Incorporation of 8Gy total body irradiation into reduced intensity conditioning does not improve allogeneic transplant outcomes for high-risk adult acute lymphoblastic leukemia: Results from the randomised prospective phase 2 UK multicentre ALL-RIC impact study

Abstract

Abstract Background Allogeneic stem cell transplant (allo-SCT) remains an important curative therapeutic modality in adults with high risk ALL. Reduced intensity conditioning (RIC) has extended the curative potential of allo-SCT to increasing numbers of older fit adults. There have been, however, no prospective randomised trials to guide the choice of the optimal RIC regimen in older adults with ALL. The UKALL14 trial previously reported outcomes for 249 adults >40y with ALL in CR1 who received fludarabine, melphalan and alemtuzumab (FMA) RIC allo-SCT. Overall survival (OS) at 4y was 55% with transplant-related mortality (TRM) 19.6%. The major cause of treatment failure was relapse (33.6% of patients by 4y) (Marks et al. 2022). Recent registry data has indicated that use of TBI is associated with improved disease outcomes (Giebel et al. 2017) in adult ALL patients receiving allo-SCT. We therefore performed a prospective, randomised comparison of the UK FMA RIC regimen with a cyclophosphamide plus 8Gy TBI RIC protocol (Cy/8TBI), with the goal of improving OS in high-risk adult ALL. Methods The FMA RIC regimen (fludarabine 30mg/m2 IV for 5d; melphalan 140mg/m2 IV single dose; alemtuzumab 30mg IV D-1 for sibling donor/20mg IV D-2 & -1 for unrelated donor) was compared to cyclophosphamide 50mg/kg for 2d plus 8Gy TBI (4# over 2d) and alemtuzumab (dosed as per FMA). Intrathecal prophylaxis was given for 2y post-SCT. Donor lymphocyte infusions were permitted for persistent minimal residual disease (MRD) or mixed donor T-cell chimerism from 3mo post-SCT. Primary endpoint was disease-free survival (DFS), with secondary endpoints including cumulative incidence of relapse (CIR), NRM, OS, GvHD rates and toxicity. Results Total 102 patients from 20 centres were randomised, with 89 proceeding to allo-SCT on study (45 FMA; 44 Cy/8TBI); 94 (92%) were in CR1 and 8 (7.8%) in CR2. Thirty patients (29.4%) had high risk cytogenetics (KMT2A-r; low hypodiploidy; complex karyotype), plus 31 (30.4%) Philadelphia positive cases. Median age of the entire cohort was 52y (IQR 46-59) with treatment arms being well balanced for pre-SCT characteristics. Total 31 (30%) and 71 (70%) patients received sibling and unrelated donor allo-SCT, respectively. Engraftment occurred in 85/89 (96%), with equivalent time to neutrophil and platelet engraftment between arms. With a median follow up of 41 months, 3y DFS by intention to treat (ITT) was 50% and 48% for FMA and Cy/8TBI, respectively (HR 1.12 [95% CI 0.64-1.95]; p=0.7). Neither cytogenetic risk nor pre-SCT MRD significantly affected DFS. At 3y, neither CIR (FMA 31% vs Cy/8TBI 36%; HR 1.23 [95% CI 0.63-2.42]; p=0.6), nor NRM (FMA 19% vs Cy/8TBI 16%; HR 0.81 [95% CI 0.32-2.07]; p=0.7) differed significantly between arms. OS at 3y (ITT) was also similar (FMA 61% vs Cy/8TBI 58%; HR 1.1; [95% CI 0.59-2.05]; p=0.8). Grade 2-4 aGvHD was seen in 2 (4.0%) and 5 (9.6%) patients with FMA and Cy/8TBI, respectively (p=0.44). No grade 3+ aGvHD was seen. Maximum grade 1 aGvHD occurred in 5 (10%) and 8 (15%) after FMA and Cy/8TBI, respectively. cGvHD developed in 4 (8%) FMA and 10 Cy/8TBI (19%) patients, with extensive in 6 (5.9%) patients, and no difference between arms (p=0.15). Total 97 and 102 SAEs were reported (in 30 and 26 patients) in the FMA and Cy/8TBI arms, respectively. Infections were not higher with Cy/8TBI (FMA 26 (27%); Cy/8TBI: 23 (23%)); nor were cardiac and pulmonary SAEs (FMA: 12 (12%); Cy/8TBI: 11 (11%)). Median total days of hospitalisation in year 1 did not differ between arms (FMA: 26 [IQR 22-33]; Cy/8TBI 27 [IQR 22-37]). During the study period, the most frequent causes of death in both arms were disease relapse (FMA 6 (32%); Cy/8TBI 9 (43%)) and infection (FMA 4 (21%); Cy/8TBI 8 (38%)). Conclusions This is the first prospective, randomised trial comparing RIC allo-SCT conditioning regimens in adult ALL. Cy/8TBI did not demonstrate superiority over a non-TBI FMA RIC protocol, achieving similar DFS and OS at 3y post-SCT. However, incorporation of 8Gy TBI into a RIC protocol designed for older adults was well tolerated, with no increases in acute or chronic GvHD, and no additional infectious or extramedullary toxicity. These data highlight the importance of performing randomised trials of innovative conditioning regimens, if outcomes for older patients receiving allo-SCT for ALL are to be improved. Importantly the ALL-RIC trial showcases feasibility and patient appetite for randomised transplant trials.