A novel UBASH3B::PVT1 fusion in B-cell acute lymphoblastic leukemia with PAX5-PTD: a case report and literature review

INTRODUCTION B-cell acute lymphoblastic leukemia (B-ALL) is a genetically heterogeneous malignancy characterized by distinct molecular subtypes resulting from recurrent gene rearrangements, chromosomal abnormalities, or specific mutations. Prognosis depends on factors including age, initial white blood cell (WBC) count, genetic alterations, and treatment response. 1, 2 The Paired Box 5 (PAX5) gene, located on chromosome 9p13, belongs to the PAX family and is essential for B-cell development. PAX5 alterations (PAX5alt) are frequently observed in B-ALL, 3, 4 and include rearrangements, sequence mutations, and focal amplifications (PAX5amp), 5 all of which contribute to disease pathogenesis. The 2025 National Comprehensive Cancer Network guidelines classify PAX5alt as a poor-risk subtype. 6 Recent studies have further identified PAX5 partial tandem duplications (PAX5-PTD) and PAX5 internal tandem duplications as rare subtypes within the PAX5alt category. The Pvt1 oncogene (PVT1) encodes a long noncoding RNA implicated in oncogenesis. Its amplification and overexpression correlate with multiple malignancies, including breast cancer and acute myeloid leukemia (AML). 7, 8 Elevated PVT1 expression may also promote the proliferation of acute promyelocytic leukemia cell. 9 The Ubiquitin-Associated and SH3 Domain Containing B (UBASH3B) protein contains an N-terminal ubiquitin-associated domain, a central SH3 domain, and a C-terminal region homologous to phosphoglycerate mutase. 10 This protein negatively regulates epidermal growth factor receptor endocytosis. 11 The UBASH3B: : PVT1 fusion has not been previously reported. Our study identified UBASH3B: : PVT1 and PAX5-PTD through high-resolution RNA sequencing (RNA-seq), and next-generation sequencing (NGS) revealed mutations in ETV6, EZH2, and PTPN11. These approaches facilitate precise molecular subtyping of B-ALL, thereby improving prognostic accuracy and enabling personalized treatment strategies. However, RNA-seq remains limited by its reliance on targeted RNA detection, which hinders the identification of structural variations such as 3q26 rearrangements that lack fusion protein products and reduces sensitivity for detecting large deletions or inversions. Optical genome mapping (OGM), a genome-wide technique, effectively detects structural variants (SVs) and copy number variants (CNVs), enabling comprehensive characterization of cytogenetic abnormalities. 12 In this study, OGM confirmed both UBASH3B: : PVT1 and PAX5-PTD. Continued advancements in genomic and transcriptomic technologies are expected to further refine B-ALL classification, ultimately supporting more targeted therapies and improved clinical outcomes. 2. CASE REPORT A 22-year-old male was admitted to a local hospital with fever (39. 1°C), epistaxis, and cough. The fever resolved after treatment with clindamycin. Hematologic tests showed a WBC count of 8. 63 × 109/L, hemoglobin level of 74 g/L, and platelet count of 39 × 109/L, indicative of anemia and thrombocytopenia. Bone marrow aspiration morphology revealed 92% blasts. Flow cytometry analysis demonstrated positivity for cCD79a, CD10, CD19, HLA-DR, CD22, CD9, cMPO, CD71, and CD11c; partial positivity for CD34, CD33, and CD15; and negativity for CD117, CD13, CD64, CD14, CD11b, CD5, CD7, CD2, CD56, CD20, and CD36. Therefore, the diagnosis was B-lymphoid leukemia with myeloid lineage expression. Karyotype analysis demonstrated the following abnormalities: 56, XY, +X, ins (1;? ) (q21;? ), +4, +8, +9, +10, +17 × 2, +18, +21, +224/46, XY6 (Fig. 1A). Multiplex polymerase chain reaction (PCR) analysis confirmed the absence of all 43 common fusion genes. Fluorescence in situ hybridization detected no BCR: : ABL1 fusion. NGS identified mutations in ETV6 p. Tyr634Cys (variant allele frequency VAF: 63. 35%), EZH2 p. Lys314Asn (VAF: 7. 23%), and PTPN11 p. Ala461Gly (VAF: 83. 33%). Based on these findings, the patient was diagnosed with hyperdiploid B-ALL. After 1 week of debulking therapy with vincristine and prednisone, he was referred to our hospital, where RNA-seq and OGM were performed. RNA-seq revealed a UBASH3B: : PVT1 fusion (split reads = 11; discordant mates = 1; junction sequence: TCCTCTCCAT GGGGTTCCCC AGAGCCCGCG CGAAACAGGT TGAGACATCA CACAATAAAT C) and PAX5-PTD (split reads = 601; discordant mates = 300; junction sequence: CCAGGCCCGC AGTCCTACCC CATTGTGACA GGACATGGAG GAGTGAATCA GCTTGGGGGG G) (Fig. 1B–C). OGM confirmed UBASH3B: : PVT1 (confidence = 1; VAF = 0. 27) (Fig. 1D) and PAX5-PTD (confidence = −1; VAF = 0. 42) (Fig. 1E) and also detected chromosomal gains, including 1q+, +4, 8q+, +9, +10, +14, +17, +18, +21, and +X (Fig. 1F). Because no previous reports have described UBASH3B: : PVT1, reverse transcription polymerase chain reaction (RT-PCR) was performed using the patient’s available complementary DNA (cDNA). The primers were F (5′–3′): GCTGCGAGAGAGGAGCTGTA and R (5′–3′): TGATGTTTAGAACCCAGGCC, with ABL1 serving as the internal reference. The PCR product was 206 base pairs in length. The results, presented in Figure 1G, confirmed the presence of UBASH3B: : PVT1. The PCR products were subsequently sequenced using the Sanger method with the same primers. As shown in Figure 1H, the Sanger sequencing results were consistent with those obtained from RNA‑seq and OGM. Because PVT1 is overexpressed in breast cancer and AML, we measured its expression in this patient’s cDNA during this period, using 2 ALL and 2 AML patients as controls. The quantitative PCR primers were: F (5′–3′): TTGGCACATACAGCCATCAT and R (5′–3′): GCAGTAAAAGGGGAACACCA. ABL1 served as the internal reference gene. The results, presented in Figure 1I, demonstrated that PVT1 expression in this patient was significantly higher than that in the 4 controls, indicating that the UBASH3B: : PVT1 fusion was associated with increased PVT1 transcription. Based on these findings, the patient was ultimately diagnosed with B-ALL harboring PAX5alt. The patient achieved complete hematologic remission after one cycle of IVP (idarubicin, vincristine, prednisone) combined with blinatumomab (a bispecific T-cell engager BiTE). Following one cycle of Hyper-CVAD chemotherapy, he attained complete molecular remission (Fig. 1J). Treatment continued with alternating Hyper-CVAD chemotherapy and immunotherapy (BiTE and chimeric antigen receptor T-cell CAR-T therapy). To date, the patient has maintained event-free survival for more than 1 year. Figure 1.: Clinical features of a patient with B-cell acute lymphoblastic leukemia with UBASH3B: : PVT1 fusion gene and PAX5-PTD. (A) Karyotype analysis of bone marrow revealed a karyotype of 56, XY, +X, ins (1;? ) (q21;? ), +4, +8, +9, +10, +17 × 2, +18, +21, +224/46, XY6. (B) RNA-seq revealed a rare UBASH3B: : PVT1 fusion. The patient’s fusion transcript resulted from a translocation between exon 1 of UBASH3B (located on chromosome 11) and exon 1-3 of PVT1 (chromosome 8). (C) RNA-seq revealed PAX5-PTD. The patient’s PAX5-PTD resulted from a duplication between exons 2 and 7 of PAX5 (located on chromosome 9). (D) OGM revealed a UBASH3B: : PVT1 fusion. (E) OGM revealed PAX5-PTD. (F) OGM revealed a karyotype of 54, XY, +X, 1q+, +4, 8q+, +9, +10, +14, +17, +18, +21. (G) Results of UBASH3B: : PVT1 detected by RT-PCR. The primers were F (5’–3’): GCTGCGAGAGAGGAGCTGTA and R (5’–3’): TGATGTTTAGAACCCAGGCC, with ABL1 serving as the internal reference. This PCR product is 206 base pairs in length. RT-PCR sequencing results show the fusion between UBASH3B exon 1 and PVT1 exon 1–3. (H) Confirmation of the breakpoint of the UBASH3B: : PVT1 fusion via Sanger sequencing. The UBASH3B transcript is ENST00000284273. 5, and the PVT1 transcript is ENST00000521951. 1. (I) The expression level of PVT1 in this patient was confirmed to be higher than that in 2 ALL patients and 2 AML patients by quantitative PCR. The primers were F (5’–3’): TTGGCACATACAGCCATCAT and R (5’–3’): GCAGTAAAAGGGGAACACCA, with ABL1 serving as the internal reference. The results were statistically significant. **p < 0. 010, ***p <. 001, ****p <. 001. (J) Previous chemotherapy regimens and treatment response. (K) OGM detected the UBASH3B: : PVT1 fusion at the genomic DNA level. The schematic illustrates this fusion event. According to National Center for Biotechnology Information (NCBI) reference sequences NM₀01363365. 2 and NC₀00008. 11, exon 1 of UBASH3B is joined to exons 1–5 of PVT1. The University of California Santa Cruz (UCSC) Genome Browser reference sequence ENST00000667305. 2, however, indicates a junction between exon 1 of UBASH3B and PVT1 exons 1–6. NCBI = National Center for Biotechnology Information, OGM = Optical genome mapping, RNA-seq = RNA sequencing, RT-PCR = reverse transcription polymerase chain reaction, UCSC = University of California Santa Cruz. 3. DISCUSSION Familiades et al13 first reported partial or complete amplification of PAX5 in 2009, noting that partial amplifications predominantly involved exons 2 to 5 or exon 5 alone. Öfverholm et al14 later corroborated these findings and introduced the term “intragenic amplifications of PAX5. ” Subsequent studies have shown that such intragenic amplifications frequently span exons 2 to 5 or 2 to 7, although other exons may also be affected. 4, 15, 16 These regions encode the paired box DNA-binding and octapeptide domains. In 2019, Gu et al17 described internal tandem duplications in PAX5, and Tsai et al18 first reported partial tandem duplications of the gene in 2023. Here, RNA sequencing identified PAX5 breakpoints at chr9: 36923355 and chr9: 37020801, generating a PAX5: : PAX5 fusion transcript joining exons 1 to 7 to exons 2 to 9. This configuration corresponds to a tandem repeat of PAX5 exons 2 to 7 (Fig. 1C). We carefully reanalyzed the PAX5-PTD results obtained by OGM, which revealed 2 possible breakpoint combinations: chr9: 36969906 and chr9: 37031741 or chr9: 36920129 and chr9: 37031741, producing a tandem repeat of exons 2 to 5 (confidence = −1, VAF = 0. 42) or exons 2 to 7 (confidence = −1, VAF = 0. 42) (Fig. 1E). Based on the operational principle of OGM, only a range interval can be determined for tandem repeats; therefore, this variation manifests as 2 distinct possibilities. The larger interval represents the final result, which was subsequently confirmed by RNA-seq analysis. Following prior nomenclature, we designate tandem repeats of PAX5 exons 2 to 5 or 2 to 7 as PAX5-PTD, with details summarized in Table 1. Table 1 - Several studies of acute lymphoblastic leukemia that exhibit focal amplifications of PAX5. Author Publication time Type Technology Target Exons Domain Reference Chang et al 2024 Focal internal tandem PAX5 amplifications WGSPCRRNA-SeqSanger sequencing DNA/RNA 52–51–7 The DNA-binding paired domain 4 Jean et al 2022 PAX5 intragenic tandem multiplication CMAOGM DNA 2–51–5 The paired box/DNA-binding domain 16 Tsai et al 2023 Partial tandem duplications of PAX5 RNA-SeqMLPA DNA/RNA 2–52–7 18 Schwab et al 2017 Intragenic amplification of PAX5 MLPAFISH DNA 1256810 The DNA-binding and octapeptide domains 15 Gu et al 2019 In-frame internal tandem duplication of PAX5 WGSRT-PCRSanger sequencingFISH DNA 2–5 The DNA-binding domain 17 Öfverholm et al 2013 Intragenic amplifications of PAX5 MLPA DNA 522–5 14 Familiades et al 2009 Partial amplification of PAX5Complete amplification of PAX5 qPCRFISHRT-PCR DNA 2–551–10 13 Present case Partial tandem duplications of PAX5 RNA-SeqOGM RNADNA 2–72–5 Our case CMA = chromosomal micro array, FISH = Fluorescent in situ hybridization, MLPA = multiplex ligation-dependent probe amplification, OGM = optical genome mapping, PCR = polymerase chain reaction, qPCR = quantitative PCR, RNA-Seq = RNA sequencing, RT-PCR = PCR with reverse transcription, WGS = whole genome sequencing. The PVT1 gene resides on chromosome 8q24. 21 and is known to form fusion genes with partners such as MYC and CCDC26. 19 Here, we identified a novel fusion between UBASH3B, located at 11q24. 1, and PVT1 (Fig. 1B and D). Because RNA-seq and OGM interrogate RNA and DNA, respectively, and reference different transcripts, the exons involved differ between assays. RNA-seq mapped the PVT1 breakpoint to chromosome 8 position 129032402 and the UBASH3B breakpoint to chromosome 11 position 122526918, producing a UBASH3B: : PVT1 fusion transcript joining exon 1 of UBASH3B to exons 1 to 3 of PVT1 (Fig. 1B). Because PVT1 is a long noncoding RNA, the resulting fusion retains only the UBA/TS-N protein domain of UBASH3B (Fig. 1B). In contrast, OGM detected the PVT1 breakpoint at chromosome 8 position 128016168 and the UBASH3B breakpoint at chromosome 11 position 122715291, generating a UBASH3B: : PVT1 transcript that covers exon 1 of UBASH3B fused to exons 1 to 5 or 1 to 6 of PVT1 (Fig. 1K). Despite differences in the exons involved, the UBASH3B: : PVT1 fusion was unequivocally present (Fig. 1G–H). Furthermore, although PVT1 is a highly unstable long noncoding RNA, its transcription rate substantially outpaces its degradation rate, maintaining a high steady-state expression level (Fig. 1I). The UBASH3B: : PVT1 fusion has not been previously reported, and its prognostic significance in B-ALL remains unknown. PVT1 is established as a key driver of leukemia cell proliferation, 9 and UBASH3B represents a potential therapeutic target in this context. 11 Consequently, the UBASH3B: : PVT1 fusion may promote leukemic proliferation; however, this hypothesis requires further experimental validation. NGS identified mutations in ETV6, EZH2, and PTPN11. ETV6 mutations result in dysfunctional ETV6 protein, which promotes inflammation and impairs platelet production, thereby contributing to leukemogenesis. 20 Mutations in EZH2 drive malignant transformation by mediating epigenetic reprogramming in B cells, thereby further promoting leukemia development. 21 In patients with AML without NPM1 mutations, PTPN11 mutations are correlated with poor prognosis, although they do not influence outcomes in NPM1-mutated cases. 22 Collectively, these genetic alterations are commonly associated with leukemia initiation and adverse prognosis. The patient was initially diagnosed with B-ALL exhibiting hyperdiploidy at a local hospital and was therefore considered to have a favorable prognosis. Although hyperdiploidy in pediatric B-ALL generally correlates with favorable outcomes, 23 it is associated with poor prognosis in adult cases. 24 Based on established age criteria, this 22-year-old individual was classified within the adult group. 24 Following referral to our center, comprehensive molecular profiling revised the diagnosis to B-ALL with PAX5alt, which is currently recognized as a poor-risk subtype. Several treatment options warrant further investigation, including intensive chemotherapy and allogeneic hematopoietic stem cell transplantation. The incidence of severe chemotherapy-related complications and transplant-related mortality remains considerable. 25 Over the past decade, immune-targeted and cellular therapies—including BiTEs (eg, blinatumomab), CAR-T therapy, and tyrosine kinase inhibitors—have progressively reshaped the treatment landscape for poor-risk B-ALL. In this study, the patient achieved sustained molecular remission for more than one year following combined treatment with chemotherapy, BiTE, and CAR-T therapy, without experiencing significant chemotherapy-related adverse effects. Accurate diagnosis and prognostic stratification are essential in the management of leukemia. Conventional chromosome karyotype analysis has limited resolution, detecting only chromosomal abnormalities larger than 5 to 10 Mb and frequently missing smaller SVs, including deletions, duplications, insertions, or translocations. Although RNA-seq reliably identifies SVs that produce fusion proteins, it often fails to detect focal CNVs and SVs located in noncoding regions. OGM can detect SVs smaller than 1 kb and reveal previously unidentified pathogenic variants, enabling more precise molecular diagnosis. Nevertheless, OGM cannot detect chromosomal abnormalities in telomeric and centromeric regions, such as Robertsonian translocations, and must therefore be supplemented with karyotype analysis. 12 Furthermore, OGM cannot identify single-nucleotide variants (SNVs) and must be used in conjunction with NGS. 26 While NGS provides high resolution for detecting SNVs, it has limited ability to detect SVs and CNVs. Owing to the inherent limitations of each individual technology—karyotyping, RNA-seq, OGM, and NGS—our results indicate that combining OGM with NGS provides comprehensive and accurate genome-wide detection of SVs, CNVs, and SNVs. In summary, RNA-seq and OGM identified both UBASH3B: : PVT1 and PAX5-PTD in a patient with B-ALL. These alterations are likely to contribute to B-ALL pathogenesis and disease progression. Further studies involving larger cohorts of patients harboring UBASH3B: : PVT1 and PAX5-PTD are required to validate these findings. ACKNOWLEDGMENTS This study was supported by a grant from the NHC Key Laboratory of Thrombosis and Hemostasis, The First Affiliated Hospital of Soochow University (KJS2420), and the Suzhou Gusu Medical Talent (GSWS2022019). All the samples were from Hematologic Biobank, National Clinical Research Center for Hematologic Diseases, Jiangsu Provincial Science and Technology Resources (Clinical Resources). Coordination Service Platform, The First Affiliated Hospital of Soochow University. ETHICS APPROVAL This study was reviewed and approved by the Ethics Committee of the First Affiliated Hospital of Soochow University. This study was conducted with the patients’ consent. AUTHOR CONTRIBUTIONS All authors contributed to the work and approved the manuscript for publication. X. Z. and Y. W. analyzed data and wrote the manuscript. X. Z. , Q. Y. , H. S. , and M. W. performed most of the experiments. M. G. , B. X. , T. L. , J. Z. , T. Y. , Q. W. , Y. K. , and X. X. helped edit the paper. M. C. and Y. Z. analyzed data. H. H. , S. C. , and Q. W. analyzed data and helped edit the paper.