Introduction: De novo BCR::ABL1+ acute myeloid leukemia (AML) is a distinct and rare entity (0.1%-3% of AML), classified as adverse-risk in the ELN 2022 classification (Döhner H et al., Blood 2022). However, we recently published the favorable outcome of 18 de novo BCR::ABL1+ AML from the French DATAML registry treated with imatinib + intensive chemotherapy (IC) (Gondran C et al., Blood Cancer Journal 2024): CR/CRi, 94.4%; 3-year OS, 77%; that compared favorably to intermediate- and adverse-risk AML. To confirm these results, we performed a multicentric European retrospective data collection of de novo BCR::ABL1+ AML characteristics, treatment data and outcome.

Methods: Inclusion criteria for this retrospective study were: adult AML with ≥ 20% bone marrow blasts, BCR::ABL1+or t(9;22)(q34.1;q11.2), with no history of previous chronic myeloid leukemia (even if < 6 months) and no prior exposition to BCR::ABL1 tyrosine kinase inhibitor (TKI), who were treated with IC +/- TKI. Acute leukemia of ambiguous lineage were excluded.

Results: We collected the data of 250 patients with de novo BCR::ABL1+ AML diagnosed from March 1999 to December 2024. 212 patients were treated with IC: 89 with IC alone and 123 with IC + TKI. Median age was 54 years and 67 patients (32%) were older than 60. Forty-three patients (28.5%) presented with splenomegaly. The median WBC was 45.109/L (29 in the IC arm vs. 53 in the IC + TKI arm). One hundred and fifteen patients (60%) had additional chromosomal abnormalities, 28 (15%) had a monosomal karyotype and 62 (32%) had complex abnormalities. BCR::ABL1 isotype was p210 in 80% and p190 in 20%. Molecular data were available for the most recent patients: the co-occurring mutations present in more than 10% of cases were: RUNX1 (N=32/100: 32%), BCOR (N=11/82: 13%), ASXL1 (N=10/79: 13%), WT1 (N=9/72: 13%), DNMT3A (N=8/73: 11%), and NPM1 (N=15/147: 10%); a TP53 mutation was found in 7% (N=5/77).

IC was a 7+3 including daunorubicin+cytarabine in 107 patients (59%), idarubicin+cytarabine in 59 patients (36%) or another scheme in 14 (8%). In the IC + TKI arm, imatinib was added in 66 patients (54%; 400-800 mg/day), dasatinib in 40 patients (33%; 100-140 mg/day), ponatinib in 11 patients (9%; 15-45 mg/day), and nilotinib in 6 patients (5%; 600-800 mg/day).

Response assessment was available in 78 patients in IC arm and 122 patients in IC + TKI arm. Composite complete remission (CR/CRi) was achieved in 52 patients (66.7%) in the IC arm, and 110 patients (90.2%) in the IC + TKI arm (P<0.0001). 28.2% and 9.0% of patients had primary refractory disease in the IC and IC + TKI arm, respectively (p<0.001). 97 patients (59.9%) were transplanted in CR1 (26 patients [50%] in IC arm and 71 patients [64.5%] in IC + TKI arm, P=0.077). The relapse rate was 28% in the IC arm and 18% in the IC + TKI arm (P=0.159). With a median follow-up of 66.7 months, the 3-year and 5-year OS were 42.1% and 38.5% in the IC arm and 70.9% and 62.9% in the IC + TKI arm (P<0.0001). The median OS was 20.5 months in the IC arm and not reached in the IC + TKI arm. The 3-year and 5-year RFS were 52.1% and 49.3% in the IC arm and 73.0% and 67.1% in the IC + TKI arm (P=0.0095). The 3-year and 5-year EFS were 32.1% and 30.4% in the IC arm and 66.7% and 61.4% in the IC + TKI arm (P<0.0001). Of note, in the IC + TKI arm, 26 out of 39 patients (67%) who were not transplanted did not relapse and had a median OS of 65.8 months.

In multivariate analyses, the addition of TKI to IC was significantly and independently associated with an improvement in CR/CRi rate (OR: 4.74 [2.17-10.35]; P<0.001), in OS (HR: 0.40 [0.27-0.62]; P<0.001), in EFS (HR 0.37 [0.25-0.55]; P<0.001) and RFS (HR: 0.42 [0.23-0.75]; P=0.004) adjusted for confounding factors including allo-HSCT as time-dependent variable. The other independent poor predictive factors were: age older than 60 years for OS and EFS; and WBC over 50.109/L, IC regimen different than 7+3 and absence of alloSCT for RFS.

Conclusions: The addition of a TKI significantly improves the outcome of de novo BCR::ABL1+ AML treated with intensive chemotherapy, and should be a standard of care for this entity. This study should lead to the revision of the current ELN AML genetic risk classification, as this entity no longer deserves to be classified as adverse.

This content is only available as a PDF.
2025
Sign in via your Institution