Post-Transplantation Cyclophosphamide (Cy) Reduces Graft Rejection and Graft-Versus-Host Disease (GVHD) after Non-Myeloablative, Partially HLA-Mismatched (Haploidentical) Bone Marrow Transplantation (BMT).

Abstract

Abstract In animal models of haploidentical BMT, a properly timed high dose of Cy post-BMT selectively eliminates host-versus-graft and graft-versus-host reactive T cells, thereby preventing graft rejection and reducing GVHD. To test the activity of post-BMT Cy in humans, 56 patients (median age 48 years) with advanced hematologic malignancies received Cy 50 mg/kg IV, either once (on day 3; n=20) or twice (on days 3 and 4; n=36) after non-myeloablative conditioning and haploidentical BMT (“mini-haploBMT”). Diagnoses included AML (n=18), MDS (n=8), ALL (n=3), CML (n=4), CMMoL (n=3), CLL (n=3), multiple myeloma (MM; n=3), Hodgkin’s disease (HD; n=5), non-Hodgkin’s lymphoma (NHL; n=8), and paroxysmal nocturnal hemoglobinuria (PNH; n=1). Most patients had advanced disease or failed a previous autologous transplant. All were conditioned as outpatients with fludarabine (30 mg/m2 IV q d, days −6 to −2), Cy (14.5 mg/kg IV on days −6, −5), and total body irradiation (200 cGy on day −1), transplanted with non-T cell-depleted marrow, and treated with G-CSF, FK-506 and mycophenolate mofetil beginning the day after the last dose of Cy. Median number of hospitalizations in the first 60 days after BMT was 1 (range, 0–3), and the median number of hospital days was 4 (range, 1–53). Neutrophil engraftment (>500/mm3) after one or two days of post-BMT Cy occurred at a median of 15 versus 17 days after BMT, respectively (p=.36), and platelet counts >20,000/mm3 without transfusion were achieved at a median of 31 and 31 days, respectively (p=.78). Graft rejection, which in all cases was followed by recovery of autologous hematopoiesis, occurred in 5/18 (28%) versus 4/35 (11%) evaluable patients receiving one versus two doses of post-transplant Cy, respectively (p=.24). Full donor hematopoietic chimerism in the peripheral blood on day 60 was seen in 10/13 versus 25/31 engrafting patients who received one versus two doses of post-BMT Cy; with two exceptions, mixed chimerism was an immediate harbinger of relapse. Compared to patients receiving a single dose of post-BMT Cy, those receiving two doses had significantly less Grade II–IV (43% vs. 78%; p=.01) and Grade III–IV acute GVHD (20% vs. 53%; p=.006) by day 200 after BMT. Death from GVHD occurred in 5/13 evaluable patients receiving one dose versus 2/28 evaluable patients receiving 2 doses of post-BMT Cy. This latter group also had a non-significantly lower incidence of non-relapse mortality at 100 days (4% vs. 15%, p=.13) and at 1 year (23% vs. 35%; p=.37). At a median follow-up of 172 days (range, 32–1440 days), 21 patients remain alive without disease progression, including all 3 patients with chronic phase CML and the patient with PNH. In summary, mini-haploBMT with 2 doses of post-transplantation Cy resulted in full donor chimerism without severe GVHD in 23/36 patients (64%). Its toxicity profile compares favorably to matched sibling BMT. Moreover, in a very high-risk patient group, the early event-free survival is encouraging.