Alternative Splicing, E7107 Compound Could Aid in Prostate Cancer Pathogenesis and Drug Development


In prostate cancer, investigators have revealed a link between disease progression and the severity of splicing dysregulation, as well as establishing intron retention as an indicator of prostate cancer stemness and aggressiveness.

In prostate cancer, investigators have revealed a link between disease progression and the severity of splicing dysregulation, as well as establishing intron retention (IR) as an indicator of prostate cancer stemness and aggressiveness, according to findings published in Nature Communications.1

These findings, the authors concluded, have the potential to lead to treatment that targets the abnormal RNA splicing in these advanced cancers and could also distinguish prostate tumors that are more aggressive from slow-growing disease.

Prevalent genomic copy number variations (CNVs) were revealed through a systematic interrogation of 274 splicing-regulatory genes (SRGs). During prostate cancer development and progression, this led to the mis-expression of approximately 68% of SRGs. Moreover, the use of the spliceosome modulator E7107 appeared to have reversed cancer aggressiveness and was shown to impede castration-resistant prostate cancer (CRPC).

“We found, surprisingly, that when prostate cancer cells become more aggressive, they have more genes that retain the introns in their messenger RNA or mRNA,” Song Liu, MD, director of Roswell Park’s Department of Biostatistics and Bioinformatics, stated in a press release.2 “We have observed this retention and also noted that retention of introns is more pronounced in advanced tumors.”

Two AS mapping algorithms, rMATS v4.0 and SUPPA v2.2.0, were used to assess the differential splicing landscape embedded in RNA-seq data. An AR activity score was established using Z-scores, calculated from the expression of 20 experimentally validated AR targets as a base. This helped to decode the AS events regulated by AR signaling in samples. These AR targets included KLK3, KLK2, TMPRSS2, ELL2, CENPN, GNMT, MAF, NNMT, MED28, EAF2, MPHOSPH9, PTGER4, HERC3, ZBTB10, ACSL3, FKBP5, C1orf116, NKX3-1, ABCC4, and PMEPA1. For each gene identified, patients were stratified into 2 cohorts consisting of relatively high and low gene expression. From there a cutoff was chosen based on which value yields the lowest P value.

When comparing either bulk or paired tumors and normal tissues analyses, it was found that primary prostate cancer possessed more AS events (~1.9-fold by rMATS; ~1.7-fold by SUPPA). This was accompanied by preferential increases in alternative 3’ (A3) and alternative 5’ (A5), and IR. Following androgen deprivation therapy (ADT) or neoadjuvant hormone therapy, prostate cancer was also shown to increase differentially spliced events (DSEs) while mCRPC showed an increase in DSEs accompanied by noticeable increases in A3, A5, SE, and IR. CRPC-neuroendocrine disease (CRPC-NE) was noted as harboring a distinct splicing landscape relative to CRPC-adenocarcinomas (CRPC-Ad). Additionally, rMATS-based AS mapping with a false discovery rate (FDR) <0.05 showed identical results to those using FDR <0.1.

Investigators found distinct AS profiles for basal versus luminal cells, with basal cells having notably more IR. A comparative gene set enrichment analysis (GSEA) was performed and researchers found that prostate cancer with aggressive phenotypes generally had a global basal-like AS profile. A high number of DSEs were observed in both LNCaP and AR cells that contained RB1 and TP53 knockdown. The analysis also indicated a deficiency in RB1 and TP53 within the AS signatures of LNCaP and AR cells, though it was also noted that both were significantly enriched in mCRPC compared with primary prostate cancer.

A greater than 18-fold increase was observed by investigators in IR in primary prostate cancer versus normal. Researchers consistently observed a preferential upregulation of IR with therapy-resistant, aggressive, and metastatic prostate cancer. Similarly, IR upregulation was seen in prostate tumors and epithelial cells with low versus high AR activity.

The IR-affected genes with a high-confidence set of human nonsense-mediated (NMD) RNA decay targeted approximately 10% of genes in all targeted groups. The average delta-PSI value for upregulated IR were found to be 0.143 in primary prostate cancer, 0.18 in CRPC-Ad, and 0.15 in CRPC-NE.

Investigators compared splicing-affected genes (SAGs) with differentially expressed genes

(DEGs)—identified in the AR-low versus AR-high genes—in order to identify the impact of AR-associated splicing on AR-regulated gene expression. Only 2% of SAGs overlapped with the DEGs in primary prostate cancer. This overlap increased to 23% in CRPC-Ad.

Researchers consulted 8 available large-scale clinical datasets in cBioportal48, which showed the mutational landscape of the top 15 altered SRGs in primary prostate cancer and CRPC. It was found that frequently deleted and amplified genes often coincided with the deletion of tumor suppressor (TS) genes and oncogene amplification. Most SRGs were mutated at low frequency; of the 274 SRGs that mutated at a rate of 5% or greater, 20 (7.3%) were noted in The Cancer Genome Atlas–prostate cancer cohort and 29 (10.6%) were from the Stand Up to Cancer CRPC cohort.

Chromosomal distribution of mutated SRGs (5% or greater) indicate that SRGs represent a frequently mutated pathway in prostate cancer. Thirty-one percent to 68% of patients with primary prostate cancer and 87% to 94% of patients with CRPC harbor at least 1 mutation of 1 SRG.

High or low tumor expressions were seen as an unfavorable signature in distinct splicing landscapes, with total DSEs (1.73-fold) and IR (18.91-fold) being upregulated in the high patient population. Additionally, researchers also found that AR-positive disease tended to have more SRG deletions, whereas AR-native disease harbored more SRG amplifications.

Prostate cancer cells exhibited preferential sensitivity to E7107 relative to nontumorigenic prostate epithelial cells RWPE1; PC3 was also noted as being more sensitive than LNCaP cells. A long-term E7107 treatment, of 6 to 7 days, induced massive cell death. Treatment of PCa cells

with 5 nanometer (nM) E7107 for 6 hours notably reshaped selected genes’ splicing pattern. E7107 was shown to hinder the AS globally in both cell types, though SE was found to be affected the most.

At the gene expression level, E7107 reshaped the transcriptomes and exhibited a slight suppressive effect on transcription, as well as activated p53. Transcript levels of AR were also upregulated in PC3 cells.

Additionally, investigators observed significant inhibition when treating Myc-driven murine prostate cancer (Hi-Myc tumors) with E7107. The majority HiMyc cases were shown to contain reduced tumor areas along with prominent benign and hyperplasic glands. These benign/hyperplastic glands in animals treated with E7107 expressed AR and MYC levels that resembled those found in the vehicle-treated tumors. Benign/hyperplastic glands in E7107-treated prostates additionally showed a reduction in heterogeneous Ki-67-positive cells compared with the vehicle-treated Hi-Myc tumors.

“Future characterizations of the origins and consequences of aberrant splicing in aggressive [prostate cancer] could enhance our understanding of disease pathogenesis and aid innovative drug development,” concluded the authors of the study.


  1. Zhang D, Hu Q, Liu X, et al. Intron retention is a hallmark and spliceosome represents a therapeutic vulnerability in aggressive prostate cancer. Nat Comm. 2020;11(2089):1-19. doi: 10.1038/s41467-020-15815-7
  2. Aggressive prostate cancer marked by abnormal RNA splicing, study reveals. News release. Roswell Park Comprehensive Cancer Center. Published April 29, 2020. Accessed May 05, 2020.
Related Videos
Neal Shore, MD, FACS, discusses current and ongoing research with radioligand therapy in patients with nonmetastatic castration-sensitive prostate cancer.
Jun Gong, MD
Neal Shore, MD, FACS
Neal Shore, MD, FACS,
Samuel U. Takvorian, MD, MS
Jun Gong, MD
Janaki Neela Sharma, MD