Abstract
Background:
Since the first publication at the John Hopkins Hospital, (Luznik et al. BBMT 2008), Haploidentical T-cell replete Stem Cell Transplantation (HaploSCT) with post transplantation cyclophosphamide (PT-Cy) has become a reproducible and feasibility therapeutic option for many patients (pts) with hematologic malignancies , notably because of the low incidence of GVHD and infections, without increased graft failure. Initially, bone marrow (BM) was considered as the favorite source of hematopoietic stem cells in order to minimize the risk of GVHD. In a retrospective study, we previously showed no difference in terms of increased events (GVHD, NRM), regardless of the hematopoietic stem cell source (PBSC or BM) (Castagna et al. BMT 2014). Some recent publication seems to confirm this observation (Sugita J. BBMT 2015 , Solomon R. Adv in Hem. 2016), although no randomized study so far has been conducted.
Here, we retrospectively analyze the incidence and the characteristics of GVHD in the setting of HaploSCT using PT-Cy, after infusion of PBSC.
Methods:
Inclusion criteria were: adult pts with hematologic malignancies receiving a PBSC HaploSCT from 2012 to 2015 in 4 centers (3 in France and 1 in Italy) with PT-Cy as part of the GVHD prophylaxis.
PBSC infusion at day 0 was followed by PT-Cy 50 mg/kg on days +3 and +4 in association with calcineurin inhibitors (cyclosporine A or Tacrolimus) and mycophenolate mofetil (MMF), started at day +5. All patients received G-CSF support from day+5 until neutrophil recovery.
Study end points were the cumulative incidences of acute (a) and chronic (c) GVHD, with a specific organ grading evaluation, non-relapse mortality (NRM), relapse (CIR) as well as progression free (PFS) and overall survival (OS). Additionally, we analyzed the composite endpoint "GVHD and relapse free survival" (GFRS) for which the occurrence of relapse, death or severe chronic GVHD was considered as relevant events.
Correlation between CD 34+ and CD 3+ and the incidence of aGVHD and cGVHD was studied by a linear continuous variable.
Results:
Between March 2012 and December 2015, 192 pts with a median age of 57 years (range: 16-73) received T-cell replete PBSC HaploSCT for hematologic malignancies (myeloid: n= 55%; lymphoid: n=45%) in 4 centers.
Patient's characteristics are shown in table 1. Pts received non myeloablative (according to Baltimore regimen) or busulfan-based reduced intensity conditioning, in 56% and 44% of cases, respectively. All, but 3 pts, engrafted, with a median time of 19 days (range, 14-47) to neutrophil recovery (ANC >500 x106/L) and 22 days (range, 14-252) to platelet recovery (PLT > 20 G x 109/L). The median CD34+ x 106/Kg and CD3+ x 106/Kg cells infused were 5.5 (range, 1.5-14.8) and 404 (range, 38-704), respectively. No relevant correlation was observed between the CD34+ and CD3+ infused cells and the incidence of GVHD, studied by linear continues variable. We noted only a trend to develop severe cGVHD with an increasing number of CD3+ cells infused. This result has to be considered with caution, because of the small events (6 pts affected by severe cGVHD).
Complete donor T cell chimerisme was evaluable in 162 pts (83%) and achieved by day +30. The incidence of aGVHD was 38% at 100 days (all grades), whereas grade II-IV and III-IV were 24% and 10%, respectively. Concerning patients with aGVHD grade 2-4, the most affected organ was skin (19%), followed by gut (9%) and liver (2%). The incidence of 3-year cGVHD according to NIH classification was 15% (all grades).
Three percent of pts developed severe cGVHD (lung n=1; liver, n=1). The most frequent involved organs were skin and mucosae (70%). No patient showed gut cGVHD. Finally, in univariate analysis, busulfan-based conditioning seems to negatively impact on severe cGVHD (p = 0.03; HR=3.37 [1.09 -10.46])
After a median follow up of 20 (range, 4 - 52) months, NRM at 100 days and 1 year was 10% and 20%, respectively. Three-year OS, PFS, CIR and GRFS were 63%, 55%, 25% and 49%, respectively.
Conclusion:
This retrospective study shows a very low incidence of severe cGVHD after HaploSCT even with PBSC as stem cell source, suggesting that the use of PT-Cy may overcome the anticipated increased incidence of cGVHD, contrary to as previously reported in the HLA identical setting (Mohty et al. Leukemia 2003).
Similar to HLA identical sibling and unrelated donor transplantation, the most frequent organs involved are skin and mucosae
No relevant conflicts of interest to declare.
