Safety of myeloablative total marrow and lymphoid irradiation versus chemotherapy-based conditioning in adults with high-risk acute lymphoblastic leukemia in a limited resources setting Free

Introduction

In patients with relapsed or refractory acute lymphoblastic leukemia (RR ALL), allogeneic hematopoietic stem cell transplant (HSCT) is a potentially curative therapeutic option, particularly in middle-income countries where novel immunotherapies are not accessible. Myeloablative total body irradiation (TBI) is considered a standard of care but it is also associated with significant toxicity. Total marrow and lymphoid irradiation (TMLI) is an emerging alternative to TBI that allows the delivery of high radiation to bone marrow and other target structures, with a better toxicity profile.

Objective: This was a retrospective multicenter study aimed at comparing the safety and efficacy of TMLI-based conditioning vs chemotherapy-based conditioning in adults with ALL.

Methods

Consecutive patients with RR ALL 16-65 years from 2023 onwards who were conditioned with a TMLI-based regimen in two institutions were included. As a historical control group, consecutive patients in the 2 years preceding the availability of TMLI were also included. TMLI-based conditioning consisted of cyclophosphamide (Cy) 350 mg/m² and fludarabine (Flu) 25 mg/m² (day -6 to day -4) followed by TMLI total dose of 12 Gy from day -3 to day -1, delivered in 6 fractions of 2 Gy every 12 hours over three consecutive days, delivered using a tomotherapy system. In patients with a history of central nervous system (CNS) infiltration, a cranial 6 Gy boost was added. The graft source was peripheral blood cells for all patients. As prophylaxis for graft-versus-host disease (GVHD), patients received post-transplant cyclophosphamide, tacrolimus and mycophenolate mofetil. Patients in the historical cohort received CyFlu plus melphalan 140 mg/m² plus 2 Gy of TBI for haploidentical recipients. Our primary outcome was 100-day non-relapse mortality. Secondary outcomes included graft failure, adverse events, overall survival (OS), relapse-free survival (RFS), and GVHD rates.

Results

Twenty-two patients were included in the TMLI group, the median age was 21 years ( range, 16-54), 59% male, all had B-cell ALL, 86% were Ph-negative. The HCT comorbidity index was 0 in 72.7%, all had 0-1 ECOG status. The median number of prior treatment lines was 2 (range 1–4) 50% were blinatumomab-exposed and n=2 had a history of prior HSCT. Twelve patients (54.5%) were in first complete remission (CR1) whereas 9 were in CR2 (40.9%). Donors were matched siblings (27.3%, n=6) or haploidentical (72.7%, n=16). The median CD34+ cell dose infused was 10 × 10⁶/kg (range, 5.2–15).

Twenty-one patients were included in the CyFluMel historical control group without significant differences in the baseline characteristics other than a higher proportion of prior exposure to blinatumomab in the TMLI group (50% vs 14.3% [p=.01]). Donors, cell source and CD34+ cells infused were similar. Median follow-up was 8.5 months (range, 1–18) in the TMLI group and 22 months (range, 2–43) in the control group.

No differences were observed in median neutrophil and platelet engraftment for TMLI vs CyFluMel group of 13 vs 14 days (p=0.89) or graft failure (4.5% vs. 0%, p=1) or grade ≥3 mucositis (4.8% vs. 0%, p=0.68). Hemorrhagic cystitis occurred at comparable rates in both groups, while CMV reactivation occurred in 45.5% in TMLI vs 28.6% in controls (p =0.34). The hospitalization rate was lower with TMLI (50% vs. 76.2% p=.076), with non-infectious causes predominated in TMLI (social support 13.6%, oral intolerance 9.1%) vs CyFluMel mainly associated with fever and neutropenia (66.7%) and mucositis (4.8%) (p=.045). The TMLI group had a 4.5% NRM at 100 days (95%CI 0.3-19) vs. 4.8% (95% CI 0.3-20) in controls.

Acute GVHD was observed in 45.5% (n=10) in TMLI vs 61.9% (n=13) in controls, with similar severity (p=0.39). Chronic GVHD in TMLI was 63.6% vs 36.4% in controls (p=0.27), severe cases in the TMLI group (18.6%, n = 4) (p = 0.98). All patients had negative MRD by flow cytometry at day +60. The 1-year OS in TMLI was 76.2% vs 93% in the control group (p=.15). RFS rates were 72% in TMLI vs 78.5% in controls (p = 0.9); 1-year NRM was 4.5% in TMLI vs 9.5% in controls (p=0.7).

Conclusions

Myeloablative doses of TMLI have a similar safety profile to reduced intensity chemotherapy-based conditioning in adults with ALL, with lower rates of hospitalization due to infectious complications. Longer follow-up is needed to establish whether TMLI is more effective than chemotherapy in preventing relapses.