Impact of Low-Dose TBI on Outcome of Reduced Intensity Allogeneic Hematopoietic Stem Cell Transplantation from HLA Identical Sibling for Acute Myeloid Leukemia

Introduction

In recent year, reduced-intensity conditioning (RIC) is increasingly used in allogeneic hematopoietic stem cell transplantation (allo-HSCT) for acute myeloid leukemia (AML) patients considered unfit for myeloablative conditioning. Several RIC regimens include low-dose total body irradiation (TBI). However, it is unknown whether addition of low-dose TBI to RIC is beneficial. Therefore, we conducted retrospective analysis to elucidate the clinical impact of low-dose TBI in RIC on outcome of allo-HSCT from HLA identical sibling for AML patients.

Patients and methods

Clinical data of patients with AML aged over 15 years, who received RIC and underwent allo-HSCT from HLA identical sibling for the first time between January 2000 and December 2012, were extracted from the database the Japan Society for Hematopoietic Cell Transplantation. Low-dose TBI was defined as 2-4Gy of TBI. Overall survival (OS), relapse, Non relapse mortality (NRM) and engraftment were compared between patients who received low-dose TBI (TBI group) and patients who did not receive TBI (Non-TBI group) using log-rank test and Gray’s test respectively. To evaluate the association of patient characteristics with low-dose TBI, Cox proportional hazards model and Fine and Gray proportional hazards model were used for multivariate analyses.

Results

There were 409 patients with AML who received RIC and underwent allo-HSCT from HLA identical sibling. Of those, 108 patients were TBI group and 301 patients were Non-TBI group. In term of graft source, a higher proportion of TBI group received bone marrow (BM) than non-TBI group (54.6% vs. 31.2% [P<0.001]). Age, gender, disease status, performance status (PS), cytogenetic risk category were comparable between both groups. The probability of 5-years OS (36.0% vs. 32.8% [P=0.656]), relapse (37.3% vs. 38.8% [P=0.933]), and NRM (23.9% vs. 25.2% P=[0.686]) were comparable between both groups. Although, the median time to engraftment of neutrophil and platelet were shorter in non-TBI group in whole patients (17d vs. 15d [P=0.031], 22.5d vs. 19d [P=0.024]), subgroup analysis according to donor source showed the median time to engraftment of neutrophil and platelet were similar between both groups in BM recipients (17d vs. 16.5d [P=0.395], 25.0d vs. 23.5d [P=0.392]) and peripheral blood recipients (14d vs. 15d [P=0.213], 20d vs. 17d [P=0.174]). In multivariate analysis, low-dose TBI was not associated with OS, relapse, NRM, and engraftment after adjusting graft source, age, gender, disease status, PS, and cytogenetic risk category.

Conclusion

This study showed that adding low-dose TBI to RIC did not affect clinical outcome of allo-HSCT from HLA identical sibling for AML patients. These results suggested RIC regimen without low-dose TBI is valid option for older or medically infirm AML patients who receive allo-HSCT from HLA identical sibling.