Transplantation from Matched Siblings Using Once-Daily Intravenous Busulfan/Fludarabine with Thymoglobulin: A Myeloablative Regimen with Low Nonrelapse Mortality in All But Older Patients with High-Risk Disease

Two hundred patients received hematopoietic stem cell transplantation (HSCT) from matched sibling donors (MSD) after myeloablative conditioning including fludarabine (Flu) and once-daily intravenous busulfan (Bu). Thymoglobulin (TG) was added to methotexate (MTX) and cyclosporine (CsA) as graft-versus-host disease (GVHD) prophylaxis. For low-risk (acute leukemia CR1/CR2, CML CP1) patients projected 5-year nonrelapse mortality (NRM) and overall survival (OS) were 4% and 76% for those ≤45 years old (n = 54) and 6% and 83% for those >45 (n = 31). For high-risk (HR) patients NRM was 6% versus 27% (18% at 1 year) (P = .04) and OS 64% versus 37% (P = .47) in younger (n = 40) and older (n = 75) patients, respectively. To correct for imbalance in HR diagnoses each of 17 younger HR patients were matched with 2 older HR (OHR) patients by diagnosis and details of stage, and thereafter for other risk factors. For the younger HR and OHR patients, respectively, OS was 70% versus 37% (P = .02) and NRM 0 versus 34% (P = .02). When outcomes of OHR patients were compared with the other 3 groups combined NRM was 27% versus 5%, respectively (P = .002). Incidence of acute graft-versus-host disease (aGVHD) grade II-IV, aGVHD grade III-IV, and chronic GVHD (cGVHD) was 23% versus 10% (P = .02), 4% versus 2% (P = ns), and 66% versus 41% (P = .001), respectively. Nine of 14 nonrelapse deaths in the OHR group were related to GVHD or its treatment compared with 3 of 6 in all others (P value for GVHD related death = .01). Multivariate analysis of OS and DFS correcting for potentially confounding pretransplant factors identified only the OHR patients as having significantly increased risk (relative risk [RR] 3.32, confidence interval [CI] 1.71-6.47, P < .0001, and RR 3.32, CI 1.71-6.43, P < .0001, respectively). The effect of age on NRM is only apparent in HR patients, and is not explained by heterogeneity in diagnoses. Older HR patients experience more GVHD and more GVHD-related death than others, but NRM is no higher than reported with many nonmyeloablative regimens.