A highly sensitive screening system to evaluate the reversibility of neuroendocrine prostate cancer to prostate adenocarcinoma

Prostate cancer grows in an androgen-dependent manner, and the standard therapy for advanced prostate cancer is endocrine therapy targeting the androgen receptor (AR) signaling pathway. However, with the widespread use of potent next-generation AR signaling inhibitors (ARSIs), the incidence of treatment-related neuroendocrine prostate cancer (t-NEPC), which is completely independent of the AR pathway, is rapidly rising.^{1, 2} Unlike its AR-dependent ancestor, t-NEPC remains a poor prognosis cancer, lacking effective treatment options, and thus necessitating the exploration and development of innovative therapies.²

Transdifferentiation of prostate adenocarcinoma into neuroendocrine prostate cancer (NEPC) occurs through lineage plasticity.^3-5 Lineage plasticity is a biological process that enhances cell survival by enabling adaptation to the environment, avoidance of stress, or tissue repair.⁴ Within cancer, lineage plasticity facilitates the development of therapy resistance in cancer cells by reprogramming into therapy-resistant phenotypes that bypass targeted therapies.⁵ This phenomenon is especially notable in cancer varieties where there are potent targeted therapies for key growth pathways, such as AR-driven prostate cancer, epidermal growth factor receptor (EGFR)-mutant lung cancer, and BRAF-mutant melanoma.⁴

Recent genomic analyses have revealed several alterations enriched in t-NEPC, with the loss of tumor suppressor genes (TP53 and RB1) being crucial genomic changes linked to t-NEPC.^{3, 6} Moreover, epigenetic genes, such as EZH2 and SOX2, induce neuroendocrine (NE) transdifferentiation.^{7, 8} A study with gene-engineered mice lacking Pten and Rb1, or all three (Pten, Rb1, and Trp53), showed reduced expression of AR and increased expression of NE-related genes, phenocopying human NEPC. Furthermore, using EZH2 inhibitors restored AR expression and sensitivity to antiandrogen therapy.⁷ These findings suggest that plasticity in t-NEPC is potentially reversible, and regulating cellular lineage could serve as a novel therapeutic strategy. However, EZH2 inhibition did not increase AR expression or activity in organoids from human NEPC.⁹ To date, the reversibility of lineage plasticity in prostate cancer has not been confirmed in human-derived t-NEPC clinical models.

One major hurdle in NEPC research is the paucity of human-derived t-NEPC cell lines suitable for genetic manipulation or large-scale compound screening. Previously, we created a novel t-NEPC cell line called KUCaP13, derived from a patient-derived xenograft (PDX) and verified its lineage originating from prostate adenocarcinoma.¹⁰ The cell line's origin as prostate adenocarcinoma is supported by the preservation of a homozygous CHD1 deletion from patient tissue to the cell line.¹⁰ Concurrent inactivation of TP53, RB1, and PTEN in KUCaP13 makes it an ideal candidate for NEPC research.¹¹ KUCaP13 cells grow suspended in culture, forming spheroids. They are dissociated into single cells to facilitate lentiviral transfection, fluorescence-activated cell sorting (FACS) sorting, and accurate cell counting. Herein, we hypothesized that t-NEPC can be transdifferentiated back to adenocarcinoma and attempted to prove the reversibility of cellular lineage by compound screening using KUCaP13 cells (Figure S1). To this end, we developed a highly sensitive screening system to detect AR re-expression in KUCaP13 cells and performed a high-throughput screening using a chemical library.

NEPC is an aggressive variant of prostate cancer. However, due to its rarity and the lack of experimental models, its biology has remained unclear. Recent advances in genomic analysis of human specimens have revealed NEPC's genomic and molecular biological characteristics. The acquisition of MYCN oncogenes, coupled with the loss of tumor suppressor genes (TP53, RB1, and PTEN), play a critical role in NEPC development.^{13, 14} However, since these genes are not direct therapeutic targets, NEPC remains a cancer with a very poor prognosis for which no effective treatment currently exists. The origin of NEPC, whether it arises from a small subset of pre-existing NE cells in the prostate gland or from the transdifferentiation of adenocarcinoma cells, has sparked controversy.^{15, 16} However, recent reports strongly suggest that t-NEPC undergoes transdifferentiation from an adenocarcinoma due to epithelial plasticity.^{6, 17, 18} The mechanism underlying lineage plasticity in t-NEPC is thought to be epigenetic and potentially reversible; however, the detailed processes have not been fully clarified.⁴

Despite recent advances in t-NEPC research, no t-NEPC cell lines are available for genetic manipulation or large-scale compound screening. The solitary widely employed cell line derived from a patient with NEPC is NCI-H660. Initially classified as originating from small-cell lung cancer, it was subsequently recognized as prostatic in nature due to the presence of TMPRSS2-ERG fusion.¹⁹ However, since the cell line originated from a lymph node metastasis in a patient who had died within a month of diagnosis without receiving treatment, NCI-H660 is likely to be a de novo NEPC, and it is uncertain whether it originally expressed AR.^{19, 20} Therefore, NCI-H660 is not a suitable model for studying the lineage plasticity of t-NEPC. In contrast, a novel t-NEPC cell line, named KUCaP13, was created from the tissue of a patient initially diagnosed with adenocarcinoma, which subsequently transformed into NEPC. KUCaP13 is the initial patient-derived t-NEPC cell line displaying the triple loss of tumor suppressors crucial for NEPC progression via lineage plasticity.¹⁰ In prostate cancer, lineage alteration can be evaluated by AR activity, unlike in other cancer types, such as lung cancer and melanoma. Hence, in the present study, we focused on identifying compounds that can re-express AR through chemical screening using KUCaP13 cells. Identifying these compounds will contribute to understanding the mechanism of lineage plasticity in t-NEPC and enable the development of innovative therapies for t-NEPC in combination with ARSIs.

AR is crucial in the development and progression of most prostate cancer cases. AR, a ligand-dependent transcription factor, is localized in the cytoplasm when inactive.²¹ Ligand binding, by the native androgens testosterone and 5α-dihydrotestosterone, induces nuclear translocation and receptor dimerization. In the nucleus, AR recruits various collaborative factors and binds ARE sequences located in the enhancer and promoter regions of target genes. Thus, it regulates the transcription of androgen-responsive genes, including prostate-specific antigens.²² Azeem et al. developed ARE-based assays that can test AR activity on a large scale, rendering them valuable for drug screening.²³ Luciferase is frequently employed as a reporter in high-throughput screening assays due to its remarkable sensitivity, extensive dynamic range, and swift measurement capabilities.²⁴ Thus, in our study, a reporter gene, AREluc, was transduced into KUCaP13 cells to detect AR activity using luciferase assay. The positive control cells, KUCaP13_AREluc overexpressing AR, exhibited enhanced luminescence upon the administration of synthetic androgens. This reporter assay system enables the assessment of plasticity reversibility by detecting AR expression in KUCaP13 and facilitates high-throughput screenings. This is a novel screening system for evaluating the reversibility of plasticity using the t-NEPC cell line.

Compounds responsible for AR re-expression in KUCaP13 cells were screened using the library of known compounds provided by our Drug Discovery Research. Previously, we identified disulfiram as a sensitizer of cisplatin in bladder cancer through high-throughput chemical screening using this library.²⁵ Our present study detected compounds that enhance luciferase luminescence and identified 30 hit compounds in the first screening. However, all hit compounds were confirmed as false positives through the second screening and qPCR. One possible reason is the inhibitory effect of the compounds on luciferase. In firefly luciferase reporter gene assays, inhibitors of luciferase can act intracellularly to prolong the lifespan of ectopically expressed firefly luciferase enzyme. This leads to heightened luciferase activity, which can be visually indistinguishable from the activation of reporter gene transcription.²⁶ In previous studies, inhibitory activity against luciferase was identified in 12% of the compounds present in a chemical library.²⁷ In our present study, 73% of the compounds that exhibited false positives shared the same root scaffolds as luciferase inhibitors that had been reported previously.²⁷

The reversibility of lineage plasticity in t-NEPC has been demonstrated in gene-engineered mice, but this has not been confirmed in models from clinical samples. Recently, the reversibility of lineage plasticity in castration-resistant prostate cancer (CRPC) in the intermediate/lineage-plastic state through the JAK/STAT pathway has been reported.²⁸ The combination of a JAK inhibitor and an FGFR inhibitor increased AR expression in organoids with enhanced JAK/STAT signaling from patients with CRPC but did not affect AR expression in organoids from patients with NEPC.²⁸ In the current situation, where the reversibility of cellular lineage of t-NEPC in clinical specimens has not been demonstrated, the establishment of a screening system to detect AR re-expression using t-NEPC cell lines and applying it in large-scale chemical screening may open a door for an innovative development.

In our study, a single compound alone was unable to re-express AR in KUCaP13 cells. However, since lineage plasticity involves multiple processes, there may be potential for AR re-expression in t-NEPC utilizing a synergistic combination of multiple compounds. Alternatively, it may be that the regulation of cellular lineage in cells that had undergone complete transdifferentiation to t-NEPC is challenging. In a previous study identifying the role of JAK/STAT signaling in NEPC transdifferentiation, it was reported that therapeutic timing was critical due to cell-state heterogeneity in a patient with CRPC and the absence of JAK/STAT activation in NEPC.²⁸ Hence, reversing the cell lineage in the earlier stages of transdifferentiation may be crucial.

Our study has several limitations. This screening system was challenged by the absence of a compound that can serve as a positive control. Before compound screening, we independently tested several drugs that could potentially impact plasticity in t-NEPC, including an EZH2 inhibitor (EPZ-6438), a bromodomain inhibitor ((+)-JQ1), an LSD inhibitor (GSK2879552 2HCl), and a histone deacetylase inhibitor (Trichostatin A); however, none of them elicited an increase in luminescence in the presence of R1881 (data not shown). Therefore, we established KUCaP13_AREluc_AR cells as the positive control. Furthermore, owing to technical limitations associated with the nature of KUCaP13 cell growth in suspension and the difficulty of refreshing culture media during the compound screening process, we measured luminescence 2 days post-exposure to the compounds. In a previous report using organoids, human organoids had been exposed to compounds for 14 days to assess the reversibility of plasticity.²⁸ In human prostate cancer-derived cells, it may take more than 2 days for AR re-expression. We set the entire screening to be completed in 3 days since luminescence in KUCaP13_AREluc_AR cells showed the most significant increase after exposure to R1881 cells for a 3-day incubation period (data not shown). Additionally, as cell viability was not assessed, it is possible that some of the compounds could have caused cell death at 10 μM, potentially resulting in false negatives.

In conclusion, we developed a highly sensitive screening system to evaluate the reversibility of plasticity in t-NEPC using KUCaP13. Despite not achieving the goal of AR-re-expression, this study paves the way for the application of high-throughput screening approaches in t-NEPC research and future exploration of alternative strategies, including targeting earlier stages of transdifferentiation or investigating synergistic combinations of compounds.

Tomohiro Fukui: Data curation (lead); formal analysis (lead); funding acquisition (equal); investigation (lead); methodology (equal); validation (lead); visualization (lead); writing – original draft (lead); writing – review and editing (equal). Kosuke Okasho: Investigation (equal); resources (lead); writing – review and editing (equal). Yukiko Okuno: Formal analysis (equal); resources (equal); visualization (equal); writing – review and editing (equal). Maki Fujiwara: Investigation (equal). Kensuke Hikami: Investigation (equal). Arinobu Fukunaga: Investigation (equal). Takuro Sunada: Formal analysis (equal); investigation (equal). Yuki Kita: Formal analysis (equal); writing – review and editing (equal). Takayuki Sumiyoshi: Formal analysis (equal); writing – review and editing (equal). Takayuki Goto: Formal analysis (equal); writing – review and editing (equal). Ryoichi Saito: Formal analysis (equal); writing – review and editing (equal). Osamu Ogawa: Project administration (equal); supervision (equal). Takashi Kobayashi: Project administration (equal); resources (equal); supervision (equal). Shusuke Akamatsu: Conceptualization (lead); formal analysis (equal); funding acquisition (lead); methodology (equal); project administration (lead); supervision (lead); writing – review and editing (lead).

This research was funded by JSPS KAKENHI Grant Number JP21K19568 to Shusuke Akamatsu and by the funding from the Japanese Urological Association to Shusuke Akamatsu. Partial financial support was received from Bayer Yakuhin, Ltd to Tomohiro Fukui.

Tomohiro Fukui received research funding from Bayer Yakuhin, Ltd.