The introduction of tyrosine kinase inhibitors (TKIs) to treat chronic myelogenous leukemia (CML), has largely replaced curative strategies based on allogeneic hematopoietic stem cell transplantation (HSCT). Allogeneic HSCT continue to be recommended for patients progressing to a more advanced phase of the disease and those in first chronic phase (CP1) failing second and third generation TKIs. Although the role of allogeneic HSCT using matched related or unrelated donor remains well established in the post-TKI era, controversies exist in transplant using alternative donors, and data in this setting is scarce.

Umbilical cord blood (UCB) can be used as an alternative stem cell source for patients in whom allogeneic HSCT is indicated, but lack an appropriate human leucocyte antigen (HLA)-matched adult donor. The advantages of UCBT include rapid availability, absence of donor risk, and the relatively lower risk of GVHD with preserved graft versus leukemia effect.

Because of the lack of large studies on the outcome of UCBT for CML patients in the TKI era, we performed a retrospective analysis to identify risk factors for outcomes after unrelated cord blood transplants in adults with CML.

Through the EBMT database, 150 CML patients were identified as UCBT recipients. Median year of UCBT was 2008 (range, 2000 to 2015) with a median follow up of 62.2 months (range, 3 to 203.1 months). Median age at the time of transplant was 40.4 years (range, 18.8 to 66.5) with a male to female ratio of 92/58. The median time from CML diagnosis to UCBT was 29.9 months (range, 3.4 to 273.6) with first (Q1) and third (Q3) quartile at 12.4 and 57.3 months respectively. Eleven patients received an UCBT as a second allogeneic HSCT. Conditioning regimen was of reduced intensity (RIC) in 44 patients while 92 received a myeloablative conditioning (MAC), and 61 out of 128 evaluable patients received T-cell depletion.

Disease status at the time of transplant was first chronic phase (CP1, n=45), second or third chronic phase (CP2 and CP3, n=48) or advanced phase (accelerated phase or blastic phase, AP/BP, n=55).

Overall survival (OS) at 12, 24 and 36 months were 52.1% (95% CI 44.0 to 60.1%), 44.5% (95% CI 36.3 to 52.6 %) and 41.2% (95% CI 33.0% to 49.4%) respectively. Median time to neutrophil and platelet engraftment were 23 and 47 days respectively in 119 patients who engrafted, while 24 and 7 patients experienced primary (16%) or secondary (4.6%) graft failure respectively.

At 24 months, the cumulative incidence of non-relapse mortality (NRM) and relapse were 38.9% (95% CI 31.0 to 46.9%) and 25.3% (95% CI 18.2% to 32.3%), while cumulative incidence of grade 2-4 acute GvHD was 34.3% (95% CI 26.1 to 42.5%) and 24 months cumulative incidence of chronic GvHD was 24.8% (95% CI 15.6 to 34.0%).

OS and disease-free survival (DFS) were significantly reduced for patients in AP/BP at the time of transplant compared to patients in CP1 or CP2/CP3 (24 months OS, 27.1% vs 55.3% and 53.7%, p=0.0026; 24 months DFS, 21.5% vs 39.4% and 47.1% respectively, p=0.0059, Figure 1). Cumulative incidence of relapse was significantly increased in patients in AP/BP at the time of HSCT compared to patient in CP1 or CP2/3 (24 months relapse incidence of 37.6% vs 20.6% and 17.1% respectively, p=0.0178).

There was no significant difference in OS and DFS in patients receiving RIC compared to MAC (24 months OS 47.1 and 48.1% respectively, p=0.8; 24 months DFS 33% vs 40.6%, p=0.36). Cumulative incidence of relapse at 24 months was 34.7% in RIC and 24.6% in MAC, although this was not statistically significant (p=0.14).

In conclusion, UCBT is more effective in CML patients who are in first CP but also in those who are in second or third CP at the time of transplant. NRM remains high, possibly related to differences in practices of UCBT over time and between centers. Comparison with alternative graft source such as haploidentical HSCT is needed, although little data is available in this setting.

Patients who are in advanced phase disease at the time of transplant have a dismal prognosis comparable to those who receive an allogeneic HSCT from HLA-matched adult donors, although pre and post-transplant intervention with 3rd generation TKI may help improve outcome of these patients.

Finally the use of RIC regimen for UCBT is an alternative for patients who cannot tolerate a MAC regimen; further analysis will help identify patients who would benefit from this approach.


De Lavallade:Incyte: Honoraria, Research Funding; BMS: Honoraria, Research Funding. Angelucci:Jazz Pharmaceuticals Italy: Other: Local ( national) advisory board; Vertex Pharmaceuticals Incorporated (MA) and CRISPR CAS9 Therapeutics AG (CH): Other: Chair DMC; Roche Italy: Other: Local (national) advisory board; Celgene: Honoraria, Other: Chair DMC; Novartis: Honoraria, Other: Chair Steering Comiittee TELESTO Protocol. Chalandon:Roche: Membership on an entity's Board of Directors or advisory committees, Other: Travel costs.

Author notes


Asterisk with author names denotes non-ASH members.