Toppling the HAT to Treat Lethal Prostate Cancer

Increased androgen receptor (AR) signaling is a hallmark of advanced castration-resistant prostate cancer (CRPC). AR is a ligand-dependent nuclear transcription factor. Normal AR function is essential for the development of the prostate gland, male sexual differentiation, and spermatogenesis. AR plays a protumorigenic role in primary prostate cancer, which is the premise for antiandrogen treatment. However, the disease invariably relapses as advanced CRPC with restored AR signaling in most cases through various mechanisms, including AR mutations, AR splice variants, and AR-gene/enhancer amplification, coregulator modulation, and intratumoral androgen synthesis. Abiraterone, a CYP17A1 inhibitor, and the second-generation antiandrogen enzalutamide can effectively block activated AR signaling and improve CRPC survival, but unfortunately most patients relapse within a few years. The antiandrogen refractory CRPC displays further increased AR transcriptional activity. The relatively lower prevalence of AR independence or neuroendocrine differentiation further supports the central role of AR signaling in the majority of advanced castration-resistant tumors (1, 2).

The oncogenic AR transcriptional activity in the CRPC setting is dependent on a number of coactivators and chromatin-associated proteins, many of which remain to be identified. Recent advances in our fundamental understanding of transcriptional control in cancer and the identification of transcriptional coactivators as druggable targets have opened up new avenues for targeting AR signaling downstream of its chromatin binding and overcoming resistance mechanisms. Targeting of proteins functionally associated with AR is now seen as a promising new approach in blocking dysregulated AR signaling. Disruption of AR interaction with its coactivators, such as BRD4, MED1, MLL-Menin, DNAPKs, etc., has shown promising results in attenuating AR signaling in CRPC models and is currently being evaluated in clinical trials (3).

The cyclic-AMP response element binding protein (CREB) binding protein (CBP) and p300 (also known as PCAF or KAT3B) are paralog histone acetyltransferases (HAT) that have high sequence homology and functional similarity. p300/CBP preferentially acetylates histone H3K18/K27—a form of chromatin modification that strongly correlates with active transcription. In addition, p300/CBP can acetylate specific lysines on other transcription regulators, recruit components of the Pol II machinery, and act as adaptors to recruit other cofactors. The p300/CBP transcription function, especially their histone acetyltransferase activity and chromatin binding, is regulated both intramolecularly by their conserved autoinhibitory loop (AIL), bromodomain (BRD), and PHD-RING region, and intermolecularly by their interacting proteins (4). p300/CBP has multiple binding pockets that enable it to interact with both basal transcription factors (e.g., TATA-binding protein, TFIIB, and RNA Pol II) and sequence-specific transcription factors (e.g., CREB, c-Jun, c-Myb, Sap-1a, MyoD, NF-κB, AR, and other nuclear receptors). In the context of prostate cancer, p300/CBP could acetylate AR, prevent AR degradation, or bind with AR on the target promoter/enhancer to increase its transcriptional activity. Furthermore, p300/CBP serves as a coactivator and transcriptional bridge that integrates and transduces the diverse regulatory signals emitted by the various component proteins within the AR transcriptional machinery. Prostate cancer cells cultured in androgen-deprived conditions show increased p300/CBP expression. This is further evidenced in patients with prostate cancer where p300/CBP is found to be upregulated. Taken together, this evidence points to the key role of p300/CBP in oncogenic AR function and its potential as a suitable therapeutic target both in androgen-sensitive prostate cancer and CRPC (5).

The development of small-molecule inhibitors against p300/CBP targeting either its catalytic HAT domain or the single chromatin binding BRD is an active area of drug discovery and development for diverse human diseases, including inflammation, autoimmune disorders, and cancer. Nearly 75 small-molecule inhibitors of p300/CBP have been reported, including natural products, bisubstrate analogues, and synthetic small molecules exemplified by the widely used small-molecule probe C646. However, these inhibitors lack drug-like properties and exhibit weak potency and poor selectivity (6). The first structure-guided inhibitor A-485, targeted to the acetyl mark depositing HAT domain of p300/CBP, decreased H3K27ac and H3K18ac, but not H3K9ac mark, and suppressed proliferation of a number of cancer cell lines, including prostate cancer. The A-485–mediated reduction in H3K27ac and H3K9ac deposition in prostate cancer cells expressing full-length AR or the AR splice variants showed a marked loss in AR transcriptional activity in vitro and in vivo (4). A-485 was also used in combination with PD-L1 inhibitors, where it was found to block p300/CBP recruitment to the CD274 promoter, which in turn blocked exosomal PD-L1 secretion (7). B026 is another HAT-targeted inhibitor with greater efficacy that also severely attenuated CRPC growth (8). p300/CBP inhibitors targeting its acetyl-lysine binding BRD have also shown comparable efficacy in prostate cancer cells. The deep hydrophobic pockets of BRDs form a noncovalent bond with the acetyl-lysine on the surface of the proteins and thus serve as an excellent target for drug development. One such compound, GNE-049, targeting the BRD of p300/CBP, preferentially blocked the growth of AR-positive prostate cancer cells in vitro and in vivo (5). CCS1477 is another p300/CBP BRD-targeting inhibitor currently being evaluated in a phase I trial for the treatment of hematologic malignancies and advanced prostate cancer and is the first of its class to enter a clinical trial stage.

In this issue of Cancer Discovery, Welti and colleagues report the targeting of p300/CBP BRD by CCS1477 as a potential strategy for treating AR-addicted CRPC (9). Here, they demonstrate the antitumor activity of CCS1477—a selective, potent, orally bioavailable p300/CBP BRD inhibitor—in multiple prostate cancer cell lines and CRPC xenograft models. Confirming earlier results, the transcriptomic analysis showed p300/CBP levels to significantly correlate with the expression of AR and androgen-resistance gene signatures in primary and CRPC tissues. Correlative analysis of paired tumor biopsies showed no significant association between p300/CBP protein levels and full-length AR and the splice variant AR-V7. However, patients with lower levels of p300 at primary diagnosis correlated with longer times for transition to CRPC. siRNA-mediated knockdown of p300 and/or CBP strongly interfered with AR-driven transcription of key AR genes, including PSA, and inhibited the growth of AR-V7– expressing CRPC cell lines. Interestingly, maximal growth inhibition was observed by the combined knockdown of p300 and CBP. Similar inhibitory effects were observed when CRPC lines were treated with CCS1477 that binds to the BRDs of CBP and p300 with K_d values 1.7 nmol/L and 1.3 nmol/L, respectively. The most significant growth-inhibitory activity of CCS1477 was observed in AR-proficient cells (IC₅₀ < 100 nmol/L), whereas AR-negative cells were less sensitive (IC₅₀ >1,000 nmol/L). Prostate cancer cells treated with CCS1477 showed significant transcriptional alterations, including the downregulation of AR and MYC signatures and positive enrichment of E2F, G₂–M checkpoint, and mitotic spindle signature. Importantly, CCS1477 treatment led to downregulation of AR-V7 and MYC expression and reduced AR, CBP, and p300 recruitment to key AR target genes (Fig. 1). Investigations into the CBP and p300 nuclear interactome and global recruitment upon CCS1477 treatment would provide more mechanistic insights into the loss of AR-V7 expression, AR signaling, and MYC regulation by this potent inhibitor. CCS1477-induced alterations differed significantly from those caused by the BET BRD protein inhibitor JQ1. Although the number of differentially expressed genes was lower in CCS1477-treated cells, its impact on AR signaling pathways was more profound than JQ1. These findings point to (i) the potency and selectivity of CCS1477 and (ii) the central role of p300 and CBP in regulating AR signaling. The efficacy of CCS1477 was also validated in 22RV1 mouse xenograft and CRPC patient-derived xenograft models where treatment with CCS1477 led to inhibition of AR signaling, remission in tumor volume, and increased survival. CCS1477 is being currently evaluated in a phase I clinical trial. Analysis of plasma PSA levels and pharmacodynamics of AR-FL, AR-V7, MYC, KLK3, CBP, p300, and Ki-67 proteins in biopsies from patients treated with CCS1477 from the dose-escalation phase of the trial showed altered AR signaling, indicating the on-target activity of the compound. However, further investigation is needed to validate these findings.

Overall, the study reported by Welti and colleagues shows how AR coactivator p300/CBP represents a potential therapeutic option in lethal prostate cancer. Inhibition of p300/CBP by CCS1477 not only promotes the turnover of crucial cancer signaling proteins, it also prevents the transcription of critical cancer-driving genes, thus signifying a potent therapeutic approach. However, studying the CBP and p300 nuclear interactome and its role in orchestrating the chromatin landscape in CRPC in response to CCS1477 treatment would provide more mechanistic insights into its role in the restoration of AR signaling and hence the development of a more refined synergistic treatment strategy. Also, as it has become increasingly clear that p300/CBP proteins are versatile transcriptional coactivators that can influence different physiologic processes, studies in a larger cohort of patients will help define any risk associated with CCS1477 clinical use. Although much work still needs to be done, a better understanding of p300/CBP function in the context of lethal prostate cancer is essential for the clinical application of CCS1477 and synergistic combination treatments. Future studies will hopefully shed light on these exciting key questions in this rapidly evolving cancer therapeutics targeting epigenetic regulators.

No disclosures were reported.