Mutations in DNMT3A, U2AF1, and EZH2 may predict poor outcomes in intermediate-risk AML patients in CR1

The optimal post-remission therapy for patients with intermediate-risk (IR) cytogenetics is still controversial. Furthermore, the benefit of screening genetic information to stratify IR AML patients who have achieved first complete remission (CR1) has not been determined yet. Caner Saygin and colleagues from the Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH, retrospectively analyzed the clinical relevance of recurrent driver mutations in IR AML patients who received treatment at their center and achieved CR1 between 2002–2016. The data was published in a recent edition of Blood Cancer Journal.

This study aimed to investigate whether driver mutations have a role in outcome prediction and an effect on clinical decision making with a distinctive focus on the performance of hematopoietic stem cell transplant (HCT).

In total, 100 IR-AML patients (median age = 58.5 years, range 24–75) who had mutational information available at diagnosis and achieved complete remission (CR) after ≥ 1 round of intensive chemotherapy at Cleveland Clinic (CC) were included in this analysis.  In order to validate the findings of this study, the authors identified 48 adult de novo ELN IR AML patients using data from The Cancer Genome Atlas (TCGA).

Key findings:

• Patients harboring mutations in DNMT3A, U2AF1, and EZH2 had worse overall and relapse-free survival (OS and RFS) compared to patients without these mutations
  • Median OS and RFS in patients with or without DNMT3A mutation respectively
    • OS: 12 vs 35 months, P = 0.002
    • RFS: 9 vs 16 months, P = 0.045
  • Median OS and RFS in patients with or without U2AF1 mutation respectively
    • OS: 9 vs 36 months, P = 0.0003
    • RFS: 7 vs 16 months, P = 0.013
  • Median OS and RFS in patients with or without EZH2 mutation
    • OS: 13 vs 26 months, P = 0.002
    • RFS: 5 vs 15 months, P = 0.038

The authors then compared the outcomes of IR AML patients in CR1 with or without mutations in DNMT3A, U2AF1, and EZH2 in both the CC and TCGA cohorts

• Survival of patients with or without mutations in DNMT3A, U2AF1 and EZH2 in the CC group respectively
  • Median OS: 13 vs 41 months, P < 0.0001
  • Median RFS: 9 vs 16 months, P = 0.003
• Survival of patients with or without mutations in DNMT3A, U2AF1 and EZH2 in the TCGA group respectively
  • Median OS: 9.8 vs 46.8 months, P = 0.0002
  • Median RFS: 9.3 vs 17 months, P = 0.02
• In both the CC and TCGA cohort, patients who did not have mutations in DNMT3A, U2AF1, and EZH2, and received HCT had superior outcomes
• In both CC and TCGA cohorts, HCT improved outcomes for patients who had mutations in DNMT3A, U2AF1, and EZH2
  • In the CC cohort, median OS and RFS for patients who underwent HCT were 14 and  11  months respectively compared to 7.5 and 5 months in patients who did not undergo HCT, P = 0.04 for OS and P = 0.002 for RFS
  • In the TCGA cohort, median OS and RFS for patients who underwent HCT were 24.6 and 12.5 months respectively compared to 6.3 and 7.8 months in patients who did not undergo HCT, P = 0.001 for OS and P = 0.003 for RFS

In summary, these findings suggest that DNMT3A, U2AF1, and EZH2 mutations have efficient predictive value to stratify patients with IR-AML after achieving CR1. Moreover, the presence of these mutations had an independent effect on survival in both CC and TCGA groups. The authors stated that HCT improved outcomes in patients who had DNMT3A, U2AF1, and EZH2 mutations.

Key limitations of the study were its retrospective nature, the lack of sequencing data, and the different types of transplantation and conditioning regimens. The study group further concluded that “a prospective validation of our findings is needed and we believe that our results may contribute to improving prognostication of patients with AML and the design of clinical trials.”