Homoharringtonine added to CLAG regimen improves outcome of pediatric Relapsed/Refractory AML: A multicenter prospective study

Abstract Induction The prognosis of refractory/relapsed (R/R) Acute myeloid leukemia (AML) is dismal. Regimens including cladribine, cytarabine, and granulocyte-stimulating factor (CLAG) resulted in excellent outcomes with a 50-62.7% complete remission (CR) rate in R/R AML children. The addition of medications with different mechanisms to CLAG can increase the efficacy for R/R AML. Homoharringtonine (HHT) is a natural plant alkaloid derived from Cephalotaxus, and has displayed promising efficacy and tolerability in pediatric and adult newly-diagnosed and R/R AML. Accordingly, we presented data of the largest sample prospective clinical data on the efficacy and safety of HHT with CLAG (CHAG regimen) in childhood R/R AML. Methods The trial in patients aged < 18 years with R/R AML (ChiCTR2000038340) is a single-arm, multicenter, open-label study conducted in China. All patients had received CHAG regimens for salvage chemotherapies as follows: cladribine (5mg/m2/day, day 1-5), HHT (1mg/m2/day, day 1-14), cytarabine (10mg/m2, quaque 12 hora, day 1-14), granulocyte colony-stimulating factor (200ug/m2/day, day1-14) (cycle 1). Patients achieving CR after cycle 1 proceeded to subsequent allogeneic hematopoietic transplantation (allo-HSCT). Those with partial response (PR) or no response (NR) were administered a second cycle of CHAG as consolidation therapy (cycle 2) and then continued to allo-HSCT once CR obtained. For patients not eligible for HSCT, 1-3 cycles of CHAG consolidations would be administered, with efforts to initiate the HSCT process. The primary objective was to determine CR rate of cycle 1, and the second endpoints included MRD rate of cycle 1, CR and MRD rate of cycle 2, and long-term survival [overall survival (OS) and event-free survival (EFS)]. Results Between December 1, 2019, and January 10, 2025, 175 fit children with R/R AML were enrolled. All subjects who received CHAG therapy had been heavily pretreated and had received a median of 3 lines (range, 1-8) of prior therapies, of whom 93 (53.1%) patients with primary refractory and 82 (46.9%) with relapsed AML [11 (6.3%) after allo-HSCT]. Among relapsed patients, 78 (44.6%) patients were first relapse, and 4 (2.3%) were second and greater relapse. The median age was 8 years old (range, 0.5-17). Twenty-four (13.7%) patients presented with extramedullary AML at diagnosis. Twenty-one were identified with FLT3-ITD and 14 with FLT3-TKD. The most common fusion gene included RUNX1::RUNX1T1 in 40 patients (22.9%) and KMT2A-rearranged in 31 (17.7%). Complex karyotype was present in 11.4% (20/175). Patients received a median of two cycles of CHAG (range, 1-5). 75 patients (42.9%) received one cycle and 100 (57.1%) received more than one cycle, including 65 (37.1%) two and 35 (20.0%) ≥ three cycles as trial treatment. The CR rate was 77.1% (135/175), with 68.6% (120/175) attaining MRD negativity via multiparameter flow cytometry (MFC) of cycle 1, and 86.1% (87/101) and 75.2% (76/101) after cycle 2. The CR and MRD negative rate for cycle 1 were 84.6% (66/78) and 76.9% (60/78) for those in first relapse, and 74.2% (72/97) and 61.9% (60/97) for patients in second and greater relapse or with refractory disease. A total of 129 (73.7%) patients consolidated with HSCT in remission, among whom 58 proceeded to HSCT after cycle 1, 55 after cycle 2 and 16 after cycle 3. 39 patients experienced relapse during treatment, including 15 relapsed during conventional chemotherapies and 24 relapsed after allo-HSCT. With a median follow-up of 20.1 months (range, 1-66.9), the median duration of EFS and OS was not reached. The estimated 36-month OS and EFS was 67.8% (95% CI, 60.5-75.9) and 60.7% (95% CI, 53.4-69.0). Acquisition of CR and MFC-MRD negativity after cycle 1, brided to HSCT showed a significant trend toward improved OS and EFS (P < 0.05). Meanwhile, fusion gene of RUNX1-RUNX1T1, CBFB-MYH11 and KMT2Ar, and complex karyotype and previous utilization of HSCT play a statistical significance on OS and EFS (P < 0.05). The common treatment emergent ≥ grade 3 AE occurring were hematologic toxicities, with no grade 5 events and related mortality observed. The most common nonhematologic toxicities were febrile neutropenia, mostly assessed as grade 3 to 4, and not life-threatening. Conclusion In summary, this trial showed that CHAG regimen is safe and highly active in R/R AML children. The CHAG regimen may represent a highly promising bridging strategy to allo-HSCT in pediatric R/R AML.