Objective: No specific therapy exists for autoimmune diseases caused by self-reactive lymphocytes. As shown in experimental animals, which led to pilot clinical studies, elimination of self-reactive lymphocytes can be accomplished with high-dose chemoradiotherapy, followed by autologous stem cell transplantation, by re-establishment of unresponsiveness to self antigens of newly generated lymphocytes, due to a mechanism of central clonal deletion. We hypothesized that self-reactive lymphocytes causing autoimmune disease may be successfully eliminated by highly immunosuppressive yet not necessarily myeloablative conditioning in conjunction with allogeneic blood stem cell transplantation, since immunocompetent alloreactive lymphocytes of donor origin can effectively eliminate residual host-type hematopoietic cells, self-reactive lymphocytes included, by a mechanism that resembles graft-vs-leukemia (GVL) effects. The present report is an attempt to confirm the existence of graft-vs-autoimmunity (GVA) effects in parallel with amplification of the alloreactive potential of donor lymphocytes following allogeneic non-myeloablative stem cell transplantation (NST).
Methods: We identified a patient with severe psoriatic arthritis who also had Philadelphia (bcr/abl) positive chronic myelogenous leukemia and therefore was fully eligible for NST. Both diseases responded initially to non-myeloablative conditioning involving fludarabine 30 mg/m2 x 6, anti-T-lymphocyte globulin 10 mg/kg X 4, and busulfan 4 mg/kg x 2.
Results: The initial NST procedure was uneventful and resulted in elimination of all signs of autoimmunity (psoriasis and arthritis). Recurrence of polyarthritis and exacerbation of psoriasis were observed in parallel with a significant increase in the proportion of male (host) DNA, and 5% of the mitoses were bcr/abl positive, indicating an increase in the clone of CML. Both bcr/abl-positive cells identified by RT-PCR and psoriatic arthritis were successfully eliminated following discontinuation of anti-GVHD prophylaxis with cyclosporine A (CSA), which resulted in activation of the alloreactive potential of donor T cells, accompanied by graft-vs-host disease (GVHD), suggesting the existence of GVA effects. RT-PCR for bcr/abl remains consistently negative for nearly 3 years, and all DNA remains donor type.
Conclusions: The response of autoimmune disease manifestations to GVA effects in parallel with elimination of all host-derived hematopoietic cells supports our working hypothesis that autoimmune diseases caused by self-reactive lymphocytes may be effectively treated by elimination of alloreactive self-reactive lymphocytes following induction of host-vs-graft tolerance, in analogy with replacement of malignant or genetically abnormal host cells following DLI. It is therefore suggested that intentional GVA effects may be inducible by DLI following a conventional or preferably safer non-myeloablative regimen in recipients with life-threatening autoimmune diseases resistant to conventional modalities. Adoptive immunotherapy of autoimmunity may thus involve a two-step procedure: first, inducing host-vs-graft and graft-vs-host transplantation tolerance through a transient stage of mixed chimerism; second, inducing controlled GVA effects, initially by discontinuation of CSA and then, if indicated, by late outpatient DLI to eradicate residual hematopoietic cells of host origin.