External validation and extended application of the transplant conditioning intensity score in acute myeloid leukemia

This study aimed to validate the utility of the transplant conditioning intensity (TCI) score in 1714 patients with acute myeloid leukemia (AML) undergoing allogeneic bone marrow or peripheral blood stem cell transplantation (BMT/PBSCT) and assess its applicability to 753 patients with AML undergoing umbilical cord blood transplantation (UCBT) both during first complete remission. Patients classified into a high TCI group accounted for 63% and 56% in the BMT/PBSCT and UCBT cohorts, respectively. In the BMT/PBSCT cohort, the risk of relapse was lower in patients in the high versus intermediate TCI group (P = 0.002), although non-relapse mortality (NRM) did not differ among the three TCI groups. In the UCBT cohort, both relapse and NRM did not differ among the TCI groups. Increasing cutoff points for intermediate and high TCI categories significantly improved the ability to predict relapse and NRM in the BMT/PBSCT cohort (P = 0.030 and 0.006, respectively), and relapse but not NRM in the UCBT cohort (P = 0.005 and 0.364, respectively). These findings highlight the difference in the threshold level of the TCI score for outcome discrimination between European and Japanese cohorts. The TCI scheme appears less effective for UCBT than for BMT/PBSCT.