Is Low Dose Total Body Irradiation Necessary in Peripheral Blood Reduced Intensity Conditioning Haploidentical Transplantation?

Background The most suitable reduced-intensity conditioning (RIC) regimen for post-transplant cyclophosphamide haploidentical HCT (haplo-HCT) has not been determined. Various combinations have been proposed, some of which include the use of total body irradiation (TBI). The reasons that may impede the use of TBI are feasibility, accessibility and affordability particularly in limited resources settings. Methods we performed an observational retrospective study of two haplo-HCT centers where the same conditioning regimens are used. We included consecutive adults with neoplastic diseases who received peripheral blood haplo-HCT from 2015 to date. Patients with donor specific antibodies were excluded. We aimed to assess the effect of TBI in RIC haplo-HCT on outcomes in comparison to MAC and RIC without TBI (Chemo-RIC). Conditioning was based on three days of cyclophosphamide and fludarabine plus either melphalan (MEL) 140-200 mg/m2 or busulfan (BU) 8-12 mg/kg. RIC was defined as MEL ≤140 mg/m2 and BU ≤8 mg/kg and was allowed for persons in complete remission and negative residual disease or PET-CT according to the underlying diagnosis. TBI 2 Gy was used on day 0 according to access and the clinical judgement of the transplant team. The primary outcome was the proportion of patients with graft failure (GF). Secondary outcomes included graft-versus-host disease (GVHD), overall (OS), event-free (EFS), and graft-versus-host-relapse free survival (GRFS) by the Kaplan Meier method. Cumulative incidence of non-relapse mortality (NRM) was analyzed considering competing risks. Results 184 patients with a median of 31.5 years (range 10-79) were included. Acute leukemia was the most common diagnosis (n=124; 67.4%). The most frequent conditioning was MAC (n=77, 41.8%), n=68 received TBI-RIC (37%) and n=39 Chemo-RIC (21.2%). Patients treated with TBI-RIC had similar baseline characteristics vs. Chemo-RIC including age, sex, disease risk index, hematopoietic-cell comorbidity index, with a slight difference in CD34+ cells infused (median 10 × 106/kg in Chemo-RIC vs. 9 × 106/kg for TBI-RIC; p=.02). Of 172 patients at risk of GF, 15 had an event, 10 primary (5.4%) and 4 secondary failure (2.2%). Patients who received TBI-RIC had significantly less GF than Chemo-RIC but not MAC (Figure 1, Panel A). A multivariate logistic regression model revealed prior treatment lines [OR 0.4 (95% CI 0.21-0.76)] and TBI-RIC to be associated with a reduced probability of GF [OR 0.14 (95% CI .02-.81)]. NRM at 100-days was 19% in Chemo-RIC, 14% in TBI-RIC and 13% in MAC (p=.015) (Figure 1, Panel B). GVHD incidence and severity, OS, EFS, and GRFS were similar accross the three groups.In conclusion 2 Gy TBI-RIC was associated with less GF and a lower cumulative incidence of NRM in haplo-HCT compared to Chemo-RIC, and resulted in similar outcomes to MAC. TBI-RIC should be preferred if it is available, accessible and affordable.