Unrelated Cord Blood Versus Non-T Cell Depleted Haploidentical Stem Cell Transplantation: Comparable Results in Adults with Acute Leukemia, a Eurocord and ALWP-EBMT Study

Unrelated cord blood transplantation (UCBT) and related Haploidentical non T-cell depleted stem cell transplantation (Haplo) are alternative options to treat patients with high risk acute leukemia. Both techniques have shown encouraging results.

The current retrospective analysis aimed to compare the outcomes of both approaches in adults with ALL and AML receiving either UCBT (n=928) or Haplo (n=518) reported to EBMT centres between 2007 and 2011.

Diagnosis was ALL for 716 patients (Haplo=158 and UCBT=558) and AML for 730 (Haplo=360, UCBT=370). Haplo grafts were not T-cell depleted, with 236 patients receiving unmanipulated bone marrow (BM), 253 receiving peripheral blood stem cells (PBSC) and 29 receiving BM+PBSC. For UCBT, 431 patients were treated with a single unit, while 497 received double unit UCBT.

Compared with recipients of UCBT, Haplo recipients were more likely to have AML (69% versus 40%, p=<0.001), and were transplanted more recently (2011 versus 2009, p=0.001), had a longer median interval from CR1 to transplant (139 versus 126 days, p=0.07), and had more frequently advanced phase of disease at time of transplant (35% versus 17%, p=0.001).

Median follow-up was 19 months and 25 months, (p=0.007) for Haplo and UCBT, respectively. The type of conditioning regimen (MAC or RIC) was similar between both groups, with MAC representing 60% and 56% for Haplo and UCBT, respectively.

For Haplo, the most frequently used regimens were Thiotepa-Busulfan-Fludarabine (TBF) for MAC and Treosulfan-based regimens for RIC. Graft versus host disease (GVHD) prophylaxis included CsA+MMF (41%), Sirolimus (15%), while 163 patients received post-transplant Cyclophosphamide (Cy). For UCBT cases, the most common MAC regimens were TBF for 33% of cases, CyFluTBI-12Gy for 27% and CyTBI12Gy for 17%. The most frequently used RIC was CyFluTBI-2Gy in 72% of the cases. GVHD prophylaxis was CsA+MMF in 70% of cases.

In univariate analysis, cumulative incidence (CI) of neutrophil engraftment was 92% vs 84% (p=<0.001) for Haplo and UCBT; CI of acute GVHD was 27% vs 32% (p=0.11) and for chronic GVHD it was 30% vs 25% (p=0.06) for Haplo and UCBT. For non-relapse mortality (NRM), relapse incidence (RI) and leukemia-free survival (LFS) outcomes were analyzed separately according to disease status, and no statistically significant differences were observed between Haplo and UCBT.

According to disease status, for patients in CR1 (n=610), 2 years RI was 27% versus 28%, p=0.67; NRM 26% versus 29%, p=0.30; and 2 years LFS 48% versus 43%, p=0.16; for Haplo and UCBT, respectively.

For patients in CR2 (n=502), 2 years RI was 33% versus 28%, p=0.51; NRM 33% versus 33%, p=0.93; and 2 years LFS 34% versus 39%, p=0.61, for Haplo and UCBT, respectively.

For patients in advanced disease status (n=334), 2 years LFS was 11% versus 14%, p=0.92 for Haplo and UCBT, respectively.

Infections and GVHD were the most common causes of transplant-related deaths in both groups, (infection 30% in both groups, GVHD 15% and 9%, after Haplo and UCBT, respectively), and 40% of deaths were due to disease recurrence in both Haplo and UCB recipients.

In multivariate analysis, UCBT was associated with lower incidence of chronic GVHD (HR= 0.62; p<0.001) and higher NRM (HR= 1.28; p=0.04) when compared to Haplo. No statistically significant differences were observed between Haplo and UCBT for RI (HR= 0.87; p=0.24) and LFS (HR= 1.06; p=0.47). Factors independently associated with lower LFS were disease status at transplant (HR 2.69, p=<0.001), and use of ATG (HR 1.28, p=0.001).

In conclusion, in this retrospective analysis, non TCD Haplo and UCBT showed similar RI, NRM, and LFS, highlighting that both approaches are valid for this population of patients. Longer follow-up, more homogeneous transplant strategies (stem cell source, type of conditioning regimen and GVHD prophylaxis) are needed for further evaluation. While waiting for the results of the BMT CTN trial on non TCD Haplo using post-transplant Cy and double UCBT both in the RIC setting, the choice of the specific approach for Haplo or UCBT should be likely based on transplant centre expertise and policy.