Impact Of Rabbit Anti-Thymocyte Globulin-Containing Reduced-Intensity Conditioning Regimens On Outcomes Of Adults Undergoing Unrelated Cord Blood Transplantation For Hematological Malignancies

To assess the impact of antithymocyte globulins (ATG), on patients' outcome after unrelated cord blood transplantation (UCBT) following a reduced-intensity regimen (RIC), we conducted a retrospective registry-based analysis on 661 adults with hematological malignancies who underwent unrelated single (s) or double (d) UCBT following RIC with the TBI/cyclophosphamide/fludarabine regimen (TCF) between 2004 and 2011 in EBMT centers. Participating centers were asked to provide additional information on type, timing and total dose of ATG used. Diagnosis was AML/ALL in 51%, MDS/CML in 19% and lymphoid malignancies in 30%; 28% of patients were transplanted in early disease status, 28% in intermediate and 44% in advanced disease. Thirty percent of patients had a previous autologous transplantation. Single UCBT was used in 226 (34%) patients, while 435 (66%) were transplanted with dUCBT. HLA matching was defined as low resolution for HLA-A and HLA-B, and high resolution for HLA-DRB1, and for dUCBT, the highest degree of HLA incompatibility was considered. Therefore, most of the HLA incompatibilities were 4/6 (n=435, 72%). Median number of collected total nucleated and CD34+ cells were 4.4x10⁷/kg and 1.6x10⁵/kg, respectively. All patients received TCF, with TBI 2Gy (86%), TBI 4Gy (12%) and TBI 6Gy (2%). Rabitt ATG (rATG) was used as part of RIC in 82 patients (12.4%) with a median total dose of rATG (Fresenius®) of 20mg/kg (5-60) and rATG (Genzyme®) of 8mg/kg (5-15). GVHD prophylaxis consisted of cyclosporine A (CsA)+ mycophenolate mofetil (MMF) in 91%, Csa alone±other in 9%. The median follow-up was 36.3 months. When compared to patients not receiving rATG-TCF, patients given rATG-TCF had more MDS/CML (30% vs 20%, p<0.01), were transplanted more recently (p=0.02) and there was a trend of being transplanted with more advanced disease (53% vs 43%, p=0.06). Table below shows overall outcomes for 661 patients and univariate analysis for outcomes by the use of rATG-TCF. Type and dose of rATG were not associated with any outcomes. Multivariate (MV) models for outcomes were built adjusting for the differences between 2 groups of rATG-TCF (yes and no) and other risk factors that impact outcomes (patients's age>51 years, positive CMV serology, MDS/CML, advanced disease status, year of transplant, HLA 4/6 and ABO incompatibility). In the final MV models, use of rATG was associated with decreased incidence of aGVHD (HR=0.31, 95%, CI=0.17-0.55, p<0.0001), higher incidence of NRM (HR=1.68, 95% CI=1.16-2.43, p=0.0009) and decreased OS (HR=1.69, 95% CI=1.19-2.415, p=0.003), however it was not associated with engraftment, chronic GVHD and relapse. In rATG-RIC-group, the main cause of death was transplantation related in 51% of cases. Death related to infections was 72% in the rATG-RIC group compared to 39% in the non rATG group. In conclusion, in this retrospective and multicentre analysis, use of rATG as part of TCF regimen, was associated with decreased incidence of acute GVHD, however with increased NRM and decreased OS, probably related to the higher incidence of infections. Despite the retrospective design of our study, we suggest that in UCBT for adults with hematological malignancies given a TCF regimen, systematic use of rATG might be a matter of concern. In addition, timing and dose of rATG are still open questions and only randomized studies may address this issue.

Yakoub-Agha:Genzyme: Honoraria, Research Funding.