AUTHORS
Mei Li, Xiaoqin Feng2, Chunfu Li, PhD3, Mincui Zheng4, Huirong Mai5, Lihua Yang, MD6, Hua Jiang, MD7, Xiangling He8, Hong Wen9, Honggui Xu10, Chun Chen, PhD11, Nainong Li, MD12, Yongzhi Zheng, PhD13
AFFILIATIONS
1 Fujian Medical University Union Hospital, Fuzhou, China
2 Nanfang Hospital, Southern Medical University, Fuzhou, China
3 Nanfang-Chunfu Children’s Institute of Hematology & Oncology, TaiXin Hospital, Dongguan, China
4 Hunan Children’s Hospital, Changsha, China
5 Shenzhen Children’s Hospital, Shenzhen, China
6 Zhujiang Hospital of Southern Medical University, Zhujiang, China
7 Guangzhou Women and Children’s Medical Center, Guangzhou, China
8 People’s Hospital of Hunan Province, Changsha, China
9 The First Affiliated Hospital of Xiamen University, Xiamen, China
10 Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China
11 The Seventh Affiliated Hospital, Sun Yat-sen University, Shenzhen, China
12 Fujian Medical University Union Hospital, Fuzhou, China
13 Fujian Medical University Union Hospital,, Fuzhou, China
ABSTRACT
Objective
To evaluate whether allogeneic hematopoietic stem cell transplantation (allo-HSCT) in first complete remission (CR1) confers a survival advantage over chemotherapy consolidation in children with non-MLLT3 KMT2A-rearranged (KMT2A-r) acute myeloid leukemia (AML) treated with FLAG-IDA induction.
Methods
We analyzed 875 children with de novo non-APL AML prospectively enrolled in the multicenter C-HUANAN-AML-15 study, including 153 with KMT2A-r AML. Outcomes were compared by KMT2A rearrangement status, fusion partner, induction regimen (FLAG-IDA vs. DAE), and postremission strategy (chemotherapy consolidation vs. CR1 allo-HSCT). Event-free survival (EFS), overall survival (OS), and cumulative incidence of relapse (CIR) were estimated using Kaplan–Meier and competing-risk methods. Propensity score matching was applied to address induction selection bias, and multivariable Cox and Fine–Gray models were used to identify independent prognostic factors and assess the effect of CR1 transplantation.
Results
KMT2A-r AML was associated with 5-year EFS, OS, and CIR comparable to those of other AML subtypes, although substantial outcome heterogeneity was observed across KMT2A fusion partners. In the five-group analysis, patients were classified as having KMT2A::MLLT3 AML, non-MLLT3 KMT2A-r AML, or non–KMT2A-r AML categorized as low, intermediate, or high risk. Outcomes in patients with KMT2A::MLLT3 were comparable to those in the non–KMT2A-r low-risk group, whereas outcomes in patients with non-MLLT3 KMT2A-r AML were comparable to those in the non–KMT2A-r intermediate-risk group. Both KMT2A-r subgroups had significantly more favorable outcomes than the non–KMT2A-r high-risk group. In the unadjusted analysis, FLAG-IDA induction was associated with superior EFS compared with DAE. However, this difference was no longer evident after propensity score matching, with comparable EFS, OS, and CIR between the two induction regimens. In the landmark analysis of children with non-MLLT3 KMT2A-r AML who achieved first complete remission (CR1), the 5-year EFS was 72.6% (95% CI, 57.2%–92.2%) among patients receiving chemotherapy consolidation and 68.1% (95% CI, 54.0%–85.9%) among those undergoing allogeneic hematopoietic stem cell transplantation in CR1 (CR1 allo-HSCT; P = .865). The corresponding 5-year OS estimates were 84.4% (95% CI, 71.5%–99.7%) and 69.3% (95% CI, 54.9%–87.5%), respectively (P = .307), while the 5-year CIR estimates were 23.5% (95% CI, 6.6%–40.4%) and 22.6% (95% CI, 8.3%–36.8%), respectively (Gray’s test, P = .898).
Conclusions
In this prospective non-randomized cohort, CR1 allo-HSCT was not associated with a detectable survival advantage over chemotherapy consolidation in children with non-MLLT3 KMT2A-r AML who achieved remission after FLAG-IDA. These data support prospective evaluation of response- and fusion partner–adapted post-remission strategies that may spare selected patients from transplantation without compromising outcomes.
DISCLOSURES
Yongzhi Zheng, PhD: No relevant conflicts of interest to declare.