Abstract B13: Leveraging cloud-based computational resources for gene fusion discovery with potential clinical implications for pediatric solid tumor patients

Abstract Introduction: Gene fusions are important oncogenic drivers with significant clinical impact in some cancer types. This is particularly true in pediatric cancers that often have low mutational burden and lack other diagnostic markers and therapeutic targets. Many gene fusions are rare or private to the individual patient and can be difficult to detect with methods optimized for common fusions. Unbiased sequencing methods and expansive computational resources are needed for expanding our ability to characterize fusions. Building a comprehensive catalog of oncogenic gene fusions will improve our understanding of their diversity and fully harness their potential for clinical impact. Methods: Patients are eligible for the GAIN/iCat2 study if they have been diagnosed with high-risk or recurrent/refractory extracranial solid tumor at age 30 or less and have a sample available for sequencing. Enrolled patients with an unclear diagnosis after standard clinical testing are nominated for transcriptome sequencing by the study investigators. We developed a computational pipeline in Google Cloud for gene fusion discovery utilizing paired end Illumina RNA-Seq data, multiple fusion callers, and a custom algorithm for integrative data analysis. The multicaller fusion detection approach enables us to address the high false-positive rate typical for gene fusion calling in transcriptomic data while improving the sensitivity to detect the more challenging fusions. After filtering, the fusions are annotated using the databases of known fusions and cancer genes. The predicted fusion transcripts are inspected visually, and the fusions are selected based on relevance to diagnostic classification or therapy to be validated by an orthogonal method. Results: 41 tumor samples were sequenced and analyzed for gene fusions. A total of 203 candidate fusions were detected by two or more fusion callers. Based on functional annotations and potential impact on diagnosis or therapeutic approaches, 12 fusion transcripts of interest were identified, 10 of which were validated by either pre-enrollment testing or an orthogonal method. Of 16 mesenchymal cases, 6 validated fusions had diagnostic relevance and 3 validated fusions had therapeutic implications (ERC1-BRAF, RBPMS-NTRK2, and VCAN-IL23R). Two patients responded to matched targeted therapy. In one case, diagnostic classification was revised. Conclusions: Whole-transcriptome sequencing in this selected patient population identified some fusion transcripts with clinical relevance. Determining the biologic significance of previously unreported fusions will require orthogonal sequencing such as whole genome, functional studies, and analysis of larger patient populations. Improved accuracy and scalability of methods for large-scale gene fusion analysis in the growing public datasets are likely to expand the landscape of gene fusions in cancer. Citation Format: Alma Imamovic, Alanna J. Church, Laura B. Corson, Deirdre Reidy, Navin Pinto, Luke Maese, Theodore W. Laetsch, AeRang Kim, Susan I. Vear, Margaret E. Macy, Mark A. Applebaum, Rochelle Bagatell, Amit J. Sabnis, Daniel A. Weiser, Julia L. Glade-Bender, Gianna R. Strand, Lobin A. Lee, R. Seth Pinches, Catherine M. Clinton, Brian D. Crompton, Neal I. Lindeman, Steven G. DuBois, Katherine A. Janeway, Eliezer M. Van Allen. Leveraging cloud-based computational resources for gene fusion discovery with potential clinical implications for pediatric solid tumor patients [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr B13.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.