Genomic and epigenomic influences in AML

AML genomes tend to be less mutated compared to other cancers. Nevertheless, a number of mutations in AML cells involve genes which are implicated in epigenetic changes affecting gene activity and expression.¹ Broadly, epigenetic functions that become dysregulated include: DNA methylation, DNA hydroxymethylation, histone acetylation, histone lysine methylation, and other histone modifications.  Examples for epigenetic modifier gene mutations are DNMT3A, ETO, TET1/2, IDH1/2 and ASXL1. As epigenetic changes do not result in changes to the DNA sequence, they can be modified to some extent by targeted drug intervention.

Qingyu Xu, Department of Hematology and BMT Center, Chinese PLA General Hospital, Beijing, CN, and colleagues, aimed to look at treatment with a hypomethylating agent, decitabine, and whether epigenetic changes predict response. They studied patients with intermediate risk AML (IR-AML) as they had noted that patients frequently have epigenetic modifier gene mutations (EMMs) such as DNMT3A, TET1/2, IDH1/2 and ASXL1.² They aimed to assess how IR-AML patients (n=83) responded to the addition of hypomethylating agent (HMA) decitabine, to CAG (cytarabine, aclarubicin and granulocyte colony-stimulating factor; G-CSF) induction therapy (DCAG).  

Study design:

• DCAG regimen:
  • Days one-five: Intravenous (IV) drip decitabine 20 mg/m²; IV injection aclarubicin 20 mg; cytarabine in standard cases received IV drip 100 or 200 mg/m² every 12 hours (19 cases), and the elderly received a hypodermic injection 10 or 20 mg/m² every 12 hours
  • G-CSF 300 µg per day was given subcutaneously from Day zero until neutrophil recovery
• “7+3” Standard induction treatment:
  • Cytarabine for seven days and one of the anthracyclines (daunorubicin or idarubicin) or mitoxantrone for three days
• Forty-eight received conventional 7+3 regimens, 35 had the DCAG regimen (Table 1 details EMM (+) and EMM (-) in each treatment group)

Table 1. Gender and age of patients on each treatment regimen who were EMM (+) and EMM (-) All patients DCAG regimen “7+3” Standard induction regimens EMM (+) EMM (-) EMM (+) EMM (-) Gender (%)         Female 10 (58.8) 12 (66.7) 11 (55) 14 (50) Male 7 (41.2) 6 (33.3) 9 (45) 14 (50) Total 17 18 20 28 Median age, year (range) 59 (46–73) 56 (14–74) 45.5 (20–64) 35.5 (15–61)

• Pre-treatment bone marrow was used to obtain mononuclear cells for next-generation sequencing of a 250 kb region of the genome, covering 128 genes frequently found to be mutated in AML

Key findings:

• 271 mutations in 107 genes were found in 92.8% of participants
  • On average, there were 3.26 mutations per case
  • There were eight genes with more than 10% mutational frequency (CEBPA, DNMT3A, FLT3-ITD, NPM1, NRAS, ASXL1, IDH2, and WT1)
  • Mutated genes could be grouped into seven pathways: cell cycle regulation, cohesin, DNA repair, epigenetic modification, RNA splicing, signal transduction, and transcription
• There were some differences in patient profiles between treatment groups
  • Patients in the DCAG treatment group with EMM (+) were significantly older than the other groups, and also had significantly lower hemoglobin that the DCAG EMM (-) group
  • DCAG EMM (+) also had lower leukocyte counts, more M5 and less co-mutations than standard treatment EMM (+), and less biCEBPA mutations than DCAG EMM (-) and standard treatment EMM (-)
  • When compared with standard treatment EMM (-), DCAG EMM (+) had a less normal karyotype, and had less patients deemed to be ‘intermediate risk’, but had more at adverse risk and underwent less allogeneic hematopoietic stem cell transplantation (allo-HSCT)
  • Comparing the standard treatment groups, EMM (+) had less normal karyotypes and intermediate risk, but more co-mutations ≥3 and more NPM1 mutations
• Multivariate analysis (Table 2)

Table 2. Multivariate analysis

Abbreviations: 95% CI, 95% confidence interval; CR, complete remission; DCAG, decitabine with cytarabine, aclarubicin and granulocyte colony-stimulating factor; EFS, event-free survival; EMM (−), epigenetic modifier gene mutations-negative; EMM (+), epigenetic modifier gene mutations-positive; HR, hazard risk; ORR, overall response rate (complete remission rate + partial remission rate); OS, overall survival; RFS, relapse-free survival; RR, relative risk.
Outcomes	DCAG vs Standard induction		EMM (+) vs EMM (-)
	EMM (+) [RR/HR, 95% CI (p)]	EMM (−) [RR/HR, 95% CI (p)]	DCAG regimen [RR/HR, 95% CI (p)]	Standard induction regimens [RR/HR, 95% CI (p)]
OS	0.12, 0.02–1.71 (0.02)	2.56, 0.60–10.88 (0.20)	0.10, 0.01–0.97 (0.047)	6.93, 1.19–40.34 (0.03)
EFS	0.05, 0.01–0.34 (0.002)	0.62, 0.16–2.50 (0.51)	0.16, 0.03–0.90 (0.04)	7.01, 1.40–35.15 (0.02)
RFS	0.02, 0.002–0.41 (0.01)	0.40, 0.06–2.82 (0.36)	0.06, 0.003–1.12 (0.06)	12.39, 1.13–135.85 (0.04)

• 5-year survival analysis also showed that EMM (+) patients on DCAG had the best OS (83.9% ± 10.4%, p = 0.03), EFS (84.6% ± 10%, p = 0.05) and RFS (90% ± 9.5%, p = 0.14) compared to all other groups. 5-year OS was 28.6 % +4% for EMM (+) patients on standard treatment, 35.1 % + 14.5% for EMM (-) patients on DCAG, and 65.4 % + 10.2% for EMM (-) patients on standard treatment
• The DNMT3A mutation was associated with a better OS in the DCAG group
• The ASXL1 mutation in EMM (+) patients was associated with an adverse RFS, OS, and EFS when compared with EMM (-) patients
• In patients undergoing allo-HSCT, this was related to a better overall survival at 5 years regardless of induction regimens. However, in EMM (+) group, DCAG only treatment had a similar OS, EFS and RFS as the standard regimen + all-HSCT

Conclusions:

• OS, EFS and RFS DCAG were superior to the standard regimen for EMM (+) patients and it was equivalent to allo-SCT in this group
• Findings were not replicated by another study³, possibly due to using HMAs (such as decitabine) alone or in combination with different treatments
• Limitations of study include; small sample, retrospective analysis, DCAG and CAG more often used in China and Japan, and bias in each arm of cohort