Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma Associated With IgM Paraprotein A Clinicopathologic Study of 26 Cases

Transcription

1 Hematopathology / CLL/SLL WITH IGM PARAPROTEIN Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma Associated With IgM Paraprotein A Clinicopathologic Study of 26 Cases C. Cameron Yin, MD, PhD, 1 Pei Lin, MD, 1 Dennis A. Carney, MD, 2 Beverly C. Handy, MD, 3 George Z. Rassidakis, MD, PhD, 1 Joan H. Admirand, MD, 1 Michael J. Keating, MD, 2 and L. Jeffrey Medeiros, MD 1 Key Words: Chronic lymphocytic leukemia/small lymphocytic lymphoma; Serum IgM; Plasmacytoid lymphocytes; ZAP-70 DOI: /FDGWB5C2MYRYXH2E Abstract We studied the clinicopathologic, immunophenotypic, and cytogenetic features of 26 patients with chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) associated with serum IgM paraprotein. The study group (16 men; 10 women; median age, 64 years; range, years) represents approximately 2.5% of CLL/SLL cases at our institution. The paraprotein level ranged from 1 to 14 g/l (median, 4 g/l). Neoplasms in bone marrow were composed of small round lymphocytes arranged in nodular (n = 6), diffuse (n = 5), interstitial (n = 5), or mixed (n = 10) patterns. All cases were positive for monotypic surface immunoglobulin light chain, IgM/IgD, CD5, CD19, CD20, and CD23. CD11c (14/20 [70%]), CD79b (11/19 [58%]), FMC-7 (11/26 [42%]), CD22 (8/20 [40%]), and ZAP-70 (6/19 [32%]) were expressed in subsets of cases. Of 17 bone marrow specimens assessed by conventional cytogenetics, 6 were abnormal and 11 were diploid. The overall survival of this group (median follow-up, 24 months) was not significantly different from that for an age-, sex-, and stage-matched group of 52 CLL/SLL patients without IgM paraprotein (P =.60). We conclude that CLL/SLL cases with serum IgM paraprotein are similar to other CLL/SLL cases in their clinicopathologic and immunophenotypic features. Chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) represents 6.7% of all non-hodgkin lymphomas, and CLL is one of the most common types of leukemia in the Western world. 1 The median age at diagnosis is 65 years, and there is a male/female ratio of 2:1. Although most patients are asymptomatic at diagnosis, some can have systemic (B type) symptoms, hepatosplenomegaly, lymphadenopathy, and cytopenias due to leukemic infiltration of bone marrow and other organs. As defined in the World Health Organization (WHO) classification, CLL/SLL is a neoplasm of small lymphocytes that are positive for monotypic surface immunoglobulin (dim intensity), CD5, CD19, CD20 (dim), and CD23 and usually negative or dimly positive for CD22, CD79b, and FMC-7. 1 Serum IgM paraprotein (so-called M component) can be detected in a subset of patients with CLL/SLL by conventional methods such as serum protein electrophoresis and can be detected in a higher number of patients by more sensitive techniques. 2 The WHO classification recognizes that a small M component can be found in some patients with CLL/SLL, but no mention is made of the frequency of this occurrence or the range of serum paraprotein levels in patients with CLL/SLL. Furthermore, the clinical significance of a serum IgM paraprotein in patients with CLL/SLL is unknown. An earlier study reported that patients with CLL/SLL with IgM paraproteinemia have an inferior survival compared with patients with CLL/SLL without serum paraprotein. 3 However, other studies have not confirmed this observation. 4 We assessed the clinical, morphologic, immunophenotypic, and conventional cytogenetic findings in a group of patients with CLL/SLL with IgM paraprotein. We also compared the clinical outcome of this group of patients with that 594 Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E

2 Hematopathology / ORIGINAL ARTICLE for a group of patients with CLL/SLL without IgM paraprotein matched for age, sex, and stage of disease. Materials and Methods Case Selection The files of the Department of Hematopathology, The University of Texas M.D. Anderson Cancer Center, Houston, from January 1997 to March 2004 were searched for cases of CLL/SLL with IgM paraprotein detected by serum protein electrophoresis and immunofixation. The diagnosis of CLL/SLL was based on morphologic and immunophenotypic findings using the criteria of the WHO classification. 1 All cases were composed of a proliferation of predominantly small lymphocytes, with or without morphologic evidence of plasmacytoid differentiation, that were positive for monotypic immunoglobulin light chain, surface IgM/IgD, CD5, CD19, CD20, and CD23. We also selected a group of 52 patients from the same period with CLL/SLL without serum IgM paraprotein that were matched for age, sex, and stage of disease. This group was used for survival analysis and for a comparison of morphologic features in a subset of cases. Serum Protein Electrophoresis and Immunofixation Serum protein electrophoresis and immunofixation were performed using the PARAGON (Beckman Coulter, Brea, CA) or the HYDRAGEL (Sebia, Norcross, GA) system. Total serum IgM and monoclonal IgM (M component) levels were quantified using densitometric scanning. Levels at the time of initial examination and the highest levels during the course of disease were recorded. Morphologic Evaluation H&E-stained slides routinely prepared from formalinfixed, paraffin-embedded bone marrow biopsy and/or aspirate clot specimens and Wright-Giemsa stained aspirate smears were reviewed. The percentage of lymphocytes, their cytologic features, and the extent and pattern of leukemic infiltration were assessed. In cases that also had lymph node biopsy specimens, the histologic features of these specimens also were reviewed. We also compared the cytologic features of the lymphocytes in CLL/SLL cases associated with serum IgM paraprotein and CLL/SLL cases from the matched control group without serum IgM paraprotein. For the latter group, bone marrow aspirate smears from 19 patients were readily available for analysis. For both groups of patients, 3 observers (P.L., J.H.A., and L.J.M.) quantified the percentage of plasmacytoid lymphocytes in bone marrow aspirate smears by performing manual counts. These smears were reviewed without knowledge of the presence or level of serum IgM paraprotein. Plasmacytoid lymphocytes were defined as cells with eccentrically situated nuclei and moderately abundant cytoplasm and/or with cartwheel-like (plasmacytoid) chromatin. The percentages of plasmacytoid lymphocytes generated by each pathologist for an individual case were averaged. Flow Cytometric Immunophenotypic Analysis Flow cytometric immunophenotypic analysis was performed using a FACScan or FACSCalibur instrument (Becton Dickinson Biosciences, San Jose, CA). Lymphocytes were gated for analysis using CD45 (peridinin chlorophyll-α protein conjugated) and side scatter and analyzed using a panel of monoclonal antibodies specific for CD3 (fluorescein isothiocyanate [FITC] or phycoerythrin [PE]), CD5 (FITC or PE), CD10 (FITC), CD11c (PE), CD19 (FITC, PE, or allophycocyanin), CD20 (PE), CD22 (FITC), CD23 (PE), CD38 (allophycocyanin), CD79b (PE), FMC-7 (FITC), immunoglobulin light chains (FITC), and surface IgM/IgD (FITC). All antibodies were purchased from Becton Dickinson Biosciences, and negative staining levels for each antibody were set by comparison with an isotype-matched control sample. By using the scoring system proposed by Moreau and colleagues, 5 cases were scored based on 5 variables: expression of dim surface immunoglobulin (1 point), CD5 (1 point), and CD23 (1 point); dim or absent CD22/CD79b (1 point); and absent FMC-7 (1 point). Cases with a score of 4 or 5 were considered to have a typical immunophenotype for CLL/SLL. Cases that deviated from this pattern of antigen expression were considered to have an atypical immunophenotype for CLL/SLL. Immunohistochemical Analysis for ZAP-70 Formalin-fixed, paraffin-embedded bone marrow biopsy tissue sections were stained with a monoclonal antibody specific for ZAP-70 (dilution 1:500, Upstate Cell Signaling Systems, Lake Placid, NY) as described previously. 6 Detection of the primary antibody was achieved using the LSAB+ kit (DAKO, Carpinteria, CA), which contains secondary biotinylated antibody and streptavidin/horseradish peroxidase complex, according to the manufacturer s instructions. The chromogen was 3,3'-diaminobenzidine/hydrogen peroxide (DAKO), and slides were counterstained with hematoxylin. Nuclear staining of nonneoplastic, reactive T cells served as an internal control. Conventional Cytogenetics Conventional G-band karyotyping was performed on metaphase cells from samples cultured for 24 or 48 hours using previously described methods. 7 Metaphase cells were obtained using a methanol acetic acid fixation method. The karyotype reports were written using the International System for Human Cytogenetic Nomenclature. 8 Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E 595

3 Yin et al / CLL/SLL WITH IGM PARAPROTEIN Survival Analysis The outcome for patients was obtained by reviewing medical records and the database of the Department of Leukemia, The University of Texas M.D. Anderson Cancer Center. By using the Kaplan-Meier method, we compared the overall survival of patients with CLL/SLL with serum IgM paraprotein with a matched group of 52 patients with CLL/SLL without serum IgM paraprotein. Analysis was performed using Statistica (StatSoft, Tulsa, OK). Results Clinical Findings We identified 26 untreated patients with CLL/SLL with IgM paraprotein from a total of 1,053 untreated patients with CLL/SLL at our institution from January 1997 to March 2004, for a frequency of 2.47%. There were 16 men and 10 women with a median age of 64 years (range, years). The clinical features are summarized in Table 1. Of the 26 patients, 4 (15%) had systemic symptoms, including fever, fatigue, night sweats, and weight loss. No patients had symptoms related to hyperviscosity. Physical examination revealed generalized lymphadenopathy in 17 patients (65%) and splenomegaly in 9 patients (35%). Laboratory evaluation showed 16 patients (62%) with an elevated WBC count, 15 (58%) with anemia, and 7 (27%) with thrombocytopenia. Serum β 2 -microglobulin and lactate dehydrogenase levels were elevated in 24 patients (92%) and 5 patients (19%), respectively. Seven patients who did not have peripheral lymphocytosis (lymphocyte count, <5,000/µL [< /L]) had generalized lymphadenopathy. The lymphocytes in this group ranged from 1,400 to 3,500/µL ( /L; median, 2,200/µL [ /L]). In addition to bone marrow aspiration biopsy specimens, lymph node biopsy specimens were available for this subset of 7 patients. The levels of serum total IgM and IgM paraprotein at initial diagnosis varied widely Table 2. During the course of disease, the serum IgM paraprotein level in 2 patients rose, from 1.1 to 21 g/l in one patient and from 3 to 14 g/l in the other patient. The serum IgM paraprotein level remained stable or decreased in the remaining 24 patients. Table 1 Clinicopathologic Features of Patients With Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma With and Without IgM Paraprotein IgM Paraprotein With (n = 26) Without (n = 52) Table 2 Bone Marrow Findings, ZAP-70 Expression, and Levels of Serum IgM in Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma With and Without IgM Paraprotein IgM Paraprotein With (n = 26) Without (n = 52) Pattern Nodular 6 10 Diffuse 5 10 Interstitial 5 13 Mixed Extent (%) Range Median ZAP-70 6/19 ND IgM Total, mg/dl (g/l) (reference range, ( ) Range 87-3,560 ( ) ( ) Median 590 (5.9) 60 (0.6) Monoclonal (g/l) Range 1-14 NA Median 4 NA NA, not applicable as no monoclonal IgM was detected; ND, not done. Range Median Range Median Reference Range Age (y) Sex M F Lymphadenopathy Splenomegaly 9 20 WBC count, /µl 11,400-82,000 25,500 (25.5) 11, ,000 24,900 (24.9) 4,000-11,000 (4-11) ( 10 9 /L) ( ) ( ) Lymphocytes, % ( ) 59 (0.59) ( ) 77 (0.77) ( ) Hemoglobin concentration, (99-135) 12.9 (129) (48-138) 11.9 (119) Men, ( ); g/dl (g/l) women, ( ) Platelet count, (71-132) 107 (107) (26-133) 115 (115) (71-132) 10 3 /µl ( 10 9 /L) Lactate dehydrogenase, U/L 669-1, β 2 -Microglobulin, 10 3 g/l Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E

4 Hematopathology / ORIGINAL ARTICLE We also detected IgG paraprotein (in addition to IgM) in the serum of 2 patients (IgG level, 6 and 8 g/l, respectively). In one patient, the IgM and IgG paraproteins had the same immunoglobulin light chain. In the other patient, the immunoglobulin light chains were different. The clinical features of the 52 patients with CLL/SLL without serum IgM paraprotein are summarized in Table 1. Morphologic Findings All cases had bone marrow involvement. Bone marrow aspirate smears in all cases showed an infiltrate of predominantly small lymphocytes, with a subset of plasmacytoid lymphocytes, representing 11% to 93% (median, 58%) of the nucleated cells. The cases with a lymphocyte percentage within the normal range, 3% to 17% ( ), were shown by flow cytometric studies to express monoclonal immunoglobulin light chain. Histologic sections of the bone marrow biopsy and clot specimens showed increased, predominantly small lymphocytes in nodular (n = 6), diffuse (n = 5), interstitial (n = 5), or mixed (n = 10) patterns (Table 2). The leukemic infiltrate replaced 20% to 90% of the bone marrow medullary space. All 7 lymph node biopsy specimens in cases without peripheral lymphocytosis displayed typical morphologic findings of CLL/SLL, with a proliferation of small lymphocytes showing condensed chromatin and vaguely nodular areas of larger, nucleolated cells corresponding to proliferation centers, as described in the WHO classification. 1 The pathologic features of the 52 patients with CLL/SLL without serum IgM paraprotein are summarized in Table 2. A The blinded comparison of the percentage of plasmacytoid lymphocytes in 20 CLL/SLL cases associated with serum IgM paraprotein (group 1) and 19 cases of CLL/SLL without serum IgM paraprotein (group 2) was as follows: In group 1, the median percentage of plasmacytoid lymphocytes was 10% (range, 2%-16%), compared with 5% (range, 2%-23%) in group 2 Image 1 and Image 2. With a cutoff of 10%, 10 cases (50%) in group 1 had at least 10% plasmacytoid lymphocytes, compared with 4 (21%) in group 2 (P =.14; Fisher exact test). With a cutoff of 15%, 2 cases (10%) in group 1 compared with 3 (16%) in group 2 had at least 15% plasmacytoid lymphocytes (P =.64; Fisher exact test). Only 1 case of CLL/SLL that was not associated with serum IgM paraprotein had more than 20% plasmacytoid lymphocytes. We also compared the percentage of plasmacytoid lymphocytes with the serum levels of total IgM and IgM paraprotein. There was no correlation between the percentage of plasmacytoid lymphocytes and serum total IgM or IgM paraprotein levels (P =.22 and.17, respectively; Spearman rank correlation test). Immunophenotypic Findings The neoplastic cells in all 26 cases were positive for monotypic immunoglobulin light chain (κ, 14; λ, 12), IgM/IgD (dim, 14; moderate, 12), CD5, CD19, CD20 (dim, 1; moderate, 14; bright, 11), and CD23. In addition, the leukemic cells were positive for CD11c (dim) in 14 (70%) of 20, CD22 in 8 (40%; dim, 6; moderate, 2) of 20, CD38 in 5 (26%) of 19, CD79b (dim) in 11 (58%) of 19, and FMC-7 in 11 (42%; dim, 6; moderate, 5) of 26. CD3 (n = 12) and CD10 (n = 20) were negative in all cases assessed. Sixteen cases were considered to have a typical immunophenotype with B Image 1 Bone marrow aspirate smears of 2 chronic lymphocytic leukemia/small lymphocytic lymphoma cases associated with serum IgM paraprotein. A, In this case, many lymphocytes had abundant cytoplasm and eccentric nuclei consistent with plasmacytoid differentiation. B, In this case, few plasmacytoid lymphocytes were present (A and B, Wright-Giemsa, 1,000). Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E 597

5 Yin et al / CLL/SLL WITH IGM PARAPROTEIN scores of 4 or 5. The other 10 cases had an immunophenotype atypical for CLL/SLL with scores of 2 or 3. The monoclonal immunoglobulin light chain expressed by the neoplastic cells was concordant with the serum monoclonal immunoglobulin light chain in 22 (85%) of 26 cases (κ, 12; λ, 10) but discordant in the other 4 cases. In 2 cases, κ light chain was expressed by the CLL/SLL cells and the serum had λ light chain. In the other 2 cases, λ light chain was expressed by the CLL/SLL cells and κ light chain was detected in serum. The level of monoclonal serum light chain of these 4 cases ranged from 2 to 5 g/l. A Formalin-fixed, paraffin-embedded bone marrow biopsy specimens were available in 19 cases for analysis of ZAP-70 by immunohistochemical studies. The neoplastic cells expressed ZAP-70 in 6 cases (32%), all with more than 90% of cells positive (Table 2) Image 3. The remaining 13 cases were negative. Cytogenetic Abnormalities Conventional cytogenetic analysis was performed on bone marrow aspirate material in 17 cases. A diploid karyotype was found in 11 cases. In 6 cases, cytogenetic abnormalities were Image 2 Bone marrow aspirate smears of 2 chronic lymphocytic leukemia/small lymphocytic lymphoma cases not associated with serum IgM paraprotein. A, In this case, many lymphocytes had abundant cytoplasm and eccentric nuclei consistent with plasmacytoid differentiation. B, In this case, few plasmacytoid lymphocytes were present (A and B, Wright-Giemsa, 1,000). A B B Image 3 A, Case of chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) associated with serum IgM paraprotein that was positive for ZAP-70. B, Case of CLL/SLL associated with serum IgM paraprotein that was negative for ZAP- 70 (A and B, ZAP-70 immunohistochemistry with hematoxylin counterstain, 400). 598 Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E

6 Hematopathology / ORIGINAL ARTICLE identified Table 3. One case had trisomy 12, 1 case had inv(2)(q13q31), 1 case had loss of chromosome Y, and 3 cases had complex karyotypes. Clinical Outcome Clinical follow-up was available for 21 of 26 patients who had CLL/SLL associated with IgM paraprotein. Of these patients, 17 received therapy after initial diagnosis; 4 were only observed. Clinical follow-up was available for all 52 patients in the matched group who had CLL/SLL not associated with IgM paraprotein. The overall survival of patients with CLL/SLL with IgM paraprotein was not significantly different from patients with CLL/SLL without IgM paraprotein (P =.60; Kaplan-Meier analysis) Figure 1. There also was no significant difference in overall survival between patients with CLL/SLL with a typical immunophenotype and those with CLL/SLL with an atypical immunophenotype (P =.22; Kaplan-Meier analysis). For the 19 cases assessed for ZAP-70 expression, 14 had clinical follow-up. ZAP-70 expression did not predict an adverse prognosis in this small subgroup. Discussion The association of CLL/SLL with serum monoclonal paraprotein, usually IgM, is well recognized in the literature. 4,9,10 However, the frequency of IgM paraprotein and the range of IgM levels in patients with CLL/SLL are not well described. Thus, we studied 26 cases of CLL/SLL with serum Table 3 Cytogenetic Findings in 17 Patients With Chronic Lymphocytic Leukemia/Small Lymphocytic Lymphoma With IgM Paraprotein Case No. Karyotype 1 47,XY,+12[8]/46,XY[12] 2 46,XY[25] 3 46,XX[20] ,XY, 6,del(9)(q21),add(10)(p15),add(14)(q32), +2-3mar[6]/46,XY[23] 5 46,XY[20] 6 46,XY[20] 8 46,XX[20] 10 46,XX[20] 14 46,XY[19] 17 46,XY[20] 18 45,X, Y[5]/46,XY[15] 19 46,XY[20] 21 45,XY,der(1)t(1;7)(p31;q33),der(3)t(3;7)(q21;p22), del(6)(q21),der(7)t(3;7)t(1;7),del(10)(q24),der(13;17) (q10;q10)[7]/47,xy,del(6)(q21),+12[2]/46,xy[11] 22 50,XX,i(2)(q10),+3,+3,+13,+18[3]/46,XX[4] 24 46,XX[20] 25 46,XY[29] 26 46,XY,inv(2)(q13q31)[20] IgM paraprotein to analyze the clinicopathologic, immunophenotypic, and cytogenetic features. For this study, we relied on immunophenotype to distinguish CLL/SLL with IgM paraprotein from other low-grade B-cell lymphomas, in particular, lymphoplasmacytic lymphoma/waldenström macroglobulinemia (LPL/WM), because these entities can resemble each other. This approach has been used by others. 11 As specified in the WHO classification, CLL/SLL is positive for monotypic immunoglobulin light chain (dim), CD5, CD19, CD20 (dim), and CD23 and is negative or dimly positive for CD22, CD79b, and FMC-7. 1 All CLL/SLL patients in this study had relatively low levels of serum monoclonal IgM. The highest serum IgM paraprotein level at time of diagnosis was 14 g/l with a median level of 4 g/l, and 1 patient had a higher serum level, 21 g/l, during the disease course. Thus, the risk of hyperviscosity is minimal in these patients, and none of the patients had symptoms related to hyperviscosity. These findings support the position expressed in the WHO classification that the levels of IgM paraprotein in patients with CLL/SLL, if present, usually are low. 1 Monoclonal serum IgM (and total IgM) levels in this study also did not correlate with tumor burden in the bone marrow or with patient survival. In the present study, the surface immunoglobulin light chain expressed by the leukemic cells was identical to that of the serum IgM paraprotein in 22 (85%) of 26 cases, suggesting that the serum paraprotein, in most cases, is the secreted product of the leukemic cells. In the literature, the concordance between the surface immunoglobulin light chain expressed by the neoplastic cells and the serum paraprotein Proportion Surviving Months Figure 1 Survival curve analysis. The solid line represents 26 patients with chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL) associated with serum IgM paraprotein; 4 of these patients died. The dotted line represents the matched group of 52 patients with CLL/SLL without serum IgM paraprotein; 5 of these patients died. P =.60. Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E 599

7 Yin et al / CLL/SLL WITH IGM PARAPROTEIN has ranged from 88% to 100% in CLL/SLL cases. 2,12 The explanation for the 4 discordant cases in the present study is unknown. Furthermore, we noted a monoclonal serum IgG protein coexisting with IgM paraprotein in 2 cases, 1 with an identical and 1 with a different immunoglobulin light chain. The presence of both IgM and IgG monoclonal proteins suggests that neoplastic transformation may have occurred at the time of isotype switching from IgM to IgG 13 or that the tumor cells retained their capability of isotype switching, a process independent of immunoglobulin heavy chain gene mutation. 14 As an alternative, these additional M components may represent biclonal or triclonal lymphoproliferative disorders as described in the literature 15 or the development of a subclone within the original tumor cells due to clonal evolution. Bernstein and colleagues, 3 in a study of 111 patients, reported that the presence of serum monoclonal protein in patients with CLL/SLL was associated with a shorter median survival: 63 months for patients with CLL/SLL with paraprotein compared with 103 months for patients without serum paraprotein (P =.012). For this reason, we compared the survival of patients with CLL/SLL associated with IgM paraprotein in the present study with another group of patients with CLL/SLL without IgM paraprotein matched for age, sex, and stage of disease. In this analysis, the overall survival of patients with CLL/SLL associated with IgM paraprotein was not significantly different from that of patients with CLL/SLL without IgM paraproteinemia. We cannot explain the discrepancy between our results and those of Bernstein et al. 3 Different therapeutic regimens used for the patients in the study by Bernstein et al 3 and the present study is one possible explanation. The shorter clinical follow-up and smaller number of patients in the present study are also possible explanations. Known prognostic indicators in CLL/SLL that predict an adverse outcome include advanced age, elevated serum levels of lactate dehydrogenase and β 2 -microglobulin, diffuse growth pattern in the bone marrow, increased prolymphocytes, rapid lymphocyte doubling time, and complex karyotype Recently, molecular studies identified 2 subtypes of CLL/SLL: those with mutated and those with unmutated immunoglobulin genes. Patients with unmutated CLL/SLL have a poorer outcome. 20,21 Gene expression profiling studies subsequently identified ZAP-70 as being expressed preferentially in unmutated CLL/SLL cases, and, thus, ZAP-70 expression has been used as a surrogate marker for the mutational status of the immunoglobulin genes. 20,21 ZAP-70 is a member of the syk tyrosine kinase family that is expressed primarily in T cells and natural killer cells. 22,23 ZAP-70 is essential for T-cell activation and for B-cell maturation from the pro-b to pre-b-cell stage. 22,23 We assessed ZAP-70 expression in a subgroup of 19 cases of CLL/SLL associated with serum IgM paraprotein, and approximately one third were positive for ZAP- 70. These results suggest that cases of CLL/SLL associated with serum IgM paraprotein are heterogeneous and that ZAP-70+ neoplasms are likely to have unmutated immunoglobulin genes. Immunophenotypically, a typical CLL/SLL profile was identified in 16 (62%) of 26 cases. The remaining 10 (38%) cases showed an atypical profile with moderate intensity of CD22, CD79b, and immunoglobulin light chain, as well as variable expression of FMC-7. Expression of CD79b and FMC-7 has been recognized in atypical CLL/SLL In the present study group, however, the morphologic features, clinical behavior, and ZAP-70 expression of cases with an atypical immunophenotype were otherwise similar to the cases that had a typical immunophenotype. In the blinded comparison of bone marrow aspirate smears of CLL/SLL cases with or without serum IgM paraprotein, all 3 observers found an overall quantitative difference in the percentages of plasmacytoid lymphocytes in these 2 groups. The median percentage of plasmacytoid lymphocytes was 10% (range, 2%-16%) in CLL/SLL cases associated with serum IgM paraprotein and 5% (range, 2%-23%) in CLL/SLL cases without serum IgM paraprotein. However, the range of the plasmacytoid lymphocyte percentage showed extensive overlap, and we were impressed by the percentage of plasmacytoid lymphocytes in some CLL/SLL cases without serum IgM paraprotein. In fact, the CLL/SLL case with the highest percentage of plasmacytoid lymphocytes, 23%, was not associated with serum IgM paraprotein. Thus, in the examination of bone marrow aspirate smears of an individual case of CLL/SLL, one cannot reliably predict the presence or absence of serum IgM paraprotein. The differential diagnosis of CLL/SLL with IgM paraprotein and LPL/WM is important clinically. Currently, patients with CLL/SLL and LPL/WM at our hospital, when they require therapy, often are treated with different therapeutic regimens. Morphologic features, immunophenotype, and serum IgM paraprotein levels are all helpful in making this distinction. Morphologically, the neoplastic small lymphocytes of CLL/SLL (with or without IgM paraprotein) can be associated with a variable number of prolymphocytes. The presence of proliferation centers, although uncommon in bone marrow unless disease is extensive, also supports the diagnosis of CLL/SLL. Immunophenotypically, CD5 expression by the neoplastic cells provides support for CLL/SLL. Other immunophenotypic findings also are helpful because CLL/SLL cells typically express dim surface immunoglobulin light chain and CD23 and are negative for FMC-7 and CD22. However, a subset of CLL/SLL cases can deviate from this typical immunophenotype and express 600 Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E

8 Hematopathology / ORIGINAL ARTICLE relatively brighter surface immunoglobulin light chain, FMC- 7, or CD22 and be negative for CD23. 24,26 Furthermore, others have reported a small subset of cases of LPL/WM that expressed CD5 or CD The level of serum IgM paraprotein also is helpful. A serum IgM paraprotein level of more than 30 g/l is very rare in patients with CLL/SLL and is presumptive evidence of LPL/WM. All patients with CLL/SLL in the present study had much lower serum IgM paraprotein levels, with a median of 4 g/l. However, in the experience of Lin et al, 30 serum IgM paraprotein levels in patients with LPL/WM also are commonly lower than 30 g/l and can also be lower than 20 g/l, the upper limit of IgM paraprotein in this CLL/SLL study group. In summary, patients with CLL/SLL with serum IgM paraprotein represent a heterogeneous group with regard to immunophenotype, ZAP-70 expression, and clinical outcome. Serum paraprotein levels usually are low. The leukemic cells can exhibit an atypical immunophenotype with variable expression of CD22, CD79b, and FMC-7. The overall survival of patients with CLL/SLL with IgM paraprotein does not seem to be significantly different from that of CLL/SLL patients without IgM paraprotein. From the Departments of 1 Hematopathology, 2 Leukemia, and 3 Laboratory Medicine, The University of Texas M.D. Anderson Cancer Center, Houston. Address correspondence to Dr Medeiros: Dept of Hematopathology, Box 72, UT M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX Acknowledgment: We thank Susan Lerner, Department of Leukemia, The University of Texas M.D. Anderson Cancer Center for assistance in statistical analysis. References 1. Muller-Hermelink HK, Catovsky D, Montserrat E, et al. Chronic lymphocytic leukaemia/small lymphocytic lymphoma. In: Jaffe ES, Harris NL, Stein H, et al, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: IARC Press; 2001: Deegan MJ, Abraham JP, Sawdyk M, et al. High incidence of monoclonal proteins in the serum and urine of chronic lymphocytic leukemia patients. Blood. 1984;64: Bernstein ZP, Fitzpatrick JE, O Donnell A, et al. Clinical significance of monoclonal proteins in chronic lymphocytic leukemia. Leukemia. 1992;6: Schaffner KF, Krause JR, Kelly RH. Biclonal IgM gammopathy in chronic lymphocytic leukemia. Arch Pathol Lab Med. 1988;112: Moreau EJ, Matutes E, A Hern RP, et al. Improvement of the chronic lymphocytic leukemia scoring system with the monoclonal antibody SN8 (79b). Am J Clin Pathol. 1997;108: Admirand JH, Rassidakis GZ, Abruzzo LV, et al. Immunohistochemical detection of ZAP-70 in 341 cases of non-hodgkin and Hodgkin lymphoma. Mod Pathol. 2004;17: Hao S, Sanger W, Onciu M, et al. Mantle cell lymphoma with 8q24 chromosomal abnormalities: a report of 5 cases with blastoid features. Mod Pathol. 2002;15: Mittleman F, ed. ISCN 1995: An International System for Human Cytogenetic Nomenclature Basel, Switzerland: Karger; Rozman C, Montserrat E, Vinolas N. Serum immunoglobulins in B-chronic lymphocytic leukemia: natural history and prognostic significance. Cancer. 1988;61: Xu HJ, Roberts-Thomson PJ. Low molecular weight IgM in the sera of patients with chronic lymphocytic leukemia. Pathology. 1993;25: Owen RG, Parapia LA, Higginson J, et al. Clinicopathological correlates of IgM paraproteinemias. Clin Lymphoma. 2000;1: Hansen DA, Robbins BA, Bylund DJ, et al. Identification of monoclonal immunoglobulins and quantitative immunoglobulin abnormalities in hairy cell leukemia and chronic lymphocytic leukemia. Am J Clin Pathol. 1994;102: Sahota SS, Garand R, Bataille R, et al. V H gene analysis of clonally related IgM and IgG from human lymphoplasmacytoid B-cell tumors with chronic lymphocytic leukemia features and high serum monoclonal IgG. Blood. 1998;91: Efremov DG, Ivanovski M, Batista FD, et al. IgM-producing chronic lymphocytic leukemia cells undergo immunoglobulin isotype-switching without acquiring somatic mutations. J Clin Invest. 1996;98: Sanchez M, Almeida J, Gonzalez D, et al. Incidence and clinicobiologic characteristics of leukemic B-cell chronic lymphoproliferative disorders with more than one B-cell clone. Blood. 2003;102: Mandelli F, De Rossi G, Mancini P, et al. Prognosis in chronic lymphocytic leukemia: a retrospective multicentric study from the GIMEMA group. J Clin Oncol. 1987;5: Montserrat E. Classical and new prognostic factors in chronic lymphocytic leukemia: where to now? Hematol J. 2002;3: Rozman C, Montserrat E, Rodriguez-Fernandez JM, et al. Bone marrow histologic pattern the best single prognostic parameter in chronic lymphocytic leukemia: a multivariate survival analysis of 329 cases. Blood. 1984;64: Vallespi T, Montserrat E, Sanz MA. Chronic lymphocytic leukaemia: prognostic value of lymphocyte morphological subtypes: a multivariate survival analysis in 146 patients. Br J Haematol. 1991;77: Crespo M, Bosch F, Villamor N, et al. ZAP-70 expression as a surrogate for immunoglobulin-variable-region mutations in chronic lymphocytic leukemia. N Engl J Med. 2003;348: Wiestner A, Rosenwald A, Barry TS, et al. ZAP-70 expression identifies a chronic lymphocytic leukemia subtype with unmutated immunoglobulin genes, inferior clinical outcome, and distinct gene expression profile. Blood. 2003;101: Chan AC, Iwashima M, Turck CW, et al. ZAP-70: a 70 kd protein-tyrosine kinase that associates with the TCR zeta chain. Cell. 1992;71: Chen L, Widhopf G, Huynh L, et al. Expression of ZAP-70 is associated with increased B-cell receptor signaling in chronic lymphocytic leukemia. Blood. 2002;100: Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E 601

9 Yin et al / CLL/SLL WITH IGM PARAPROTEIN 24. Asplund SL, McKenna RW, Howard MS, et al. Immunophenotype does not correlate with lymph node histology in chronic lymphocytic leukemia/small lymphocytic lymphoma. Am J Surg Pathol. 2002;26: Huh YO, Pugh WC, Kantarjian HM, et al. Detection of subgroups of chronic B-cell leukemias by FMC7 monoclonal antibody. Am J Clin Pathol. 1994;101: Schlette E, Medeiros LJ, Keating M, et al. CD79b expression in chronic lymphocytic leukemia: association with trisomy 12 and atypical immunophenotype. Arch Pathol Lab Med. 2003;127: Lin P, Bueso-Ramos C, Wilson CS, et al. Waldenström macroglobulinemia involving extramedullary sites: morphologic and immunophenotypic findings in 44 patients. Am J Surg Pathol. 2003;27: Owen RG, Treon SP, Al-Katib A, et al. Clinicopathological definition of Waldenström s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenström s Macroglobulinemia. Semin Oncol. 2003;30: San Miguel JF, Vidriales MB, Ocio E, et al. Immunophenotypic analysis of Waldenström s macroglobulinemia. Semin Oncol. 2003;30: Lin P, Hao S, Handy BC, et al. Lymphoid neoplasms associated with IgM paraprotein: a study of 382 patients Am J Clin Pathol. 2005;123: Am J Clin Pathol 2005;123: DOI: /FDGWB5C2MYRYXH2E