Genetic characterization of <scp>AML</scp> defined by differentiation shows a high frequency of <i>DDX41</i> mutations

Acute myeloid leukaemia (AML) is a biologically heterogeneous disease of haematopoietic stem and progenitor cells characterized by clonal expansion of undifferentiated myeloid precursors and impaired haematopoiesis. Although AML is genetically diverse with a complex underlying genetic landscape, more than 95% of AML patients have driver mutations.1, 2 Accordingly, AML classification either based on the 5th edition of the World Health Organization Classification of Haematolymphoid Tumours (WHO-HAEM53) or International Consensus Classification (ICC4) guidelines as well as prognostication in AML is based mainly on genetics.5 Currently, the majority of AML cases (80%–90%) are classified based on defining genetic aberrations (DGAs) in both classification systems,6 which improve patient management by targeted treatment approaches and genetic risk classification. WHO-HAEM53 classifies cases lacking DGA phenotypically (criteria based on cytomorphology, cytochemistry staining and immunophenotyping) as AML with minimal differentiation, AML without maturation, AML with maturation, acute basophilic leukaemia, acute myelomonocytic leukaemia, acute monocytic leukaemia, acute erythroid leukaemia or acute megakaryoblastic leukaemia summarized as AML defined by differentiation (AML-diff). These cases pose challenges in patient management and conceptually if genetic aberrations are completely absent. Our study aimed to characterize AML-diff and AML cases lacking clonal markers genetically using whole genome and whole transcriptome sequencing (WGS/WTS). We analysed 2833 AML patients (median age: 71 years [18–96]; male: female = 1:0.88 Table S1) whose samples were sent to our laboratory between January 2019 and July 2024. All samples were analysed by morphology, cytogenetics and molecular genetics for 55 genes associated with myeloid malignancies (Methods S1). AML cases lacking clonal markers by full standard diagnostic work-up were additionally analysed by WGS/WTS (Methods S1) given available material. All patients were diagnosed according to WHO-HAEM5 (Table S1; Figure S1) and gave their written informed consent for genetic analyses and to the use of laboratory results and clinical data for research purposes according to the Declaration of Helsinki. The study was further approved by the laboratory's institutional review board. According to the WHO classification, DGA (excluding myelodysplasia-related AML [AML-MR]) was found in 39% (1106/2833), while AML-MR accounted for 51% (n = 1447) (Table S1; Figure S1). Of the latter, the majority (99%; n = 1438) displayed MR-defining genetics, while nine patients were classified as AML-MR solely due to a documented history of myelodysplastic neoplasm (MDS) or myelodysplastic/myeloproliferative neoplasm (MDS/MPN). Ten per cent (280/2833) of the entire cohort remained as AML-diff. We then genetically characterized AML cases without DGA or MR genetics, that is, AML-diff and AML-MR solely defined by clinical history (in total n = 289; from now on named AML-diff*). Of AML-diff*, 96% (277/289) showed genetic aberrations with 24% (70/289) having both molecular and cytogenetic abnormalities, 68% (197/289) mutations only and 3% (10/289) cytogenetic abnormalities only (Figure 1; Figure S2) in line with a previous study.7 The most frequently mutated genes in AML-diff* were DNMT3A (30%), FLT3 (25%) and DDX41 (24%) (Figure S3). Mutations in IDH1 and/or IDH2 were detected in 23%. Overall, in almost half of AML-diff* patients (140/289; 48%), at least one mutation in FLT3, IDH1 or IDH2 was found making them eligible for targeted treatment approaches (reviewed in Turkalj et al.8). In addition, 18% of AML-diff* showed mutations in RUNX1 (Figure 1), an inclusion criterion of the ICC4 for myelodysplasia-related AML but not of the WHO, recapitulating previous studies.7, 9 In 10 patients, 11 TP53 mutations were detected of which 5 had VAFs above 10% affecting 4 patients that would have been classified as AML with mutated TP53 according to ICC guidelines (Table S2). The most frequent cytogenetic abnormality in AML-diff* was trisomy 8 (10%; 35% of cases with aberrant karyotype) (Figure S4), another inclusion criterion of the ICC for myelodysplasia-related AML. In cases showing cytogenetic abnormalities only (without co-occurring mutations, n = 10), chromosome 8 abnormalities were detected in the majority of cases (7/10; Table S3). Overall, within AML-diff*, 77 patients (27%) would have been classified as AML with myelodysplasia-related gene mutation due to the presence of RUNX1 mutations or as AML with myelodysplasia-related cytogenetic abnormality as having +8 or del(20q) (Figure 1) according to ICC as shown previously.6, 7, 9 In AML-diff*, 62/289 (21%) showed mutations confined to a single gene and a normal karyotype (Figure 1, Figure S5). In these cases, DDX41 was most frequently affected (49%; 30/62). While putative germline mutations prevailed (26/30), four cases showed solely putative somatic DDX41 variants with VAFs below 35% (Figure 2A). Overall, DDX41 variants were detected in 6% of the entire AML cohort, which is in line with other studies.10, 11 Interestingly in our study, in all 289 AML-diff* cases, 24% (n = 68) were DDX41-mutated leading to an odds ratio of 7.4 compared to the remaining AML cases (101/2523 [4%], p < 0.001; Figure 2B; odds ratio of 9.6 after excluding cases with DGA based on ICC) and an unexpected high overlap to the DDX41-mutated cohort described by Maierhofer et al.11 In our study, DDX41 variants were detected in 6% of AML-MR (84 of 1424 evaluable for DDX41) and in 2% of AML-DGA (17 of 1099 evaluable for DDX41). In both, AML-diff* and the remaining AML cohort, the majority of cases had DDX41 variants of putative germline origin (84% and 95%, respectively), while only a small proportion of cases had solely putative somatic DDX41 variants (Figure 2B). The most frequent putative somatic variant in both AML sub-cohorts was p.Arg525His (63% and 48% of DDX41-positive cases), while the most frequent putative germline variant affected p.Met1 (25% and 17% of DDX41-positive cases; Tables S4/S5; Figures S6/S7) similar to previous studies.10, 11 In this line, Gurnari et al. showed that p.Arg525His as well as germline truncating hits had the largest prognostic impact, although current prognostic tools were not able to adequately assess survival in DDX41-mutant myeloid malignancies.12 Within AML-diff*, only 12/289 cases (4%; 0.4% of the entire AML cohort) lacked any genetic aberration by standard cytogenetic and molecular diagnostics (median bone marrow blast count: 28% [20–98]; Figure 1). Eleven cases were evaluable by WGS/WTS for extended genetic analysis (Table S6). In two cases, a UBTF-TD, a recurrent alteration in paediatric AML and rarely described in adult AML,13 was detected (with additional MYC mutations in one case). Genetic variants of unknown significance (VUS) in BCOR, SETBP1, KMT2D or ID3 were found in additional three cases. Structurally, WGS/WTS identified a t(7;14)(q36.3;q24.3) affecting PTPRN2 in one case and a microdeletion on chromosome (Chr) 4 affecting PTPN13 combined with a translocation between Chr 5 and Chr 21 resulting in HNRNPH1::ERG fusion in another case. In the literature, the last mentioned fusion has so far only been described in five patients.14 Lu et al. showed that HNRNPH1::ERG-positive AML cases were associated with high blast cell counts fitting with our case showing 58% bone marrow blasts. Only four cases (1% of AML-diff*, 0.1% of the entire cohort; median bone marrow blast count: 24% [20–31]) lacked any genetic aberration after WGS/WTS analysis for structural variations and for mutations in 122 genes associated with myeloid or lymphoid malignancies observing no technical concerns. In summary, our analysis of an unselected cohort of AML patients confirms that only a minority of cases (10%) are classified as AML-diff according to WHO-HAEM5. Even within this subcohort, nearly all patients showed genetic aberrations in extended genetic analysis including WGS/WTS. Clinically, the significant proportion (24%, odds ratio: 7.4) of candidates for germline testing for DDX41 was the most crucial finding in AML defined by differentiation. This has direct implications for patient care, including decisions regarding allogeneic haemopoietic stem cell transplantation, donor selection and genetic counselling. Our findings highlight the importance of including DDX41 in genetic AML screening and germline testing, especially in cases lacking AML defining genetic abnormalities. Gregor Hoermann and Christoph Kornauth designed the study and interpreted the data. Sandra Huber analysed the data. Christoph Kornauth and Sandra Huber wrote the manuscript. Isolde Summerer was responsible for cytogenetic analyses, Wolfgang Kern for immunophenotypic analyses, Manja Meggendorfer and Stephan Hutter for molecular and bioinformatic analyses and Torsten Haferlach and Christian Pohlkamp for cytomorphologic analyses. All authors read and contributed to the final version of the manuscript. The authors would like to thank all co-workers at the MLL Munich Leukemia Laboratory for their dedicated work. The authors would also like to thank all physicians for providing samples and caring for patients as well as collecting data. No financial assistance was received in support of the study. Torsten Haferlach and Wolfgang Kern declare part ownership of MLL Munich Leukemia Laboratory. Christoph Kornauth, Stephan Hutter, Sandra Huber, Isolde Summerer, Christian Pohlkamp, Manja Meggendorfer and Gregor Hoermann are employed by the MLL. The study adhered to the tenets of the Declaration of Helsinki and was approved by the laboratory's institutional review board and the Ethics Committee of the Bavarian Medical Association for the use of archived DNA/RNA samples and associated clinical information. The clinical data were retrieved and the samples were collected and analysed with the endorsement of the Ethics Committee of the Bavarian Medical Association. All patients gave their informed consent for genetic analyses and the use of laboratory results for research purposes. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Figure S1. Distribution of WHO entities within the entire AML cohort. diff, defined by differentiation; MR, myelodysplasia-related. Figure S2. Clonality detection in AML-diff*. CYTO, cytogenetic abnormality; MUT, mutation; NEG, negative/no clonality detected (neither MUT nor CYTO). Figure S3. Genes mutated in AML-diff* (n = 289). Figure S4. Cytogenetic abnormalities in AML-diff* (n = 289). Figure S5. Mutations detected in cases with normal karyotype and only a single gene mutated (n = 61). Figure S6. DDX41 putative somatic variants. Figure S7. DDX41 putative germline variants. Table S1. Cohort overview. Table S2. TP53-mutated cases in AML-diff*. Table S3. Cytogenetic abnormalities of AML-diff* cases without co-occurring mutations. Table S4. DDX41 variants in AML-diff* (n = 68). Table S5. DDX41 variants in the remaining AML cohort (n = 101). Table S6. Aberrations from extended genetic analysis. 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.