Lowering Conditioning Intensity in Fit Adolescents and Adults with Lymphoblastic Leukemia without Residual Disease Undergoing Transplantation: A Proof of Concept

Background

Myeloablative conditioning (MAC) is considered standard for fit persons with acute lymphoblastic leukemia (ALL) undergoing allogeneic transplantation (allo-HCT). Next generation flow cytometry allows for the detection of persons without measurable residual disease (MRD) who may not need MAC. Since 2019 Reduced-intensity conditioning (RIC) has been used for persons with undetectable MRD (MRD^neg) regardless of their fitness or comorbidities. In this pre-post study we compare the outcomes of persons with ALL treated before and after the implementation of this strategy.

Methods

This was a prospective study including adolescents and adults with ALL from 2015-2024 in two institutions. Before 2019 patients received MAC based on cyclophosphamide 350 mg/m2 plus fludarabine 25 mg/m2 (CyFlu) for 3 days and melphalan 200 mg/m2 or busulfan (Bu)≥12 mg/kg if considered fit. After 2019 patients with pre-transplant MRD^neg could receive RIC (CyFlu plus melphalan 100-140 mg/m2 or Bu 8-9 mg/kg with 2 Gy of total body irradiation), regardless of comorbidities or fitness in an effort to reduce short-term complications. MRD was centrally assessed using next generation flow cytometry using Euroflow protocols before conditioning. Peripheral blood was used in all cases. Donors included matched siblings or haploidentical. Graft-versus-host disease prophylaxis was based on post-transplant cyclophosphamide or calcineurin inhibitor plus methotrexate.The primary outcome was overall survival (OS). Secondary outcomes were the cumulative incidence of relapse (CIR), non-relapse mortality (NRM) and GVHD-relapse-free survival (GRFS).

Results

Eighty-eight patients received an allo-HCT for ALL during the study period, n=41 RIC and n=31 MAC, the remainder received alternative conditioning were excluded. The median age was 27 years (15-62), 66.7% were men, mostly with B-ALL (87.5%) without comorbidities (HCT-CI score 0; 84%) and a median of 2 treatment lines (range, 1-6). Fifty-seven had MRD^neg, while 15 had detectable measurable residual disease (MRD^pos), 16 were not assessed. Donors were haploidentical in n=56 and matched siblings in n=16. No differences were observed in the baseline characteristics of RIC-MRD^neg vs MAC-MRD^neg groups. The median follow-up was 35.6 months (range 2,7-134). Median day of neutrophil engraftment was longer in MAC-MRD^neg (16 vs. 14 days; p=.016). No differences in platelet engraftment, graft failure, acute or chronic GVHD incidence or severity were observed. Three year OS in RIC-MRD^neg was 77.1% vs. 54.3% for MAC-MRD^neg (p=.2). The 2-year CIR was 44% in RIC-MRD^neg (95% CI 22-64%) and 51% in MAC-MRD^neg (95% CI 23-74%) (p=.6) with a trend towards worse cumulative incidence of 2-year NRM in MAC-MRD^neg 9.8% (95% CI 1.6-27%) vs RIC-MRD^neg 5.9% (95% CI 1-18%) (p=.06); 2 year GRFS RIC-MRD^neg 40.4% and MAC-MRD^neg 28.2%, 3-year GRFS was 34.6% in RIC-MRD^neg and 28.2% in MAC-MRD^neg(p=.5).

The 2-year OS in RIC-MRD^pos 60% vs MAC-MRD^pos 57.1%. Meanwhile the 2-year RIC-MRD^pos CIR was 20% (95% CI 0.35-63%) and MAC-MRD^pos CIR was 33% (6.2-64%). Also, the 2-year NRM in RIC-MRD^pos was 20% (0.43-62%) and the 2-year NRM in MAC-MRD^pos patients was 20% (95% CI 2.6-49%). In a multivariate analysis of OS, pre-transplant MRD was associated with OS (HR 2.9; 95% CI 1.4-5.8) but the intensity of conditioning did not.

Conclusion

For patients with ALL and MRD^neg before transplant, RIC is an acceptable alternative and offers comparable outcomes to MAC. No differences in outcomes were observed, validating our strategy.