Mutational profile of ZBTB16-RARA-positive acute myeloid leukemia

Abstract

Background: The ZBTB16-RARA fusion gene, resulting from the reciprocal translocation between ZBTB16 on chromosome 11 and RARA genes on chromosome 17 [t(11;17)(q23;q21)], is rarely observed in acute myeloid leukemia (AML), and accounts for about 1% of retinoic acid receptor-α (RARA) rearrangements. AML with this rare translocation shows unusual bone marrow (BM) morphology, with intermediate aspects between acute promyelocytic leukemia (APL) and AML with maturation. Patients may have a high incidence of disseminated intravascular coagulation at diagnosis, are poorly responsive to all-trans retinoic acid (ATRA) and arsenic tryoxyde, and are reported to have an overall poor prognosis.

Aims: The mutational profile of ZBTB16-RARA rearranged AML has not been described so far.

Materials and methods: We performed targeted next-generation sequencing of 24 myeloid genes in BM diagnostic samples from seven ZBTB16-RARA+AML, 103 non-RARA rearranged AML, and 46 APL. The seven ZBTB16-RARA-positive patients were then screened for additional mutations using whole exome sequencing (n = 3) or an extended cancer panel including 409 genes (n = 4).

Results: ZBTB16-RARA+AML showed an intermediate number of mutations per patient and involvement of different genes, as compared to APL and other AMLs. In particular, we found a high incidence of ARID1A mutations in ZBTB16-RARA+AML (five of seven cases, 71%). Mutations in ARID2 and SMARCA4, other tumor suppressor genes also belonging to SWI/SNF chromatin remodeling complexes, were also identified in one case (14%).

Discussion and conclusion: Our data suggest the association of mutations of the ARID1A gene and of the other members of the SWI/SNF chromatin remodeling complexes with ZBTB16-RARA+AMLs, where they may support the peculiar disease phenotype.

Keywords: AML; ARID1A; NGS; ZBTB16-RARA.

FIGURE 1

Mutational profiles of ZBTB16‐RARA+, as compared to APL and to other AML subtypes. (A) Mutational profile of AML patients using targeted NGS including 24 myeloid‐specific genes. Data from seven ZBTB16‐RARA+is shown and compared to that of 46 APL and 103 non‐RARA rearranged AML, grouped according to their karyotype. Complex karyotype: 24 (three or more abnormalities); RA: 4 inv(16)(p13.1q22) or t(16;16)(p13.1;q22), 4 inv(3)(q21.3q26.2) or t(3;3)(q21.3;q26.2) or 3q26‐rearrangements, and 1 t(6;9)(p23;q34.1). Other karyotype: 22; Normal karyotype: 42; karyotype not available (NA): 6. (B) Comparison in the number of mutations (mean ± SD; p‐value) and the most frequently mutated genes in the three AML subtypes. AML, acute myeloid leukemia; APL, acute promyelocytic leukemia; NGS, next‐generation sequencing; RA, recurrent abnormalities (2); SD, standard deviation

FIGURE 2

Extended mutational landscape of ZBTB16‐RARA+AML. Studying 417 genes, we identified 37 mutations, with a mean of 5.29 (SD ±1.60) mutations per patient. The most frequently mutated genes were ARID1A (71%), TET2 (57%), RUNX1, and CSF3R (28% each)

FIGURE 3

Schematic representation of ARID1A protein domains. Domain information was derived from UniProt database (https://www.uniprot.org/). Location and type of somatic mutations are reported. ARID, AT‐rich interactive domain; LXXLL, leucine‐rich steroid receptor binding motif; NLS, nuclear localization domain