CD8 Expression on B Cells in Chronic Lymphocytic Leukemia.
The analysis of the expression of CD antigens on cells from peripheral blood, bone marrow, and lymph nodes of patients with leukemia and lymphoma (ie, immunophenotyping) has become routine and has been used to classify some of these hematologic malignant neoplasms. A few CD antigens are strictly lineage specific (eg, CD3 is normally expressed only on T cells) but "lineage promiscuity" and "lineage infidelity" have been proposed. The former refers to a normal period in the differentiation of hematopoietic cells in which they coexpress multiple lineage markers, whereas the latter pertains to malignant transformation (clonal expansion) that leads to aberrant gene expression.
Aberrant expressions of CD antigens have been found mainly in acute but also in chronic leukemias. Although these entities sometimes are called biphenotypic leukemias, their prognosis is often not different, although in a few cases it is worse than that of the "common" leukemias.[2-4] The common B chronic lymphocytic leukemia (B-CLL) is characterized by an increased number of B lymphocytes that express CD19 and CD20 (so-called pan-B-cell markers), CD21, HLA-DR, and surface immunoglobulins. The latter are weakly expressed, giving a "dim fluorescence" when tested by flow cytometry. There is a restriction in the light chains of the surface immunoglobulins, which represents the clonal excess, a feature of malignancy. We describe a patient with B-CLL whose B cells expressed CD8, a predominantly T-cell marker also expressed in natural killer cells, and we propose that patients with this uncommon feature of B-CLL might have an increased sensitivity to fludarabine phosphate, a substance recently often used to treat patients with CLL.
REPORT OF A CASE
A 57-year-old white man was diagnosed as having stage 0 B-CLL in December 1991, after a routine physical examination and laboratory tests. His laboratory tests revealed the following values: hemoglobin, 153 g/L; platelets, 150 x [10.sup.9]/L; white blood cells (WBC), 80 x [10.sup.9]/L; and lymphocytes, 72 x [10.sup.9]/L (more than 90% of the WBC were mature lymphocytes). Lymph nodes, liver, and spleen were not palpable. A flow cytometric analysis (2-color analysis, FACScan, Becton Dickinson Biosciences, San Jose, Calif) of peripheral blood cells showed that 95% of lymphocytes were [CD5.sup.+] B cells with surface immunoglobulins of only [Lambda] type (ie, evidence of clonal excess, consistent with a diagnosis of B-CLL). Surprisingly, CD8 was expressed on virtually all B cells. The patient did not require any treatment and was followed up every 6 months in the clinic. In January 1997, he developed tiredness, recurrent respiratory tract infections, and cervical lymphadenopathy; both his spleen and liver were palpable 2 and 3 cm below the respective costal margin. A complete blood cell count showed the following values: hemoglobin, 120 g/L; platelets, 70 x [10.sup.9]/L; and WBC, 50 x [10.sup.9]/L. More than 90% of the WBC were lymphocytes. Bone marrow aspiration and biopsy (diffuse infiltration by mostly mature looking lymphocytes) also showed features consistent with the diagnosis of B-CLL. Immunophenotyping of peripheral blood and bone marrow aspirate (3-color analysis, FACSCalibur, Becton Dickinson) showed that virtually all lymphocytes expressed CD19 and CD20 and also coexpressed CD5 and CD23 (Table 1). There was a dim surface fluorescence for immunoglobulin D, and immunoglobulin M was barely detected on a few cells. The [Lambda] light chain restriction of the surface immunoglobulins was again noted, whereas B cells still showed aberrant expression of CD8 (Figure). Genotypic studies performed on peripheral blood lymphocytes demonstrated (by polymerase chain reaction) rearrangement of the [Mu] heavy chain gene.
[Figure ILLUSTRATION OMITTED]
Table 1. Flow Cytometric Analysis of Peripheral Blood and Bone Marrow Lymphocytes(*)
Peripheral Bone Blood, Marrow, Lymphocyte Markers % Positive % Positive CD4 and CD3 <1 2 CD7 <1 1 CD8 and CD3 <1 <1 CD19 99 96 CD20 99 98 CD19 and CD4 <1 <1 CD19 and CD8 99 98 CD19 and CD10 <1 <1 CD19 and CD11c 87 92 CD19 and CD5 99 97 CD19 and CD23 99 97 CD19 and CD25 <1 <1 CD19 and [Kappa] 5 3 CD19 and [Lambda] 58 95 Surface immunoglobulin G 1 8 Surface immunoglobulin M 3 6 Surface immunoglobulin D 91 86 HLA-DR 99 96
(*) Cells were gated for lymphocytes (forward scatter [FSC] vs side scatter [SSC] and CD45 vs CD14). Most antibodies used for analysis were monoclonal and purchased from Becton Dickinson Immunocytochemistry Systems.
The patient was then given a course of fludarabine phosphate therapy (25 mg/[m.sup.2] daily for 5 days). He responded well to this regimen; the WBC count decreased to 10 x [10.sup.9]/L ([is greater than] 90% lymphocytes), the spleen and liver were not palpable, and the lymphadenopathy disappeared. However, throughout the ensuing days, his blood cell count continued to decrease, and he became severely pancytopenic (hemoglobin, 80 g/L; WBC, 2.0 x [10.sup.9]/L, mostly lymphocytes; and platelets, 31 x [10.sup.9]/L). Because of the fear of further myelosuppression, he was not administered any additional fludarabine until 6 months later, when his complete blood cell count returned to normal. He was then given 2 more courses of fludarabine at 3- and 5-month intervals. Following each course, he became rapidly pancytopenic, and further treatment with fludarabine was withheld for some time. He received his fourth and last course of fludarabine in November 1998, after which he again became pancytopenic (hemoglobin, 90 g/L; WBC, 2.9 x [10.sup.9]/L, mostly lymphocytes; absolute neutrophil count, [is less than] 3 x [10.sup.9]/L; platelets, 42 x [10.sup.9]/L), and his respiratory tract infections (gram-positive cocci) recurred, for which he was treated with antibiotics but without much success. He was also given daily granulocyte colony-stimulating factor (300 [micro]g), granulocyte-macrophage colony-stimulating factor (500 [micro]g), epoietin alfa (10 000 U), and interleukin 11 (50 [micro]g/kg) subcutaneously, but he failed to respond to these substances and remained pancytopenic and severely neutropenic. He also required red blood cell and platelet transfusions almost daily. The patient then declined any further hospital treatment and returned home, where he died a few days later.
B chronic lymphocytic leukemia is the most common form of leukemia in western countries, typically afflicting persons older than 50 years. B chronic lymphocytic leukemia has a characteristic immunophenotype, with expression of pan-B-cell antigens (CD19, CD20), HLA-DR, low-intensity expression of a monoclonal surface immunoglobulin together with expression of CD5, an antigen expressed on T cells and also on a small subset of normal B cells ([B.sub.1] lymphocytes). The expression of at least one T-cell antigen (in addition to CD5) in B-CLL cells was observed in 28% of the patients described by Kurec et al. However, CD8, a glycoprotein typically expressed on the surface of thymocytes, cytotoxic and suppressor T cells, and natural killer cells, was found on B cells only in 1 of their 61 patients with B-CLL. Another retrospective analysis of 72 patients with B-CLL showed only 1 patient with CD8 expression on B cells, suggesting that this is a distinctly rare event. Indeed, only a small number of patients with B-CLL whose B cells expressed CD8 have been described, mostly as individual case reports (Table 2). CD8 expression on B cells has also been reported in lymphosarcoma cell leukemia and lymphoma.[7-9] In a recent study, Mulligan et al concluded that CD8 expression in B-CLL is perhaps more common than generally recognized. They found that this occurred in 0.5% of their 2000 cases of B-CLL. Of their 10 cases, 8 had stable stage A disease, mostly stage A(0), and 2 had progressive disease. All had moderate to high-intensity CD8 expression on the B-CLL cells, weak expression of CD22, and very weak expression of FMC-7, whereas CD23 was strongly expressed. The percentage of B cells that expressed CD8 was not given. An increased expression of CD25 and a higher-than-expected frequency of [Lambda] light chain was found. They speculated that the aberrant expression of CD8 may have resulted from alterations of the regulatory mechanisms when the [C.sub.[Kappa]] locus was deleted.
Table 2. [CD8.sup.+] B Cells in Patients With B Chronic Lymphocytic Leukemia(*)
Cases Reported/ Source, y Total Cases Therapy Perl et al, 1986 1/1 BACOP Porwit et al, 1987 1/1 None Kurec et al, 1992 1([dagger])/62 NA Ghosh et al, 1993 1/72 Chlorambucil and COP Brohee et al, 1994 1/20 NA Koelliker et al, 1994 2/2 None Attadia et al, 1996 1/1 None Avila-Carino et al, 1997 1/1 NA Fernhout et al, 1997 2/4 None Mulligan et al, 1998 10/NA NA Follow-up Prognosis After Compared Source, y Diagnosis, y With Average Perl et al, 1986 NA NA Porwit et al, 1987 >1 Good Kurec et al, 1992 NA NA Ghosh et al, 1993 1.8 Poor Brohee et al, 1994 2 Good Koelliker et al, 1994 NA Good Attadia et al, 1996 >3.5 Good Avila-Carino et al, 1997 >3 NA Fernhout et al, 1997 >5 Good Mulligan et al, 1998 NA Good
(*) BACOP indicates bleomycin, doxorubicin, cyclophosphamide, vincristine, and prednisone; COP, cyclophosphamide, vincristine, and prednisone; and NA, not available.
([dagger]) This case was only mentioned briefly.
In our laboratory's experience of more than 100 cases of B-CLL, we had never before noticed the presence of CD8 on B cells. As in most reported cases of [CD8.sup.+] B-CLL cells, our patient showed only [Lambda] light chains (dim) on the surface of B cells. This selectivity of [Lambda] light chains in this rare type of B-CLL, in contrast to the typical B-CLL, is unexplained. Our patient did not show expression of CD25 on is B cells, in contrast to most of the published cases.[10,12] In one report of B-CLL, CD8 expression on B cells was suggested as a marker of aggressive disease; however, other reports did not show any difference in presentation or survival of these patients when compared with patients with classic B-CLL, strongly suggesting that CD8 antigen is not a marker of aggressive disease.
Our patient showed increased sensitivity to fludarabine therapy. According to the drug's manufacturer, myelosuppression manifested as neutropenia (neutrophils, [is less than] 0.5 x [10.sup.9]/L), thrombocytopenia, and anemia may be severe and cumulative and can be found in 59% of patients treated with this drug, although only 11% had infections. Because this treatment was more recently introduced than other treatments for CLL, the suggestion that patients with [CD8.sup.+] B-CLL cells are more sensitive to fludarabine therapy remains to be substantiated by other reports. Indeed, in most other reports this treatment was not used. In the series of Mulligan et al, only one patient was treated with fludarabine, and this patient was followed up for only 1.5 years. Although it is not clear whether this apparent sensitivity to fludarabine is related to the aberrant [CD8.sup.+] B cells, we proposed that institution of fludarabine as a first line therapy for patients with similar phenotype should be followed up carefully for increased susceptibility to myelotoxicity and the dose of the drug should be decreased.
The generous grant from the John R. Oishei Foundation is gratefully acknowledged.
[1.] Greeves MF, Chan LC, Furley AJW, Watt SM, Molgaard HV. Lineage promiscuity in hemopoietic differentiation and leukemia. Blood. 1986;67:1-11.
[2.] Smith LJ, Curtis JE, Messner HA, Senn JS, Furthmayr H, McCulloch EA. Lineage infidelity in acute leukemia. Blood. 1983;61:1138-1145.
[3.] Ferrara F, DeRosa C, Fasanaro A, et al. Myeloid antigen expression in adult acute lymphoblastic leukemia: clinicohematological correlations and prognostic relevance. Hematol Pathol. 1990;4:93-98.
[4.] Ball ED, Davis RB, Griffin JD, et al. Prognostic value of lymphocyte surface markers in acute myeloid leukemia. Blood. 1991;77:2242-2250.
[5.] Kurec AS, Threatte GA, Gottlieb AJ, Smith JR, Anderson J, Davey FR. Immunophenotypic classification of chronic lymphocytic leukaemia (CLL). Br J Haematol. 1992;81:45-51.
[6.] Ghosh K, Sivakumaron M, Wood JK. Aberrant CD8 antigen expression in a patient with chronic lymphocytic leukaemia showing unusual disease progression. Br J Haematol. 1993;85:205-206.
[7.] Schroff RW, Foon KA, Billing RJ, Fahey JL. Immunologic classification of lymphocytic leukemias based on monoclonal antibody-defined cell surface antigens. Blood. 1982;59:207-215.
[8.] Aisenberg AC, Bloch KJ, Wilkes BM. Malignant lymphoma with dual B and T cell markers: analysis of the neoplastic cells with monoclonal antibodies directed against T cell subsets. J Exp Med. 1981;154:1709-1714.
[9.] Hsu CCS, Marti GE, Schrek R, Williams RC. Lymphocytes bearing B- and T-cell markers in patient with lymphosarcoma cell leukemia. Clin Immunol Immunopathol. 1975;3:385-395.
[10.] Mulligan SP, Dao LP, Francis SE, et al. B-cell chronic lymphocytic leukemia with CD8 expression: report of 10 cases and immunochemical analysis of the CD8 antigen. Br J Haematol. 1998;103:157-162.
[11.] Peri A, Szigeti A, Gergely P, Feher J, Magyarosi E. Abrogation by chemotherapy of T-cell antigen expression in B-cell chronic lymphocytic leukemia. N Engl J Med. 1986;314:186-187.
[12.] Porwit A, Borgonovo L, Osby E, Lenkei R, Smith CI, Hammarstrom L. B-cell chronic lymphocytic leukaemia with aberrant expression of CD8 antigen. Eur J Haematol. 1987;39:311-317.
[13.] Brohee D, Cauchie P, Neve P. Co-expression of CD2 or CD8 antigens in B-cell chronic lymphocytic leukaemia: a flow-cytometry analysis of 3 cases. Acta Clin Belg. 1994;49:183-186.
[14.] Koelliker DD, Steele PE, Hurtubise PE, Flessa HC, Sheng YP, Swerdlow SH. [CD8.sup.+] B-cell chronic lymphocytic leukemia: a report of two cases. Am J Clin Pathol. 1994;102:212-216.
[15.] Attadia V, Alosi M, Improta S, Baccarini M, De Paoli P. Immunophenotypic and molecular genetic characterization of a case of [CD8.sup.+] B cell chronic lymphocytic leukaemia. Leukemia. 1996;10:1544-1550.
[16.] Avila-Carino J, Lewin N, Tomita Y, et al. B-CLL cells with unusual properties. Int J Cancer. 1997;70:1-8.
[17.] Fernhout F, Dinkelaar RB, Hagemeijer A, Groeneveld K, van Kammen E, van Dongen JJM. Four aged siblings with B cell chronic lymphocytic leukemia. Leukemia. 1997;11:2060-2065.
Accepted for publication January 11, 2000.
From the Division of Hematology/Oncology, Departments of Medicine (Drs Islam, Akhter, and Ambrus) and Pathology (Drs Vladutiu and Sands), State University of New York at Buffalo, School of Medicine and Biomedical Sciences, and Kaleida Health/Buffalo General Hospital (Drs Islam, Vladutiu, Akhter, Sands, and Ambrus and Ms Donahue), Buffalo, NY.
Reprints: Anwarul Islam, MD, PhD, FRCPath, Division of Hematology/Oncology, Department of Medicine, Buffalo General Hospital/Kaleida Health, 100 High St, Buffalo, NY 14203 (e-mail: aislam@acsu. buffalo.edu).
|Printer friendly Cite/link Email Feedback|
|Author:||Islam, Anwarul; Vladutiu, Adrian O.; Donahue, Theresa; Akhter, Selina; Sands, Amy M.; Ambrus, Julian|
|Publication:||Archives of Pathology & Laboratory Medicine|
|Date:||Sep 1, 2000|
|Previous Article:||Malignant Craniopharyngioma.|
|Next Article:||Glandular Malignant Peripheral Nerve Sheath Tumor.|