(P 193) TARGETING THE PAX-FOXO FUSION VIA HDAC INHIBITION IN RHABDOMYOSARCOMA

Objective: Approximately 70% of patients who present with metastatic rhabdomyosarcoma(RMS) and 50-80% of patients who develop recurrent disease will not survive. There are no FDA approved targeted therapies for RMS and currently, there are only a handful of trials exploring novel targeted agents in refractory RMS. Epigenetic regulation, including post-translational modifications of histones, has been shown to play an important role in the pathogenesis of multiple types of cancer, including RMS, a disease that is driven by Pax-Foxo fusions. Further, the chromatin remodeling NuRD complex which contains HDAC1,2 interacts with the fusion protein and may have a role in RMS pathogenesis. HDAC inhibitors have been shown to inhibit tumor growth and migration in RMS cell lines and tumor models and are promising potential therapies for RMS. Herein, we explored the therapeutic possibility of HDAC inhibition in RMS.

Methods: The IC50G of the class 1 HDAC inhibitor mocetinostat (moce) was determined in 50 human soft tissue and bone sarcoma cell lines representing several histologies by exposing the lines to six drug concentrations over a 5-day incubation and using direct cell counts. The IC50s were corroborated in a 2nd assay for the RMS cell lines. These compounds were assayed alone and in combination with several agents and synergy was determined by calculating cell growth inhibition. The most effective combinations were assayed in fusion+ and fusion- RMS xenografts to corroborate in vitro findings. Xenografts were treated with 100mg/kg daily moce, 0.5-1.5mg/kg weekly bortezomib, 10mg/kg weekly vinorelbine or combinations of moce with bortezomib or vinorelbine. Coimmunoprecipitation of the Pax-Foxo fusion complex was carried out and western blots were done to assay for the presence NuRD complex members. Selected fusion complex members were selectively knocked down and the IC50 was determined to verify target inhibition. ChipSeq analysis of mocetinostat treated (48 hours, 1uM moce) fusion+ and fusion- RMS lines was carried out.

Results: 7 RMS cell lines were found to be relatively more sensitive to moce (IC50G range: 0.03-0.25µM) as compared to other bony and soft tissue sarcomas (Fig. 1) and this was corroborated in a second growth inhibition assay(IC50 range: 0.08-0.92µM). Two lines were considered “relatively resistant.” Vinorelbine was the most potent combinatorial agent identified and, when combined with moce was more potent at growth inhibition than either drug alone. These findings extended to JR and SJCRH40 fusion+ xenografts where combination therapy was more effective than single agent moce. Coimmunoprecipitation and western blots showed the presence of NuRD complex subunits. sgRNA knockdown of Foxo1, Pax3, HDAC1 and HDAC2 decreased sensitivity to moce in the JR cell line, but not in the A204 (fusion negative) cell line. Pathway analysis revealed that moce treated fusion+ cell lines had a similar signature to lines that underwent Pax-Foxo knockdown. The most common genes upregulated in the mocetinostat treated JR line were involved in muscle differentiation and function and neural genes.

Conclusion: The HDAC inhibitor mocetinostat has selective activity against RMS. This sensitivity is augmented by vinorelbine in vitro and in vivo. Preliminary evidence indicates that the RMS fusion complex can be targeted by mocetinostat, putatively by targeting HDAC subunits in the NuRD complex. Treatment with mocetinostat was shown to lead to upregulation of genes involved in muscle and nerve differentiation. An ongoing clinical trial is looking at the combination of mocetinostat and vinorelbine in RMS (NCT04299113).

IC50s.pdf