Identification of BET inhibitors (BETi) against solitary fibrous tumor (SFT) through high-throughput screening (HTS)

Abstract

Cancers, especially fusion oncoprotein (FO)-driven hematological cancers and sarcomas, often develop from a low number of key mutations. Solitary Fibrous Tumor (SFT) is a rare mesenchymal tumor driven by the NAB2-STAT6 oncofusion gene. Currently, the treatment options for SFT remain limited, with anti-angiogenic drugs providing only partial responses with an average survival of two years. We constructed SFT cell models harboring specific NAB2-STAT6 fusion transcripts using the CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) technology, and we used these cells as models of SFT. High-throughput drug screens demonstrated that the BET inhibitor Mivebresib can differentially reduce proliferation in SFT cell models. Subsequently, BET inhibitors Mivebresib and BMS-986158 efficiently reduced tumor growth in an SFT patient-derived xenograft (PDX) animal model. Furthermore, our data showed that NAB2-STAT6 fusions may lead to high levels of DNA damage in SFTs. Consequently, combining BET inhibitors with PARP (Poly (ADP-ribose) polymerase) inhibitors or with ATR inhibitors significantly enhanced anti-proliferative effects in SFT cells. Taken together, this study establishes BET inhibitors Mivebresib and BMS-986158 as promising anti-SFT agents.

Keywords: BET inhibitor (BETi); High-throughput screen (HTS); Mivebresib; Solitary fibrous tumor (SFT).

Fig. 1

Identification of Mivebresib as an efficient and specific anti-SFT (Solitary Fibrous Tumor) agent. (a) Secondary high-throughput screening (HTS) was performed using the Moffitt-ns and immortalized lung fibroblast cells. The CTG effects of both cell lines under all four doses were plotted. The green line indicated the differences in suppression effects between Moffitt-ns and Lf > 40 %. The blue line indicated that the suppression effects at the highest dose should be > 50 % in Moffitt-ns cells. 3 compounds were shown (Mivebresib: green, Zinc Pyrithione: red, and TAK-901: yellow). (b-f) Confirmatory MTS assays were performed in two primary SFT cells (INT-SFT and IEC139) and two control LMS cells (SKUT-1 and CP0024). Only Mivebresib showed efficient and selective cell proliferation-suppressing effects in SFT cells. (): INT-SFT; (): IEC139; (): SKUT-1; (): CP0024.

Fig. 2

In vitro evaluation of activities and specificities of additional BET inhibitors in SFT cells. a)In vitro evaluation of activities of additional BET inhibitors (BMS-986158, Pelabresib, ABBV-744, PLX51107, GSK778, and GSK046) in INT-SFT cells using Flow cytometry-based apoptosis analysis. b)In vitro evaluation of activities of additional BET inhibitors (BMS-986158, Pelabresib, ABBV-744, PLX51107, GSK778, and GSK046) in IEC139 cells using Flow cytometry-based apoptosis analysis. For both a) and b), cells were treated with candidate chemicals (50 nM) for 72 hours, and the drug activities were expressed as the percentage of non-viable cell subpopulation to the total cell population. For statistical analysis, two-tailed t-tests were conducted. ** denotes p<0.01, *** denotes p<0.001, n.s. denotes no significant difference. c) Dose-response curves for BMS-986158 at 72 hours in primary SFT cells (INT-SFT and IEC139) and control leiomyosarcoma (CP0024 and SK-UT-1) cells using MTS cell viability assays. Cell viability was normalized to untreated conditions (n=4).

Fig. 3

BET inhibitors (BETi) Mivebresib and BMS-986158 exerted anti-proliferative effects in SFTs via cell cycle and DNA damage response pathways. a) Protein levels of apoptosis marker (cleaved PARP-1) and DSBs (double-strand breaks) marker (γ-H2AX) increased in SFT cells after 72-hour treatment with 50 nM of Mivebresib or BMS-986158. b) Protein levels of phosphorylated ATR at serine 428 (p-ATR, Ser428), p21, Cyclin D1, RAD51, and Wee-1 in SFT cells (INT-SFT and IEC139) during a 24-hour treatment with 50 nM Mivebresib. The α-tubulin was used as a loading control. c) Protein levels of phosphorylated ATR at serine 428 (p-ATR, Ser428), p21, Cyclin D1, RAD51, and Wee-1 in SFT cells (INT-SFT and IEC139) during a 24-hour treatment with 50 nM BMS-986158. The α-tubulin was used as a loading control. d) Ratios representing proliferating (G2 and S) vs non-proliferating (sub-G1 and G1) cells upon Mivebresib or BMS-986158 treatment. Bar plots represent mean values with standard deviations. For statistical analysis, two-tailed t-tests were conducted. * denotes p < 0.05; *** denotes p < 0.001.

Fig. 4

Combinatorial effects between BET inhibitors (Mivebresib and BMS-986158) and PARPi Rucaparib. a) Flow cytometry-based apoptosis assays showed that combining Mivebresib and Rucaparib increased apoptotic and necrotic cell populations in INT-SFT and IEC139 cells (n=3). b) Western blot assays showed that combining Mivebresib and Rucaparib increased cleaved PARP-1 and γ-H2AX protein levels in INT-SFT and IEC139 cells after 72-hour treatment (n=3). The α-tubulin was used as a loading control. c) Flow-cytometry-based apoptosis assays showed that combining BMS-986158 and Rucaparib increased apoptotic and necrotic cell populations in INT-SFT and IEC139 cells (n=3). d) Western blot assays showed that combining BMS-986158 and Rucaparib increased cleaved PARP-1 and γ-H2AX protein levels in INT-SFT and IEC139 cells after 72-hour treatment (n=3). The α-tubulin was used as a loading control. For statistical analysis, two-tailed t-tests were conducted. * denotes p < 0.05; ** denotes p < 0.01.

Fig. 5

Combinatorial effects between BET inhibitors (Mivebresib and BMS-986158) and ATRi Berzosertib. a) Flow-cytometry-based apoptosis assays showed that combining Mivebresib and Berzosertib increased apoptotic and necrotic cell populations in INT-SFT and IEC139 cells (n=3). b) Western blot assays showed that combining Mivebresib and Berzosertib increased cleaved PARP-1 and γ-H2AX protein levels in INT-SFT and IEC139 cells after 72-hour treatment (n=3). The α-tubulin was used as a loading control. c) Flow-cytometry-based apoptosis assays showed that combining BMS-986158 and Berzosertib increased apoptotic and necrotic cell populations in INT-SFT and IEC139 cells (n=3). d) Western blot assays showed that combining BMS-986158 and Berzosertib increased cleaved PARP-1 and γ-H2AX protein levels in INT-SFT and IEC139 cells after 72-hour treatment (n=3). The α-tubulin was used as a loading control. For statistical analysis, two-tailed t-tests were conducted. ** denotes p < 0.01.

Fig. 6

Evaluation of in vivo anti-tumor effects of Mivebresib in IEC139 PDX models. a) Schematic illustration of Mivebresib dosing regimen in IEC139 PDX models. b) Representative images of xenografts harvested at the end of the treatment period. (left) Control, (right) Mivebresib. c) Tumor volume progression throughout the treatment period for both Control- and Mivebresib-treated groups (n=3-4 per group). d) Body weight measurements throughout the treatment period for Control- and Mivebresib-treated groups (n=3-4 per group). Statistical analysis was conducted using two-way ANOVA with multiple daily comparisons (Sidak). * denotes p < 0.05; ** denotes p < 0.01; *** denotes p < 0.001; non-significant results are not shown.

Fig. 7

Transcriptomic response to BET inhibition in SFT cells. a) Heatmap of differentially expressed genes (DEGs). b) Venn diagrams illustrate a substantial overlap of up- and downregulated genes between treatments. c) GSEA reveals consistent downregulation of inflammatory, MYC, KRAS, and EMT-related pathways, and upregulation of DNA repair pathways following BET inhibition.