Evidence and Efficacy
In 2003, Niederwieser et al. presented the findings of a phase I clinical trial utilising low dose TBI (200cGy on day 0) and fludarabine (30 mg/m2 days -4 to -2) based conditioning in HLA matched and mismatched unrelated transplants.r The study included patients with haematological diseases who were ineligible for conventional allogeneic SCT because of age (> 50 yrs) and/or significant concomitant disease or preceding extensive therapies, such as failed SCT. Exclusion criteria included CrCl ≤ 50 mL/min, LVEF ≤ 30%, bilirubin > 2 X ULN and/or transaminases > 4 X ULN, DLCO < 35%, and Karnofsky performance score < 50. Post transplant immunosuppression was with cyclosporin (CSP) and mycophenolate (MMF). CSP levels were targeted at ~500 ng/mL and tapered from +64 or +100 through +180. Oral MMF 15 mg/kg was given until +27 or +40 and tapered through +96. 52 patients, median age 48 yrs, 88% of whom had undergone previous conventional transplantation or had refractory/advanced disease, were included in this study. Primary endpoint was mixed chimerism at +28, with complete chimerism achieved for NK cells at +28, granulocytes +56 and +180 for CD3 cells. Primary and secondary graft rejection occurred in 12% pts. Acute graft versus host disease (aGVHD) rates were grade II 42%, grade III 8% and grade IV in 13%. GVHD was fatal in 9%. Day 100 transplant-related mortality (TRM) was 11%. 45% achieved complete remission (CR), including molecular remissions. 1 year mortality from disease progression was 27%. 6 patients had donor leukocyte infusion (DLI) for relapsed or persistent disease. At a median follow up of 19 months, overall survival (OS) was 35%, with disease-free survival (DFS) 25%.r
Three phase II multicentre clinical trials and a randomised phase III trial have subsequently been reported. Kerbauy et al.r reported the outcomes of 24 patients with CML, median age 58 yrs (27–71 yrs.), 14 in first chronic (CP1), 4 in second chronic (CP2) and 6 in accelerated phase (AP) who received HLA-matched sibling transplants. 8/24 patients were conditioned with 200cGy TBI alone and the remainder (16/24) received fludarabine 90 mg/m2 with TBI. All patients initially engrafted. However, 4/8 patients in the TBI-alone arm experienced graft rejection between 3-6 months post SCT despite DLI while all others had sustained engraftment. At a median follow up of 36 months, non-relapse mortality (NRM) was 21% and DFS 54%. 2 yr estimated OS for CP1 70% and CP2 56%, with DFS at 36 months.
Hegenbart et al.r presented the outcomes of Flu TBI 200 cGy based conditioning in 122 pts AML patients transplanted using both related and unrelated donors. Median age 57 yrs, with 51/122 pts in CR1, 39/122 in CR2 and 32 patients had more advanced stages of AML. Cumulative incidence of aGVHD grades II-IV at 6 months was 35% after related and 42% after unrelated HCT, respectively. Cumulative NRMs were 10% and 22%, and cumulative mortalities from disease progression were 47% and 33% at 2 years for related and unrelated recipients, respectively. 2-yr OS was 48%, DFS was 44%. Patients transplanted in CR1 had 2-yr OS of 44% after related and 63% after unrelated transplant, respectively.
More recently a phase II multicentre randomised study of 94 patients compared Flu-TBI to 8Gy TLI + ATG for various haematological malignancies.r Flu-TBI was associated with a higher incidence of chronic GVHD (40.8% vs 17.8%, p=0.017), reduced incidence of relapse/progression at 4 years (22% vs 50%, p=0.07) but no difference in 4-year overall (53% vs 54%) or progression free survival (54% vs 37%)r. The incidence of infection (bacterial, fungal and CMV) in the first 100 days was lower in the Flu-TBI group, but statistical significance was not achieved. Similarly, there was no statistical difference in cumulative incidence of grade 2-4 aGVHD at day 180 or NRM at 4 years between the 2 groups.
A randomised, phase III trial compared the addition of fludarabine to 2 Gy TBI (n=41) versus TBI alone (n=44). The incidence of acute and chronic GVHD was higher in the combined-modality group, but statistical significance was not achieved. There was a trend towards a higher progression/relapse rate in the TBI alone group, leading to a worse progression free survival (3yr 36% v 53%, p=0.05). Furthermore, there was a trend towards higher OS in the Flu-TBI arm (54% v 65%, p=0.09).r A further randomised study compared FluBu-ATG (Arm A) with Flu-TBI (Arm B)r Whilst the FluBu-ATG was associated with a reduction in cumulative incidence of relapse (1yr CIR 14% v 37%, HR 2.49 CI 1.45-2.49, P<0.01), this was balanced by an increase in NRM (1yr NRM HR 0.48; 95% CI, 0.25-0.92, P=0.027), leading to no detectable difference in overall survival at 1 year and 3 years. A prospective phase II study compared the Flu-TBI protocol (n=25 )with FluBu-ATG (n=41).r Scheduling was similar to previous studies, except for the use of tacrolimus, rather than ciclosporin, for GVHD prophylaxis. Overall survival, relapse-free survival, and cumulative incidences of acute and chronic GVHD were similar in both cohorts.