Feasibility of Reduced-Intensity Cord Blood Transplantation (RICBT) as a Salvage Therapy for Graft Failure (GF): Results of a Nationwide Survey of 63 Patients.

Background: The outcome of patients who suffer from GF after allogeneic transplantation is poor, primarily due to an increased risk of infectious complications and toxicities. We evaluated whether rescue with RICBT could improve survival after GF.

Patients and Methods: We retrospectively surveyed the data of 63 patients (median age, 51 years: range, 17–68 years) with hematological disorders who received RICBT within 3 months of GF between Jan, 2000, and Apr, 2006. The diagnosis included AML (n=28), MDS (12), CML (2), ALL (13), lymphoma (4), and AA (4), and 28 patients were not in remission before the preceding allogeneic HCT, which had been performed with either myeloablative (29) or reduced-intensity conditioning regimens (34). Twelve patients (19%) had received unrelated BM, while 46 (73%) received CB. Fifty-one patients had developed primary GF (failure of the ANC to surpass 500/mm³ or <5% donor T-cells) at a median of 28 days (17–56), and 12 had secondary GF (decrease in the ANC <100/mm³ or <5% donor T-cells after initial engraftment) at a median of 38 days (20–101).

Results: The median time interval between the preceding HCT to salvage RICBT was 47 days (27–162), and that between the diagnosis of GF to salvage RICBT was 16 days (5–61). Prior to RICBT, 21 patients (33%) were in PS 3–4, 8 (13%) had grade 3 organ toxicities according to CTCAE v3.0, and 53 (84%) had documented infections or febrile neutropenia. All patients were conditioned with fludarabine-based regimens with L-PAM (n=20), BU (15), CY (12) or thiotepa (2), with or without additional 2–4 Gy TBI. GVHD prophylaxis included CyA/FK alone (37) and a combination of MTX plus CyA/FK (10). The median number of infused TNC was 2.3 (1.0–4.3) x10⁷/kg. Among the 46 evaluable patients who survived more than 28 days after RICBT, 33 (72%) engrafted, but 10 (22%) again experienced GF. Multivariate analysis disclosed that infusion of TNC ≥2.0x10⁷/kg and a fludarabine/L-PAM-based regimen were associated with a higher probability of engraftment. The cumulative incidence of grade II–IV acute GVHD was 28%. The median follow-up of surviving patients was 236 days (39–937) after rescue RICBT. The 1-year overall and progression-free survival rates were 27% and 20%, respectively. Forty-five patients (70%) died at a median of 37 days (2–485) after RICBT, with a day-100 TRM of 54%. The cause of death included bacterial (14), fungal (10) and viral infections (7), and relapse (6). Multivariate analysis showed that high-risk disease, grade 3 organ toxicity before RICBT, and the use of MTX were associated with an increased risk of mortality.

Conclusions: Our data support the feasibility of RICBT for patients suffering from GF, with an engraftment rate of >70%, when they receive a graft containing ≥2.0x10⁷/kg cells. The earlier application of RICBT while patients still have better PS without infections or organ toxicities will be the subject of continued clinical research.