Allo- HSCT for Advanced Myelodysplastic Syndrome:MSKCC Vs MDACC

Allogeneic hematopoietic stem cell transplantation (Allo-HSCT) is the only curative treatment for myelodysplastic syndrome (MDS). Relapse, infections and graft-versus-host disease (GvHD) are the main causes of treatment failure. We compared the outcomes of patients receiving T cell depleted (TCD) grafts at Memorial Sloan-Kettering Cancer Center (MSKCC) with patients receiving unmodified grafts at MD Anderson Cancer Center (MDACC) for advanced MDS (RAEB-1 and 2).

Adult patients transplanted between 2001 -2012 were included in this retrospective analysis. All recipients of TCD grafts (N=60) received myeloablative conditioning (MAC) and antithymocyte globulin (ATG) to prevent graft rejection. None of them received post-transplant GvHD prophylaxis. Of the 129 recipients of unmodified grafts, 87 received MAC and 42 reduced intensity conditioning (RIC); GvHD prophylaxis consisted of tacrolimus and mini-dose methotrexate in the majority of patients (N=113). ATG was given to all matched unrelated donor (MUD) recipients. Patients in the unmodified group had more therapy-related MDS (MDS-t), very poor risk cytogenetics by IPSS-R at diagnosis and bone marrow (BM) blast count >5% at transplant. Only the TCD group had mismatched donors (Table 1).

Univariate analysis identified a lower incidence of grade II-IV acute GvHD in the TCD group with 100-day cumulative incidence (CI) of 13.3% vs. 34.1% in the unmodified group (p=0.031). There was no difference in grade III-IV acute GvHD with a 10% CI in both groups at day-100 (p=0.546). The incidence of chronic GvHD was lower in the TCD group with a CI at 3-yrs of 3.4% vs. 44.3% in the unmodified group (p < 0.001). The non-relapse mortality (NRM) in both groups was similar. CIs at day 100, 1yr, and 3 yrs in the TCD group were 8.3%, 20.2% and 32.7% vs. 12.4%, 22.5% and 28.1% in the unmodified group (p=0.628). Relapse was lower in the TCD group, with CI at 1 and 3 yrs of 8.5% and 15.5%, vs. 31.0% and 39.4% in the unmodified group (p=0.002). Since the unmodified recipients had worse disease characteristics, further analyses in patients with good/intermediate risk cytogenetic showed that the relapse incidence was similar between these subgroups, with 3-yr CIs of 7.9% in TCD vs. 18% in unmodified group (p=0.185). The most common causes of death in the TCD group were infections (32%) and relapse (28%), while in the unmodified group it was relapse (55%), GVHD (20%) and infections (13%). Considering the differences in disease characteristics between the groups, multivariate regression models were performed for relapse-free survival (RFS) and overall survival (OS) adjusting for MDS-t, high-risk cytogenetics at diagnosis and high blast count at HSCT. No significant differences were observed between the groups for RFS (HR=1.44, p=0.128) and OS (HR= 1.35, p=0.236) (Table 2). High-risk cytogenetics at diagnosis (very poor risk) was the only significant prognostic factor for RFS (HR=5.32, p<0.001) and OS (HR=4.81, p<0.001).

Allo-HSCT is an effective treatment for patients with MDS with similar long term survival with either unmodified or TCD allografts. TCD is associated with a lower incidence of acute and chronic GVHD and without an increased risk of relapse.