Identification of a novel MYO1F::MLLT10 fusion in adult acute monocytic leukemia

Acute myeloid leukemia (AML) is a heterogeneous malignancy characterized by clonal proliferation of myeloid precursors, often driven by recurrent cytogenetic and molecular abnormalities. Translocations involving the KMT2A (MLL) gene at 11q23.3 occur in AML, particularly in infant and pediatric cases, and are associated with monocytic differentiation and poor prognosis.^1-3 The MLLT10 gene at 10p12.31, a frequent KMT2A fusion partner, is implicated in AML and T-cell acute lymphoblastic leukemia (T-ALL) with monocytic features.^3,4 MYO1F at 19p13.2, encoding an unconventional myosin protein, is a rare KMT2A fusion partner reported in a few infant and pediatric AML cases (2 infants with KMT2A::MYO1F, 1 with a complex rearrangement and 1 pediatric case with a complex translocation involving chromosomes 7, 11, 19, and 22).^1,2,5,6 To our knowledge, MYO1F::MLLT10 fusion has not previously been described in AML. We report the first adult case of acute monocytic leukemia (World Health Organization/ International Consensus Classification 2022) with reciprocal in-frame MYO1F::MLLT10 and MLLT10::MYO1F fusions, identified by RNA sequencing and supported by interphase fluorescence in situ hybridization (FISH). A 58-year-old man with diabetes mellitus, chronic kidney disease, hypertension, and renal stones presented with right lower quadrant abdominal pain, nausea, vomiting, and fatigue. Imaging revealed left staghorn calculus, atrophic right kidney, splenic infarct, hepatosplenomegaly, and non-bulky multi-station lymphadenopathy. Laboratory findings revealed rapidly progressive leukocytosis (white blood cell count [WBC] 33.7x10⁹/L initially, rising to 104x10⁹/L pre-percutaneous nephrostomy [PCN]), anemia (hemoglobin [Hb] 8 g/dL), thrombocytopenia (platelets 128x10⁹/L), and acute kidney injury. Bone marrow aspiration and biopsy demonstrated a markedly hypercellular marrow (>95%) with >80% blasts/blast equivalents. Flow cytometry confirmed 73% blasts, positive for CD117, HLA-DR, CD11b, CD11c, CD13, CD15, CD33, CD38, partial CD14 and CD64, and negative for CD34, CD19, TdT, MPO. Abnormal monocytes (11%) expressed CD56 with decreased CD14. Peripheral smear revealed predominantly monoblasts and promonocytes with characteristic cytomorphology.

Bone marrow cells cultured for 24 hours revealed a karyotype of 45,X,-Y/46,XY (75% of cells with -Y), confirmed by chromosomal microarray analysis (-Y in ~70% of cells). Standard FISH was negative for common AML rearrangements (MECOM, DEK::NUP214, RUNX1T1::RUNX1, BCR/ABL1, KMT2A, PML/RARA, and CBFB) and TP53 deletion. Next-generation sequencing (NGS)-based targeted DNA sequencing (275-gene panel) revealed no pathogenic mutations, including FLT3-internal tandem duplication (ITD)/TKD or NPM1, with a low tumor mutation burden (2.4 mutations/Mb). RNA fusion panel (TruSight 507-gene panel) detected two novel in-frame chimeric fusion transcripts: MYO1F::MLLT10 and MLLT10::MYO1F (Figures 1 and 2). Metaphase spreads lacked a discernible t(10;19)(p12.31; p13.2). Therefore, an MLLT10/KMT2A dual fusion probe on destained G-banded metaphase spreads (Figure 3A, B), showed MLLT10 signals only on chromosome 10 short arms; no signals on either chromosome 19. However, 49% of interphase nuclei (98/200) displayed three MLLT10 signals - one large, two smaller - often close together (82/98 nuclei, 83.7%; Figure 3C) -indicating a split MLLT10 allele, potentially involving an insertion of MYO1F sequences from 19p13.2 into the MLLT10 gene region at 10p12.31, suggesting MLLT10 rearrangement. Initial management included leukapheresis (WBC reduced to 39.3x10⁹/L) and hydroxyurea (1 g twice a day for 4 days), followed by azacitidine/venetoclax (ven) induction (7+21 days, VIALE-A protocol). The patient had a partial response, but persistent disease (20% blasts) on bone marrow biopsy 2 weeks later. He later received FLAG-IDA-Ven salvage therapy and achieved short-lived complete remission. He relapsed (43% blasts, WBC 183.6x10⁹/L) 3 weeks after first consolidative high-dose cytarabine (HiDAC) (~4 months from the diagnosis) and was treated with leukapheresis (×2) and hydroxyurea. At the latest follow-up, 2 weeks post-relapse, he was on supportive care only, without AML-directed therapy, amid multiorgan failure (renal, ocular), awaiting hospice care.

The biospecimens and clinical data used in this study were collected previously as part of the patient’s routine diagnostic work-up and clinical care. No separate specimens were collected for the study. The institutional review board of Fox Chase Cancer Center provided ethical approval for this work.

This case represents the first report of MYO1F::MLLT10 and reciprocal MLLT10::MYO1F fusions in AML (Online Supplementary Table S1), identified by RNA sequencing, confirmed by multiple fusion-calling algorithms (using FusionCatcher, STAR-Fusion, and Arriba; Online Supplementary Table S2), and supported by orthogonal testing using FISH (Figure 3A-C). MLLT10 (also known as AF10), a known fusion partner in AML and T-ALL (e.g., KMT2A::MLLT10, PICALM::M-LLT10), drives leukemogenesis via its 3’ domains (plant homeodomain [PHD]-finger, leucine zipper), promoting monocytic differentiation through chromatin remodeling and transcriptional dysregulation (e.g., DOT1L recruitment, H3K79 methylation).^3,4 In MYO1F::MLLT10 fusion protein (443 amino acids [aa]), 5’ MYO1F exon 1 (start codon, promoter/enhancer elements) fuses to the C-terminal exons 15–23 (442 aa) of the 3’ MLLT10 partner, potentially up-regulating MLLT10’s oncogenic domains. In the reciprocal MLLT10::MYO1F fusion protein (839 aa), 5’ MLLT10 exons 1-18 (816 aa; N-terminal domains including PHD-finger, zf-HC5HC2H) fuses to 3’ MYO1F exons 25-28 (23 aa, partial myosin tail). The MYO1F::MLLT10 fusion transcript (breakpoints: 5′ MYO1F, chr19:8,642,191 [- strand]; 3′ MLLT10, chr10:22,015,173 [+ strand]) showed high expression levels and strong read support: STAR-Fusion (62 fragments: 39 junction, 23 spanning; FFPM=13.4276), FusionCatcher (61 fragments: 19 unique, 42 spanning), and Arriba (42 fragments: 6/27 split reads, 9 discordant mates). The fusion maintains the reading frame and encodes a chimeric protein (MANTLSGSS…, 443 aa; Online Supplementary Table S1). The reciprocal MLLT10::MYO1F fusion (breakpoints: 5′ MLLT10, chr10:22,022,016 [+ strand]; 3′ MYO1F, chr19:8,590,446 [strand]) was detected by STAR-Fusion (43 fragments: 40 junction reads, 3 spanning; FFPM=9.3126), FusionCatcher (54 fragments: 12 unique reads, 42 spanning pairs), and Arriba (24 fragments: 14/8 split reads, 2 discordant mates; coverage 309/85). This chimeric transcript was predicted to be in-frame according to STAR-Fusion and FusionCatcher, but out-of-frame according to Arriba due to selection of a shorter MYO1F isoform (507 bp in ENST00000596245.1 vs. 4,303 bp in ENST00000338257.8). It encodes a chimeric fusion protein (MVSSDR…, 839 aa; Online Supplementary Table S1). STAR-Fusion and FusionCatcher confirm canonical GT/AG splice sites for both fusions with Arriba’s CDS/splice site annotations reflecting minor junction mapping differences. High expression levels (FFPM 13.4276 for MYO1F::MLLT10 vs. 9.3126 for MLLT10::MYO1F) and Arriba’s coverage support MYO1F::M-LLT10 as the driver (Online Supplementary Table S2), with C-terminal domains of 3’ MLLT10 driving leukemogenesis, akin to KMT2A::MLLT10 fusions. In the reciprocal MLLT10::MYO1F fusion, the retained N-terminal PHD-finger and SH3 domains from 5′ MLLT10 suggest potential functionality with a secondary regulatory role, possibly contributing to stabilization of the rearrangement/translocation; however, it may be less oncogenic due to the lack of C-terminal oncogenic leucine zipper. Only 23 aa (part of the myosin tail) from 3’ exons 25-28 of MYO1F are included, indicating truncation of the fusion protein by a stop codon (EPTRKGMAKGKPRRSSQAPTRAA*) and translation of only part of MYO1F exon 25 sequence (chr19:8590446 to 8590363 [- strand], 84bp, 28 aa). This region lacks significant functional domains (e.g., the motor domain encoded by earlier exons), suggesting a primarily structural rather than functional role in the fusion protein. The asymmetric breakpoints in MLLT10 and MYO1F in both fusion products likely indicate an unbalanced reciprocal translocation. However, the presence of MLLT10 exons 15-18 in both derivative chromosomes raises the possibility of a duplicated segment and thus cannot entirely exclude a complex rearrangement from additional events. FISH with an MLLT10/KMT2A probe showed no detectable MLLT10 signal splitting in metaphase spreads (limited by suboptimal banding quality and insufficient metaphase spreads), but 49% of interphase nuclei displayed three MLLT10 signals (1 large, 2 smaller), indicating a split MLLT10 allele, likely due to a cryptic insertion of MYO1F sequences from 19p13.2 into the MLLT10 gene at 10p12.31. Interphase FISH confirmed the MLLT10 rearrangement, while the absence of visible chromosomal alterations suggests a submicroscopic cryptic insertion or underrepresentation of the malignant clone in metaphase spreads. The cryptic nature of the 10p;19p rearrangement, without other AML-defining cytogenetic changes, highlights the value of RNA fusion analysis for identifying rare gene fusions.

Figure 1.Schematic of MYO1F::MLLT10 fusion in acute monocytic leukemia. This diagram illustrates the MYO1F::MLLT10 fusion generated by a chromosomal rearrangement between 10p12.31 and 19p13.2. The breakpoint in the MYO1F occurs at chr19:8642191 (exon 1, start codon, negative strand) from chromosome 19, which fuses to the MLLT10 gene at chr10:22015173–22015284 (exon 15/16, positive strand). The resulting in-frame chimeric transcript, confirmed by RNA sequencing, encodes a protein that combines promoter/enhancer elements and the start codon of MYO1F with the transcriptional regulatory domains of MLLT10, driving acute monocytic leukemia. Exons are represented as boxes, introns as lines, and the fusion junction is indicated by a dashed line. Chromosomal orientations and breakpoints are annotated for clarity.

Figure 2.Schematic of MLLT10::MYO1F fusion in acute monocytic leukemia. This diagram shows MLLT10::MYO1F reciprocal fusion from the same chromosomal translocation. The breakpoint in the MLLT10 is at chr10:22022016 (exon 18/19, positive strand), retaining regulatory domains including the plant homeodomain (PHD)-finger, zinc finger (zf)-HC5HC2H, and Jnk-SapK_ap motifs, which fuses to the MYO1F at chr19:8590363-8590446 (exon 25, negative strand). RNA sequencing confirmed the fusion transcript, which combines the N-terminal region of MLLT10 with the C-terminal exons of MYO1F, encoding an in-frame chimeric protein according to STAR-Fusion and FusionCatcher and an out-of-frame protein according to Arriba. While MLLT10’s retained regulatory domains may contribute to fusion stability or minor functions, the fusion likely has reduced oncogenic potential. Exons are represented as boxes, introns as lines, and the fusion junction is indicated by a dashed line. Chromosomal orientations and breakpoints are annotated for clarity.

MYO1F, a rare KMT2A partner in infant AML, may contribute to cytoskeletal or signaling abnormalities, a novel role when fused with MLLT10.^5,6 The MYO1F::MLLT10 fusion in this case is likely driving the monocytic phenotype (>80% blasts, CD117⁺, CD34^-, CD56⁺, partial CD14⁺), consistent with MLLT10-rearranged AML.^3,4 Although the absence of FLT3/ NPM1 mutations was favorable, the relapse after first consolidative HiDAC indicates poor prognosis.⁷ The patient’s complex clinical course, including retinal hemorrhages and renal failure, underscores the systemic impact of AML. This case expands the spectrum of MLLT10 rearrangements in AML and contributes to the evolving understanding of rare AML-associated fusions. Its identification by targeted RNA sequencing - with robust read support, high expression, and in-frame predictions across multiple fusion calling algorithms - underscores the importance of integrated molecular and cytogenetic profiling in AML. Because MY-O1F::MLLT10 was identified only through RNA sequencing, similar cryptic events may currently be under-recognized. Broader screening of AML cohorts using targeted RNA sequencing or polymerase chain reaction analysis of cDNA from diagnostic bone marrow aspirates will be essential to ascertain its prevalence. The MYO1F::MLLT10 fusion is the probable leukemogenic driver, analogous to KMT2A::MLLT10 or cytoskeletal perturbations from MYO1F’s functional domains. In line with the approaches for novel fusion genes such as NUP98-NSD1, the transforming activity of MYO1F::M-LLT10 can be assessed using in vitro assays - including foci formation or anchorage-independent growth by transfection of a plasmid expressing MYO1F::MLLT10 into NIH-3T3 fibroblasts. Additional insights could be obtained by evaluating proliferation rates and/or altered differentiation in normal or immortalized myeloid cell lines.⁸ Furthermore, assessing therapeutic vulnerabilities - such as to DOT1L inhibitors that target chromatin-remodeling pathways implicated in MLLT10-driven malignancies - could reveal actionable targets for this rare AML subtype.

Figure 3.Hybridization of MLLT10 probe in patient bone marrow cells. (A) Schematic of orange-labeled MLLT10 probe hybridizing to the entire MLLT10 locus and flanking sequences at sub-band 10p12.3 (adapted from the MetaSystems XL t(10;11) MLLT10/KMT2A Dual Fusion Probe [MetaSystems, Medford, MA]), which comprises an orange-labeled probe targeting the MLLT10 gene region at 10p12.31 and a green-labeled probe targeting the KMT2A gene region at 11q23.3; the latter probe is not shown). (B) G-banding (left) and hybridization of MLLT10/KMT2A probe (right) to a representative 4′,6-diamidino-2-phenylindole (DAPI)-stained metaphase spread from the patient’s bone marrow. The MLLT10 probe hybridizes to the short arm of chromosome 10 only (red arrows), with no hybridization to chromosome arm 19p (100x objective lens; total magnification of 160x). (C) Representative fluorescence in situ hybridization (FISH) images showing nuclei with: (top panel) normal two MLLT10 signal pattern; (bottom panel) abnormal pattern, exhibiting 1 larger MLLT10 signal (thick vertical arrow on far right) and 2 smaller MLLT10 signals (thin red arrows in same nucleus) (20x objective lens; total magnification of 32x). Images of interphase nuclei were captured using a MetaSystems Metafer microscope workstation, and the raw images were extracted and processed to depict MLLT10 signals in magenta and DAPI-stained nuclei in blue, with blue outlines marking nuclear boundaries.

Footnotes

• Received May 22, 2025
• Accepted November 7, 2025

Correspondence

R. NEJATI - Reza.Nejati@fccc.edu

Disclosures

No conflicts of interest to disclose.

Contributions