Impact of gene mutations on the efficacy of CPX-351 versus 7+3 in older patients with newly diagnosed, high-risk/secondary acute myeloid leukemia

In newly diagnosed acute myeloid leukemia (AML), risk stratification relies mostly on cytogenetics, genetics, age, fitness, and other causal factors. Based on its clinical evolution, AML can be classified into three categories: secondary AML (s-AML), arising after an antecedent myelodysplastic syndrome (MDS); therapy-related AML (t-AML) developing after chemotherapy; or de novo AML arising without an identifiable cause. s-AML and t-AML are associated with inferior prognosis compared with de novo AML, and are clinically and genetically heterogeneous, with gene mutations in s-AML closely associated with MDS.¹ 

CPX-351, a novel dual-drug liposomal encapsulation of cytarabine and daunorubicin, is approved by the U.S. Food & Drug Administration and EMA for the treatment of adults with newly diagnosed t-AML, and AML with myelodysplasia-related changes, based on the results of a randomized phase III trial (NCT01696084). Association of cytogenetic factors with outcome after CPX-351 use has so far not been studied.

At the 61^st American Society of Hematology  Annual Meeting & Exposition, Coleman Lindsley, Dana-Farber/Harvard Cancer Center, Boston, US, presented the results of a study evaluating the genetic characteristics of patients enrolled in the phase III trial and the impact of gene mutations on outcomes.²

Study design and patients characteristics²

• In total, 309 adults were enrolled and randomized 1:1 to receive CPX-351 or 7+3 (standard therapy) induction and consolidation
• Targeted sequencing of 112 leukemia-related genes was performed from pre-treatment blood or bone marrow samples in 184 patients, 93 in the CPX-351 group and 91 in the 7+3 group
• Patients and disease characteristics were similar in each treatment arm
• In the CPX-351 group vs the 7+3 group, the frequency of AML subtypes was:
  • s-AML 59% vs 52%
  • de novo AML 23% vs 29%
  • t-AML 18% vs 29%

Results²

• In the entire cohort, the overall frequency of gene mutations was:
  • TP53 mutations: 33%
  • MDS-associated secondary-type mutations: 57%
  • NPM1 mutations (more common in de novo AML): 6%
• The range and frequency of gene mutations were similar between treatment arms, with the most frequently mutated genes being TP53, ASXL1, TET2, RUNX1, and DNMT3A
• The distribution of gene mutations in s-AML, de novo AML, and t-AML was:
  • TP53 mutations: 18%, 51%, and 54% (p< 0.0001)
  • Mutations in genes encoding for RNA splicing factors and chromatin modifiers: 69%, 45%, and 37% (p= 0.0008)
  • RUNX1 mutations: 35%, 13%, and 20% (p= 009)
• The association between individual gene mutations and clinical outcomes was evaluated, focusing on gene mutations present in ≥20% of patients. Gene mutations that were evaluated include; TP53, ASXL1, DNMT3A, RUNX1, and TET2, as well as mutations which activated signaling (FLT3, NRAS, KRAS, PTPN11, NF1, CBL, RIT1), and MDS associated secondary-type mutations (SRSF2, U2AF1, SF3B1, ZRSR2, ASXL1, BCOR, EZH2, STAG2). Among patients with these mutations, the rates of complete remission (CR) or CR with incomplete neutrophil or platelet recovery), and the rates of transplantation were similar between treatment arms
• Median overall survival (OS) based on treatment arm in patients with or without common gene mutations is shown in Table 1. Median OS was longer in the CPX-351 arm versus 7+3 arm among patients with DNMT3A and TET2 mutations
• In multivariate analysis, clinical variables such as age, ECOG performance status, prior hypomethylating therapy, cytogenetic risk, presence of TP53 or RUNX1 or activated signaling mutations, were independently associated with shorter OS. In contrast, CPX-351 treatment was independently associated with prolonged OS, compared with 7+3

Table 1. Median overall survival (months)

CI, confidence interval; NR, not reached

 

Gene

CPX-351

7+3

Mutated

Unmutated

Mutated

Unmutated

TP53

(95% CI)

5.7

(3─7.6)

10.9

(7─26.3)

5.1

(2.6─7.4)

9.4

(4.6─12.2)

SRSF2, U2AF1, SF3B1, ZRSR2, ASXL1, BCOR, EZH2, STAG2 (secondary-type mutations)

(95% CI)

10.1

(6.4─18.5)

6

(4.3─12)

6.6

(3.6─9.5)

7.3

(4.3─9.8)

FLT3, NRAS, KRAS, PTPN11, NF1, CBL, RIT1 (activated signaling mutations)

(95% CI)

7

(3.8─9.3)

11.3

(5.6─NR)

5.7

(3.2─8.6)

7.6

(4.3─10.9)

ASXL1

(95% CI)

8.9

(5.4─18.7)

7.6

(5.6─12)

6.6

(3.3─11.7)

7.3

(4─9.2)

TET2

(95% CI)

9.3

(5.3─18.7)

7.5

(5.5─12)

4.1

(1.9─5.9)

7.8

(4.9─10.9)

RUNX1

(95% CI)

8.9

(3.5─11.9)

7.7

(5.7─13)

4.7

(1.5─6.6)

8.2

(4.6─10.9)

DNMT3A

(95% CI)

26.3

(7─NR)

7.4

(5.3─10.1)

6.4

(2.3─10.9)

7.4

(4─9.6)

Conclusion²

• Patients enrolled in the phase III study had high-risk genetic characteristics, including 33% with TP53 mutations and 57% with secondary-type gene mutations
• TP53 mutations were associated with a poor prognosis, irrespective of treatment arm
• Median OS was significantly longer for CPX-351 vs 7+3 among patients with two of the most common mutations: DNMT3A and TET2
• In a multivariate analysis, incorporating clinical and genetic characteristics, CPX-351 was associated with prolonged OS compared with 7+3
• Conclusions were limited by post-hoc study design and a small sample size