Abstract 1873: SpliceCore® a novel ML tool for identifying disease-specific alternative splicing uncovered a promising therapeutic target for triple negative breast cancer
Miguel A. Manzanares,Alyssa Casill,Kendall Anderson,Priyanka Dhingra,Vanessa Frederick,Adam Geier,Martin Akerman,Gayatri Arun
DOI: https://doi.org/10.1158/1538-7445.am2022-1873
IF: 11.2
2022-06-16
Cancer Research
Abstract:Accumulating evidence highlights aberrant splicing as a disease-driving event in cancer, affecting tumor progression, metastasis, and therapy resistance. Multiple studies have described the tumor-promoting activity of specific aberrant cis-splicing variants and misregulation of trans-acting splicing factors in patients. Hence, the identification of aberrant splicing patterns has the potential to translate into actionable biomarkers and novel therapeutic targets, particularly for treatment insensitive cancers such as Triple Negative Breast Cancer (TNBC). Envisagenics' SpliceCore is an innovative cloud-based platform that integrates machine learning (ML) algorithms with high performance computing to analyze large RNA-seq datasets to predict biologically relevant, novel, and highly prevalent tumor-specific alternative splicing (AS) changes. Using SpliceCore, we have analyzed >2500 RNA-seq samples from different breast cancer subtypes as well as normal breast tissue and identified several AS derived therapeutic targets with the potential to translate into therapeutic candidates for TNBC. Here, we report a novel, alternatively spliced isoform that is present in 60.5% of TNBC patients and correlates with poor overall survival. This cytoplasmic protein was previously described as a regulator of the TGFβ pathway; therefore, we have validated the isoform switching action in a TGFβ dependent tumor progression model and demonstrated the activity of Splice Switching Oligos (SSO) to modulate its effect. We observed that this AS was correlated to a differential response on subcellular localization for phSmad2/3, Smad2/3 and TGFβRI. Pretreatment of the TNBC cells with SSO before TGFβ pathway activation modulates its proliferation response by affecting gene expression of p21, c-Myc and Smad7, leading to a decrease of cells in the G2 mitotic phase of the cell cycle and a loss of cell viability after the SSO treatment. More importantly, migratory response induced by TGFβ in TNBC cells was significantly inhibited by SSOs. Pre-treatment with SSO-0205 before TGFβ activation downregulated gene expression of Angiopoietin-like 4, Integrins α5 and β3, TGFβ associated migration markers, which was followed by a 55% decrease in cell migration. Our state-of-the-art ML technology, SpliceCore, has proven its ability to uncover novel disease-specific AS targets and to design splice correcting oligonucleotides for subsequent therapeutic development. Our data provides experimental proof of concept that has uncovered a novel therapeutic target for TNBC, whose aberrant splicing contributes to TNBC pathogenesis by misregulation of the TGFβ pathway. Moreover, SSOs predicted by SpliceCore were able to correct the aberrant splicing downstream effect, thereby uncovering a novel alternative splicing targeted approach for treating TNBC patients. Citation Format: Miguel A. Manzanares, Alyssa Casill, Kendall Anderson, Priyanka Dhingra, Vanessa Frederick, Adam Geier, Martin Akerman, Gayatri Arun. SpliceCore® a novel ML tool for identifying disease-specific alternative splicing uncovered a promising therapeutic target for triple negative breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl) nr 1873.
oncology