Alemtuzumab for Giant Cell Hepatitis With Autoimmune Hemolytic Anemia

Abstract

A 10-month-old boy was diagnosed with celiac disease (CD) on the basis of positive antiendomysial antibody immunofluorescent test and human recombinant anti-tissue transglutaminase autoantibodies and duodenal histology. He was started on a gluten-free diet. At 14 months of age, the child, previously untransfused, developed severe anemia (hemoglobin 3.7 g/dL, mean blood cell volume 104 fL, reticulocytes 33%); blood smear examination revealed schistocytes and direct antiglobulin test was strongly positive for immunoglobulin G and C3d. Warm-type autoimmune hemolytic anemia (AHA) was diagnosed. Steroids obtained a good control of the hemolytic disease, but at 17 months of age, still on steroids, the patient was admitted with failure to thrive, vomiting, and jaundice (hemoglobin 11.8 g/dL, mean blood cell volume 82 fL, total bilirubin 6.5 mg/dL, conjugated bilirubin 1.6 mg/dL, ALT 4025 U/L, AST 5972 U/L). Laboratory investigations excluded all known causes of liver disease and a liver biopsy (Fig. 1A) showed giant-cell hepatitis (GCH). Initial treatment consisted of steroids only (prednisone, 2 mg/kg orally per day), but the evident steroid dependency and the lack of an adequate clinical response made us attempt, during the following 12 months, different immunosuppressive treatments in various combinations, including high-dose steroids (methylprednisolone, 500 mg/m2 intravenously twice per day for 10 days), cyclosporine (5–8 mg/kg orally per day for 2 months), azathioprine (0.5 mg/kg orally per day for 2 months), mycophenolate (40 mg/kg orally per day for 4 months), and anti-C20 monoclonal antibody (rituximab, 375 mg/m2 intravenously once per week for 3 consecutive weeks). None of these attempts was successful in obtaining a complete remission; the clinical trend showed a persistent steroid-dependent partial control of AHA and GCH, with progressively worsening side effects (marked obesity, severe growth retardation, osteoporosis). During this period, the child continued with a gluten-free diet, but because of the worsened clinical condition, the severity of immunosuppression, and the immunoglobulin-replacement therapy, serology and intestinal biopsy were not repeated.FIG. 1: A, A giant cell hepatitis-diagnostic liver biopsy showing massive lymphocyte and granulocyte inflammatory infiltrate, intraacinar necrosis, and spot giant-cell transformation. Some portal spaces show the disappearance of biliar duct; others show peripheral ductal regeneration without evidence of liver fibrosis. B, Liver biopsy 2 y after treatment with alemtuzumab shows normal hepatic structure, no hepatic inflammatory infiltrate, and no evidence of giant-cell transformation.Because the clinical progression of a refractory GCH is usually aggressive and relentlessly fatal, the child was considered eligible for innovative therapies such as hematopoietic cell transplantation (HCT) and referred to our transplant centre. Orthotopic liver transplantation was not taken into consideration because of the GCH recurrence reported in most cases. The high-resolution human leucocyte-antigens typing of the patient, performed for donor searching, also showed that he had a CD-associated haplotype (DQA1*05/DQB1*02). The search was successful, leading to the identification of a suitable unrelated marrow donor. Because transplant-related mortality is high in this setting, due to veno-occlusive and graft-vs-host disease, a further immunosuppressive attempt with alemtuzumab (anti-CD52 monoclonal antibody) was started, keeping HCT as an alternative in case of no response or failure of immunohematological reconstitution. Informed consent was obtained from both parents. At the age of 28 months he received alemtuzumab 0.6 mg/kg intravenously per day for 6 consecutive days; steroids were tapered and interrupted 2 months later (Fig. 2). No acute side effects attributable to alemtuzumab were observed.FIG. 2: Changes over time of hemoglobin and aminotransferases following the initial diagnosis of Coombs-positive autoimmune hemolytic anemia.Liver function test normalized within 2 months. Liver biopsies showed complete resolution of giant cell transformation 4 months and 2 years later (Fig. 1B). Isolated AHA, which initially had settled rapidly, recurred 2 months later, but resolved with rituximab (375 mg/m2 intravenously once per week for 3 weeks). Lymphocyte subset reconstitution (ie, absolute counts and relative frequencies of peripheral blood lymphocyte subpopulations) was assayed monthly during the first year following alemtuzumab and every 3 months during the second year. The absolute counts of lymphocyte subpopulations reached the fifth percentile of age-matched reference values around 1 year after alemtuzumab. During the year following treatment, the only infectious complications observed were asymptomatic adenovirus-reactivation and central-line infections. On his parents' initiative, the boy resumed a normal diet including gluten at 40 months of age without developing symptoms. At 48 months of age, serology for CD resulted negative and an intestinal biopsy showed normal mucosa. At 5 years of age, the patient is well and in complete clinical remission. DISCUSSION GCH associated with Coombs-positive AHA (GCH-AHA) is a rare entity with an aggressive course usually leading to hepatic failure and death. GCH-AHA is a distinct disorder from the more common GCH diagnosed in the neonatal period in association with infections, metabolic disorders, and cholestasis. The etiology of GCH-AHA is unknown (1–3). Despite the association with Coombs-positive AHA and responsiveness to immunosuppressive treatments, no direct evidence exists for an autoimmune pathogenesis. Autoantibodies are usually negative and histological features characteristic of autoimmune hepatitis are missing. It has been suggested that the disease process is due to immune dysfunction and unrestrained release of cytokines from stimulated T lymphocytes and Kupffer cells (2,4). The clinical course is usually aggressive and the outcome poor. Conventional immunosuppressive treatments (steroids, azathioprine, cyclophosphamide, cyclosporine, mycophenolate) are associated with high toxicity and often produce only partial or short-lasting remission. Orthotopic liver transplantation had a poor outcome in most of the few reported cases, because of the high rate of GCH recurrence (1–3). Results of HCT for GCH-AHA, if ever performed, have never been reported (5). Alemtuzumab is a humanized immunoglobulin G1 monoclonal antibody directed against CD52, a glycoprotein expressed on circulating T and B lymphocytes and natural-killer cells. It rapidly depletes circulating lymphocytes. Although alemtuzumab is used as an immunosuppressive agent for solid-organ transplantation (6) and for reduced-intensity HCT (7), it also has been successfully used for the treatment of immune-mediated diseases, such as rheumatoid arthritis, scleroderma, and multiple sclerosis (8,9). Monoclonal antibodies may have severe side effects. Immunohematological toxicity (cytopenia and marrow hypoplasia) has been reported as the most important adverse effect following alemtuzumab, although it is rarely severe and long lasting (10). Another common problem following the treatment is viral reactivation, but regular virological surveillance and prompt treatment when necessary, as routinely performed in a hematological or transplant unit, can minimize the risks (11). Some recent reports seem to confirm that the use of alemtuzumab for the treatment of autoimmune diseases can be considered fairly safe (12). To our knowledge, our patient, who is the first reported in the literature with GCH-AHA to be treated with alemtuzumab, has achieved a complete and long-lasting remission after temporary CD52+ cell depletion, with discontinuation of steroid treatment and sustained normalization of liver function and histology. The drug was well tolerated and, despite long-lasting immunodeficiency (approximately 1 year), no major infectious events were observed (13). After full immune recovery, the disease did not relapse, suggesting that the treatment was able to eradicate the aberrant clones possibly involved in the pathogenesis of GCH. The recurrence of AHA following alemtuzumab may be attributed to the uncontrolled growth of a surviving B cell clone in the presence of a highly impaired T cell immunosurveillance; however, a second treatment with rituximab was effective in abolishing hemolysis (14,15). An association between CD and other autoimmune diseases, including AHA (16,17), has been reported in several studies. Although CD can be a false-positive diagnosis at serological screening testing in patients with other autoimmune disorders (18,19), our patient had a CD-associated human leucocyte antigens-haplotype and a positive small bowel biopsy. CD in our patient appeared as the less severe component of a complex immune disorder prevalently evident as GCH-AHA. The intestinal biopsy was not performed after gluten withdrawal, but it was normal 8 months after resuming a normal diet and the achievement of immunological reconstitution. We can speculate that alemtuzumab also eradicated the aberrant lymphocyte clones potentially involved in the pathogenesis of CD. Recently, a possible role of alemtuzumab for refractory CD disease has been reported by other authors (20). In conclusion, treatment with alemtuzumab can lead to long-term remission of GHC with AHA, and should be considered as a therapeutic option as soon as it becomes apparent that conventional immunosuppressive therapy is ineffective, avoiding the higher risks associated with human stem-cell transplantation.