Aplastic anemia and pure red cell aplasia associated with large granular lymphocyte leukemia

Aplastic anemia and pure red cell aplasia associated with large granular lymphocyte leukemia

Aplastic Anemia and Pure Red Cell Aplasia Associated With Large Granular Lymphocyte Leukemia Ronald S. Go, John A. Lust, and Robert L. Phyliky Aplasti...

132KB Sizes 0 Downloads 51 Views

Aplastic Anemia and Pure Red Cell Aplasia Associated With Large Granular Lymphocyte Leukemia Ronald S. Go, John A. Lust, and Robert L. Phyliky Aplastic anemia (AA) and pure red cell aplasia (PRCA) are two of the various types of immune-mediated cytopenias that can be associated with large granular lymphocyte (LGL) leukemia. We review the experience on LGL leukemiaassociated AA and PRCA in the published literature. In the setting of LGL leukemia, AA is found rarely, while PRCA is frequent. However, the diagnosis of LGL leukemia in the presence of pancytopenia is very challenging. In general, the clinical findings and treatment outcome are similar to the idiopathic or primary forms of AA and PRCA. Serial trials of immunosuppressive agents usually result in relatively durable remissions. In refractory cases, studies using newer immunosuppressive agents active against T cells and natural killer (NK) cells are necessary. There are many similarities in the currently known pathophysiologic mechanisms among these three disorders. As LGL leukemia is underdiagnosed, it is interesting to speculate that perhaps a significant proportion of idiopathic AA and PRCA may be, in fact, secondary to LGL leukemia. Semin Hematol 40:196-200. © 2003 Elsevier Inc. All rights reserved.

I

MMUNE-MEDIATED peripheral blood cytopenias, whether involving an isolated hematopoietic cell line or a combination of different cell lines, have well-known associations with large granular lymphocyte (LGL) leukemia.29 This is not surprising because similar mechanisms involving T cells or natural killer (NK) cells are in operation in nonclonal disorders responsible for immune cytopenias. These cytopenias may be due to the presence of cell line–specific antibody or direct cytotoxic effect of clonal LGL. The list of immune-mediated hematologic disorders associated with LGL leukemia is growing, as the recognition of the latter becomes more prevalent and now includes autoimmune hemolytic anemia, immune neutropenia, adult-onset cyclic neutropenia, immune thrombocytopenia, aplastic anemia (AA), pure red cell aplasia (PRCA), myelodysplastic syndrome, and paroxysmal nocturnal hemoglobinuria.13,15,22,29,32,45 We have previously described the Mayo Clinic series of AA or PRCA in association with LGL leukemia.12,15 The current review summarizes the collective experience of these two disorders based on published reports in the medical literature.

From the Sections of Hematology and Oncology, Gundersen Lutheran Medical Center, La Crosse, WI; and Divisions of Hematology and Hematopathology, Mayo Clinic and Foundation, Rochester, MN. Supported in part by the Gundersen Lutheran Medical Foundation (R.S.G.). Address correspondence to Robert L. Phyliky, MD, Mayo Clinic, Division of Hematology, 200 First St SW, Rochester, MN 55905. © 2003 Elsevier Inc. All rights reserved. 0037-1963/03/4003-0004$30.00/0 doi:10.1016/S0037-1963(03)00140-9

196

AA Associated With LGL Leukemia Pathogenesis The underlying mechanism of AA in the setting of LGL leukemia has not been established. On the other hand, significant progress has been made in understanding the pathophysiology of primary immunemediated AA. Central to this process is the activation and polyclonal expansion of cytotoxic T cells by specific antigens, leading to the elaboration of tissuedamaging cytokines and dysregulation of the Fas/ Fas-ligand pathway and apoptosis of hematopoietic stem cells. It is reasonable to infer that similar events occur in AA associated with LGL leukemia, as the key pathogenic cellular elements are the same and immunosuppressive agents provide similar efficacy. The role of NK cells in AA is less clear. NK cells may suppress hematopoiesis by direct cellular cytotoxicity or elaboration of cytokines such as interferon gamma and tumor necrosis factor-alpha. In AA patients, studies of NK cell activity in the bone marrow have shown conflicting results.33,51

Incidence LGL leukemia is rarely associated with AA. Review of the published literature showed that only a total of 12 patients with AA were concomitantly diagnosed with LGL leukemia.15,21,46 Nine of these cases were associated with T-cell LGL leukemia,15 while three were related to an underlying chronic NK cell lymphocytosis or NK cell LGL leukemia.21,46 Very likely LGL leukemia is underdiagnosed in this setting, especially when pancytopenia, rather than lymphocytosis, is the initial laboratory manifestation. Four percent of patients in our institutional series of T-cell LGL leukemia presented with AA.14

Seminars in Hematology, Vol 40, No 3 ( July), 2003: pp 196-200

AA and PRCA Associated With LGL Leukemia

Clinical Findings Nine of these 12 patients were males and the median age was 63 years (range, 39 to 72). In all cases, AA was present at the initial diagnosis of LGL leukemia. Absolute lymphocytosis was not observed, and all but patient one manifested relative increase in LGL in the peripheral blood smear. Immunohistochemical staining of the bone marrow consistently showed an increased proportion of activated cytotoxic T cells or NK cells. Clonal T-cell receptor (TCR) gene rearrangement was expressed in all cases of T-cell LGL. Five patients had abnormal cytogenetic studies of the bone marrow, including loss of the Y chromosome, t(1;7), and 21q⫹. Some of the pertinent associated medical conditions were rheumatoid arthritis (one patient), inflammatory bowel disease (one patient), and giant cell arteritis (one patient). One patient with NK-cell LGL leukemia transformed into an acute lymphoblastic leukemia with a similar immunophenotype after 12 months.46 In general, patients with LGL leukemia-associated AA have similar demographic characteristics, presenting signs and symptoms, and laboratory or bone marrow findings compared to patients with idiopathic AA. In our experience, the diagnosis of LGL leukemia, when AA is the presentation, can be easily missed if the blood and marrow are not carefully scrutinized.

Treatment As the number of cases reported in the literature is small, it is not possible to make specific therapeutic recommendations. In general, the treatments used were agents known to be effective in either AA or LGL leukemia, including antithymocyte globulin with cyclosporine (five patients), prednisone (five patients), cyclophosphamide (four patients), methotrexate (one patient), azathioprine (one patient), and cyclosporine as a single agent (one patient). Antithymocyte globulin with cyclosporine produced the best response with two patients each achieving complete remissions (CRs) and partial remissions (PRs). Prednisone therapy resulted in one CR and two PRs. Most responses were durable with a median duration of 3⫹ years (range, 0.8⫹ to 8⫹). However, improvements obtained with prednisone usually require maintenance therapy. No patient who received cyclophosphamide responded.

PRCA Associated With LGL Leukemia Pathogenesis Fisch et al proposed two possible mechanisms of LGL leukemia-mediated PRCA in a recent review.11 In the first, the TCR, either ␣␤ or ␥␦, recognizes a peptide expressed by erythroid precursors presented in the context of a human leukocyte antigen (HLA) class I

197

molecule, which triggers a kill signal. Normally, this killing signal is counteracted by a negative signal when killer-inhibitory receptors on the T cells bind to HLA class I molecules on the surface of erythroid precursors. However, the expression of HLA class I molecules is physiologically downregulated in erythroid precursors as they mature. This relatively low expression of HLA class I in the red cell precursors would allow LGL to cause cytolysis. The other mechanism involves major histocompatibility (MHC)-unrestricted cytotoxicity by NK or T-cell LGL, again enhanced by the deficiency of HLA class I expression in erythroid precursors, which prevents the transmission of adequate inhibitory signals. NK cell activation is dependent on the presence of circulating antibody against erythroid precursors. A similar mechanism can be operational in cytotoxic T cells, which would not directly involve the TCR (Fig 1).

Incidence Although also uncommon, PRCA is more frequently associated with LGL leukemia than with AA. We have previously reported a 7% incidence of PRCA in our cohort of LGL leukemia patients.7,14 Further review of the literature uncovered at least 78 other reports of PRCA associated with LGL leukemia.1-6,9,10,16-20,23-27,30,31,34-44,47-50 LGL leukemia was the disorder most frequently associated with PRCA in a series of 47 patients at the Mayo Clinic, diagnosed in 19% of the cases. Almost all of the LGL leukemia-associated PRCA cases reported to date involved T-cell LGL leukemia, with only three reported cases of PRCA in the setting of NK cell LGL leukemia.43,48,41

Clinical Findings There was an equal distribution between men and women. The median age at diagnosis was 56 years (range, 17 to 88). In the majority of the reviewed cases (86%), PRCA was the presenting problem and recognized in conjunction with or shortly after LGL leukemia diagnosis. Only about 10% of patients presented with hepatosplenomegaly. Lymphadenopathy was rare. A significant proportion (22%) did not have absolute lymphocytosis at diagnosis, even though most had relative increase of LGL in the peripheral blood. In cases in which more specific immunophenotype of T-cell LGL leukemia was reported, all were found to involve cytotoxic (CD8⫹) T cells. TCR gene rearrangement studies revealed most clonal lymphocytes to be of the ␣␤⫺ type. In only three cases were clonal T cells confirmed to be of the ␥␦ type.16,38,44

Associated Disorders Parvovirus B19 infection was reported in four cases when PRCA was discovered.9,24 Three of these pa-

198

Go, Lust, and Phyliky

Figure 1. Possible mechanisms of LGL leukemia-mediated PRCA. (A) In the MHC-restricted type of cytotoxicity, T-cell LGLs recognize an antigen presented by an HLA class 1 molecule. Myeloid precursors are prevented from T-LGL–induced cytolysis by virtue of having an adequate number of HLA-1 molecules on the cell surface. This is because killer-inhibitory receptors are activated. Red cell precursors are preferentially lysed as they express low concentrations of HLA-1 molecules. (B) In the non–MHCrestricted type of cytotoxicity, NK cells may be activated by the binding of an antibody to a specific cellular surface antigen. Again, cells with adequate or deficient concentrations of surface HLA-1 molecules are spared from or susceptible to cytolysis, respectively. Adapted with permission from Blackwell Synergy Publishing Co.11

tients had years of smoldering LGL leukemia not requiring therapy.9 All four patients had no other reasons to be immunosuppressed. This observation suggests that patients with LGL leukemia are at increased risk to develop parvovirus B19 infection and underscores the need to screen routinely for this infection when PRCA develops. Some other uncommon disorders have been associated with PRCA in the setting of LGL leukemia: thymoma (three patients),12,36 polyglandular autoimmune syndrome (two patients),10,48 and amegakaryocytic thrombocytopenic purpura (two patients).10,26

Treatment The use of either chemotherapeutic or immunosuppressive agents was adequately described in 65 patients. The three most commonly used treatments

were cyclophosphamide with or without concurrent low-dose corticosteroids (33 patients), cyclosporine (30 patients), and single-agent prednisone (27 patients) resulting in the following response rates (overall response/CR), respectively: 64%/64%, 73%/ 50%, and 37%/33%. Other treatments included methotrexate, antithymocyte globulin, intravenous immunoglobulin, azathioprine, chlorambucil, androgens, interferon, and combination chemotherapy regimens used in lymphoma. Cyclophosphamide and cyclosporine both produced responses in more than half of the patients treated. Responses were generally durable for several years. In some cases, molecular remission was achieved, although this was not a prerequisite for long-term clinical remission. It is important to note that it may take a few months to see a response. There should be a therapeutic trial for at least 2 to 3 months before a specific agent is deemed ineffective. In about 5% of the cases, there was no response to various immunosuppressive agents and chronic red cell transfusion was necessary. However, none of these reports described the use of newer immunosuppressive agents such as tacrolimus, mycophenolate, or alemtuzumab. In the four patients with documented parvovirus B19 infection, the bone marrow biopsies showed typical features such as giant pronormoblasts and cytoplasmic vacuolations.9,24 Two patients spontaneously recovered, even though persistence of parvovirus B19 DNA was documented in one.24 Administration of intravenous immunoglobulin resulted in CR in one patient but did not provide clinical benefit in another. Several immunosuppressive agents were subsequently utilized in the latter patient, all without response. In the rare event that thymoma and LGL leukemia are simultaneously diagnosed with PRCA, a first attempt at immunosuppressive therapy may precede thymectomy.36 One case of PRCA associated with LGL leukemia was reported to occur 6 years after thymectomy.12

Conclusion AA and PRCA can occur in association with T-cell or NK cell LGL leukemia. In general, the clinical course of AA or PRCA corresponds to the status of LGL leukemia. Remissions can be achieved in most cases with sequential immunosuppressive therapy. More intensive cytotoxic therapy such as combination chemotherapy for lymphoma is usually unnecessary. In refractory cases, study of newer immunosuppressive agents may provide new treatment options. As the clinical findings and treatment outcome of LGL leukemia-associated AA and PRCA are very similar to the idiopathic forms, and LGL leukemia remains an underdiagnosed disorder, we speculate that a signif-

AA and PRCA Associated With LGL Leukemia

icant proportion of idiopathic AA and PRCA may be, in fact, secondary to LGL leukemia.

16.

Acknowledgment We acknowledge and appreciate Drs C.Y. Li and A. Tefferi for their review of the manuscript and their past contributions; and Brenda Speltz and Carrie Speltz for their preparation of the manuscript.

17.

18.

References 1. Abkowitz JL, Kadin ME, Powell JS, et al: Pure red cell aplasia: Lymphocyte inhibition of erythropoiesis. Br J Haematol 63: 59-67, 1986 2. Akard LP, Brandt J, Lu L, et al: Chronic T cell lymphoproliferative disorder and pure red cell aplasia: Further characterization of cell-mediated inhibition of erythropoiesis and clinical response to cytotoxic chemotherapy. Am J Med 83:10691074, 1987 3. Akashi K, Shibuya T, Taniguchi S, et al: Multiple autoimmune haemopoietic disorders and insidious clonal proliferation of large granular lymphocytes. Br J Haematol 107:670-673, 1999 4. Callard RE, Smith CM, Worman C, et al: Unusual phenotype and function of an expanded subpopulation of T cells in patients with haemopoietic disorders. Clin Exp Immunol 43:497-505, 1981 5. Charles RJ, Sabo KM, Kidd PG, et al: The pathophysiology of pure red cell aplasia: Implications for therapy. Blood 87:48314838, 1996 6. Coutinho J, Lima M, dos Anjos Teixeira M, et al: Pure red cell aplasia associated to clonal CD8⫹ T-cell large granular lymphocytosis: Dependence on cyclosporin A therapy. Acta Haematol 100:207-210, 1998 7. Dhodapkar MV, Li CY, Lust JA, Tefferi A, et al: Clinical spectrum of clonal proliferations of T-large granular lymphocytes: a T-cell clonopathy of undetermined significance? Blood 84:1620-1627, 1994 8. Dhodapkar MV, Lust JA, Phyliky RL: T-cell large granular lymphocytic leukemia and pure red cell aplasia in a patient with type I autoimmune polyendocrinopathy: Response to immunosuppressive therapy. Mayo Clin Proc 69:1085-1088, 1994 9. Ergas D, Resnitzky P, Berrebi A: Pure red blood cell aplasia associated with parvovirus B19 infection in large granular lymphocyte leukemia. Blood 87:3523-3524, 1996 10. Ergas D, Tsimanis A, Shtalrid M, et al: T-gamma large granular lymphocyte leukemia associated with amegakaryocytic thrombocytopenic purpura, Sjogren’s syndrome, and polyglandular autoimmune syndrome type II, with subsequent development of pure red cell aplasia. Am J Hematol 69:132-134, 2002 11. Fisch P, Handgretinger R, Schaefer HE: Pure red cell aplasia. Br J Haematol 111:1010-1022, 2000 12. Go RS, Li CY, Tefferi A, et al: Acquired pure red cell aplasia associated with lymphoproliferative disease of granular T lymphocytes. Blood 98:483-485, 2001 13. Go RS, Tefferi A, Li CY, et al: Immune thrombocytopenia associated with lymphoproliferative disease of granular T lymphocytes. Int J Hematol 72:183, 2000 (suppl 1, abstr) 14. Go RS, Tefferi A, Li CY, et al: Lymphoproliferative disease of granular T lymphocytes: A single institution experience with 203 cases. Blood 94:515a, 1991 (suppl 1, abstr) 15. Go RS, Tefferi A, Li CY, et al: Lymphoproliferative disease of

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

32.

199

granular T lymphocytes presenting as aplastic anemia. Blood 96:3644-3646, 2000 Handgretinger R, Geiselhart A, Moris A, et al: Pure red-cell aplasia associated with clonal expansion of granular lymphocytes expressing killer-cell inhibitory receptors. N Engl J Med 340:278-284, 1999 Hansen RM, Lerner N, Abrams RA, et al: T-cell chronic lymphocytic leukemia with pure red cell aplasia: Laboratory demonstration of persistent leukemia in spite of apparent complete clinical remission. Am J Hematol 22:79-86, 1986 Hirayama Y, Nagai T, Ohta H, et al: A case of pure red cell aplasia accompanied with granular lymphocytic leukemia the tumor cells of which suppressed colony formation of BFU-E, and which was successfully treated by cyclophosphamide and cyclosporin A. Rinsho Ketsueki 38:1206-1211, 1997 Hocking W, Champlin R, Mitsuyasu R: Transient response of pure red cell aplasia to anti-thymocyte globulin in a patient with T-cell chronic lymphocytic leukemia. Am J Hematol 24:285-291, 1987 Kagami Y, Murate T, Ichikawa A, et al: Chronic cytopenia associated with T8 lymphocytosis successfully treated with glucocorticoids. Rinsho Ketsueki 31:1950-1954, 1990 Kaito K, Otsubo H, Ogasawara Y, et al: Severe aplastic anemia associated with chronic natural killer cell lymphocytosis. Int J Hematol 72:463-465, 2000 Karadimitris A, Li K, Notaro R, et al: Association of clonal T-cell large granular lymphocyte disease and paroxysmal nocturnal haemoglobinuria (PNH): Further evidence for a pathogenetic link between T cells, aplastic anaemia and PNH. Br J Haematol 115:1010-1014, 2001 Kasten-Sportes C, Zaknoen S, Steis RG, et al: T-cell receptor gene rearrangement in T-cell large granular leukocyte leukemia: Preferential V alpha but diverse J alpha usage in one of five patients. Blood 83:767-775, 1994 Kondo H, Mori A, Watanabe J, et al: Pure red cell aplasia associated with parvovirus B19 infection in T-large granular lymphocyte leukemia. Leuk Lymphoma 42:1439-1443, 2001 Kondo H, Narita K, Iwasaki H, et al: Effectiveness of cyclosporin A in a patient with pure red cell aplasia associated with T cell-lineage granular lymphocyte proliferative disorders resistant to cyclophosphamide therapy. Eur J Haematol 64: 206-207, 2000 Kouides PA, Rowe JM: Large granular lymphocyte leukemia presenting with both amegakaryocytic thrombocytopenic purpura and pure red cell aplasia: Clinical course and response to immunosuppressive therapy. Am J Hematol 49: 232-236, 1995 Kwong YL, Wong KF: Association of pure red cell aplasia with T large granular lymphocyte leukaemia. J Clin Pathol 51:672-675, 1998 Lacy MQ, Kurtin PJ, Tefferi A: Pure red cell aplasia: Association with large granular lymphocyte leukemia and the prognostic value of cytogenetic abnormalities. Blood 87:30003006, 1996 Lamy T, Loughran TP Jr: Current concepts: Large granular lymphocyte leukemia. Blood Rev 13:230-240, 1999 Linch DC, Cawley JC, MacDonald SM, et al: Acquired pure red-cell aplasia associated with an increase of T cells bearing receptors for the Fc of IgG. Acta Haematol 65:270-274, 1981 Lipton JM, Nadler LM, Canellos GP, et al: Evidence for genetic restriction in the suppression of erythropoiesis by a unique subset of T lymphocytes in man. J Clin Invest 72:694706, 1983 Loughran TP Jr, Clark EA, Price TH, et al: Adult-onset cyclic

200

33.

34. 35.

36.

37.

38.

39.

40.

41.

42.

Go, Lust, and Phyliky

neutropenia is associated with increased large granular lymphocytes. Blood 68:1082-1087, 1986 Maciejewski JP, Hibbs JR, Anderson S, et al: Bone marrow and peripheral blood lymphocyte phenotype in patients with bone marrow failure. Exp Hematol 22:1102-1110, 1994 Mamiya S, Itoh T, Miura AB: Acquired pure red cell aplasia in Japan. Eur J Haematol 59:199-205, 1997 Masuda M, Arai Y, Nishina H, et al: Large granular lymphocyte leukemia with pure red cell aplasia in a renal transplant recipient. Am J Hematol 57:72-76, 1998 Masuda M, Arai Y, Okamura T, et al: Pure red cell aplasia (PRCA) with thymoma: A possible distinct clinical entity distinct from large granular lymphocyte (LGL) leukemia. Am J Hematol 63:102, 2000 Matrai Z, Lelkes G, Milosevits J, et al: T-cell large granular lymphocytic leukemia associated with pure red cell aplasia, successfully treated with cyclophosphamide. Orv Hetil 138: 2075-2080, 1997 Mori S, Suzushima H, Nishikawa K, et al: Smoldering gamma delta T-cell granular lymphocytic leukemia associated with pure red cell aplasia. Acta Haematol 94:32-35, 1995 Nagasawa T, Abe T, Nakagawa T: Pure red cell aplasia and hypogammaglobulinemia associated with Tr-cell chronic lymphocytic leukemia. Blood 57:1025-1031, 1981 Newland AC, Catovsky D, Linch D, et al: Chronic T cell lymphocytosis: A review of 21 cases. Br J Haematol 58:433446, 1984 Oshimi K, Yamada O, Kaneko T, et al: Laboratory findings and clinical courses of 33 patients with granular lymphocyteproliferative disorders. Leukemia 7:782-788, 1993 Palomera Bernal L, Garcia Diez I: Pure red cell aplasia and neutropenia associated with chronic monoclonal T-lymphocytosis. Med Clin (Barc) 94:262-265, 1990

43. Partanen S, Ruutu T, Vuopio P, et al: Acquired pure red-cell aplasia: A consequence of increased natural killer cell activity? Leuk Res 8:117-122, 1984 44. Saitoh T, Karasawa M, Sakuraya M, et al: Improvement of extrathymic T cell type of large granular lymphocyte (LGL) leukemia by cyclosporin A: The serum level of Fas ligand is a marker of LGL leukemia activity. Eur J Haematol 65:272-275, 2000 45. Saunthararajah Y, Molldrem JL, Rivera M, et al: Coincident myelodysplastic syndrome and T-cell large granular lymphocytic disease: Clinical and pathophysiological features. Br J Haematol 112:195-200, 2001 46. Schiller GJ, Naeim F, Champlin RE: Bone marrow aplasia associated with proliferation of large granular lymphocytes and subsequent transformation to acute lymphoblastic leukemia. Am J Hematol 32:153-154, 1989 47. Shionoya S, Amano M, Imamura Y, et al: Suppressor T cell chronic lymphocytic leukaemia associated with red cell hypoplasia. Scand J Haematol 33:231-238, 1984 48. Tefferi A, Windebank KP, Veeder MH, et al: Steroid-responsive pure red cell aplasia associated with natural killer cell lymphocytosis. Am J Hematol 31:211-212, 1989 49. Yamada O, Mizoguchi H, Oshimi K: Cyclophosphamide therapy for pure red cell aplasia associated with granular lymphocyte-proliferative disorders. Br J Haematol 97:392-399, 1997 50. Yamada O, Motoji T, Mizoguchi H: Selective effect of cyclosporine monotherapy for pure red cell aplasia not associated with granular lymphocyte-proliferative disorders. Br J Haematol 106:371-376, 1999 51. Yoda Y, Kawakami Z, Abe T: Decreased frequency of bone marrow NK progenitors in aplastic anemia. Br J Haematol 71:545-549, 1989