Impact of secondary‐type mutations in acute myeloid leukemia with CEBPA mutation

To the Editor, Secondary-type mutations (STM: ASXL1, BCOR, EZH2, SF3B1, SRSF2, STAG2, U2AF1, and ZRSR2) in acute myeloid leukemia (AML) were reported to be highly specific for secondary disease and associated with inferior event-free survival (EFS) [1, 2]. In light of these findings, the newly-revised 5th edition of the WHO Classification of Haematolymphoid Tumours (WHO-HAEM5) includes STM in the criteria defining AML, myelodysplasia-related (AML-MR) [3]. In the newly-published International Consensus of Classification of myeloid neoplasms and acute leukemia (ICC), RUNX1 mutation together with STM as described by WHO-HAEM5 define the entity of AML with myelodysplasia-related gene mutations [4]. To date, several studies have examined the impact of STM in other molecularly defined entities of AML such as NPM1+ AML [5-9]. However, the impact of STM in AML with CEBPA mutation has not been studied extensively and remains unclear. Thus, we sought to investigate the clinical impact of STM in AML with CEBPA mutation. We conducted a retrospective analysis with a single center cohort of 38 cases of AML with CEBPA mutation diagnosed at our institution from 2015 to 2024. Patients were included if they met WHO-HAEM5 criteria of blasts ≥ 20% and either biallelic CEBPA mutation or single CEBPA mutation in the basic leucine zipper (bZIP) domain. Cytogenetic testing, molecular genetic testing, and variant calling in next-generation sequencing were performed according to previously described procedures [10]. Gene panel for targeted sequencing and exon coverage for hotspot genes are listed in Tables S1 and S2. The baseline clinicopathological characteristics and co-mutation landscape of the study cohort are summarized in Table 1 and Figure 1A. Of the 38 patients with AML with CEBPA mutation, 11 (29%) had STM. SRSF2 and STAG2 were the most common STMs (both 8/38, 21%), followed by ASXL1 (6/38, 16%), BCOR (2/38, 5%), and U2AF1 (1/38, 3%). RUNX1 mutations were found in two (5%) patients, both of whom had concurrent WHO-HAEM5-defined STMs. EZH2, ZRSR2, and SF3B1 mutations were not detected. The most common recurrent (>10%) co-mutated genes in the cohort were TET2 (10/38; 26%), GATA2 (9/38, 24%), WT1 (8/38, 21%), NRAS (6/38, 16%) and FLT3-ITD (4/38, 11%). Compared to patients without STM, patients with STM were older, had less proliferative disease, had lower hemoglobin levels, and had significantly lower variant allele frequency of mutant CEBPA and infrequent in-frame bZIP CEBPA mutation. Consistent with the older age of the STM+ group, only five (46%) patients received intensive chemotherapy compared to all 27 (100%) patients in the STM-ve group. Of note, complete remission rates after intensive induction therapy in patients with or without STM did not significantly differ. The transplant rate was higher in patients without STM but did not reach statistical significance. In the overall cohort, patients with STM had inferior 2-year overall survival (OS) (18% vs. 96%; p < 0.001) and EFS (18% vs. 88%; p < 0.001) (Figure 1B,C). In the intensively treated subgroup, STM remained a predictor for inferior 2-year OS (40% vs. 96%, p < 0.001) and EFS (40% vs. 88%; p = 0.013) (data not shown). We previously reported that STM had limited prognostic value in patients with NPM1+ AML [6]. Here, we show using a single-center cohort that AML with CEBPA mutations with concurrent STM is associated with distinct clinicopathological features and inferior outcome. To the best of our knowledge, this is the first study to evaluate the implications of concurrent STM and CEBPA mutations in disease classification. In 2022, the new WHO-HAEM5 classification revised the definition of AML with CEBPA mutation to include biallelic as well as single mutations located in the bZIP domain [3]. In contrast, the ICC only recognizes a subset of those mutations (i.e. in-frame bZIP mutation) to be disease defining [4]. In our cohort, 12 (32%) patients did not meet diagnostic criteria to be included in the ICC group of AML with in-frame bZIP CEBPA mutations. Importantly, STM was enriched in patients who were excluded from the ICC group (10/12, 83% vs. 1/26, 4%; p < 0.001). This suggests that the ICC is superior to WHO-HAEM5 at excluding CEBPA-mut AML patients who have distinct clinicopathological features and who may be more appropriately classified otherwise as AML-MR. Our data indicates an adverse prognostic impact of STM in CEBPA-mut AML patients. However, this finding may be confounded by the negative association between STM and in-frame bZIP CEBPA mutations which have been shown to confer favorable outcomes and which have been included in ELN classification of AML as a favorable genetic factor [11-13]. The current study is a single-center study with a limited sample size, and it was not powered to perform multivariable analysis to identify whether STM remained an independent predictor of inferior outcomes. Future prospective studies with larger cohorts are needed to confirm whether STM or in-frame bZIP mutations confer prognostic significance in AML patients with concurrent STM and CEBPA mutations. In conclusion, our data indicates that patients in the WHO-HAEM5 defined group of AML with CEBPA mutation with concurrent STM may be more appropriately classified as AML-MR. Classification according to ICC, which only selects for in-frame bZIP CEBPA mutations, identifies a homogenous cohort without STM and is effective at excluding patients who may be otherwise better classified as AML-MR. Davidson Zhao collected and analyzed the data and wrote the manuscript. Musani Rumina collected data and wrote the manuscript. Mojgan Zarif and Cuihong Wei collected data. Hong Chang designed the study and analyzed the data. All authors read, critically reviewed, and approved the manuscript. The authors thank the clinicians, nurses, and allied health professionals for their dedication to patients with leukemia at the University Health Network/Princess Margaret Cancer Centre. The authors declare no conflict of interest. The authors have confirmed ethical approval statement is not needed for this submission. The authors have confirmed patient consent statement is not needed for this submission. The authors have confirmed clinical trial registration is not needed for this submission. Data is available upon reasonable request. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.