3

  B-CELL MONOCLONAL LYMPHOCYTOSIS IN THREE INDIVIDUALS LIVING NEAR A HAZARDOUS WASTE SITE   GE MARTI, J MULLER, M STETLER-STEVENSON, AND N CAPORASO   ABSTRACT  
The blood from three ATSDR study participants initially identified by CDC as having abnormal B-cell patterns was redrawn and analyzed at CBER FDA. A B-cell monoclonal lymphocytosis (BCML) with differing surface immunophenotypes was confirmed in all three cases even though the complete blood counts (CBC) were normal. Cell cycle analysis showed that there was an increase in G2M B-cells. A review of blood films revealed that each participant had a differing lymphocyte morphology. The first individual (CDC Case Number 11, see Chapters 1 and 2) had a typical small B-cell chronic lymphocytic leukemia (B-CLL) with the presence of smudge cells. The second individual (CDC Case Number 7) appeared to have an atypical variant of B-CLL with NK (natural killer) marker expression. The third individual (CDC Case Number 9 had some lymphocytes with plasmacytoid features consistent with Waldenstrom's disease. Morphometric analysis of the size, mean density, density variability and shape of lymphocyte nuclei showed significant differences in the mean values for each individual. There was an inverse relationship between nuclear area and density for all subjects; however, values for each individual occupied a unique range along a common continuum. These initial findings stress the importance of cytomorphology in the differential diagnosis of a BCML. Its relevance to clinical presentation, management and prognosis is unknown at this time. Although all these subjects showed unique characteristics, they may represent a continuum of the same lymphoproliferative disease (LPD) which over time could develop into specific, clinically, recognizable LPD. These findings also suggest that the incidence of B-CLL may be higher than currently realized, and that environmental exposure to organic solvents needs to be further explored as a risk factor for this family of LPD.  The incidence of BCML in an adult reference population remains to be determined.
  INTRODUCTION  
    During the past two years in collaboration with the CDC, we have discovered several individuals in apparently good health with a B-CLL like immunophenotype.  Initially we noted isolated cases of increased B-cells or a CD5+  B-cell lymphocytosis. Review of the blood film and CBC revealed a slight lymphocytosis.  From these single isolated findings, we had an opportunity to restudy  three individuals. This is a report of that study.
  METHODS AND MATERIALS   Blood Films  
    Unstained blood films were prepared in the field and fresh blood in heparin was redrawn and sent by overnight mail to CBER, FDA. The blood films were stained in the Hematology Laboratory, Clinical Center, NIH and stained blood films were then photomicrographed using a 100x oil objective and Kodak slide film (ASA 40). An attempt was made to photograph as many lymphocytes as possible from a single blood film. The slide film was developed commercially and transferred (digitized) to a Kodak Photo CD. Adobe Photoshop was used to prepare a composite (gallery or catalogue) of the individual lymphocytes.
  Image Analysis  
    NIH Image, v. 1.58 was used to process and analyze the composite images. This program was created by Wayne Rasband, NCI, NIH and is available as freeware by anonymous FTP from ftp://zippy.nimh.nih.gov. After digitization, the area, mean, maximum and minimum densities, perimeter, and major and minor axes were measured for each lymphocyte nucleus. The number
of cells analyzed from each slide was 66, 66 and 76 respectively.
  Flow Cytometry  
    An aliquot of the heparinized blood was washed, cytoshed for 40 min at 37° C and stained with the following panel of reagents: unstained cells, Simultest  reagents, isotype control, Leucogate, CD3/CD19, CD4/CD8, NK cells CD3/CD16/CD56, CD19/CD5, CD20/CD5, and CD20/CD23 (All are from BDIS.). The kappa-lambda analysis consisted of three sets of reagents: CD19 (B4 RD1, Coulter) and B-cell kappa FITC marker and CD19 (B4 PE) and B-cell lambda marker FITC from Ortho Diagnostic Systems, Inc. (Raritan, NJ).  Simultest anti-kappa FITC and anti-lambda PE to which was added CD19 (Leu 12 PerCP) were all from BDIS. Anti-kappa R-PE, anti-lambda FITC, and CD19 TriColor were obtained from Caltag (Burlingame, CA). It should be noted that in the two three color analyses the flurochromes are switched on the kappa and lambda reagents. Whole blood lysis was carried out after a 30 minute incubation in the cold using FACSLyse. After washing twice, they were fixed and resuspended in 0.5 ml of 0.5% buffered paraformaldehyde. A FACScan flow cytometer using Cellquest software was used to acquire and analyze 20,000 events. In the three color analysis, scatter and the third color CD19 PerCP (FL3) was used for gating on B-cells. FACSComp and Calabrite beads (BDIS) were used to standardize the setup of the instrument. For cell cycle analysis, hypaque ficoll cell suspensions (1 x 10 6 cells per ml final concentration) were stained with the FITC conjugated anti-CD20 (B1, Coulter) or anti HLA-DR FITC reagent (BDIS), fixed in 70% ETOH and then stained with propidium iodine (PI). At the time of analysis, the ethanol fixed cells are incubated with PI (50 mg/ml) and ribonuclease (1 mg per ml) for 30 min at 37°C and analyzed immediately on a FACScan flow cytometer using CELLFIT software for both the acquisition and analysis of 20,000 events. Modfit LT was also used. FDAplot was used to prepare composite displays.
  RESULTS   Image Analysis  
    Figure 1 shows a representative selection of lymphocytes from digitized images of subjects’ blood films. A summary of the morphometric data obtained for each individual is given in Table 1. This data shows that the mean values of each of the measured parameters are different for each subject. Moreover, when the data is arranged in the order CDC Case Numbers 9, 7, 11, several strong trends are evident: nuclear area decreases, nuclear density increases, variability in nuclear density decreases, the ratio of major to minor axes decreases. This confirms the morphological review of the blood films that indicates a trend to smaller, rounder, more uniformly dense lymphocytes in Case Number 11 compared to Case Number 9. Plots of nuclear area and density for each nucleus (Figure 2) showed an inverse relationship for lymphocytes in all three individuals.  However values for each subject occupied a different range along a continuum.
      Flow Cytometry  
    A composite of the immunophenotyping results for each individual are show in Figures 3, 4 and 5. The absolute lymphocyte count for these three individuals was less than 5000 cells per ml. They all had 23% to 30% B-cells implying 1150 to 1500 B-cells per ml. The percent kappa-positive cells were 79%, 88% and 93% with kappa/lambda ratios of 8.1, 12.1 and 40.5 respectively.  Under the best of circumstances, this would mean that one in four or one in three cells is a B-cell and there is an even higher chance that the B-cell is monoclonal. In all three individuals the larger cells, particularly the ones with granules, can be disregarded as NK cells. In all three individuals a population of normal B-cells remain. This is quite evident in case numbers 7 and 11. It should also be noted that case 11 is the most B-CLL like in that the CD20 CD5 clone shows a decrease in fluorescence intensity for CD20 expression.  This is classical for B-CLL whereas case 7 shows a distinct CD20 CD5 clone, the CD20 fluorescence intensity is not yet decreased. Cases 7 and 11 both show expression of CD23, while it is absent in case 9. In addition, case 9 does not show the same pattern for CD5 expression. Thus in conclusion, the immunophenotypes are distinctly different for each subject.
            The cell cycle analysis is summarized in Figure 6 and Table 2. In Figure 6 the data on these three individuals is summarized in multiple two parameter contour and dot plots and single parameter DNA histograms. We wish to draw attention to an increase in G2M cells best seen in the dotplots and suggested in the single parameter DNA histograms and contour plots. This is usually not seen in normal peripheral blood lymphocytes and is usually not seen in B-CLL samples unless a large number of cells are analyzed using a wide gate. The increased G2M cells appear to be in the CD20 positive population.  For these individuals, the S + G2M CD20 positive cells are 6.8%, 6.8%, and 6.9% respectively. The number of HLA-DR positive cells tended to be lower compared to the CD20 data; however, Case Number 11 shows 10% S + G2M HLA-DR positive cells.
  DISCUSSION  
    In this study, we found three individuals with a BCML with differing or variable immunophenotypes, associated with evidence of increased G2M CD20 positive cells. All three had varying lymphocyte morphology by light microscopy which was confirmed by morphometric analysis. The difficulties encountered in the technical determination of a kappa lambda analysis have been raised and discussed elsewhere (1). Ironically the B lymphocytes of these individuals would make excellent positive controls for these reagents. In addition to the kappa/lambda analysis, the CD5/CD20 or CD5/CD19 pattern of expression constitutes a characteristic pattern of normal remnant T cells when B-cells predominate. Residual normal B-cells are seen when the lymphocytosis is not very large. The characteristic signature of a CD5 positive B-cell clone
is well known. In an earlier report we suggested a progression from a negative CD5 BCML, to a CD5 CD20 bright BCML to a CD5 CD20 dim BCML (2-4).  We see this same exact progression in these three individuals and it seems to parallel their ages. This suggests the need for longitudinal and cross sectional studies.
 
    The finding of increased G2M cells is not totally unexpected. Aneuploidy is rarely seen flow cytometrically in B-CLL and is probably limited by the number of channels used for detection but this has not been formally tested.  Based on our experience, the use of 2048 channels might be expected to resolve some of the subtle cytogentic translocations including trisomy 12.  And in B-CLL if the gates are set wide enough, large B-cells show not only increased G2M cells but some are probably CD10 positive (2). It is suspected that in early preclinical B-CLL, it is easier to find G2M cells because they are not diluted out by the clone.
          While the measured, morphological parameters also suggest a progression and a shift in the lymphocyte profile, the exact role of morphometric analysis remains to be discerned. More data needs to be accumulated to determine the usefulness of such numerical data. We envision that the combination of flow cytometric analysis and sorting, morphometric analysis and molecular markers will permit us to discern not only the early events of LPD but differences between closely related LPD.
 
    The first and foremost question raised by these findings is what is the incidence of a BCML with or without CD5 expression in a normal adult population?  In the study of Jack et. al. (Chap. 10), approximately 1000 samples sent for routine blood counts in patients over the age of 40 years who had no clinical reason to suspect a hematological malignancy were screened. At least 1.7% had evidence of a clonal B-cell population using an IgH PCR assay and most of these patients were found to have a B-CLL phenotype by flow cytometry.  It is also common for clinical flow cytometry laboratories to find B-CLL clones in normal donors or in patients with slightly raised or normal absolute lymphocyte counts (personal observation). However, the frequency of this observation is unknown. Of interest, Liu et al., using a nested PCR assay showed that the incidence of BCL2 translocations in both blood and spleen are 55% and 35% respectively and increase with age (5). BCL2 translocations occur predominantly in follicular lymphomas and is thought to be a major step in the lymphomagenesis of this particular B-NHL. This incidence of BCL2 mutations is higher than the lifetime risk of NHL. Therefore other events must be required for the full expression of this lymphoma. There are earlier reports showing a similar phenomenon in tonsils (6, 7). A similar situation would probably exist for a BCML. In fact these two assays should probably be combined along with other markers of human lymphoproliferative disease such as cyclin D1/PRAD in future studies of BCML.
 
    A second question concerns the natural history of a BCML. There are essentially three possibilities. It is transient and will disappear; remain stable or progress to a clinically recognizable LPD. Given the increase incidence in NHL over the last decade and an aging population, the presence of a BCML may be an important predisposing factor.
 
    A third question concerns the differential diagnosis. At first glance, this concern may seem artificial, academic and trivial. At this point a CD5+ BCML is a finding and not a diagnosis. It could be the cellular equivalent of a monoclonal gammopathy of unknown significance (MGUS). There is certainly no harm to be done in reviewing the blood film and attempting to correlate the flow cytometric data. In fact, flow cytometry is becoming an useful adjunct for the differential diagnosis of closely related chronic lymphoid leukemia and low grade lymphomas. Unfortunately we do not yet have the tools to predict which BCML will progress to a clinically recognizable LPD. In fact we do not even have an assay to predict which Rai Stage 0 or I or Binet Stage A or B, B-CLL patients will progress to an advanced clinical stage.
 
    Related to the natural history and the differential diagnosis of a BCML is what to tell patients and their families. Faguet et al., have described and followed 28 patients with early or pre clinical B-CLL (8). Over a five year period at least 8 of these patients have progressed to overt B-CLL. They were followed annually with CBCs and microscopic surface immunofluoresence. It seems reasonable to anticipate that at least some individuals with BCML will progress. However early B-CLL can be an indolent disease and in the absence of therapy these patients can have a normal life span (9). A benign monoclonal lymphocytosis has also been recognized (10, 11). Follow up of these individuals should probably be no different than for patients with early or pre-clinical B-CLL. Reassurance and an interval CBC will suffice in the majority of individuals. A flow cytometric analysis of lymphocyte subsets with or without a PCR based assay for monoclonality should probably be carried out in an investigative protocol setting. A CT scan would be expected to non revealing while a bone marrow biopsy and aspirate should be considered invasive at this stage. If the latter is done, the biopsy should be done in conjunction with an immunperoxidase assay to aid in the detection of an early interstitial B-cell infiltrate. Also the aspirate should be submitted for the flow cytometric analysis of lymphocyte subsets and a kappa lambda analysis.
 
    There is one clinical setting where these considerations might assume greater significance. Familial B-CLL is defined as two or more first degree relatives with the disease. Quite often two or more siblings or two generations will have the disease. Unaffected family members from B-CLL kindred’s might benefit from these assays in terms of an early diagnosis. But at this time there would still be no way of predicting if an early diagnosis in such a family member would progress to overt B-CLL. This remains experimental, requires clinical correlation and is best carried out in an investigative protocol with the availability of genetic counseling unless the testing is done anonymously.
 
    In conclusion, pilot studies have been described in three individuals with BCML. Before a causal association with environmental exposure to organic solvents can be made, the incidence of BCML in a normal adult reference population is needed. Such a study should combine flow cytometry with a PCR based assay of monoclonality and other markers associated with LPD. Protocols to investigate family members should include these assays in order to characterize their significance regarding risk. If testing is not done anonymously, some provision for genetic counseling and future follow up would be necessary. Regardless of the course taken, these finding suggest that we may be getting closer to the underlying molecular and cellular lesions of the most common LPD in aged people: B-CLL and low grade B-NHL.
  ACKNOWLEDGMENTS  
    The technical expertise of PA Carter, GC Washington, F Abassi, V Zenger and D Cragg are gratefully acknowledged.
  REFERENCES  
1. Marti GE. Quantitative flow cytometry (QFCM) in B-chronic lymphocytic leukemia (B-CLL).
    Clin Immunol Newsletter 15: 1995.
2. Marti GE, Faguet GB, Stewart C, Branham P, Carter PH, Washington GC, et al., Evolution
   of leukemic heterogeneity of human B-CLL lymphocytes between and within patients. Curr
   Top Microbiol Immunol 1992; 182:303-11.
3. Marti GE, Faguet G, Bertin P, Agee J, Washington G, Ruiz S, et al., CD20 and CD5
    expression in B chronic lymphocytic leukemia (B-CLL). Proc NY Acad Sci USA 1992;
    651:480.
4. Marti GE. An edited transcription of an oral presentation in Mechanisms of B cell neoplasia.
    In: Potter M, Melchers F, eds. Basel:Editiones Roche 1995; 129-36.
5. Liu Y, Hernandez AM, Shibata D, Cortopassi GA. BCL2 translocation frequency rises with
    age in humans. PNAS 1994; 8910-14.
6. Limpens J, de Jong D, van Kreiken JHM, Price CGA, Young BD, van Ommen GB, and
    Kluin PM. Bcl-2/JH rearrangements in benign lymphoid tissues with follicular hyperplasia.
    Oncogene 1991; 6:2271-76.
7. Aster JC, Kobayashi Y, Shiota M, Mori S, Sklar J. Am J Pathol 1992; 141:291-99.
8. Faguet GB, Agee JF, Marti GE. Clone emergence and evolution in chronic lymphocytic
    leukemia: characterization of clinical, laboratory and immunophenotypic profiles of 25 patients.
    Leukemia and Lymphoma 1992; 6:345-56.
9. Rozman C and Montserrat E. Chronic lymphocytic leukemia current concepts. NEJM 1996;
    333:1052-53.
10. Han T, Ozer H, Gavigan M, Gajera R, Minowada J, Bloom ML, et al., Benign monoclonal
      B-cell lymphocytosis - a benign variant of CLL: clinical, immunologic, phenotypic, and
      cytogenetic studies in 20 patients. Blood 1984; 64(1):244-52.
11. Bassan R, Buzzetti M, Marini B, Rambaldi A, Allavena P, Barbui T. Investigation of chronic
      lymphocytosis in adults. Am J Clin Pathol 1988; 89:783-87.
   Return To Contents