Ptcy + ATG Compared to Ptcy Alone As Gvhd Prophylaxis for Peripheral Blood Stem Cells Allotransplant Is Associated with Significant Lower Incidence of Acute Gvhd without Increasing Relapse or Death By Infection

Backgound: High-dose post-transplant cyclophosphamide (PTCY) is currently used for GVHD prophylaxis both in the matched and the haplo transplant settings. However, peripheral blood stem cells (PBSC) as source of graft are significantly associated with a higher incidence of GVHD after transplant with PTCY compared to the use of bone marrow, especially with haplodonors. New strategies are needed to decrease GVHD incidence in this setting.

Methods: This retrospective study included 115 adults (males n=68, median age: 60 yo (range: 23-71) allotransplanted with PTCY or PTCY+ATG between March 2014 and March 2019 in a single institution. Three conditionings were used: the Baltimore regimen (n=40, period: 11/2013-05/2017) consisting of Fludarabine 30 mg/m²/day (d), d -6 to -2, cyclophosphamide 14.5 mg/kg day -6, low dose total body irradiation 2 Grays d-1; the Clo-Baltimore regimen (n=37, period: 03/2014-04/2017), consisting of Clofarabine 30 mg/m²/d, d -6 to -2, cyclophosphamide 14.5 mg/kg d -6, low dose total body irradiation 2 Grays d-1; and the CloB2A1 regimen (n=38, period: 05/2017-03/2019) consisting of Clofarabine 30 mg/m²/d, d -6 to -2, busulfan 3,4 mg/kg d -3 and d-2, ATG 2,5 mg/kg d-1. The three groups shared most characteristics (gender, disease risk index, type of donor [matched vs haplo], ABO compatibility, donor/recipient CMV status, previous allograft, median age of the donor, median CD34⁺ and CD45⁺ graft cells counts). Baltimore patients were younger (median 56 years vs Clo-Baltimore: 61 years and CloB2A1: 60 years, p=0.02) and mostly transplanted for lymphoid diseases (65% vs 8% vs 19%, p<0.001). The median number of CD3⁺ T cells infused was significantly higher for CloB2A1 patients (27.5 10⁷/kg vs Baltimore 22.65 vs Clo-Baltimore 20.8, p=0.008). All patients received PBSC as stem cell source at d 0 and PTCY 50 mg/kg/d, d +3 and +4 with cyclosporine and mycophenolate mofetyl as GVHD prophylaxis. All patients provided informed consent for data collection before the graft.

Results: The median follow-up was 42 (29-64), 33 (7-63) and 15 (3-25) months for the Baltimore, Clo-Baltimore and CloB2A1 groups, respectively. Primary graft failure was significantly higher in the CloB2A1 group (at day+60: 22% (6 haplo, 1 10/10 and 1 9/10) vs Baltimore 5.2% (1 haplo, 1 10/10) vs Clo-Baltimore 0% (p=0.003). Similar 18-month survivals were observed for the three groups (Baltimore/Clo-Baltimore/CloB2A1): OS: 65% (51-81) vs 67.1% (53-84) vs 51.2% (35-73), p=0.27; DFS: 60% (46-77) vs 56.7% (42-75) vs 37.8% (23-62) p=0.20); GRFS (50% (36-68) vs 37.8% (25-57) vs 36.7% (22-60), p=0.63). The incidences of relapse and death as well as the causes of deaths (no more death by infection in the CLOB2A1 sub-group) were also comparable. Conversely, considering engrafted patients, the incidence of grade 2-4 acute GVHD was significantly lower in the CloB2A1 sub-group (19% vs Baltimore: 50% vs Clo-Baltimore: 51.4%, p=0.009) and despite higher numbers of CD3⁺ T cells infused in this group. The incidence of grade 3-4 acute GVHD and chronic GVHD (considering engrafted patients alive at day+100) was lower also in the CloB2A1 sub-group (6.2% vs 16.6% vs 13.5%, p=0.32 and 16% vs 29.4% vs 31%, p=0.51). Considering only haplotransplants, 18-month OS (p=0.95), DFS (p=0.98) and GRFS (p=0.39) were similar between the three groups (Baltimore: 57.1%, 50% and 35.7%; Clo-Baltimore: 60%, 53.5%, 35.7%; CloB2A1: 57.9%, 48.3%, 46.6%). The incidence of grade 2-4 acute GVHD remained significantly lower in the CloB2A1 haplo sub-group (p=0.01). Considering only matched transplants, 18-month survivals were significantly lower for the CLoB2A1 sub-group (OS: 44.4% vs Baltimore 83.3% vs Clo-Baltimore: 88.8%, p=0.05; DFS: 33% vs 83.3% vs 66.6%, p=0.008; GRFS: 22.2% vs 75% vs 44.4%, p=0.008).

Conclusion: The CloB2A1 + PTCY regimen does not appear to be a good platform, neither for haplotransplant nor for matched transplant since 1) the incidence of primary graft failure is relatively high for the former and 2) survivals are lower for the latter. However, this study shows that the addition of one day of ATG to PTCY decreases the incidence of GVHD after transplant without increasing the incidence of relapse or death by infection. The use of ATG +PTCY should be tested in patients receiving PBSC haplo-transplant but with other conditioning regimens allowing for a good engraftment, such as Baltimore-based regimens. These data have to be validated prospectively.