Abstract
Background
Splenomegaly is an event occurring in a variable range between 10–40% of de novo acute myeloid leukemia (AML), recently linked to poorer prognosis. Studies in murine models have shown that loss of the additional sex combs-like 1 (ASXL1) gene function leads to a significantly enlarged spleen volume, due to an increased infiltration of myeloid cells into the spleen.
Methods
In 58 de novo AML patients presenting with splenomegaly at diagnosis, we evaluated the occurrence of ASXL1 somatic mutations, deepened the molecular profile and conducted high-throughput RNA sequencing, with the aim of unveiling possible peculiar aspects of this rare clinical scenario.
Results
ASXL1 mutations (ASXL1mut) were detected in 23/58 (40%) patients, being the most frequently mutated gene, followed by TET2 and NRAS. ASXL1mut cases were significantly older than ASXL1wt (71 vs 64 years old, p = 0.003), showed a significantly higher white blood cells count (31,970/uL vs 17,810/uL, p = 0.044) and a higher platelet count (177,700/uL vs 67,700/uL, p = 0.0006). In contrast, the median bone marrow blasts percentage was lower in the ASXL1mut subset compared to ASXL1wt (36.4% vs 72,1%, p = 0.002). Comparing the gene expression profile of the ASXL1mut and ASXL1wt groups, we found the upregulation of PCDHB2 and LURAP1L/LURAP1L-AS1 (all involved in mechanisms of cellular interaction and migration) genes in the former group, unveiling a role in the splenic infiltration of ASXL1mut leukemic cells.
Conclusions
Overall, our data paves the way for further studies of an AML subgroup with a distinctive phenotype, whose prompt identification could improve patient management and therapeutic decision making.
Supplementary Information
The online version contains supplementary material available at 10.1186/s40364-025-00833-8.
Keywords: Acute myeloid leukemia, Splenomegaly, ASXL1, PCDHB2, LURAP1L/LURAP1L-AS1
To the Editor
Splenomegaly is a pathological enlargement of the spleen, whose prevalence in the context of acute myeloid leukemia (AML) varies, with studies reporting its occurrence in 10–40% of cases [1, 2]. A thorough characterization of biological and clinical features of de novo AML patients presenting with an enlarged spleen volume at diagnosis is lacking. Studies in murine models have shown that loss of ASXL1 function leads to significant splenomegaly; histological analyses of spleens from ASXL1-mutant mice demonstrate a disorganized architecture and increased presence of myeloid cells, highlighting the ASXL1 role in maintaining normal spleen function and its contribution to pathological spleen enlargement in myeloid malignancies [3].
In 58 de novo AML patients with splenomegaly at diagnosis (Supplementary Table 1), we evaluated the occurrence of ASXL1 somatic mutations, with the aim of unveiling possible biological and phenotypical peculiar aspects of this uncommon clinical scenario (see methodological approaches in Supplementary File 1).
Splenomegaly was detected in 58/560 (10,4%) patients diagnosed with de novo AML at our institution from 2005 to 2022. A certain degree of both myeloid and erythroid maturation was observed alongside the presence of blasts (Fig. 1A-D). Although this finding may occasionally occur in AML in general, in our series it represents a recurrent cytomorphological signature.
Fig. 1.
Morphological assessment and mutational analysis. Bone marrow smears from an ASXL1mut patient showing the coexistence of both normal mature and pathologic (dysplastic, blast) cells. Red arrows indicate normal myeloid and lymphoid cells. Blue arrows indicate dysplastic cells. Black arrows indicate blasts. A from left to right, red arrows indicate a monocyte, a myelocyte, a neutrophilic granulocyte; blue arrow indicates a dysplastic myelocyte; black arrows indicate a monoblast and a myeloblast. B red arrows indicate a neutrophilic granulocyte, a late erythroblast, a monocyte; black arrows indicate three myeloblasts. C red arrows indicate a band cell, a promyelocyte, another band cell, a basophilic erythroblast, a lymphocyte and a late erythroblast; the blue arrow indicates a dysplastic megakaryocyte. D red arrows indicate two monocytes, a late erythroblast and a myelocyte; black arrows indicate two myeloblasts. E Map of ASXL1 variants identified on a linear schematization of the protein (lollipop plots) (hg19, RefSeq: NM_015338). F Oncoprinter visualization of all variants identified. For all cases (columns), main clinical parameters are reported. The percentage value reported for each gene indicates its variants occurrence in the cohort analyzed. G NRAS mutational profile. FAB: French-American-British classification, WBCs: white blood cells, PLTs: platelets
ASXL1 mutations were detected in 23/58 (40%) patients, with a variant allele frequency (VAF) ranging from 10 to 80%; all of them had a truncating effect: 20 (87%) were frameshift insertions/deletions and 3 (13%) nonsense variants affecting exons 12 and 13 (as widely described in AML) [4]. The most frequent variant was p. G646fs*12 (8/23, 35%) followed by p. E635fs*15 (4/23, 17%) (Fig. 1E). Most mutations (21 out of 23) produced truncated ASXL1 proteins that lack the C-terminal PHD domain but can exert a neomorphic activity by disrupting Polycomb repression and enhancing BAP1 deubiquitinase activity [4].
Upon molecular characterization of our AML cases with splenomegaly, a total of 146 variants were identified across 22 of the 26 genes commonly mutated in myeloid neoplasms [5, 6] (Fig. 1F). ASXL1 was the most frequently mutated gene (23/45, 51%), followed by TET2 (18/45, 40%) and NRAS (16/45, 36%). Interestingly, the occurrence of NPM1 and TP53 mutated cases (representing two distinct AML subclasses) is largely under-represented in our cohort (Fig. 1F).
Moreover, a statistically significant association was observed between ASXL1 and NRAS mutations. Indeed, 12/23 (52%) ASXL1mut cases had concurrent NRAS mutations, compared to 4/22 (18%) ASXL1wt cases (p = 0.028 – Fig. 1G). The cooperative leukemogenic interaction between ASXL1 mutations and activating NRAS mutations was reported in myeloid malignancies, and this co-occurrence was associated with poorer outcomes [7]. In mice, ASXL1 loss with oncogenic NRAS G12D mutation led to hyperactivation of RAS-MAPK signalling via FLT3 upregulation, induction of AP1 oncogenic programs, immune microenvironment suppression, increased spleen weights and promoted progression from chronic myelomonocytic leukemia to AML [8].
Our cohort was consequently divided into two subsets: those bearing ASXL1mut (23/58, 40%) and those ASXL1wt (35/58, 60%). ASXL1mut cases were significantly older than ASXL1wt (71 vs 64 years, p = 0.003) (Fig. 2A), had higher total white blood cells count (31,970/uL vs 17,810/uL, p = 0.044) (Fig. 2B) and platelet count (177,700/uL vs 67,700/uL, p = 0.0006) (Fig. 2C). The median bone marrow blasts percentage was lower in the ASXL1mut subset compared to the ASXL1wt (36.4% vs 72.1%, p = 0.002) (Fig. 2D). A trend towards a myelomonocytic and monocytic differentiation was observed in the ASXL1mut patients [M4-5 = 16/23 (70%) in the ASXL1mut subset vs 14/34 (41%) in the ASXL1wt subset, p = 0.057 – Fig. 2E]. Survival analysis showed no differences between the two groups (p = 0.11, Fig. 2F).
Fig. 2.
ASXL1mut vs ASXL1wt comparisons. Comparisons between ASXL1mut and ASXL1wt subgroups. Main clinical and biological parameters are reported: (A) Age, (B) WBCs, (C) PLTs, (D) BM blasts percentage, (E) FAB subtypes, (F) Overall survival. A-D Boxplots representing the distribution. The boxes extend from the 25th to 75th percentiles. The line in the box represents the median value. The whiskers range from the smallest value to the largest. G Principal Component Analysis (PCA) showing the separation between the two groups of samples (ASXL1wt and ASXL1mut) and highlighting the homogeneity of samples within each group. H Top ten differentially expressed protein-coding genes [upregulated (in red) and downregulated (in green)]. I Illustration of a possible mechanism connecting ASXL1 mutations and splenomegaly in AML, based on RNA seq results and recent literature data. In leukemic cells, the ASXL1 loss of function could induce the upregulation of PCDHB2 and of the couple LURAP1L/LURAP1L-AS1 (all involved in mechanisms of cellular interaction and migration). This could play a role in the splenic infiltration of ASXL1mut leukemic cells
Gene expression profiles from 13/23 ASXL1mut cases (with splenomegaly) were compared with 7 ASXL1wt patients (4 with and 3 without splenomegaly). According to the Principal Component Analysis (PCA) results, carried out on read counts relative to the expressed genes, a clear separation was observed between the two groups (Fig. 2G). A detailed description of RNAseq data analysis is reported in Supplementary File 1. PCDHB2 gene resulted the TOP UPregulated between differentially expressed genes (DEGs) (Log2FC = 8.211 and p-adjusted value 4.74e-07) and both the couple LURAP1L and LURAP1L-AS1 resulted upregulated (7.71 and 7.97 times) in ASXL1mut patients (Fig. 2H).
Interestingly, PCDHB2, LURAP1L, and LURAP1L-AS1 are all involved (via distinct mechanisms) in processes of adhesion/migration and cell-stroma/immunity interactions [9–11]. Their aberrant overexpression in ASXL1mut blasts might therefore contribute to a biological phenotype favoring their splenic infiltration (Supplementary Fig. 1). To date, a role in promoting extramedullary infiltration in AML has been demonstrated only for miR-29c&b2 [12] but the lack of data on microRNA expression in our samples does not allow us to verify its possible role in our cohort. Acknowledging the limitations of this study, we recognize that no functional analyses were performed to validate the mechanistic impact of ASXL1 mutations or the involvement of candidate effector genes. However, our primary objective was to identify and characterize a distinct AML subgroup with splenomegaly and a unique clinical-molecular profile. Future studies will be required to functionally dissect the pathogenic mechanisms suggested by our findings and to better define the biological and clinical relevance of NRAS mutations in the context of ASXL1-mutated AML patients with splenomegaly.
In conclusion, we describe a subset of AML patients characterized by splenomegaly, with reproducible clinical features and a genomic profile enriched for ASXL1 mutations. The possible pathogenic involvement of novel players paves the way for deeper investigation. Future studies will be needed to clarify or refine this small but interesting piece of the AML puzzle.
Supplementary Information
Acknowledgements
This work was supported by “Associazione Italiana contro le Leucemie (AIL)-BARI”.
Abbreviations
- AML
Acute myeloid leukemia
- VAF
Variant allele frequency
- PCA
Principal Component Analysis
- DEGs
Differentially expressed genes
Authors' contributions
Conception and design of the study: FT and FA. Acquisition of data and/or analysis and interpretation of data: FT, NC, CC, IR, LA, AZ, EP, MRC, GT, CFM, AM, AM, GS, PM and FA. Clinical data acquisition: FT, VPG, MD, PC and MGM. RNA-sequencing, bioinformatic analysis and data interpretation: MFC, FM, CT, BB and AT. Drafting of the manuscript: FA. All authors revised the manuscript for important intellectual content and approved the final version submitted for publication.
Funding
Not applicable.
Data availability
Data is provided within the manuscript or supplementary information files.
Declarations
Ethics approval and consent to participate
The local Ethics Committee of “Azienda Ospedaliero Universitaria Policlinico di Bari” approved the study. Informed consent was obtained from all patients before study inclusion, in accordance with the Declaration of Helsinki. Patients' records/information were anonymized and de-identified before analysis.
Consent for publication
Consent for publication was obtained from patients before their enrolment in the present study.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally to this work.
References
- 1.Byrd J, Mrózek K, Dodge R, Carroll A. are predictive of induction success, cumulative incidence of relapse, and overall survival in adult patients with de novo acute myeloid leukemia: results from Cancer …. Blood. 2002;100:4325–36. Available from: http://www.bloodjournal.org/content/100/13/4325.short. [DOI] [PubMed]
- 2.Asif N, Hassan K. Acute Myeloid Leukemia amongst Adults. J Islam Med Dent Coll. 2013;2:58–63. [Google Scholar]
- 3.Bera R, Chiu MC, Huang YJ, Lin TH, Kuo MC, Shih LY. RUNX1 mutations promote leukemogenesis of myeloid malignancies in ASXL1-mutated leukemia. J Hematol Oncol. 2019;12:1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gao X, You X, Droin N, Banaszak LG, Churpek J, Padron E, et al. Role of ASXL1 in hematopoiesis and myeloid diseases. Exp Hematol. 2022;115:14–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cumbo C, Tarantini F, Anelli L, Zagaria A, Redavid I, Minervini CF, et al. IRF4 expression is low in Philadelphia negative myeloproliferative neoplasms and is associated with a worse prognosis. Exp Hematol Oncol. BioMed Central; 2021;10:1–4. Available from: 10.1186/s40164-021-00253-y. [DOI] [PMC free article] [PubMed]
- 6.Cumbo C, Orsini P, Tarantini F, Anelli L, Zagaria A, Tragni V, et al. TNFRSF13B gene mutation in familial acute myeloid leukemia: a new piece in the complex scenario of hereditary predisposition? Hematol Oncol. 2023;41:942–6. [DOI] [PubMed] [Google Scholar]
- 7.Medina EA, Delma CR, Yang FC. ASXL1/2 mutations and myeloid malignancies. J Hematol Oncol. BioMed Central; 2022;15:1–18. Available from: 10.1186/s13045-022-01336-x. [DOI] [PMC free article] [PubMed]
- 8.You X, Liu F, Binder M, Vedder A, Lasho T, Wen Z, et al. Asxl1 loss cooperates with oncogenic Nras in mice to reprogram the immune microenvironment and drive leukemic transformation. Blood. 2022;139:1066–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Vishnubalaji R, Awata D, Alajez NM. LURAP1L-AS1 long noncoding RNA promotes breast cancer progression and associates with poor prognosis. Non-coding RNA Res. 2025;12:1–9. 10.1016/j.ncrna.2025.01.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lee ICJ, Leung T, Tan I. Adaptor protein LRAP25 mediates myotonic dystrophy kinase-related Cdc42-binding kinase (MRCK) regulation of LIMK1 protein in lamellipodial F-actin dynamics. J Biol Chem. 2014;289:26989–7003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chen WV, Maniatis T. Clustered protocadherins. Development. 2013;140:3297–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wei Y, Lu W, Yu Y, Zhai Y, Guo H, Yang S, et al. miR-29c&b2 encourage extramedullary infiltration resulting in the poor prognosis of acute myeloid leukemia. Oncogene. Springer US; 2021;40:3434–48. Available from: 10.1038/s41388-021-01775-9. [DOI] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data is provided within the manuscript or supplementary information files.


