Use of HLA-mismatched unrelated donors after myeloablative conditioning is an effective treatment of blood disorders when HLA-matched donors are not available. However, the feasibility of HLA-mismatched HCT following nonmyeloablative conditioning is not well established. We previously demonstrated in over 1200 patients (pts) that a nonmyeloablative conditioning regimen consisting of 2 Gy TBI and fludarabine (90 mg/m2) followed by post-grafting immunosuppression with MMF/CSP was an effective approach for HCT from 10/10 HLA-matched donors. Here we report the results of a clinical trial aimed at determining whether stable hematopoietic grafts from HLA-class I mismatched donors could be safely established after nonmyeloablative conditioning. Pts were conditioned with fludarabine and 2 Gy TBI, followed by extended immunosuppression with CSP (day −3 to +180; tapered by +365) and MMF (tid days 0 to +100; tapered by +150). 46 pts ineligible for myeloablative conditioning were enrolled on this multi-institutional study through April/07. 36 unrelated and 3 related donor-recipient pairs were mismatched in both the HVG and GVH vectors. 43 unrelated pairs were mismatched for at least one antigen (AG) (15 at HLA-A, 5 at HLA-B, 22 at HLA-C) except one pt who had two allele mismatches (HLA-A and B). All 3 related pairs were mismatched for one HLA-A AG. 13 pts, in addition to the AG mismatch, had an allelic mismatch (3 at HLA-A, 4 at HLA-B, 6 at HLA-C). All pts were matched for HLA-DRB1 and HLA-DQB1 at the allele level. Diagnoses included NHL (n=17), HD (n=5), AML (n=11), ALL (n=5), CLL (n=3), MM (n=4) and MDS/AML (n=1). The median age of pts was 56 (range 13–69) years and median time from diagnosis to HCT was 32 (range 5–205) months. Disease status at time of HCT included CR (n=23), PR (n=11), stable disease (n=2), and refractory relapse (n=10). Before HCT, pts had received a median of 7 (range 2–15) courses of therapy. 43% had failed myeloablative HCT (autologous n=18; allogeneic n=2). All pts received G-CSF mobilized PBSC containing median doses of 7.3 ×106 CD34 and 2.4 × 108 CD3 cells /kg. 12 pts were not evaluable for the graft rejection outcome (3 early death, 7 mismatched in GVH vector only, 2 tandem auto-allo). The remaining pts showed median percentages of donor peripheral blood T cell, granulocyte and marrow chimerism of 100% at all time points (on days 28, 56, and 84). Two pts experienced graft rejection but had a successful 2nd HCT from the same or an alternative donor. 35 of the 46 pts developed acute GVHD (2 grade I, 21 grade II, 5 grade III, 7 grade IV) and 15 developed chronic GVHD. At 2 years, the cumulative probabilities of relapse/progression and non-relapse mortality were 30% and 45%, respectively. The 2 year Kaplan-Meier estimates of OS and PFS were 27% and 25%. This study demonstrates that nonmyeloablative conditioning using fludarabine and low-dose TBI can be successfully used for grafts from pts with HLA- class I mismatched donors.

Author notes

Disclosure:Consultancy: Bayer. Research Funding: Bayer. Off Label Use: Fludarabine, cyclosporine, mycophenolate mofetil.