Total body irradiation correlates with chronic graft versus host disease and affects prognosis of patients with acute lymphoblastic leukemia receiving an HLA identical allogeneic bone marrow transplant

Purpose: To investigate whether different procedure variables involved in t he delivery of fractionated total body irradiation (TBI) impact on prognosi s of patients affected by acute lymphoblastic leukemia (ALL) receiving allo geneic bone marrow transplant (BMT). Methods and Materials: Ninety-three consecutive patients with ALL receiving a human leukocyte antigen (HLA) identical allogeneic BMT between 1 August 1983 and 30 September 1995 were conditioned with the same protocol consisti ng of cyclophosphamide and fractionated TBI. The planned total dose of TBI was 12 Gy (2 Gy, twice a day for 3 days). Along the 12-year period, variati ons in delivering TBI schedule occurred with regard to used radiation sourc e, instantaneous dose rate, technical setting, and actual total dose receiv ed by the patient. We tested these different TBI variables as well as facto rs related to patient, state of disease, and transplant-induced disease to investigate their influence on transplant-related mortality, leukemia relap se, and survival. Results: At median follow-up of 7 years (range 3-15 years) the probabilitie s of leukemia-free survival (LFS) and overall survival (OS) for the 93 pati ents were 60% and 41%, respectively. At univariate analysis, chronic graft versus host disease (cGvHd) (p = 0.0005), age (p = 0.01), and state of dise ase (p = 0.03) were factors affecting LFS whereas chronic GvHd (p = 0.0005) , acute GvHd (p = 0.03), age (p = 0.0001), and GvHd prophylaxis (p = 0.01) were factors affecting overall survival. The occurrence of chronic GvHd was correlated with actually delivered TBI dose (p = 0.04). Combined stratific ation of prognostic factors showed that patients who received the planned t otal dose of TBI (12 Gy) and were affected by chronic GvHd had higher proba bilities of LFS (p = 0.01) and OS (p = n.s.) than patients receiving less t han 12 Gy and/or without occurrence of chronic GvHd. Moreover, TBI dose had a significant impact on LFS in patients transplanted in first remission (p = 0.05). At multivariate analysis, TBI dose was an independent factor affe cting overall survival (p = 0.05) as well as chronic GvHd (p = 0.001) and a ge (p = 0.04). Conclusions: This retrospective analysis showed that different variables in volved in TBI delivery may influence the occurrence of cGvHd and affect pro gnosis. of patients with ALL receiving allogeneic BMT. The total dose of 12 Gy, administered in six fractions over 3 days, appears to be an effective and low toxic regimen for ALL patients transplanted in first remission. (C) 1999 Elsevier Science Inc.