70.2) and 41 (range: 29.9 to 60.8), respectively. Individual variation of 3 normal subjects tested repeatedly over a period of 14 to 20 weeks showed an average coefficient of variation of 3.88%. When the blood from one of these subjects was admixed with a known quantity of neoplastic B-cells, the sensitivity of detection of lymphoma cells was less than 5% of the total leukocyte count. Various gating techniques could even improve this level of detection. In a subsequent study of non-neoplastic lymph nodes using this approach, similar results were obtained (5).
      Present Observations  
    Employing methodologies similar to those of the previous studies (2, 5), we have detected clonal B-cell populations representing a minor component of the sample cellularity in over 100 cases that were otherwise morphologically unremarkable. To confirm the validity of such findings, an alternative, molecular genetic-based approach for B-cell clonality assessment was used. Such
analysis takes advantage of the recombination of the immunoglobulin heavy chain gene that takes place in both normal and neoplastic B lymphocytes. The rearranged immunoglobulin genes of polyclonal B-cells are of different sizes, whereas those of monoclonal B-cells are identical in size (Figure 1). PCR amplification of these genes was selected because of the its known high degree
of sensitivity (3, 4). We analyzed 31 specimens (including bone marrow, blood, and pleural fluids) that contained minor (£10% of sample cellularity) immunoglobulin light chain-restricted B-cell populations detected by the previously described flow cytometric analysis but lacked overt morphologic abnormalities. These cases were chosen because appropriately preserved DNA
was available. DNA extraction and PCR amplification of Ig heavy (H) chain genes using variable and joining region primers were performed on the cases using a strategy similar to previously described approaches (3, 6-13) (Figure 4). The primers used selectively amplified DNA segments resulting from IgH chain gene rearrangements. The products resulting from the amplification reactions varied in size in individual normal (polyclonal) B-cells, whereas these products were of the same size in monoclonal B-cells that carried the same rearranged IgH chain genes. Because of their identical size, the amplification products in monoclonal B-cell populations demonstrated a distinct band using electrophoretic gel analysis, while the products of polyclonal B-cells typically produced a “smear.” In this study, PCR amplified products consistent with the presence of clonal B-cell populations were found in a majority of the cases studied (87%) (unpublished data). It should be noted that 4 of the patients with negative PCR results have biopsy-proven B-cell lymphoma.  Overall, our rate of clonality detection is similar to that of previous investigations (3, 9-13). One reason for the PCR-negative cases may be that despite the extreme sensitivity of this analysis, a clonal B-cell population that is a minor fraction of the total population is frequently undetectable because of the high background caused by the amplification of coexisting polyclonal B-cells (14).
    CONCLUSION  
    Flow cytometry is a useful adjunct to microscopy in the diagnosis and classification of lymphoproliferative disorders. It provides objective and quantitative data, allowing a higher level of sensitivity of detection and better characterization of B-cell lymphomas/leukemias than conventional diagnostic techniques. PCR has a similar level of sensitivity for the detection of B-cell monoclonality but is not as informative. Nevertheless, these two approaches can be exploited to detect small numbers of monoclonal B-cells. In this study, we have found that morphologically unremarkable bone marrow and other clinical samples from patients suspected to have lymphoproliferative disorders may contain small numbers of monoclonal B-cells by flow cytometric analysis.  The significance of the presence of a small clonal population of B lymphocytes in an otherwise morphologically unremarkable specimen, however, is uncertain. (14) At the present time, therapeutic decisions based solely on such a finding, without consideration of all available clinical and laboratory data, may be inappropriate. Randomized prospective studies are necessary to determine the impact of these observations.
  REFERENCES  
1. Braylan RC, Benson NA: Flow cytometric analysis of lymphomas. Arch Pathol Lab Med
    1989; 113:627-633.
2. Braylan RC, Benson NA, Iturraspe J: Analysis of lymphomas by flow cytometry: Current and
    emerging strategies. Ann NY Acad Sci 1993; 677:364-377.
3. Achille A, Scarpa A, Montresor M, et al.: Routine application of polymerase chain reaction in
    the diagnosis of monoclonality of B-cell lymphoid proliferations. Diagn Mol Pathol 1995;
    4:14-24.
4. Campana D, Pui C: Detection of minimal residual disease in acute leukemia: Methodologic
    advances and clinical significance. Blood 1995; 85:1416-1434.
5. Maiese RL, Segal GH, Iturraspe J, Braylan RC: The cell surface antigen and DNA content
    distribution of lymph nodes with reactive hyperplasia. Mod Pathol 1995; 8:536-543.
6. Miller SA, Dykes DD, Polesky HF: A simple salting out procedure for extracting DNA from
    human nucleated cells. Nucleic Acids Res 1988; 16:1215.
7. Yamada M, Hudson S, Tournay O, et al.: Detection of minimal disease in hematopoietic
     malignancies of the B-cell lineage by using third-complementarity-determining region
     (CDR-III)-specific probes. Proc Natl Acad Sci 1989; 86:5123-5127.
8. Deane M, Norton JD: Immunoglobulin heavy chain variable region family usage is independent
    of tumor cell phenotype in human B lineage leukemias. Eur J Immunol 1990; 20:2209-2217.
9. Deane M, Norton JD: Immunoglobulin gene 'fingerprinting': an approach to analysis of B
    lymphoid clonality in lymphoproliferative disorders. Brit J Haematol 1991; 77:274-281.
10. Diss TC, Peng H, Wotherspoon AC, Isaacson PG, Pan L: Detection of monoclonality in
      low-grade B-cell lymphomas using the polymerase chain reaction is dependent on primer
      selection and lymphoma type. J Pathol 1993; 169:291-295.
11. Segal GH, Wittwer CT, Fishleder AJ, Stoler MH, Tubbs RR, Kjeldsberg CR: Identification
      of monoclonal B-cell populations by rapid cycle polymerase chain reaction: A practical
      screening method for the detection of immunoglobulin gene rearrangements. Am J Pathol
      1992; 141:1291-1297.
12. Segal GH, Jorgensen T, Masih AS, Braylan RC: Optimal primer selection for clonality
      assessment by polymerase chain reaction analysis: I. Low grade B-cell lymphoproliferative
      disorders of nonfollicular center cell type. Hum Pathol 1994; 25:1269-1275.
13. Sioutos N, Bagg A, Michaud GY, et al.: Polymerase chain reaction versus Southern blot
      hybridization: detection of immunoglobulin heavy chain-chain gene rearrangements. Diagn
      Mol Pathol 1995; 4:8-13.
14. Chan WC, Greiner TC: Diagnosis of lymphomas by the polymerase chain reaction
      [editorial]. Am J Clin Pathol 1994; 101:273-274.
15. Sklar J: Antigen receptor genes: Structure, function, and techniques for analysis of their
      rearrangements. In: Knowles DM, ed. Neoplastic Hematopathology. Baltimore MD:
      Williams and Wilkins, 1992; 219.
   Return To Contents