RUNX1 Mutations in Acute Myeloid Leukemia (AML): Correlation with Distinct Cytogenetic Subgroups and Clinical Outcome. Results of the AML Study Group (AMLSG)

Background: The runt-related transcription factor 1 (RUNX1) gene encodes a transcription factor that is required for hematopoietic stem cell emergence during development and that functions as a key regulator of hematopoiesis at several steps. Mutations in RUNX1 have been identified in sporadic myeloid leukemia through translocations [e.g., RUNX1-RUNX1T1 in t(8;21) or RUNX1-EVI1 in t(3;21)], point mutations or amplifications. In addition, germline mutations in RUNX1 result in familial platelet disorder with propensity for the development of myeloid leukemia. More recent data suggest that RUNX1 mutations are not strongly associated with MDS, secondary AML (s-AML) or therapy-related AML (t-AML) but seem to be more related to distinct cytogenetic subgroups such as trisomy 13, trisomy 21, loss of 7q, and trisomy 8.

Aims: To evaluate the incidence and clinical impact of RUNX1 mutations in a large cohort of younger (16 to 60 years of age) adult AML patients who were entered on AMLSG treatment protocol AML HD98A.

Methods: RUNX1 mutation screening was performed in 349 consecutive AML patients (de novo AML, n=282; s-AML, n=49; t-AML, n=18) using a DNA-based PCR assay for amplification of exons 1 to 8 followed by direct sequencing. The only criterion to include patients was the availability of a bone marrow or peripheral blood sample from diagnosis for gene mutation analysis.

Results: RUNX1 mutations were identified in 32 of 349 (9.2%) AML; mutations clustered in exon 3 (11/32) and exon 8 (11/32), but also occurred in other regions of the gene (10/32). With regard to cytogenetic subgroups, the incidence of RUNX1 mutations was 9.7% (20/206) in AML with normal karyotype, 8.3% (3/36) in core-binding factor leukemia, 14.2% (4/28) in various cytogenetic abnormalites including trisomy 8 in three cases, 7.6% (2/26) in AML with high-risk aberrations, whereas only 1 of 8 pts with t(11q23) and none of 29 cases with t(15;17) leukemia revealed a RUNX1 mutation; in 2 of 16 cases cytogenetics was not available. RUNX1 mutations were significantly associated with MLL-PTD (p=0.003), whereas concurrent NPM1 mutations were less frequent in the RUNX1 mutated group (p=0.0002); there was no association of RUNX1 mutations with FLT3-ITD/TKD, CEBPA, RAS and WT1 mutations. There were no differences in patients characteristics such as age, WBC counts, LDH levels, platelet counts, and distribution of de novo AML, s-AML, and t-AML between the RUNX1 mutated and RUNX1 wildtype group. Compared to RUNX1 wildtype AML, those with RUNX1 mutations had a significantly higher rate of resistant disease following induction therapy (38% and 20%, respectively; p=0.03) which translated into a significantly inferior event-free survival (p=0.004). There was no difference in relapse-free and overall survival between the two groups.

Conclusions: In younger adult patients with AML, RUNX1 mutations are found in approximately 10% of cases and are associated with cytogenetic subgroups. RUNX1 mutations appear to be associated with a higher rate of induction failure.