Thiotepa-Based Vs TBI-Based Myeloablative Conditioning Prior To Allogeneic Stem Cell Transplantation (HSCT) For Acute Myeloid Leukemia (AML) In First Complete Remission (CR1): A Retrospective Analysis From The ALWP Of The EBMT

Thiotepa (N,N'N'-triethylenethiophosphoramide), which is an alkylating compound, has an antineoplastic activity and has been used in oncology (e.g. breast-, ovarian- and bladder cancer) for decades. In the recent years, and because of its good safety profile, Thiotepa has been increasingly used both for autologous and allogeneic hematopoietic stem cell transplantation conditioning. Interestingly, this agent has a very active myeloablative activity but also its mechanism of action can mimick the effect of radiation.

With this background, the aim of this study was to compare outcome of patients receiving a myeloablative conditioning consisting of either high dose TBI or Thiotepa-based chemotherapy.

Inclusion criteria were adults with AML, first allograft in CR1 from an HLA-matched sibling donor (MSD) or an unrelated donor (UD) between 2000 and 2011 and myeloablative conditioning. We first compared patient and transplant characteristics between the two types of conditioning, and then performed a pair-matched analysis.

The number of patients was 2833 in the TBI group and 102 in the Thiotepa group. Patients who received Thiotepa were older (49y vs 40y, p<10^-4), transplanted more recently (2009 vs 2006, p<10^-4) and later after the diagnosis of AML (183 days vs 143 days, p<10^-4). The percentage of secondary AML was also higher in the Thiotepa group (14% vs 6%; p=0.0002). There was no difference regarding patient/donor gender, type of donor and source of stem cells.

In this cohort, we were able to match 96 patients who received Thiotepa with 185 patients who received high dose TBI. Matching factors were: age at transplantation (10 years classes), year of transplant, interval from diagnosis to transplant (less or more than median day), secondary AML and type of donor (MSD/UD). The characteristics of the 2 groups are summarized in Table 1.

Median dose of TBI was 12 Gy (range, 8-16). In this group, TBI was combined with Cyclophosphamide (84% of cases), Fludarabine (14%) or other compounds (2%). On the other hand, Thiotepa was administered with Cyclophosphamide (45%), Fludarabine (54%) with/without Busulfan and other combinations (1%).

Engraftment occurred in 96% of patients using Thiotepa-based conditioning versus 99% after TBI (p=0.11). The interval from transplant to neutrophils count>500/µL was 16 days (range, 9-42) versus 17 days (range, 9-81) in the 2 groups, respectively (p=0.23). Acute GvHD grade II+ was observed in 25 patients (27%) after Thiotepa-containing regimen versus 42 patients (25%) after TBI (p=0.78). 2-years cumulative incidence of chronic GVHD was 48±4% and 41±6% in the 2 groups, respectively (p=0.15). The 2-year cumulative incidences of non-relapse mortality (NRM) was 21±4% versus 27±4% (p=0.57) and relapse incidence (RI) was 18±4% versus 21±3% (p=0.71) in the Thiopeta and TBI groups, respectively. The 2-year leukemia-free survival (LFS) and overall survival (OS) were 61±5% and 64±5% in the Thiotepa group versus 51±4% and 52±4% in the TBI group (LFS: p=0.40; OS: p=0.25).

This pair-matched analysis suggests that a Thiotepa-based myeloablative conditioning regimen prior to allogeneic HSCT in AML in first CR, can allow achieving similar results to high-dose TBI-based myeloablative conditioning. Also, given the deleterious long term side effects of TBI, it is likely that a Thiotepa-based myeloablative conditioning would represent an attractive and valid alternative to TBI. Prospective trials are currently planned in this setting.

Bacigalupo:ADIENNE : Speakers Bureau. Mohty:Riemser : Research Funding.