Background: Current risk stratification of AML is based on the presence or absence of specific cytogenetic abnormalities and somatic mutations. Although the European LeukemiaNet (ELN) risk stratification system incorporates the most common gene fusions with prognostic significance that are observed in AML (i.e., t(8;21), inv(16)/t(16;16), KMT2A and MECOM rearrangements, and several others), recent pediatric reports have suggested that RNA fusion sequencing panels detect other, less common but clinically relevant gene fusions in a high proportion of children with acute leukemia. However, there are minimal data on the utility of this technology in adult patients with AML. Here, we analyze results and clinical impact of a 557-gene RNA fusion panel in adults with newly diagnosed AML.

Methods: This study included adults age >= 18y with newly diagnosed AML who were treated at the University of Colorado. Patients with high grade myelodysplastic syndromes (MDS) (>=10% blasts) were also included. All patients had the comprehensive myeloid neoplasm panel performed, which includes analysis of DNA for 50 recurrently mutated genes in myeloid neoplasms and an RNA fusion panel that evaluates 557 genes for known and novel RNA fusions. Target enrichment is performed via a hybridization-capture-based method followed by sequencing on an Illumina NovaSeq6000 instrument, and a custom in-house bioinformatics pipeline is applied. Clinically significant fusions and high confidence novel fusions of unclear significance are reported.

Results: The RNA fusion panel identified gene fusions and/or a KMT2A partial tandem duplication (PTD) in 25/110 patients (22.7%). Of these patients, 24 had AML and 1 had MDS. The fusions identified included 7 KMT2A partial tandem duplications (PTDs), 4 KMT2A rearrangements, 3 core-binding factor (RUNX1::RUNX1T1) rearrangements, 2 PML/RARA rearrangements (in patients with known APL), and 15 other less common fusions. Since the KMT2A, core-binding factor, and PML/RARA rearrangements are readily detected by routine clinical FISH and molecular assays, we focused our analysis on the 14 other gene fusions that were identified in 13 patients (1 patient had 2 fusions identified). For these patients, the median age was 72.5y (range 30 to 88y).

Two NPM1 fusions (NPM1::ARPC1B and NPM1::MLF1), 2 RUNX1 fusions (RUNX1::SP3 and RUNX1::DIP2A), and 2 NUP98 fusions (NUP98::NSD31 and NUP98::KDM5A) were detected. Of note, the RUNX1::DIP2A and both NUP98 fusions were cryptic fusions that were not identified on chromosome analysis. The NPM1::ARPC18 fusion has not previously been reported, although other NPM1 fusions including NPM1::MLF1 have been reported in AML and have been associated with a poor prognosis.

Additional fusions with known or potential clinical significance that were each identified in a single patient included KMT2C::LHFPL3-AS2, NF1::SUZ12, PCM1::JAK2, SMC4::MECOM, EWSR1::FLI1, RFX2::MLLT1, TBL1XR1::TP63, and SSBP2::GALNTL6. Of this list, the only fusion that has been well-described in leukemias and other myeloid malignancies is the PCM1::JAK2 fusion. MECOM rearrangements are a well-known adverse risk finding in AML, but this is the first report of a MECOM fusion event with SMC4 as a partner. TBL1XR1::TP63 has not been reported in AML, however it is a recurrent event in diffuse large B cell lymphoma and other TP63 fusions have been reported in pts with MDS and AML. To our knowledge, this is also the first report of the EWSR1::FLI1 fusion in AML, which is a hallmark of Ewings sarcoma. In this case, the patient was 39 years old and his AML was refractory to induction with 7+3. The patient subsequently achieved a complete remission with venetoclax/azacitidine, suggesting sensitivity to venetoclax-based regimens with this particular fusion.

Conclusions: The RNA fusion panel detected known clinically significant fusions with prognostic and therapeutic importance in adults with AML. This included NPM1 and NUP98 fusions, which have been reported to have a poor prognosis but have been associated with HOXA gene overexpression and possible sensitivity to menin inhibition. The RNA fusion panel also detected multiple novel fusions that are potential leukemia drivers and have not previously been reported in AML, including an EWSR1::FLI1 fusion.

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