Significant and Rapid Improvement in Survival After Unrelated Donor (URD) Hematopoietic Cell Transplantation (HCT): Analysis of National Marrow Donor Program (NMDP) Facilitated Transplants From 2000 to 2009

Abstract 234

Allogeneic HCT using an URD can be curative therapy for patients with high-risk hematologic diseases. Over the past few decades, changes in practice have led to steady improvements in HCT outcomes, but patients and physicians may forgo URD HCT due to safety and efficacy concerns. To determine if outcomes have improved in the current era, we conducted an analysis of 15,509 patients undergoing first URD HCT using either peripheral blood stem cells or bone marrow facilitated by the NMDP from 2000 to 2009. Recipients of myeloablative (MA) and reduced intensity conditioning (RIC) were included. Two time periods were studied (2000–04 and 2005–09) and analysis was stratified by age at HCT: <18 years (y) (N=1255 and 1600), 18–59y (N=3951 and 6035) and ≥ 60y (N=412 and 1806). Acute leukemia was the most common indication for HCT for both time periods (53% and 47% for <18y, 53% and 56% for 18–59y and 53% and 53% for ≥ 60y). Notable changes in practice from 2000–04 to 2005–09 for all age groups included HCT for earlier stage disease, shorter time from diagnosis to HCT, increased use of PBSC as a graft source, decreased use of total body irradiation based conditioning and increased use of HLA 8/8 matched donors. Overall survival (OS) was 12–13% better at 1y for patients transplanted in the more recent era for all three age groups, a highly significant finding (see Table). The large improvement in 1y survival rates was mostly sustained through 3y of follow-up. Survival gains were due to across-the-board reductions in transplant related mortality (TRM) and, in children and the elderly, fewer post-HCT relapses. One contributor to reduced TRM was a significant decline in the incidence of grade 3–4 acute graft-versus-host disease (GVHD) for <18y (21 vs 15%, P <0.001) and 18–59y groups (23 vs 19%, P<0.001) although there was no change for patients ≥ 60y (17 vs 16%, P=0.58). To partially control for variables that routinely affect URD HCT outcomes such as relapse risk, conditioning intensity and HLA match, we conducted additional subset analyses in more homogenous populations when sufficient numbers of patients permitted. In children with malignant diseases, the major contributor to improvement in 1y OS in patients with both standard risk (63 vs 74%, P=<0.001) and advanced malignancies (47 vs 54%, P=0.08) was a significant reduction in TRM (23 vs 15%, P=0.002 and 27 vs 18%, P=0.01); small improvements in relapse rates did not meet statistical significance in the subset analyses. There were sufficient numbers of adults 18–59y for analysis by both disease risk and conditioning intensity. One year OS improvements for MA HCT for standard risk malignancies (55 vs 66%, P<0.001) were accompanied by significant decreases in both TRM and relapse. Improved OS for adults with advanced malignancies receiving either MA (35 vs 46%, P<0.001) or RIC HCT (44 vs 59%, P<0.001) was mainly due to reduced TRM. Adults receiving RIC HCT for standard risk diseases did not have improved OS (58 vs 60%, P=0.54). The large number of adult HCT also permitted subset analyses by HLA match status, which was restricted to patients with myeloid leukemia and MDS to minimize other variables. Better 1y OS in recipients of HLA 8/8 matched donors (52 vs 59%, p<0.001) and 7/8 matched donors (43% vs 50%, P=0.009), was accompanied by significant reduction in TRM in both HLA match groups. There were too few patients ≥ 60y in the earlier period to permit subset analyses. In summary, survival following URD HCT has improved significantly and rapidly over the last decade for all age groups. Reduced early TRM and, in certain populations, fewer relapses were important contributors to survival improvements. Increased availability of better matched donors contributes to but does not completely explain better survival in the most recent era. Multivariate analyses to better understand the practice changes associated with improved outcomes are ongoing.