Graft Failure In Reduced Intensity Conditioning Allogeneic Hematopoietic Stem Cell Transplantation

A decade ago reduced intensity conditioning (RIC) allogeneic hematopoietic stem cell transplantation (allo-HSCT) was successfully introduced to facilitate transplantation in patients with co-morbidities. In contrast to myeloablative conditioning regimens RIC transplants are associated with less toxicity and thus preferred in the elderly population. However, there are also disadvantages with RIC such as the increased risk of graft failure caused by rejection by remnant recipient T-cells at transplant, which often results in mixed chimerism during the first months post-transplant. Here, we present risk factors for graft failure in all first RIC allo-HSCTs performed at our center from 1995 to mid-2010 (n=357). Graft failure was defined as >95% recipient T-cells any time after engraftment or reinfusion of donor cells due to >50% recipient cells with no clinical or laboratory signs of relapse. Thirty-six patients (10%) experienced graft failure. The median age of the patients was 51 and 57% were males. In univariate analysis, graft failure was 3-fold increased in male compared to female patients (15% versus 4%; P<0.01). Furthermore, patients with non-malignant disorders (26% versus 2%; P<0.01) and solid tumors (20% vs. 2%; P<0.01) had an increased risk of graft failure when compared to acute leukemia. Graft versus host disease (GVHD) prophylaxis using cyclosporine A (CSA) in combination with mycophenolate mofetil (MMF) resulted in increased numbers of graft failure when compared to our standard regimen with CSA and methotrexate (18% vs. 5%; P<0.01). The graft failure rate was the lowest among HLA-identical sibling donors (6%), slightly higher in 8/8 matched unrelated donors (10%), and the highest in 6/6 matched unrelated donors (18%; P<0.01). Total nucleated cell (TNC) dose was also important with less graft failures when the TNC dose was >10 x 10⁸/kg, whereas CD34+ cell dose had no impact on graft failure (P=0.15). Moreover, in univariate analyses year of transplant, donor sex, cell source, CMV mismatch, ABO mismatch, or pre-planned G-CSF did not influence graft failure rates (P>0.05). In multivariate analysis (Table 1), solid tumors were associated with markedly increased risk of graft failure (RR=10.95; P=0.03), whereas there was a tendency towards increased graft failure risk in non-malignant disorders (RR=8.01; P=0.05). Graft failures were also increased in male recipients (RR=3.27; P<0.01), in grafts with a TNC dose less than 10 x 10⁸/kg (RR=2.17; P=0.03), when using MMF containing GVHD prophylaxis (RR=3.61; P<0.01), and in transplants with HLA match less than 8/8 (RR=4.94; P<0.01). In conclusion, graft failure is augmented in male recipients, and associated with diseases which usually do not receive high-dose chemotherapy pre-transplant such as solid tumors. HLA mismatch as well as TNC dose and GVHD prophylaxis were also important risk factors for graft failure in the present study.