Second Allogeneic Transplant with Cord Blood for Treatment of Failed First Allogeneic Transplant Due to Primary Graft Failure or Disease Recurrence

Background: For patients with graft failure or disease relapse following allogeneic hematopoietic cell transplant (HCT), a second HCT may be lifesaving. Prior studies have shown a survival advantage of this approach over non-HCT treatment options, although data remains limited. In the setting of a failed first allogeneic transplant, cord blood (CB) as the donor source for the second HCT provides the important advantage of prompt availability of donor cells without risk to the donor during cell procurement. In the event of second HCT to treat disease relapse, CB offers the ability to use highly mismatched grafts with the potential of enhancing the graft versus leukemia effect. Herein, we report the outcomes of patients undergoing second HCT with CB grafts after relapse or graft failure following their first allogeneic HCT.

Methods: Twenty-three patients median age 33 yrs (range 4-69 yrs) received a CB transplant (CBT) as a second allogeneic HCT at the Fred Hutchinson Cancer Research Center between 2006 – 2013 for MDS/AML (n = 12), ALL (n =8), Hodgkin Lymphoma (n = 2) and DLBCL (n = 1). Eight patients received their initial transplant from an HLA-matched sibling, 6 from a matched and 2 from a mismatched unrelated donor, 4 from a haploidentical donor and 3 patients had a prior double CBT (dCBT). Primary indications for a second allograft were relapse (n = 19), graft failure (n = 3) and donor derived MDS (n = 1). Conditioning regimens for the first HCT were myeloablative (MAC) in 15 patients and nonmyeloablative (NMA) in 8 patients. Median time between first and second transplant was 505 days (range, 55 - 3515). At the time of second HCT, 19 patients were in CR (5 with minimal residual disease) and 4 patients had persistent disease. For their second transplant, patients were conditioned with NMA or MAC regimens based on their age, history of prior TBI and clinically determined comorbidity score. Ten patients received NMA conditioning with fludarabine (Flu) 200mg/m², cyclophosphamide (Cy) 50 mg/kg and 200-400 cGy TBI and 13 patients received MAC conditioning with either treosulfan 42mg/kg, Flu 150-200mg/m² and TBI 200 cGy (n = 11) or Flu 75mg/m², Cy 120mg/kg and 1200 - 1320cGy TBI (n = 2). All patients received GVHD prophylaxis with cyclosporine and mycophenolate mofeteil. Twenty of the 23 patients received a dCBT with a median TNC of 4.8 x 10⁷ cells/kg and a post-thaw median CD34+ of 2.2 x 10⁵ cells/kg. 91% of UCB grafts were 1-2 HLA antigen mismatched with the recipient.

Results: Median time to engraftment was 22 days (range day 6 - 49) among 21 evaluable patients; 2 patients died prior to engraftment. Median time to platelet engraftment was 55 days (range 23 - 83). Death prior to day 100 was seen in 5 patients (infection/organ failure (n=3), encephalopathy (n =1), and relapse (n =1). Acute grade II-IV and III-IV GVHD was diagnosed in 18 of 21 and 6 of 21 evaluable patients respectively, and 9 of 16 evaluable patients developed chronic GVHD (6 mild, 2 moderate, 1 severe). At a median follow-up of 356 days (range among survivors 176 - 1876), the cumulative incidence of 1-year disease free survival (DFS) for all patients was 65% (CI 95%: 42-84) with patients receiving MAC vs NMA demonstrating significantly different results: 84% (51-95) vs 40% (12-79) (p=0.011), respectively. Overall one-yearnon-relapse mortality (NRM) was 22% (95% CI: 8-40): 15% (3-39) in the MAC group vs. 30% (7-58) in the NMA group (p=0.32). Relapse at 1 yearwas 21% (95% CI: 8-40), with no relapses in the MAC group compared to 30% in the NMA group (7-57) (p =0.034).

Conclusions: Our results demonstrate quite favorable outcomes in patients undergoing CBT as a second allogeneic HCT, with a one year DFS of 65%. The outcomes were particularly good for patients undergoing MAC CBT in which there were no relapses and the one-year DFS was 84%. This small study supports the use of CB as a donor source for second allogeneic HCT in the treatment of primary graft failure or disease recurrence after a first HCT. The use of CB as the stem cell source is particularly attractive in this setting as it eliminates the need to put a conventional donor at risk. While MAC may be preferable based on this limited data set, one must weigh the risk of the added toxicity of this approach and select conditioning regimens based on each patient’s pre-transplant disease and comorbidities.