Long-Term Follow-up of Allogeneic Hematopoietic Stem Cell Transplantation after Chimeric Antigen Receptor T-Cell Therapy in Relapsed/Refractory B-Cell Acute Lymphoblastic Leukemia

Abstract

Introduction: Chimeric antigen receptor T-cell (CART) Therapy has induced high complete remission (CR) in the patients with relapsed/refractory (r/r) B-cell acute lymphoblastic leukemia (B-ALL), but the median event-free survival is only 6.1 months (Park JH. et al. NEJM 2018). To achieve durable leukemia-free survival (LFS), it may helpful to receive allogeneic hematopoietic stem cell transplantation (allo-HSCT) after CR with CART therapy. Objective: In current study, safety and long-term outcomes of allo-HSCT after CART therapy in r/r B-ALL were assessed and risk factors for overall survival (OS) and LFS were analyzed. Methods: Between June 2017 to June 2020, 137 patients with r/r B-ALL who achieved CR with CD19 or CD22 CART then bridge to allo-HSCT in our hospital were evaluated. The median age was 9.7 years (1.8-56). Bone marrow (BM) samples were collected to detect somatic gene mutations of leukemic cells. Blood samples from patients, parents, and potential related donors were collected to analyze germline mutations of hematological and immunological hereditary predisposition genes. The disease status before transplant was minimal residual disease (MRD) negative in 114 (83.2%) patients and MRD positive in 23 (16.8%) patients. The median time from CART infusion to allo-HSCT was 51 days (29-94). The types of transplant were haploidentical (105, 76.6%), unrelated (26, 19.0%), and identical siblings (6, 4.4%). Myeloablative reduced-toxicity conditioning (RTC) with either total body irradiation (TBI)/fludarabine-based (84, 61.3%) or busulfan (BU) /fludarabine-based (53, 38.7%) were applied. Cyclosporine, mycophenolate mofetil and short-term methotrexate were employed for graft-versus-host disease (GVHD) prophylaxis. Results: Full donor engraftment was achieved in all patients. The median time for neutrophil and platelet engraftment was 15 days (9-21) and 13 days (5-33). With a median follow-up of 653 days (481-783), two-year OS was 74.1% (64.4-81.5) and two-year LFS was 64.4% (54.6-72.6). The incidences of grade Ⅱ to Ⅳ and severe acute GVHD (aGVHD) were 27%, 14.5%, respectively. Twenty-five (18.2%) cases developed chronic GVHD (cGVHD) (limited 18, extensive 7). The incidences of CMV and EBV viremia were 20.1%, 12.5%. Hemorrhagic cystitis was found in 32 (22.2%) cases. Thirty-six (26.3%) cases relapsed after transplant, 28/114 (24.6%) in MRD negative cohort and 8/23 (43.5%) in MRD positive cohort. Thirty (21.9%) patients died (relapsed 22, infection 5, multiple organ failure 2, thrombotic microangiopathy 1). Transplant-related mortality was only 5.8%. Total 137 patients with 99 kinds of somatic mutations, the most frequent 15 mutations were TP53, CYP2C19, KRAS, NRAS, PTPN11, CREBBP, KMT2D, KIT, NR3C1, STK11, FLT3, NF1, NUDT15, ABL1, ADAMTS13. LFS was significantly reduced in patients with TP53, KIT, NF1 mutation (P < 0.0001; P < 0.025; P < 0.003). Total 114 evaluable patients with 93 kinds of hematological and immunological hereditary predisposition gene mutations, the most frequent 15 mutations were BLTA, SERPINE1, CYP2C19, F7, TP53, DPYD, MLH1, TNFAIP3, YARS2, HIFIA, ATM, EP300, KIT, ADAMTS13, BRCA2. LFS was significantly reduced in patients with germline EP300 mutation (P < 0.00027). Univariate analysis showed that BU-based regimen (P=0.0194), somatic TP53 mutation (P= 0.002), fungal infection (P= 0.0217), viral infection (P=0.0187) and MRD positive before transplant (P= 0.0227) were adverse factors of OS. The negative factors of LFS were BU-based regimen (P=0.0139), germline EP300 mutation (P<0.001) and fungal infection (P =0.0002). Multivariate analysis showed that TBI-based regimen (P=0.019) and aGVHD (P=0.047) increased LFS remarkably, but fungal infection (P=0.002), somatic TP53 mutation (P<0.001) and germline EP300 mutation (P<0.001) decreased LFS significantly. Conclusions: Our large series and long-term outcomes have demonstrated that LFS and OS have been improved remarkably in r/r B-ALL who achieved CR with CART then quick bridge to RTC allo-HSCT. Our study has shown the somatic and germline mutation profiles in r/r/ B-ALL and found that somatic TP53 mutation and germline EP300 mutation are negative factors for LFS. TBI-based regimen , MRD negative before transplant and aGVHD have positive impact on LFS, and fungal infection, somatic TP53 and germline EP300 have negative impact on LFS. Disclosures No relevant conflicts of interest to declare.