Chimeric Antigen Receptor T Cell Therapy Bridged to the Second Allogeneic Hematopoietic Stem Cell Transplantation with Reduced-Toxicity Conditioning for B-Cell Acute Lymphoblastic Leukemia Relapsed after the First Transplant

Introduction: The prognosis of relapsed B-cell acute lymphoblastic leukemia (B-ALL) after allogeneic hematopoietic stem cell transplantation (allo-HSCT) with conventional therapies is very poor. CD19 or CD22-targeted chimeric antigen receptor T cell (CAR-T) therapy has achieved high complete remission (CR) in this setting. The second allo-HSCT could be as consolidation therapy for some patients with high-risk features for relapse. High relapse and transplant-related mortality (TRM) are major obstacles for the second allo-HSCT. Some strategies such as reduced-toxicity conditioning (RTC), maintenance therapy post-transplant may improve the outcomes of the second allo-HSCT.

Objective: In current study, the outcomes of CAR-T therapy bridged to the second RTC allo-HSCT for B-ALL relapsed after the first transplant were evaluated. The risk factors for prognosis were also analyzed.

Methods: Between September 2018 and December 2023, 48 patients with relapsed B-ALL after the first allo-HSCT who received CD19 or CD22 CAR-T therapy then bridged to the second allo-HSCT in our hospital were analyzed retrospectively. The median age was 22 (5–55) years old. Before the second allo-HSCT, all patients were in CR with minimal residual disease (MRD) negative in 46, and MRD positive in 2. The more frequent genetic abnormalities were TP53 mutation (n=8, mutation in 6, deletion in 2), RAS mutation (n=8) and BCR/ABL1⁺ (n=8). The median interval between two transplants was 16.3 (6–67.5) months. The types of the second allo-HSCT included haploidentical (n=35), unrelated (n=11) and identical sibling (n=2). For the second allo-HSCT, all donors were changed except one recipient with the same donor due to no other donor available. RTC regimens were mainly total body irradiation (TBI, fractionated, 8Gy in 30, 10Gy in 7)/fludarabine (FLU, 30mg/m² for 5 days)-based (n=37) or busulfan (BU, 0.8mg/kg q6h for 3 days)/FLU-based (n=8).Two patients were with total marrow irradiation (TMI, fractionated, 10-12Gy)/FLU-based regimen and one patient was with BU/Cladribine (5mg/m² for 5 days)-based conditioning. ATG was applied in haploidentical and unrelated transplants. Cyclosporine, mycophenolate mofetil and short-term methotrexate were employed for graft-versus-host disease (GVHD) prophylaxis. Some patients received maintenance therapy post-transplant with targeted medicine based on their fusion genes or gene mutations.

Results: All patients became full donor chimerism. The median time for neutrophil and platelet recovery was 15 (9-22) days and 13 (6-58) days. The incidences of grade II-IV acute GVHD (aGVHD) and chronic GVHD (cGVHD) were 14.6%, 47.9% (limited in 33.3%, extensive in 14.6%), respectively. The incidences of CMV and EBV reactivation were 39.5% and 2%. The median follow-up was 17.8 (1.6-68.6) months. One-year overall survival (OS) and leukemia-free survival (LFS) of all patients were 69.6% and 67.8 %, and 3-year OS and LFS were 63.8% and 65.3%. Sixteen patients died (relapse in 11, infection in 4 and TA-TMA in 1). One-year and 3-year cumulative relapse rate were 22.5% and 25.4%. Three-year TRM was 10.9%. The patients with Ph⁺ or Ph^- had similar OS (p=0.7) and LFS (p=0.76). The patients with RAS or TP53 mutation had lower OS and LFS than that without RAS or TP53 mutation but no significant differences (RAS: 3-years OS 50% vs. 67.2%, p=0.36; 3-years LFS 50% vs. 68.5%, p=0.32. TP53: 3-years OS 50% vs. 66.2%, p=0.25; 3-years LFS 50% vs. 68.3%, p=0.19). Univariate analysis identified that TBI-based regimen improved OS (p=0.0004) and LFS (p=0.0002), but the interval more than 12 months between two transplants (OS: p=0.142; LFS: p=0.122), haploidentical compared with unrelated (OS: p=0.637; LFS: p=0.620), older than 14 year (OS: p=0.896; LFS: p=0.0.872), more than 6 months from first HSCT to relapse (OS: p=0.438; LFS: p=0.451) had not significant impact on OS and LFS.

Conclusions: With our strategies, the outcomes of CAR-T bridged to the second RTC all-HCST in B-ALL relapsed after the first allo-HSCT are very encouraging. Both 3-year TRM and relapse rate were much lower (10.8%, 25.4%) which translated into much higher OS and LFS. Relapse was the main cause of death. TBI-based regimen was a good factor for improved OS and LFS. The patients with Ph⁺ or Ph^- had similar OS (p=0.7) and LFS (p=0.76). Patients with RAS or TP53 mutations had lower OS and LFS but no statistical significance.