* Context.-CD79b is a relatively newly characterized Bcell marker that is expressed in a minority of chronic lymphocytic leukemia (CLL) cases.
Objective.-To systematically correlate CD79b expression with specific morphologic and immunophenotypic findings and trisomy 12.
Design.-We assessed CD79b expression in 100 consecutively accrued CLL cases that were also analyzed by conventional cytogenetics. Based on the association between trisomy 12 and CD79b expression, we then assessed 43 additional CLL cases with trisomy 12. CD79b expression was correlated with morphology and expression of other immunophenotypic markers.
Results.-Eighteen (18%) of 100 consecutively accrued
cases were CD79b positive. No significant association was found between CD79b expression and atypical morphology. CD79b expression correlated with CD22 and FMC7 positivity. Eight (8%) cases had trisomy 12; 4 (50%) of these were CD79b positive, suggesting an association with trisomy 12. Examination of a second group of 51 CLL cases with trisomy 12 (including 8 cases from the initial study group) showed that CD79b was positive in 26 cases (49%), a frequency significantly higher than that of the consecutively accrued CLL cases without trisomy 12 (P
Conclusions.-We conclude that CD79b immunoreactivity is positive in approximately 20% of CLL cases and that expression correlates with trisomy 12 and atypical immunophenotypic findings.
(Arch Pathol Lab Med. 2003;127:561-566)
CD79b, a B-cell-specific polypeptide that dimerizes with CD79a to form part of the immunoglobulin antigen receptor, plays a role in modulating signal transduction. Antibodies to this antigen also have been shown to be useful in distinguishing chronic lymphocytic leukemia (CLL) from other B-cell neoplasms, inasmuch as most cases of CLL are CD79b negative and the majority of other types of B-cell neoplasms are CD79b positive.1-5 Inclusion of CD79b in the CLL scoring system, originally proposed by Matutes et al,6 has been shown to enhance the diagnostic utility of this scoring system in distinguishing CLL from other B-cell lymphoproliferative diseases.7 In view of its diagnostic utility, recommendations have been made to add anti-CD79b to antibody panels used in flow cytometry immunophenotypic analysis of B-cell lymphoproliferative disorders.4 The exact frequency of CD79b expression in CLL is not clear, since the proportion of CLL cases showing CD79b expression has been reported to vary from 5% to 27% of cases in various studies.1-4,8,9
It also remains unclear whether CD79b expression correlates with any specific morphologic or immunophenotypic features in CLL. A small number of studies have assessed the relationship between atypical morphology and CD79b, and the conclusions are contradictory.9,10 Similarly, a few studies have correlated CD79b with the presence of atypical immunophenotypes in CLL, but no comprehensive studies are available in the literature. We also were unable to find any studies that systematically assessed CD79b expression in CLL cases with trisomy 12.
In this study, we first correlated the overall frequency of CD79b expression in a group of 100 CLL cases consecutively accrued at our institution. We then correlated CD79b expression with various morphologic and immunophenotypic features, as well as with trisomy 12, and found that expression of CD79b is commonly positive in CLL cases with an atypical immunophenotype or trisomy 12.
MATERIALS AND METHODS
The initial study group included 105 consecutively accessioned cases of CLL that were studied at The University of Texas M. D. Anderson Cancer Center in a 6-month period (January 25, 2001 to July 2, 2001). For inclusion in this study, all cases had to have bone marrow aspiration and biopsy samples with flow cytometric immunophenotypic analysis and conventional cytogenetic studies performed at the time of initial evaluation. The diagnosis of CLL in all cases was supported by a combination of morphologic and immunophenotypic features, as described in the World Health Organization (WHO) classification.5 Five cases were excluded because insufficient metaphases were obtained for cytogenetic analysis.
Our initial results also led us to analyze a second group of 51 cases of CLL with trisomy 12 identified in the files of the Cytogenetics Laboratory in our department. These cases were accessioned between February 25, 1999 and December 10, 2001 and included 8 cases in the initial study group. This group consisted of 48 cases of CLL with trisomy 12 abnormality in 2 or more metaphases and 3 cases that had only 1 metaphase with trisomy 12. In the latter group of 3 cases, trisomy 12 was confirmed to be clonal by fluorescence in situ hybridization studies.
Ninety-six cases from the consecutive CLL group, including the 8 cases with trisomy 12, and 43 cases of the trisomy 12 CLL group were examined for the presence of atypical versus typical CLL morphology (Figure 1). Wright-Giemsa-stained bone marrow aspirate smears were reviewed independently by the authors. The criteria used for atypical morphology in CLL, described by Matutes et al,2 included 2 subtypes: cases with more than 10% prolymphocytes and cases with more than 15% lymphoid cells that have a lymphoplasmacytoid appearance or clefted nuclear contours in smears. In each case, the percentage of atypical lymphocytes was determined by a manual 100-cell differential count of the neoplastic cells.
We applied the following morphologic criteria to define atypical lymphoid morphology. Lymphoplasmacytoid cells were defined as variably sized lymphocytes (1.5-2.5 times the size of an erythrocyte) with moderately abundant, mildly basophilic cytoplasm and eccentrically placed nuclei. The nuclear chromatin ranged from condensed to slightly open, and occasional cells had small nucleoli (Figure 2). Clefted or irregular lymphocytes were generally small (2 times the size of an erythrocyte) and displayed irregular to deeply clef ted nuclei and scant cytoplasm (Figure 3). Prolymphocytes were of intermediate size (2-3 times the size of an erythrocyte) with a prominent, single, central nucleolus and moderately abundant pale cytoplasm and did not have plasmacytoid features. Discrepancies among observers were resolved by consensus using a multiheaded microscope.
Flow cytometric immunophenotypic studies were performed from bone marrow aspirate material at our institution during the initial evaluation of the patient. All samples were assessed using 3-color flow cytometric analysis and a FACScan (Becton Dickinson, San Jose, Calif) instrument. Lymphocytes were gated for analysis using CD45 expression and right-angle light scatter as described by others.11 Fluorescein isothiocyanate- and phycoerythrin-conjugated IgGI and IgG2 antibodies were used as negative controls, and cursors were set to include more than 95% of events as negative. The panel of antibodies, conjugated to fluorescein isothiocyanate, phycoerythrin, or allophycocyanin, included reagents specific for CD3, CD5, CD11c, CD19, CD20, CD22, CD23, CD38, CD79b (CB3-1), FMC7, and immunoglobulin Kappa and lambda light chains (Becton Dickinson). The intensity of CD20 and immunoglobulin light chain expression was also recorded.
As described in the WHO classification, a typical CLL immunophenotype was defined as a monotypic B-cell neoplasm positive for CD5 and CD23 with dim expression of CD20 and surface immunoglobulin. Immunoreactivity for CD22, CD79b, and FMC7 is negative.5
Presence or absence of antigen expression was based on visual inspection of scattergrams. In a positive case, the antigen was clearly expressed (either dim or bright) by a substantial subset of the gated events, at least 25%, and usually by the majority of the events. Cursors were not used in this assessment to assess percentages.
If the original scattergrams were unavailable for review, a conservative estimate of 40% or more of the positive events within the gate was used to establish the presence of antigen expression. Bright or dim surface immunoglobulin expression and CD20 expression were scored from 1 to 3, based on the log scale of the mean channel intensity. In other words, cases with a mean channel intensity of 1, 2, or 3 were judged as weak, moderate, or strong, respectively.
To quantify immunophenotypic features in CLL we used the scoring system established by Matutes et at.6 The score is based on expression of CD5 (1 point), CD23 (1 point), low-level (dim) surface immunoglobulin (1 point), and lack of CD22 (1 point) or FMC7 (1 point). Cases with scores of 4 or 5 are considered typical for CLL. Scores of 3 or less are considered atypical for CLL.
Conventional G-band karyotype analysis was performed on bone marrow aspirate specimens from all patients. Cells were placed in 10 mL Ham F10 medium with 20% fetal calf serum at a concentration of 2 to 4 x 10^sup 6^ nucleated cells/mL. The culture was incubated overnight at 37 deg C (approximately 24 hours). Standard harvesting procedures were used. Colcemid (0.1 mL/10 mL) was added to the culture for 30 minutes. For the hypotonic treatment, 0.075M potassium chloride was used for 30 minutes at room temperature. The fixation procedure consisted of 3 changes of methanol-glacial acetic acid (3:1) with a 10-minute waiting period between each change. The Thermaton drying chamber (Thermaton Industries, Holland, Mich) was used for slide preparation. Slides were placed in a 60 deg C oven overnight in preparation for Giemsa trypsin G banding. Twenty to 30 metaphases were analyzed for each case. The karyotypes were written using the International System for Human Cytogenetic Nomenclature.12
Statistical analysis was performed using a 2-tailed Fisher exact test. A P value of .05 or less was considered statistically significant.
Consecutively Accrued CLL Cases (n = 100)
Frequency of CD79b Expression.-We evaluated the overall frequency of CD79b expression using 100 consecutively accrued cases of CLL, all of which were positive for monotypic immunoglobulin light chain and CDS. CD79b expression was detected in 18 (18%) cases.
Clinical Data (Table 1).-Our study included 64 men and 36 women; the median age was 59 years (range, 2480 years). Thirty-eight patients received chemotherapy before arriving at our institution. Seven of the 18 patients with CD79b-positive CLL had been treated previously. No significant association was found between CD79b positivity and age, gender, or prior chemotherapy.
Morphologic Results.-Bone marrow aspirate smears of 96 consecutively accrued CLL cases were available for review and were examined for atypical morphology, which was detected in 10 cases (10%). No correlation was detected between atypical morphology and CD79b expression. However, 5 of 10 cases with atypical morphology had trisomy 12, which was significant (P
Association Between CD79b Expression and Atypical Immunophenotype.-Ten (55.6%) of 18 CD79b-positive CLL cases had a typical immunophenotype with scores of 5 (n = 7) or 4 (n = 3). The remaining 8 cases had an atypical immunophenotype with scores of 3 (n = 5), 2 (n = 2), or 1 (n = 1). In comparison, 76 (93%) of 82 CD79bnegative CLLs had a typical immunophenotype with scores of 5 (n = 52) or 4 (n = 24), and the remaining 6 (7%) cases had a score of 3. The higher frequency of lower scores in CD79b-positive CLL compared with CD79b-negative CLL was significant (P
CD79b positivity also correlated with expression of individual markers. The presence of CD79b significantly correlated with CD22 (P
Cytogenetic Results.-Twenty-six cases (26%) of CLL had cytogenetic abnormalities detected by conventional cytogenetics. Twelve cases had complex karyotypic abnormalities, and 14 cases had a single karyotypic abnormality. Trisomy 12 was the most frequent abnormality and was seen in 8 cases. Abnormalities involving region 11(q13-14) were the next most common cytogenetic defect, affecting 6 cases (24%); 4 of these were deletions. Other chromosomes that were frequently abnormal included chromosomes 17 (6 cases; 24%) and 13 (5 cases; 19%). Deletion of 13(q14) was seen in 1 case, in association with other karyotypic abnormalities. The frequency of cytogenetic abnormalities in CD79b-positive and CD79b-negative cases was not statistically significant. However, the association of CD79b with trisomy 12 was significant (P = .04).
Cases of CLL With Trisomy 12 (n = 51)
Frequency of CD79b Expression.-Based on the initial correlation between CD79b and trisomy 12 in consecutively accrued CLL cases, we further assessed CD79b frequency in 51 cases of CLL with trisomy 12, all of which expressed monotypic immunoglobulin and CDS. This group included 8 cases initially identified in the consecutively accrued study group. CD79b expression was detected in 26 (51%) of 51 CLL cases with trisomy 12 (Figure 4). This frequency of CD79b expression was significantly higher than that of CLL cases without trisomy 12 (14/92; 15%; P
Clinical Data (Table 1).-The trisomy 12 CLL group included 36 men and 15 women. The median age was 56 years (range, 27-78 years). Twenty-five patients received chemotherapy prior to evaluation at our institution. Twelve (46%) of 26 patients with CD79b-positive CLL had been previously treated. There was no significant correlation between CD79b expression and prior chemotherapy.
Morphologic Results.-Twenty-three CLL cases with trisomy 12 (45%) had atypical morphology. Nine (35%) of 26 CD79b-positive cases compared with 14 (56%) of 25 CD79b-negative cases had atypical morphology. Thus, CD79b expression did not significantly correlate with atypical morphology in this group.
Frequency of Atypical Immunophenotype (Tables 3 and 4).-Comparing CLL cases with (n = 51) or without (n = 92) trisomy 12, trisomy 12 CLL cases were more likely to express CD22 (P = .05), FMC7 (P
Association Between CD79b Expression and Immunophenotypic Results.-Eleven (42%) of 26 CD79b-posifive CLL cases with trisomy 12 had a typical immunophenotype with scores of 5 (n = 3) or 4 (n = 8), whereas 15 (58%) cases had scores of 3 (n = 10), 2 (n = 4), or 1 (n = 1). In comparison, 18 (72%) of 25 CD79b-negative CLLs with trisomy 12 had a typical immunophenotype with scores of 5 (n = 9) or 4 (n = 9), and the remaining 7 cases (28%) had scores of 3 (n = 6) or 2 (n = 1). The higher frequency of low scores in CD79b-positive cases compared with CD79b-negative CLL cases was significant (P
CD79b is a useful marker for distinguishing CLL from other B-cell lymphoproliferative disorders, and antiCD79b antibodies have been incorporated into various antibody panels for flow cytometry immunophenotypic analysis of those disorders. While its diagnostic utility is well documented, the frequency of CD79 expression in CLL and the clinical or pathologic meaning of its expression remain unclear. We found that 18% of CLL cases express CD79b, which is within the range of 5% to 27% reported in the literature. We believe that our estimate of the frequency of CD79b positivity is accurate, inasmuch as this value was derived from a relatively large number of consecutively accrued CLL cases at our hospital. Because our institution is a referral center, there may be some selection bias influencing our estimate of the frequency of CD79b expression. Although approximately half of the patients had a history of chemotherapy, we did not identify any significant association between CD79b expression and prior treatment. Thus, if selection bias is present, it appears to be unrelated to prior therapy.
The immunophenotype of most cases of CLL is distinctive. In the WHO classification for hematologic neoplasms, the typical immunophenotype for CLL includes expression of CDS, CD19, CD23, CD43, CD79a, and relatively low levels of surface immunoglobulin and CD20. In addition, CD10, CDllc, CD22, FMC7, and CD79b and cyclin Dl are typically absent.5 Cases of CLL with immunophenotypic features that differ from this typical profile have been described previously.13 As variations in the immunophenotypic profile of CLL can lead to difficulty in establishing the diagnosis, a scoring system was proposed by Matutes and colleagues6,14 to more accurately distinguish CLL from other B-cell lymphoproliferative disorders. This system is based on the expression pattern of 5 antigens, namely, CDS, CD23, CD22, FMC7, and the intensity of surface immunoglobulin. Typical cases of CLL have scores of 4 or 5. However, it is unclear whether cases with scores of 3 or less should be classified differently or simply designated as CLL, with acknowledgment of the atypical immunophenotype.5 It also has been recognized that atypical immunophenotypes are more often detected in cases with trisomy 12 and/or atypical morphology,13,15,16 In our series, CLL cases with low scores were associated with trisomy 12, usually with additional karyotypic abnormalities. In addition, the majority of CLL cases with low scores were positive for CD79b.
One may challenge the diagnosis of CLL for the 10 cases with scores of 2 or less. However, all patients presented with leukocytosis, minimal lymphadenopathy, and splenomegaly. Morphologically, 7 of 10 cases had features typical of CLL and all were CD5 positive. The lack of substantial lymphadenopathy or splenomegaly is more consistent with CLL than other entities, such as splenic marginal zone lymphoma or mantle cell lymphoma. Nor did these patients appear to have a more aggressive disease course, since the median follow-up for the group was 51 months (range, 12-84 months), with only 1 patient death at 53 months postdiagnosis. Thus, we believe that while these cases have an atypical immunophenotype, the clinical and morphologic features are more consistent with CLL than other lymphoproliferative disorders.
One of the main objectives of this study was to correlate CD79b expression with other parameters in CLL. We did not find a statistically significant correlation between CD79b and atypical morphology. Several recent studies have compared CD79b expression and atypical morphology, and the conclusions are contradictory. Some studies found a positive correlation between CD79b and atypical morphology.9,17 In contrast, no correlation was identified by others.10 We believe that at least 2 factors may account for this discrepancy. First, some of the previous studies involved a relatively small number of cases. Second, the discrepancy may be related to the use of different diagnostic criteria for atypical morphology.15,18-20 In this study, we employed the criteria proposed by Matutes et al,14 in which cases with more than 10% prolymphocytes (CLL/ PL) or more than 15% of cells with lymphoplasmacytoid features and/or cleaved nuclear contours are classified as having atypical morphology in CLL. In addition, we believe that the definitions of the different cell types is prone to interobserver variation. The definition of prolymphocytes and degree of lymphoplasmacytoid differentiation or irregularity of the nuclear contours has varied between studies.
We noted an association between CD79b expression and trisomy 12. Previous studies of trisomy 12 in CLL have shown that these cases are characterized by a high frequency of atypical immunophenotypes and morphology. Using 51 cases of CLL with trisomy 12, we confirmed that the frequency of CD79b expression is significantly higher in trisomy 12 CLL cases than in CLL cases without trisomy 12. CD79b expression and bright surface immunoglobulin expression were the most common atypical immunophenotypic markers observed in trisomy 12 CLL cases. However, CD79b expression or an atypical immunophenotype is not predictive of trisomy 12. It is also important to note that more than half (57%) of trisomy 12 CLL cases had a CLL score of 4 or 5. Therefore, while low CLL scores raise the likelihood of a CLL case having trisomy 12, a high score does not exclude trisomy 12 in an individual case.
In keeping with earlier studies, we found that atypical morphology is generally infrequent in CLL; we identified atypical morphology in 10% of CLL cases in our consecutively accrued group, but it was more frequent in our trisomy 12 CLL cases.2,13,15,20,21 Unlike other series, we did not identify cases with more than 10% prolymphocytes (CLL/PL) in either group. 11,21 Interestingly, 28 CLL cases with trisomy 12 did not have atypical morphology as defined in this study. Although this group did not show specific morphologic changes, the neoplastic lymphocytes in these 28 cases generally had finer chromatin or more evidence of lymphoplasmacytoid features than cases without trisomy 12. However, these findings were subtle and difficult to appreciate when the cases were assessed individually (as would occur in normal practice) and when not directly compared with CLL cases without trisomy 12.
To conclude, CD79b expression is relatively uncommon in CLL, accounting for 18% of cases in a consecutively accrued series of cases at our institution. A higher frequency of CD79b expression is found in CLL cases that have trisomy 12 and/or an atypical immunophenotype.
The authors thank Kimberly J. Hayes, BS, and Susan Lerner, MS, BA, for their assistance with cytogenetic and clinical data collection, and Patrick McNeil for his assistance with digital imaging.
1. Zomas AP, Matutes E, Morilla R, Owusu-Ankomah K, Seon BK, Catovsky D. Expression of the immunoglobulin-associated protein B29 in B cell disorders with the monoclonal antibody SNB (CD79b). Leukemia. 1996;10:1966-1970.
2. Matutes E, Oscier D, Garcia-Marco J, et al. Trisomy 12 defines a group of
CLL with atypical morphology: correlation between cytogenetic, clinical and laboratory features in 544 patients. Br) Haematol. 1996;92:382-388.
3. Thompson AA, Do HN, Saxon A, Wall R. Widespread 829 (CD79b) gene defects and loss of expression in chronic lymphocytic leukemia. Leuk Lymphoma. 1999;32:561-569.
4. McCarron KF, Hammel JP, Hsi ED. Usefulness of CD79b expression in the diagnosis of B-cell chronic lymphoproliferative disorders [published correction appears in Am I Clin PathoL 2000;1 14:4891. Am J Clin Pathol. 2000;114:4891. Am J Clin Fathol. 2000;113:805813.
5. Muller-Hermelink HK, Catovsky D, Montserrat E, Harris NL. Chronic lymphocytic leukaemia/small lymphocytic lymphoma. In: Jaffe ES, Harris NL, Stein H, Vardiman JW, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of Haematopoietic and Lymphoid issues. Lyon, France: International Agency for Research on Cancer (]ARC) Press; 2001:127130.
6. Matures E, Owusu-Ankomah K, Morilla R, et al. The immunological profile of B-cell disorders and proposal of a scoring system for the diagnosis of CLL. Leukemia. 1994;8:1640-1645.
7. Moreau EJ, Matutes E, A'Hern RP, et al. Improvement of the chronic lymphocytic leukemia scoring system with the monoclonal antibody SN8 (CD79b). Am J Clin Pathol. 1997;108:378-382.
8. Molica S, Vitelli G, Levato D, et al. CD27 in B-cell chronic lymphocytic leukemia: cellular expression, serum release and correlation with other soluble molecules belonging to nerve growth factor receptors (NGFr) superfamily. Haematologica. 1998;83:398-402.
9. Garcia Vela J, Delgado 1, Benito L, et al. CD79b expression in B cell chronic lymphocytic leukemia: its implication for minimal residual disease detection. Leukemia. 1999;13:1501-1505.
10. Frater JL, McCarron KF, Hammel JP, et al. Typical and atypical chronic lymphocytic leukemia differ clinically and immunophenotypically. Am I Clin Pathol. 2001;116:655-664.
11. Borowitz MJ, Guenther KL, Shults KE, Stelzer GT. Immunophenotyping of acute leukemia by flow cytometric analysis: use of CD45 and right-angle light scatter to gate on leukemic blasts in three-color analysis. Amj Clin PathoL 1993; 100:534-540.
12. Mitelman F, ed. ISCN 1995: An International System for Human Cytogenetic Nomenclature. Basel, Switzerland: Karger; 1995.
13. Criel A, Wlodarska I, Meeus P, et al. Trisomy 12 is uncommon in typical chronic lymphocytic leukaemias. Br] Haematol. 1994;87:523-528.
14. Matutes E, Polliack A. Morphological and immunophenotypic features of chronic lymphocytic leukemia. Rev Clin Exp Hematol. 2000;4:22-47.
15. Criel A, Verhoef G, Vlietinck R, et al. Further characterization of morphologically defined typical and atypical CLL: a clinical, immunophenotypic, cytogenetic and prognostic study on 390 cases. Br Haematol. 1997;97:383-391.
16. Su'ut L, O'Connor SJ, Richards SJ, et al. Trisomy 12 is seen within a specific subtype of B-cell chronic lymphoproliferative disease affecting the peripheral blood/bone marrow and co-segregates with elevated expression of CD1 a. Br Haematol. 1998;101:165-170.
17. D'Arena G, Cascavilla N, Musto P, Colella Bisogno R, Pistoiese G, Carotenuto M. CD79b expression in B-cell chronic lymphocytic leukemia. Haematologica. 2000;85:556-557.
18. Hjalmar V, Kimby E, Matutes E, et al. Trisomy 12 and lymphoplasmacytoid lymphocytes in chronic leukemic B-cell disorders. Haematologica. 1998;83:602609.
19. Bennett JM, Catovsky D, Daniel MT, et al. Proposals for the classification of chronic (mature) B and T lymphoid leukaemias: French-American-British (FAB) Cooperative Group. I Clin Pathol. 1989;42:567-584.
20. Finn WG, Thangavelu M, Yelavarthi KK, et al. Karyotype correlates with peripheral blood morphology and immunophenotype in chronic lymphocytic leukemia. Am I Clin Pathol. 1996;105:458-467.
21. Que TH, Marco JG, Ellis J, et al. Trisomy 12 in chronic lymphocytic leukemia detected by fluorescence in situ hybridization: analysis by stage, immunophenotype, and morphology. Blood. 1993;82:571-575.
Ellen Schlette, MD; L. Jeffrey Mederios, MD; Michael Keating, MB, BS; Raymond Lai, MD, PhD
Accepted for publication December 5, 2002.
From the Department of Hematopathology, The University of Texas M. D. Anderson Cancer Center, Houston.
Reprints: Ellen Schlette, MD, Department of Hematopathology, University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Unit 72, Houston, TX 77030 (e-mail: firstname.lastname@example.org).
Copyright College of American Pathologists May 2003
Provided by ProQuest Information and Learning Company. All rights Reserved