Abstract Alveolar rhabdomyosarcoma (aRMS), characterized by poor overall survival and limited advancements in therapy over the past four decades, poses a great challenge in childhood cancer treatment. More than 60% of high-risk aRMS is driven by the oncofusion protein PAX3-FOXO1. The direct targeting of transcription factor fusions, however, is quite difficult. Given that PAX3-FOXO1 is reported to drive disease by controlling transcriptional programs to maintain an undifferentiated myoblastic state, an alternative approach would be to target the transcriptional machinery recruited by the fusion. To identify potential novel targets in aRMS, we utilized the Broad Institute’s Cancer Dependency Map dataset, a collection of genome-wide CRISPR-Cas9 screens performed in over 1000 cancer cell line models. We focused on druggable dependencies that are selective to fusion-positive aRMS and intersected this data with PAX3-FOXO1-associated transcriptional complexes. Among these candidates, we identified CDK8 as a top selective dependency. Importantly, other components of the CDK8 kinase module in the mediator complex, such as CCNC, also scored as a strong selective dependency in fusion-positive aRMS. CDK8 expression is notably elevated in aRMS tumors compared to other pediatric tumors, and aRMS cell lines exhibit significantly higher CDK8 protein levels than non-transformed primary skeletal muscle cells. We next validated these in silico results through the application of shRNA knockdown, CRISPR knockout, and the use of CDK8 kinase inhibitory small molecules, including BI-1347, SEL-120-34A, and JH-XII-178, which demonstrated on-target activity as assessed by STAT1Ser727 phosphorylation. Both genetic loss of CDK8 function and kinase inhibition impaired aRMS cell line growth in vitro. In vivo studies demonstrated a significant inhibition of xenograft tumor growth with both doxycycline inducible CDK8-directed shRNAs and with pharmacologic inhibition of CDK8 with SEL120-34A treatment. To understand the mechanisms by which CDK8 inhibition impairs aRMS growth, we evaluated the transcriptional consequences of CDK8 inhibition by RNA-sequencing in aRMS, as well as the chromatin accessibility by ATAC-sequencing. We discovered that loss of CDK8 led to gain of chromatin accessibility and the upregulation of hallmark genes involved in myogenesis, such as MYH3, MYOG, MEF2C, and MYBPH, as well as alteration of PAX3-FOXO1 gene expression programs. We next validated this differentiation phenotype both in vitro through detection of increased myogenin staining by immunofluorescence and in vivo by visualization of myofibrils in the xenograft tumors and increased myoglobin quantified by qRT-PCR after CDK8 inhibition. In conclusion, this work identifies CDK8 as a novel and readily translatable therapeutic target in fusion-positive aRMS laying a robust foundation for further mechanistic and preclinical studies. Citation Format: Susu Zhang, Clare Malone, Brian Guedes, Kathleen Engel, Marissa Just, Kenneth Ross, Assil Fahs, Giulia Digiovanni, Diyana Granum, Rex Bentley, Leen Barbar, Christian Cerda-Smith, Elizabeth Mendes, Ozgun Erdogan, Alexander Kovach, Kristianne Oristian, Seth Zimmerman, Jack F. Shern, Kris Wood, Chris Counter, Corinne M. Linardic, Kimberly Stegmaier. Targeting CDK8: A translatable therapeutic approach for fusion-positive aRMS [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 5960.
Read full abstract