Lineage-specific transcriptional networks drive cellular differentiation and development. Disruption of these specific cell programs can result in cancer and create a subset of tumors that are “transcriptionally addicted.” Sarcomas, for example, are characterized by an oncogenic fusion protein consisting of a FET family RNA-binding protein fused to a transcription factor (TF). The oncogenic fusions result in a restructuring of the transcriptome promoting cancer. In chordoma, TBXT (brachyury)—normally involved in notochord differentiation—is aberrantly expressed, resulting in promotion of cancer. As these oncogenic TFs are difficult to target directly, we and others have proposed targeting associated transcriptional regulators to inhibit their activity. Cyclin-dependent kinase 9 (CDK9) interacts with TFs to promote activation of target genes and promotes transcription elongation through phosphorylation of RNA polymerase II. We have developed a potent, selective, and orally bioavailable CDK9 inhibitor, KB-0742. Here we present the preclinical activity of KB-0742 in models of sarcoma and chordoma. We first evaluated KB-0742 activity in sarcoma in a 300-immortalized cell line screen containing 18 soft tissue sarcoma cell lines. The median IC50 of the cell lines in the study was 0.8705 µM. Sarcoma cell lines were enriched for sensitivity to KB-0742 with 61% (11/18) of the lines with an IC50 below the median. The 11 cell lines had a median IC50 of 0.679 µM. We then evaluated the activity of KB-0742 in 5 patient-derived cell line (PDC) models, with all 5 showing a cytotoxic response to treatment as measured by negative GRmax values. Last, a single patient-derived organoid (PDO) model of adult rhabdomyosarcoma was evaluated. KB-0742 treatment resulted in an IC50 of 2.75 µM and a max inhibition rate of 98.61%. For chordoma we examined the activity of KB-0742 in vivo using 2 patient-derived xenograft (PDX) models. In model CF466, we observed a dose-dependent response with increased TGI and greater reductions in RNA polymerase II phosphorylation in tumors treated with KB-0742 at 60 mg/kg as compared to 30 mg/kg. We then evaluated KB-0742 as a single agent and in combination with afatinib in the model CF539. KB-0742 as a single agent showed similar TGI activity as afatinib, whereas the combination showed increased response as compared to the two single-agent arms.

These data show that transcriptionally addicted tumors are sensitive to CDK9 inhibition via KB-0742 treatment, and support the continued development of our compound to potentially treat sarcoma and chordoma. KB-0742 is currently being evaluated in a phase 1/2 clinical trial (NCT04718675) for relapsed or refractory solid tumors or non-Hodgkin lymphoma. Once a recommended phase 2 dose is established, expansion cohorts for patients with sarcoma, chordoma, and other transcriptionally addicted tumors may be opened.

Citation Format: Melinda A. Day, Douglas C. Saffran, Tressa Hood, Nikolaus Obholzer, Akanksha Pandey, Charles Y. Lin, Pavan Kumar, Daniel M. Freed, Jorge DiMartino. CDK9 inhibition via KB-0742 is a potential strategy to treat transcriptionally addicted cancers [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):Abstract nr 2564.