Influence of Donor Type, Stem Cell Source and Conditioning Regimen on Transplant Outcomes after Haploidentical Transplant with Post Transplant Cyclophosphamide for Lymphoma: A Report of the EBMT Lymphoma Working Party

Allogeneic stem cell transplantation (SCT) is potentially curative for patients with lymphoma who progress after autologous SCT. For patients with no HLA identical donor, haploidentical transplant is becoming the major source of alternative donors, particularly with the use of post transplant cyclophosphamide (ptCy). However, there are little data on whether outcomes are affected by the characteristics of the haploidentical donor, the stem cell source or the conditioning. To address this important subject, we identified 474 adult patients (35% females; median age 41 years; range 18-72) with Hodgkin lymphoma (HL-240; 51%), peripheral T cell lymphoma (PTCL-88; 19%), diffuse large B cell lymphoma (DLBCL-77; 16%), mantle cell lymphoma (MCL-40; 8%) or follicular lymphoma (FL-29; 6%), who received a haploidentical SCT (haploSCT) with ptCy between 2010 and 2016 at EBMT participating centers. Patients, donors and transplant characteristics are summarized in Table 1. Median follow-up of alive patients was 32 months (range 3-93).

Engraftment by day 100 was successful in 95% of patients. In multivariate Cox analysis (MVA), the use of peripheral blood stem cells (PBSC) positively affected engraftment (HR=1.53; p<0.001). Day 100 acute GVHD (aGVHD) grade II-IV and grade III-IV were diagnosed in 32% and 8% of patients, respectively, and were significantly affected in multivariate analysis by PBSC (HR=2.1 p<0.001 for grade II-IV and HR=4.5 p=0.001 for grade III-IV). The use of sibling haploidentical donors increased the risk of aGVHD grade II-IV (HR=1.87 p=0.01) whereas a CMV negative donor in a positive recipient increased the risk of aGVHD grade III-IV (HR=5.3 p=0.04). The 2-year cumulative incidence of chronic GVHD (cGVHD) and extensive cGVHD were 25% and 9%, respectively. On MVA, male patients had a higher risk of cGVHD and extensive cGVHD (HR =1.7 p=0.03; and 2.8 p=0.04, respectively), whereas a male donor was protective (HR= 0.6 p=0.02; and 0.3 p=0.008, respectively). The risk of cGVHD and extensive cGVHD was also higher for patients in PR (HR 1.7 p=0.03; and 2.4 p=0.04, respectively). The 2-year cumulative incidence of relapse (CIR) and non-relapse mortality (NRM) were 29% and 19%, respectively. In MVA, both CIR and NRM were negatively affected by the lymphoma subtype and by disease status at transplant, both being highest for DLBCL and for advanced disease. Older patient age was associated with a higher NRM but with a lower CIR. The type of conditioning regimen and CMV compatibility also influenced NRM, whereas ABO incompatibility influenced CIR. The 2-year progression free survival (PFS) and overall survival (OS) for HL were 57% and 72% respectively; 54% and 62% for PTCL; 35% and 35% for DLBCL; 52% and 61% for MCL and 56% and 56% for FL. In MVA, CR at SCT significantly improved PFS and OS whereas a diagnosis of DLBCL as well as a CMV donor positive/recipient positive transplant negatively affected PFS and OS.

In conclusion, this is the largest study on the influence of donor characteristics, stem cell source and conditioning in haploSCT with ptCy for lymphoma. These results provide critical information to help selecting the best donor in the setting of haploSCT for lymphoma. Our results indicate that the use of PBSC significantly improves engraftment but increases the risk of acute GVHD. Conditioning only influenced NRM but had no effect on PFS or OS. The use of female donors increases the risk of chronic GVHD. Of note, PFS and OS are mostly influenced by disease characteristics (i.e disease status and lymphoma subtype), whereas, with the exception of CMV compatibility, no other donor characteristics (donor age and gender, relationship of the donor to the recipient, degree of HLA mismatch or ABO incompatibility, prior donor pregnancy) impact on PFS or OS, supporting the use of any haplo-identical family member as a donor.

Table.

View largeDownload slide

Table.

View largeDownload slide

Close modal