Evans syndrome complicating fludarabine treatment for advanced B-CLL.

L. Shvidel,M. Shtarlid,A. Klepfish,E. Sigler,A. Berrebi

DOI: https://doi.org/10.1046/J.1365-2141.1997.D01-3688.X

1997-12-01

British Journal of Haematology

Abstract:Fludarabine is a potent agent for the treatment of early as well as advanced chronic lymphocytic leukaemia (CLL). Major side-effects include cytotoxicity and immunosuppression directed against T cells which may induce autoimmune disorders, mainly autoimmune haemolytic anaemia (AIHA) (Di Raimondo et al, 1993; Myint et al, 1995). We report an unusual occurrence of severe Evans syndrome in a CLL patient treated with fludarabine. A 49-year-old male was diagnosed with stage Rai I BCLL in 1991. Absolute lymphocyte count was 15 × 10/l, 65% of the lymphocytes expressed HLA-DR, CD20, CD5 and 30% weak surface IgG. Total T cells were 20%. Quantitative Ig serum levels revealed normal amounts of IgG and IgA and a mildly decreased IgM 0·8 g/l (normal 1·2 6 0·35 g/l). Haemoglobin (Hb) level and platelet count were normal. Coombs test was negative. No treatment was given. 2 years later the lymphocyte count increased to 45 × 10/l, macrolymphadenopathies and splenomegaly developed, and chemotherapy with chlorambucil and prednisone was initiated with mild response. Then fludarabine treatment was started (six cycles of 50 mg/d for 5 d) with significant improvement. The lymph node masses were reduced in all areas, the spleen was not palpable, the lymphocyte count decreased to 10 × 10/l, and the Hb level and platelet count remained stable. This response lasted for a year. In April 1996 the patient presented again with macropolyadenopathies, mild splenomegaly and increase in lymphocyte count to 60 × 10/l. Hb and platelets were unchanged. Two cycles of fludarabine (50 mg/d for 5 d) combined with 500 mg cyclophosphamide were given. 6 weeks after this treatment the patient presented with severe gingival bleeding, purpura and subicterus. The blood count showed Hb 11 g/dl, reticulocytes 2·5%, lymphocyte count 40 × 10/l, platelets 12 × 10/l. DAT (Coombs test) was positive with IgGþC (warm antibodies), platelet antibodies were detected with a strong IgG specificity by immunofluorescence. Bone marrow showed diffuse lymphocytosis, normoblastosis and several immature megakaryocytes. The patient was diagnosed as suffering from acute Evans syndrome and was treated successively with prednisone, Ig i.v. and vincristine. For the next 2 weeks the platelet count remained <20 × 10/l, accompanied by conjunctival and mucosal bleeding, which required continuous care. Then cyclosporine 200 mg/d was added, and a response was noted 3 weeks later, by a progressive increase in Hb level to 13 g/dl and platelets to 115 × 10/l. Cyclosporine and prednisone were given for 3 months and the patient recovered from the Evans syndrome. The Coombs test remained positive, but with no evidence of active haemolysis. Interestingly, the IgG level, which was normal before fludarabine treatment, decreased to 5·41 g/l, and the absolute number of CD4 cells decreased from 2000/ml to 150/ml, when Evans syndrome developed. We believe that our patient developed an acute Evans syndrome as a complication of exposure to fludarabine. The clinical picture was dramatic and life-threatening. De novo autoimmune haemolytic anaemia or relapse of preexisting AIHA have often been described in series of CLL patients received fludarabine (Di Raimondo et al, 1993; Myint et al, 1995). Immune thrombocytopenia (ITP) or recurrence of ITP have occasionally been reported (Montillo et al, 1994). However, we have been unable to find any published reports on the occurrence of Evans syndrome complicating fludarabine therapy. As in our case, the most important toxicity of fludarabine is Tlymphopenia, which may induce dysregulation in T-cell function and produce autoantibodies directed against red cells and platelet antigens. Sequential fludarabine should be given carefully in order to reduce the incidence of such dramatic immune complications.