Abstract 1996

The outcomes of unrelated HSCT have markedly improved with the advent of high resolution HLA-typing. However, graft-versus host disease (GvHD) remains a limiting factor, particularly in mismatched transplants. Several studies have demonstrated that TCD reduces the incidence of acute and chronic GvHD, potentially allowing for improved outcomes in the mismatched setting. We have observed excellent long term survival in our early experience performing matched related TCD HSCT in patients with advanced MDS, particularly in patients who achieve a complete remission (CR) or second refractory cytopenia phase (RCy2) prior to transplant. We report here our experience performing unrelated TCD HSCT in 85 consecutive patients with advanced MDS.

From 1989–2011, 85 patients with advanced MDS (IPSS Intermediate risk [IR]-1 or higher) and AML transformed from MDS underwent TCD HSCT (18 bone marrow [BM], 67 mobilized peripheral blood [PB]) from unrelated donors following conditioning with a total body irradiation-based (25 patients) or a busulfan-based (60 patients) myeloablative regimen. 49 donors were fully matched and 36 were partially matched (9/10 HLA matched: 23; 8/10 HLA matched: 8; 7/10 HLA matched: 1; and 5/6 HLA matched: 4 [before high resolution typing]). The median age was 55 (range 4–73). Prior to conditioning, 80 patients received chemotherapy (28 with a hypomethylating agent, 65 with intensive chemotherapy, and nine with both) and five patients did not receive chemotherapy. Prior to transplant, 34 of the 80 patients who received chemotherapy were in CR, 30 were in RCy2, and 15 failed to achieve remission (10 with RAEB, 5 with AML). Of the five patients who did not receive chemotherapy, two had refractory anemia and three had RAEB. The BM grafts were depleted of T-cells using the soybean agglutinin method followed by sheep RBC rosetting, and the G-CSF mobilized PB stem cell grafts were depleted of T-cells using immunomagnetic CD34+ selection (Isolex initially and CliniMACS after 09/2011). Rejection prophylaxis with ATG was used in all patients. No post-transplant pharmacologic prophylaxis for GvHD was given.

82 of 85 patients engrafted (92%). Three died before engraftment (3.5%, all <28 d after transplant) and two developed late graft failure (2.4%). The day-100 cumulative incidence (CI) of grade II-IV aGVHD was 19% (95% confidence interval [95%CI] 11%-28%), and the 1-year CI of aGvHD, including the late onset form, was 28% (95%CI 18%-38%). When only including grade III-IV aGVHD, the day-100 CI was 9.4% (95%CI 4.3%-17%) and the 1-year CI was 16% (95%CI 9.4%-25%). The 2-year CI of cGVHD was only 3.5% (95%CI 0.9%-9.1%). The overall survival (OS) was 53% (95%CI 43%-63%) at two years and 44% (95%CI 35%-75%) at five years. The relapse free survival (RFS) was 46% (95%CI 36%-57%) at two years and 41% (95%CI 31%-52%) at five years. There was no significant difference in OS/RFS among patients transplanted with fully HLA-matched, 9/10 HLA-matched, or 7–8/10 HLA-matched grafts. There was a trend towards worse OS in patients who had a poor risk (HR) IPSS score at any time prior to transplant; the 2-year OS in this group was 43% (95%CI 32%-60%) versus 64% (95%CI 49%-82%) in the IR-1/IR-2 IPSS groups (p=0.08). Likewise, there was a trend towards worse OS in patients who failed to achieve CR or RCy2 prior to transplant (2-year OS 32% [95%CI 16%-64%]), as compared with patients who achieved CR or RCy2 (2-year OS 58% [95%CI 48%-71%], p=0.25). The overall 1-year CI of relapse was low at 13% (95%CI 7%-21%). The 1-year CI of relapse was significantly higher in patients with IPSS poor risk cytogenetics (27%, 95%CI 5.2%-55%) as compared with intermediate (18%, 95%CI 0.1%-65%) and good (8.4%, 95%CI 0.04%-45%) risk cytogenetics (p=0.03). The 1-year NRM was 36% (95%CI 23%-49%) in those with HR disease and only 18% (95%CI 7%-33%) in those with IR-1/IR-2 risk disease. 5-year OS was superior in transplants done from 2000–2011 (48%, 95%CI 36%-59%) compared with 1989–1999 (25%, 95%CI 6%-50%, p=0.01), reflecting the availability of high resolution HLA-typing and improvements in supportive care.

These results indicate that patients with advanced MDS can achieve durable remissions and long term survival after unrelated TCD HSCT with low rates of acute and chronic GVHD even with mismatched donors. Selecting patients for HSCT before progression to IPSS HR disease and induction into CR or RCy2 prior to transplant may maximize the efficacy of this approach.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.