A majority of cases of high-risk neuroblastoma, an embryonal childhood cancer, are driven by MYC or MYCN-driven oncogenic signaling. While considered to be directly “undruggable” therapeutically, MYC and MYCN can be repressed transcriptionally by inhibition of Bromodomain-containing protein 4 (BRD4) or destabilized posttranslationally by inhibition of Aurora Kinase A (AURKA). Preclinical and early-phase clinical studies of BRD4 and AURKA inhibitors, however, show limited efficacy against neuroblastoma when used alone. We report our studies on the concomitant use of the BRD4 inhibitor I-BET151 and AURKA inhibitor alisertib. We show that, in vitro, the drugs act synergistically to inhibit viability in four models of high-risk neuroblastoma. We demonstrate that this synergy is driven, in part, by the ability of I-BET151 to mitigate reflexive upregulation of AURKA, MYC, and MYCN in response to AURKA inhibition. We then demonstrate that I-BET151 and alisertib are effective in prolonging survival in four xenograft neuroblastoma models in vivo, and this efficacy is augmented by the addition of the antitubule chemotherapeutic vincristine. These data suggest that epigenetic and posttranslational inhibition of MYC/MYCN-driven pathways may have significant clinical efficacy against neuroblastoma.

A majority of cases of high-risk neuroblastoma, an embryonal childhood cancer, are driven by MYC or MYCN-driven oncogenic signaling. While considered to be directly "undruggable" therapeutically, MYC and MYCN can be repressed transcriptionally by inhibition of Bromodomain-containing protein 4 (BRD4) or destabilized posttranslationally by inhibition of Aurora Kinase A (AURKA). Preclinical and early-phase clinical studies of BRD4 and AURKA inhibitors, however, show limited efficacy against neuroblastoma when used alone. We report our studies on the concomitant use of the BRD4 inhibitor IBET-151 and AURKA inhibitor alisertib. We show that, in vitro, the drugs act synergistically to inhibit viability in three models of high-risk neuroblastoma. We demonstrate that this synergy is driven, in part, by the ability of IBET-151 to mitigate reflexive upregulation of AURKA, MYC, and MYCN in response to AURKA inhibition. We then demonstrate that IBET-151 and alisertib are effective in prolonging survival in three xenograft neuroblastoma models in vivo, and this efficacy is augmented by the addition of the antitubule chemotherapeutic vincristine. These data suggest that epigenetic and posttranslational inhibition of MYC/MYCN-driven pathways may have significant clinical efficacy against neuroblastoma.

Background: Ewing sarcoma (ES) is an undifferentiated cancer of the bone and soft tissue canonically defined by expression of EWS-FLI1 or similar fusion proteins. The cell of origin remains undefined but is likely either a human mesenchymal stem cell (hMSC) or neural crest stem cell. The ES cells are depending on EWS-FLI1 for viability, but paradoxically exogenous EWS-FLI1 expression causes cell death in virtually all other cells. Expression of EWS-FLI1 in hMSCs has been variably successful and unreliable, and the two prior reports of stable expression did not elucidate why those efforts were successful, preventing reproducibility. Our work has shown that expression of the lncRNA HOTAIR allows for subsequent stable expression of EWS-FLI1 in hTERT-immortalized hMSCs and primary hMSCs but without induction of tumorigenesis in vivo. Methods: We used established ES cell lines, hTERT-hMSCs and primary hMSCS, to evaluate mRNA and protein expression at baseline and in the context of modulation of expression of HOTAIR, EWS-FLI1, MYC, and MYCN. Results: We demonstrated by RT-qPCR that HOTAIR is highly expressed across all ES cell lines as compared to hTERT-hMSCs and primary hMSCS. We also found that MYC is highly expressed in most ES cells lines and hMSCs. In contrast, MYCN is overexpressed in ES cell lines as compared to hMSCs. Repression of MYC in ES cell lines had no significant effect on proliferation of ES cells in vitro; repression of MYCN by shRNA in ES cell lines decreased proliferation in vitro and slowed tumor growth in vivo. Exogenous expression of HOTAIR in hMSCs allowed for viable subsequent expression of EWS-FLI1 in those cells, but without tumorigenesis in vivo (0/10 for hTERT- or primary hMSCs). Overexpression of MYC (0/6) or MYCN (1/6) in hMSCs expression HOTAIR did not reliably induce tumorigenesis. Overexpression of MYC or MYCN in hMSCs did not allow for viable EWS-FLI1 expression. Overexpression of HOTAIR, then EWS-FLI1 and MYCN, to levels comparable with ES cell lines, induced tumorigenesis from hTERT-hMSCS and primary hMSCs (6/6 for each), when cells were implanted subcutaneously into SCID mice with Matrigel. The resultant tumors were positive by immunohistochemistry for CD99 and NKX2.2, clinical markers of ES. RNA-Seq demonstrated the HOTAIR-EWS-FLI1-MYCN expressing hMSCs had gene expression profiles comparable to ES tumors. Conclusions: We have generated the first genetically defined tumorigenic models of ES from non-neoplastic human cells. This model can be used to elucidate the biology of EWS-FLI1 in malignant transformation and tumor formation. This work supports our prior data demonstrating a necessary role for HOTAIR in ES biology and suggests that MYCN may also be a driver of tumorigenesis in ES. Additional work is needed to define the underlying factors that induce HOTAIR and MYCN expression in ES (i.e., are they basally or aberrantly expressed in the cell of origin?) and if they are therapeutically targetable. This abstract is also being presented as Poster B77. Citation Format: Joshua Felgenhauer, Laura Tomino, Hasan Siddiqui, Emily Bopp, Julia Selich-Anderson, Jessica Mormol, Cenny Taslim, Nilay Shah. Generation of the first genetically defined tumorigenic model of Ewing sarcoma expressing EWS-FLI1 [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 PR02.