Scholarlink

Background <p>Gliomas comprise a heterogeneous group of Central Nervous System (CNS) tumors in which gene fusions (GFs) are important oncogenic drivers with emerging diagnostic and therapeutic relevance. However, a portion of glioma samples remain negative on targeted clinical GF panels, leaving underlying molecular drivers unresolved.</p> Methods <p>To investigate the fusion landscape of these diagnostically challenging samples, we analyzed 49 high- and low-grade gliomas previously classified as fusion-negative by the Children’s Hospital of Philadelphia Fusion Panel using whole-transcriptome long-read sequencing with Oxford Nanopore Technologies (ONT).</p> Results <p>We identified numerous candidate oncogenic GFs beyond panel constraints, including fusions involving COSMIC Cancer Gene Census genes and recurrent CNS fusion partners not captured by the clinical panel. Long-read sequencing further enabled direct resolution of full-length fusion transcripts and associated isoform structures. Integrating GF detection with isoform-level transcript analysis identified fusion-associated isoforms with alternative splicing patterns near reported GF breakpoints, including <i>ZNF254</i>::<i>GNAS</i> and <i>PTPRK</i>::<i>NOX3</i>, which have not been reported in literature or existing fusion databases. To assess functional relevance, 15 candidate GFs were evaluated using the <i>Drosophila melanogaster</i> model, with ventral nerve cord (VNC) morphology serving as a quantitative in vivo readout of fusion-induced disruption of glial regulation. Eight candidate GFs induced significant VNC abnormalities relative to wild-type controls, including glial overgrowth and tumor-like structural disruption <i>in viv</i>o. Notably, <i>CLDND1</i>::<i>WRN</i> and <i>DUSP22</i>::<i>APOE</i> produced the most pronounced VNC phenotypes.</p> Conclusions <p>Together, these findings demonstrate that transcriptome-wide long-read sequencing can uncover previously undetected candidate GFs with potential functional and clinical relevance in panel-negative gliomas.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Long-read transcriptome sequencing and Drosophila-based functional validation reveal novel gene fusions in fusion panel-negative gliomas

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月31日

查看全文
  • Karleena Rybacki,
  • Emily Na Young Cha,
  • Hannah M. Deutsch,
  • Eloise Gaudet,
  • Mian Umair Ahsan,
  • Feng Xu,
  • Joe Chan,
  • Marilyn Li,
  • Yuanquan Song,
  • Kai Wang

摘要

Background

Gliomas comprise a heterogeneous group of Central Nervous System (CNS) tumors in which gene fusions (GFs) are important oncogenic drivers with emerging diagnostic and therapeutic relevance. However, a portion of glioma samples remain negative on targeted clinical GF panels, leaving underlying molecular drivers unresolved.

Methods

To investigate the fusion landscape of these diagnostically challenging samples, we analyzed 49 high- and low-grade gliomas previously classified as fusion-negative by the Children’s Hospital of Philadelphia Fusion Panel using whole-transcriptome long-read sequencing with Oxford Nanopore Technologies (ONT).

Results

We identified numerous candidate oncogenic GFs beyond panel constraints, including fusions involving COSMIC Cancer Gene Census genes and recurrent CNS fusion partners not captured by the clinical panel. Long-read sequencing further enabled direct resolution of full-length fusion transcripts and associated isoform structures. Integrating GF detection with isoform-level transcript analysis identified fusion-associated isoforms with alternative splicing patterns near reported GF breakpoints, including ZNF254::GNAS and PTPRK::NOX3, which have not been reported in literature or existing fusion databases. To assess functional relevance, 15 candidate GFs were evaluated using the Drosophila melanogaster model, with ventral nerve cord (VNC) morphology serving as a quantitative in vivo readout of fusion-induced disruption of glial regulation. Eight candidate GFs induced significant VNC abnormalities relative to wild-type controls, including glial overgrowth and tumor-like structural disruption in vivo. Notably, CLDND1::WRN and DUSP22::APOE produced the most pronounced VNC phenotypes.

Conclusions

Together, these findings demonstrate that transcriptome-wide long-read sequencing can uncover previously undetected candidate GFs with potential functional and clinical relevance in panel-negative gliomas.