Direct Co-Targeting of Bcl-xL and Mcl-1 Exhibits Synergistic Effects in AR-V7-Expressing CRPC Models

Abstract

There is an unmet need to develop novel treatment options for patients with metastatic castration-resistant prostate cancer (mCRPC). Patients often develop resistance to next-generation hormonal therapies that target the androgen receptor (AR) axis (e.g., abiraterone and enzalutamide). A splice variant of AR, AR-V7, is associated with resistance to these inhibitors as well as mCRPC progression and poor prognoses. We embarked upon a high-throughput screen to identify synergistic combinations of targeted therapies using two CRPC cell lines, LNCaP95 and VCaP-CR. Combinations targeting BCL2L1 (Bcl-xL) (A-1331852 and navitoclax) and MCL1 (S63845) synergistically decreased cell viability and induced apoptotic activity via cleavage of PARP, caspase 3, and caspase 7 across AR-V7-expressing CRPC cell lines (LNCaP95, VCaP-CR, and 22Rv1) and a patient-derived organoid model (LuCaP 167CR). We also explored the use of a Bcl-xL-specific proteolysis-targeting chimera degrader (PROTAC) to minimize platelet toxicity associated with Bcl-xL inhibitors. We showed similar synergistic efficacy with the Bcl-xL-targeting PROTAC in combination with S63845 in the three-dimensional spheroid models. Our findings support further preclinical development of Bcl-xL and Mcl-1 inhibitors for mCRPC.

Significance: Using an unbiased, combinatorial, high-throughput drug screen, we identified the combination of co-targeting Bcl-xL and Mcl-1 to be highly synergistic across AR-V7-expressing CRPC models. We showed efficacy in higher-order models through validation across in vitro models spanning two-dimensional cell culture, three-dimensional cell culture, and a patient-derived organoid model. These findings identify a promising therapeutic strategy for patients with AR-V7-expressing CRPC.

Figure 1

Identification of Synergistic BH₃-mimetic–based combinations in LNCaP95 and VCaP-CR Cells. A, Correlation plot depicting DBSumNeg (synergy score) for drugs that pair well with S63845 across LNCaP95 and VCaP-CR lines. B, All pairings of BH₃ mimetics ranked by average DBSumNeg across LNCaP95 and VCaP-CR lines. C to F, Example drug screen data from A-1331852 paired with S63845 in VCaP-CR (C and D) and LNCaP95 (E and F) cells using CellTiter-Glo to measure response. Cell viability (C and E) and Bliss synergy were calculated for each point. Z-plots depict Bliss synergy score at each concentration (D and F). G, Patient mRNA sequencing data from the SU2C 2019 cohort of patients with mCRPC reveals the expression of AR, BCL2, Bcl-xL, and MCL1. Cohort expression was displayed utilizing GraphPad Prism 10, and data are expressed as fragments per kilobase of transcript per million mapped reads (FPKM).

Figure 2

Co-targeting Mcl-1 and Bcl-xL enhances the inhibitory effect on cell viability. The effect on cell viability of A-1331852 (targeting Bcl-xL) in combination with S63845 (targeting Mcl-1; A–C) and navitoclax (targeting Bcl-xL and Bcl-2) in combination with S63845 (D–F). Cell viability was determined following either 48-hour (22Rv1 and LNCaP-95) or 72-hour treatment (VCaP-CR) by CellTiter-Glo luminescent assay. The data provided represent the mean ± SD from three independent experiments with at least three technical replicates per experiment.

Figure 3

Optimal dose evaluation for co-targeting Bcl-xL and Mcl-1. Dose–response heatmaps (A–C and G–I) and Bliss synergy plots (D–F and J–L) for A-1331852-based (A–F) and navitoclax-based (G–L) combinations with S63845 were obtained using the SynergyFinder+ R package. Plots were generated using data from 2D cell culture experiments in LNCaP95 (A, D, G, and J), VCaP-CR (B, E, H, and K), and 22Rv1 (C, F, I, and L). Data provided represent mean ± SD from three individual experiments.

Figure 4

Live-cell imaging reveals dynamics of apoptosis and proliferation dynamics after dual treatment of A-1331852 and S63845. A to C, Count of cells with notable caspase 3/7 cleavage measured by a green fluorescent signal after dual treatment. D to F, Proliferation of cells after dual treatment as measured by the confluency of each well. All experiments contained three biological and three technical replicates. Data are plotted as mean ± SD of the means from each experiment.

Figure 5

Dual targeting of Bcl-xL and Mcl-1 leads to synergistic decreases in cell viability in 3D LNCaP95 spheroids and the LuCaP 167CR organoid model. Dual treatment of navitoclax and A-1331852 with S63845 in 3D prostate cancer models. A, Representative images for each treatment condition before treatment and 24, 48, and 72 hours after treatment. B and C, Viability of spheroids measured by 3D CellTiter-Glo after 72 hours of treatment. Plots show mean ± SD. ****, P < 0.0001. LuCaP-167CR organoids were treated with the combination of A-1331852 and S63845 (D–F) or navitoclax and S63845 (G–I). Drug curves (D and G) show mean ± SD. Dose–response matrices (E and H) and Bliss synergy plots (F and I) were generated using the Synergyfinder+ R package to evaluate synergy. All spheroid experiments were completed in triplicate. Organoid experiments had six technical replicates with three biological replicates.

Figure 6

XZ739 in combination with S63845 has synergistic antitumor effects in CRPC cell lines. A to I, CellTiter-Glo viability data from 2D cultures of LNCaP95, VCaP-CR, and 22Rv1 cell lines. Cell viability data for the dual treatment (A–C) were generated by CellTiter-Glo. Dose–response matrices (D–F) and Bliss synergy plots (G–I) were created using the SynergyFinder+ R package. J to M, 3D spheroid model and cell viability of LNCaP95 (J and K) and 22Rv1 cells (L and M). Data are presented as mean ± SD. **, P < 0.01; ****, P < 0.0001. All experiments were completed in triplicate. Spheroid images are representative of three replicates.