Aplastic anemia (AA) and other types of bone marrow failure have clinical and laboratory features consistent with an autoimmune pathophysiology, with a diversity of inciting antigens, including viruses, chemicals, and drugs. Whatever its specific etiology, a majority of patients respond with hematologic improvement after immunosuppressive therapies. One important clinical feature of AA is its evolution, sometimes years after normalization of blood counts, to other hematologic diseases such as paroxysmal nocturnal hemoglobinuria (PNH), which derive from clones of hematopoietic stem cells. In clinical research, we have completed analyses of long-term follow-up of a large cohort of 122 patients with severe disease who were treated with antithymocyte globulin (ATG) and cyclosporine (CSA), which shows a response rate of about 70% but with a high risk of relapse and a more modest risk of development of myelodysplasia or chromosomal abnormalities. A surprising and important result is that there is a strong, highly significant correlation between not only the presence of a hematologic response but the robustness of blood count improvement at 3 or 6 months post-treatment, which is predictive of outcome, both survival and evolution (but not relapse). In analysis of long-term results from our randomized trial of high dose cyclophosphamide and ATG, we observe relapse, cytogenetic changes, and persistence of abnormal glyocsylphosphoinositol-anchored protein-deficient cells occurring in both arms, indicating that the more toxic chemotherapeutic regimen does not prevent or alleviate these clinical complications. In an effort to induce tolerance in this immune-mediated disease, we have instituted a protocol that incorporates the novel immunosuppressive drug mycophenolate mofetil in early treatment and delays adminsstration of cyclosporine. To date, 54 patients have been enrolled; at 3 months the hematologic response rate is 62%, and at 6 months the rate is 74%. These values are equivalent to our experience in the treatment of severe aplastic anemia with ATG plus cyclosporine. One patient has died and 4 have relapsed; the study will need to mature in order to determine the advantage of this approach compared to standard regimens. Other clinical studies have characterized the relationship of AA to paroxysmal nocturnal hemoglobinuria (PNH): these are concurrent in almost half of AA patients,and evidence of an expanded clone is present at disease presentation, not as a late event. Cytogenetic abnormalities do occur after diagnosis and treatment but have widely different outcomes, with continued responsiveness or dependence on immunosuppressive therapies and good survival in trisomy 8 compared to refractory pancytopenia, leukemic evolution, and death in monosomy 7. Our laboratory studies have focused on the immune pathophysiology of AA, identification of a viral antigen, and the mechanism of late clonal evolution. Studies of etiology focus on an unknown hepatitis virus in the post-hepatitis AA syndrome (see Z01 HL 02319-14 HB). In efforts to better characterize the immune response and target antigens, we have isolated and immortalized helper and cytotoxic T cell clones and determined their T cell receptor repertoire based on the pattern of TCR skewing. Specific V-beta TCR were found to dominate within a patient and among patients with disease, declined with successful immunosuppression, and were not detectable in normal T cells. By immunophenotyping of patient samples, we have measured increased numbers of effector cytotoxic lymphocytes (CD28-negative CD8 cells) which correlate with disease status. Conversely, AA is characterized by a marked deficiency of an important population of regulatory NK-T cells, a decrease seen also in many other autoimmune disease. Other studies have examined the mechanism of action of ATG: in vitro stimulatory effects on hematopoiesis have been shown to nonspecific. Laboratory investigations of the mechanisms responsible for late clonal disease, both PNH and myelodysplasia (MDS), have used sensitive flow cytometric and fluorescent in situ hybridization (FISH) assays. For PNH, a marked difference in normal and deficient CD34 cells obtained from patient bone marrow has been observed: the normal CD34 cells have poor in vitro survival due to apoptosis associated with Fas expression, consistent with in vivo damage, probably from immune attack. These results are complementary to our finding of a strong association between a histocompatability antigen, HLA-DR2, and the presence of an expanded PNH clone in bone marrow failure syndromes. Additional ongoing studies of PNH include clinical observations of in vivo transfer of glycosylphosphoinositol-linked proteins from normal transfused erythrocytes to deficient red and white blood cells in PNH patients. Examination of pathways of GPI-linked protein degradation in PNH compared to normal cells are ongoing, including studies of PrPc, the normal analogue of the prion protein. Trisomy 8, but not monosomy 7, also is associated with immune abnormalities related to hematopoiesis, as cytogenetically abnormal cells are more likely than normal cells to express Fas and to be apoptotic.
Showing the most recent 10 out of 104 publications
