Single- or double-unit UCBT following RIC in adults with AL: a report from Eurocord, the ALWP and the CTIWP of the EBMT

This survey is a retrospective, multicenter registry-based study performed by the Acute Leukemia Working Party (ALWP) of the European Society for Blood and Marrow Transplantation (EBMT) and by Eurocord. EBMT registry is a voluntary working group of more than 500 transplant centers, participants of which are required once a year to report all consecutive stem cell transplantations and follow-up. Audits are routinely performed to determine the accuracy of the data. Eurocord collects data on CBT performed in >50 countries worldwide and >500 transplant centers, mainly EBMT centers. Inclusion criteria were adult (≥18 years) patients, AML or ALL, complete remission (CR) at the time of transplantation, first single (with a cryopreserved TNC ≥2.5 × 10⁷/kg) or double CBT between 2004 and 2014, and RIC conditioning. RIC was defined as use of fludarabine associated with <6 Gy TBI, or busulfan ≤8 mg/kg, melphalan ≤140 mg/m² or other nonmyeloablative drugs, as previously reported [11, 20, 21]. HLA-compatibility requirements followed the current practice of antigen level typing for HLA-A and -B and allele level typing of HLA-DRB1. CB units were 4–6/6 HLA-A, -B, and -DRB1 matched to the recipient and to the other unit in case of dCBT in most patients. However, more recently, some centers are no longer matching the CB units between them with regard to HLA based on the study by Avery et al. [22]. HLA disparities between each unit and the recipient and between the two units were not necessarily at the same loci. Grading of acute and chronic GVHD was performed using established criteria [23].

For the purpose of this study, all necessary data were collected according to EBMT and Eurocord guidelines.

Data from all patients meeting the inclusion/exclusion criteria were included in the analyses. Start time was date of transplant for all endpoints. Neutrophil engraftment was defined as first of three consecutive days with a neutrophil count of at least 0.5 × 10⁹/L. Platelet engraftment was defined as the first of seven consecutive days of an unsupported platelet count of at least 20 × 10⁹/L [2].

To evaluate the relapse incidence, patients dying either from direct toxicity of the procedure or from any other cause not related to leukemia were censored. NRM was defined as death without experiencing disease recurrence. Patients were censored at the time of relapse or of the last follow-up. Cumulative incidence functions were used for relapse incidence and NRM in a competing risk setting since death and relapse were competing together.

For estimating the cumulative incidence of chronic GVHD, death was considered as a competing event. OS and LFS were estimated using the Kaplan-Meier estimates. GRFS was defined as being alive with neither grade III–IV acute GVHD, severe chronic GVHD nor disease relapse [24]. Univariate analyses were done using Gray’s test for cumulative incidence function and log rank test for OS and LFS. Associations between single or double CBT and transplantation outcomes (chronic GVHD, relapse, NRM, LFS, and OS) were evaluated in multivariable analyses, using Cox proportional hazards. Variables introduced in the Cox models included recipient age (in decades), disease type (AML versus ALL), disease status at CBT, type of conditioning regimen (TBI, fludarabine, and cyclophosphamide (TCF) versus other), cytogenetic risk, and the use of ATG or not. Exploratory analyses of the heterogeneity of sCBT versus dCBT among pre-transplant subgroups for relapse, NRM, OS, LFS, and GRFS were performed using Cox models. The results of these Cox models were presented graphically using forest plots [25].

All tests were two sided. The type I error rate was fixed at 0.05 for determination of factors associated with time to event outcomes. Statistical analyses were performed with SPSS 19 (SPSS Inc., Chicago, IL), and R 2.13.2 (R Development Core Team, Vienna, Austria) software packages.

Patients and disease characteristics are described in Table 1. Briefly, data from 534 patients with AML (n = 408) or ALL (n = 126) receiving a first single (n = 172) or double (n = 362) CBT were included in the analyses. Among dCBT recipients, 47 received two CB units containing less than 2.5 × 10⁷ TNC/kg each. In comparison to sCBT patients, dCBT recipients had a shorter follow-up (34 versus 54 months, P = 0.0005), were more frequently male (56 versus 41%, P = 0.002), received a conditioning combining TBI, cyclophosphamide and Flu (TCF regimen, 83 versus 66%, P < 0.001) more frequently, and received ATG less frequently (16 versus 37%, P < 0.001). The two groups were not different for recipient age at transplantation (52 versus 50 years, P = 0.17) as well as for other important factors such as disease type (76% of the patients with AML in both group), disease status at transplantation (CR1 in 57 versus 53% of the patients, P = 0.6), time from diagnosis to transplantation (9.4 versus 9.5 months, P = 0.8) and cytogenetic risks (P = 0.85). Finally, as expected, TNC (median 5.1 versus 3.8 × 10⁷ TNC/kg, P < 0.001) and CD34⁺ cell (median 4.0 versus 3.1 × 10⁵ cell/kg, P = 0.003) doses at cryopreservation were significantly higher in dCBT than in sCBT recipients (Table 2).

Overall, the cumulative incidence of neutrophil engraftment at day 60 was not different in sCBT (median 77%; 95% CI 70–83%) and dCBT (median 83%; 95% CI 78–86%) recipients (P = 0.4). The median time to neutrophil engraftment was 19 and 24 days, for sCBT and dCBT, respectively, (P < 0.001) (Additional file 1: Figure S1). Similarly, the cumulative incidence of platelet engraftment at 6 months was not different in sCBT (median 65%; 95% CI 57–73%) and dCBT (median 71%; 95% CI 65–76%) recipients (P = 0.9).

There was a trend for a lower incidence of grade II–IV acute GVHD (28 versus 36%, P = 0.08) in sCBT recipients, but this was no longer the case after adjusting for confounding factors (HR = 1.1, P = 0.22). In contrast, incidences of grade III–IV acute (11 versus 13%, P = 0.6), chronic (28 versus 36% at 2 years, P = 0.2) and extensive chronic (10.6 versus 12% at 2 years, P = 0.69) GVHD were comparable in sCBT and dCBT recipients.

At 2 year, in comparison to sCBT recipients, dCBT had a similar cumulative incidence of relapse (32 versus 35%, P = 0.5) and of NRM (22 versus 29%, P = 0.2). GRFS (37 versus 31%, P = 0.13), and LFS (46 versus 36%, P = 0.06) were similar according to the type of graft. DCBT showed a significantly better OS (51 versus 41%, P = 0.03) (Additional file 1: Table S1 and Figure S2). Of note, outcomes of the 47 patients given 2 CB units containing <2.5 × 10⁷ TNC/kg each were at least as good as those observed in sCBT recipients with 2-year OS and LFS of 56.4 and 42.9%, respectively, (Additional file 1: Figure S3).

After adjusting for potential confounding factors in multivariate analyses, dCBT and sCBT recipients had a similar risk of relapse (HR = 0.9; 95% CI 0.6–1.3, P = 0.5) and NRM (HR = 0.8; 95% CI 0.5–1.2, P = 0.3), and similar GRFS (HR = 1.0; 95% CI 0.9–1.0, P = 0.3), LFS (HR = 0.8; 95% CI 0.7–1.1, P = 0.2), and OS (HR = 0.8; 95% CI, 0.6–1.1 P = 0.17) (Fig. 1) (Table 3). The only factor associated with lower OS in multivariate analysis was the use of ATG (HR = 1.8; 95% CI 1.2–2.8, P = 0.01). This was due to a significantly higher NRM in ATG in comparison with non-ATG recipients (HR = 2.4; 95% CI 1.3–4.3, P < 0.001) while relapse incidence was not affected by ATG (HR = 1.0, 95% CI 0.6–1.9, P = 1.0).

As shown in the Table 4 and in the Additional file 1: Table S2, causes of death were not statistically different between sCBT and dCBT recipients. However, there was a suggestion for more deaths from GVHD (13/362 (3.5%) versus 3/172 (1.7%)) in dCBT than in sCBT recipients in the first 100 days after transplantation, while the incidence of death from infection in the first 100 days after CBT was comparable between the 2 groups (22/362 (6.1%) in dCBT versus 13/172 (7.6%) in sCBT recipients, respectively).

To further dissect the impact of sCBT versus dCBT, we performed additional (univariate) Cox analyses separately for various pre-transplant/transplant variables. The results of these analyses are presented graphically using Forest plots in Figs. 2 and 3. There were no interactions between patient age at transplantation, patient gender, number of cells infused, disease type, disease status, HLA-matching and conditioning type (TCF versus other), and the association between sCBT versus dCBT and GRFS or OS. Further multivariate Cox models assessing possible interactions between ATG and sCBT versus dCBT demonstrated the absence of statistically significant interactions for relapse incidence (P = 0.27), NRM (P = 0.37), LFS (P = 0.97), and OS (P = 0.59). Similarly, there was no interaction between disease status (CR1 versus other) and the impact of sCBT versus dCBT on GRFS (P = 0.61).

We finally assessed what was the combined impact of cell dose and sCBT versus dCBT. In order to address this issue, we performed multivariate Cox models including four graft type groups: sCBT and TNC above median, dCBT and TNC above median, sCBT and TNC below median, and dCBT and TNC below median. As observed in the Table 5, in comparison to the reference group (sCBT and TNC above median), patients given sCBT with TNC below the median had a higher risk of relapse (HR = 2.0, 95% CI 1.0–3.9, P = 0.04) and a suggestion for a worse LFS (HR = 1.5, 95% CI 0.9–2.3, P = 0.11), while outcomes were comparable between patients receiving sCBT and TNC above median and those given dCBT (irrespective of the cell dose received).