Aurora A Kinase Inhibition Is Synthetic Lethal With the Activation of MYCN in Retinoblastoma

Abstract

Purpose

RB1 inactivation and MYCN activation have been documented as common oncogenic alterations in retinoblastoma (RB). Direct targeting of RB1 and MYCN has not yet been proven to be feasible. The current treatment options for RB mainly consist of conventional chemotherapy, which inevitably poses health-threatening side effects. Here, we aimed to screen an in-house compound library to identify potential drugs for the treatment of human RB.

Methods

Aurora A kinase (AURKA) inhibitors were identified by differential viability screening with a tool compound library, and the pharmacological safety and efficacy of candidate drugs were further validated in zebrafish and RB patient-derived xenograft (PDX) models in vivo. Further CUT & Tag assay, ChIP-qPCR and RNA seq performances showed that MYCN binds to the AURKA promoter and upregulates its transcription, suggesting that AURKA inhibition induces synthetic lethality in RB.

Results

In this study, we revealed that AURKA inhibitors exhibited high therapeutic efficacy against RB both in vitro and in vivo. Mechanistically, we found that MYCN could bind to the AURKA promoter region to regulate its transcription, thereby promoting AURKA expression and consequently driving RB progression. Interestingly, AURKA inhibition exhibited synthetic lethality with RB1-deficient and MYCN-amplification in RB cells.

Conclusions

Collectively, these findings demonstrate that AURKA is crucial for RB progression and further expanded the current understanding of synthetic lethal therapeutic strategies. Our study indicates that AURKA inhibitors may represent a new therapeutic strategy for selectively targeting patients with RB with RB1-deficient and MYCN-amplification to improve the prognosis of aggressive types of patients with RB.

Retinoblastoma (RB) is the most common pediatric intraocular malignancy, has a familial genetic tendency, and accounts for 3% of malignancies in infants.¹ Treatment modalities for RB include local therapies, chemotherapy, and even enucleation.² It is universally acknowledged that chemotherapy is still the main choice for managing RB. Chemotherapy options include intravenous chemotherapy (IVC), intra-arterial chemotherapy (IAC), periocular chemotherapy (POC), and intravitreal chemotherapy (IVitC).^3,4 Current chemotherapy drugs are effective for intraocular RB control.^2,5,6 However, exposure to conservative chemotherapy drugs, including carboplatin, etoposide, vincristine, and melphalan, is associated with pharmaceutical toxicity and side effects.^7,8 Therefore, the exploitation of novel therapeutic targets and the development of new compounds are worthy of increased attention.

In 1971, Knudson proposed the two-hit hypothesis, the classic genetic basis of RB, and suggested that RB is initiated by biallelic RB1 inactivation.⁹ Notably, a minority of nonhereditary RB (<2%) can be driven by amplification of MYCN without RB1 deletion.^10,11 It is universally acknowledged that MYCN is one of the most common proto-oncogenes in human tumors and encodes a basic helix-loop-helix (bHLH) transcription factor. Moreover, it is also particularly overexpressed in high-risk tumor subgroups, especially in several childhood tumors, including RB, neuroblastoma, and medulloblastoma.^12,13 Therefore, MYCN also plays a pivotal role in the tumorigenesis and development of RB.¹⁴ In a retrospective study of a cohort of 1068 patients with unilateral nonfamilial RB, 2.7% of patients had no detectable mutation in RB1, and approximately 50% of these patients had high-level MYCN amplification and activation.³ Besides, previously study has revealed that the emergence of MYCN independence in an initially MYCN-driven RB.⁵ RB1 inactivation and MYCN activation play pivotal roles in RB tumorigenesis, making them important targets for treatment.¹⁵ However, direct targeting of RB1 and MYCN has yet to be proven feasible. Moreover, effective therapeutic treatments targeting RB1 and MYCN require further exploration.^11,16 Unfortunately, the difficulties in targeting RB1 and MYCN with drugs have limited pharmacological breakthroughs in the targeted treatment of RB.^15–17

A recent study revealed that in RB1-deficient tumor cells, the activity of Aurora A kinase (AURKA; a kinase that promotes cell cycle progression) might be upregulated enough to overcome the hyperactivation of the spindle assembly checkpoint (SAC) during mitosis.¹⁸ In addition, AURKA inhibition exacerbated the imbalance of microtubule dynamics and effectively interfered with mitosis, which promoted tumor cell death.¹⁹ This could represent a disguised blessing for patients with RB1-deficient tumors, as AURKA inhibition could induce synthetic lethality, presenting a potential therapeutic opportunity.^18,19 A series of AURKA inhibitors (including alisertib, ENMD-2076, and LY3295668) are being assessed in clinical trials based on synthetic lethality with RB1 deficiency.^18,20 However, the mechanism of the function of AURKA inhibitors in RB remains an enigma. One study revealed that AURKA inhibitors could destabilize MYCN expression, thereby promoting its degradation.²¹ Thus, AURKA inhibition appears to kill RB cells by triggering synthetic lethality with RB1 and MYCN.

Our study reveals that AURKA inhibition effectively induces RB tumor cells’ death by triggering synthetic lethality with RB1 and MYCN. Mechanistically, we found that MYCN binds to the AURKA promoter and upregulates its transcription, then consequently driving RB progression. Overall, AURKA inhibition could be a potential targeted therapy for patients with RB1-deficient tumors and may improve the treatment response rates for MYCN-amplification aggressive types of patients with RB.

Methods

Cell Culture

The human Y79 and HEK293T cell lines were obtained from ATCC, and the human WERI-Rb1 and ARPE-19 cell lines were obtained from the Cell Bank/Stem Cell Bank (Chinese Academy of Sciences). The Y79, WERI-Rb1, and ARPE-19 cells were cultured in RPMI-1640 medium (Gibco). HEK293T cells were cultured in Dulbecco's modified Eagle's medium (Gibco). All media were supplemented with 10% FBS (Gibco), and the cells were incubated at 37°C with 5% CO₂.

Drug Screening

Using an automated platform, we tested an in-house compound library with a total of 3075 compounds in 2 RB cell lines (Y79 and WERI-Rb1). The detailed description of these compounds are listed in Supplementary Table S1. We planted 3000 cells/well in a 384-well plate and treated cells with individual compounds at a concentration of 2 µM before evaluation of cell viability after 72 hours (2 doubling times) using a CellTiter-Glo Luminescent assay (Promega). The log₂ fold change (compound/DMSO) in cell viability for each compound was calculated, and a value below −2 was defined as sensitive.

Cell Viability

To determine cell viability, cells were seeded in 96-well plates at a density of 5000 cells per well. After incubation with 10 µL of CCK-8 reagent (Dojindo) per well, the absorbance was measured at a wavelength of 450 nm at the indicated time points. The data were recorded and analyzed. The results were presented as the mean ± SEM. Cell viability was measured after 72 hours using a CCK-8 assay (Dojindo). The half maximal inhibitory concentration (IC₅₀) values of the candidate compounds were calculated using GraphPad version 9.0 software.

Western Blot Analysis

Western blot was performed as previously described.²² Briefly, the nitrocellulose (NC) membranes (Millipore) with the transferred proteins were incubated with primary antibody at 4°C overnight after blocking with 5% milk for 1 hour at room temperature. The membranes were then incubated with secondary antibodies for 1 hour at room temperature. The antibodies used for Western blot analysis were against MYCN (CST, 84406S, 1:1000), AURKA (CST, 14475T, 1:1000), and GAPDH (Proteintech, 60004-1-IG, 1:10000). The immunoblots were visualized with an Odyssey infrared imaging system (LI-COR).

Lentivirus Packaging and Generation of Stable Cell Lines

For lentivirus production, HEK293T cells were cotransfected with the indicated pLKO.1-derived lentiviral vectors (System Biosciences) and the packaging vectors psPAX2 and pMD2 G using Lipofectamine 3000 reagent (Invitrogen) and incubated with Opti-MEM I Reduced Serum Medium (Gibco). The lentivirus-containing supernatant was collected at 48 hours and 72 hours and filtered with 0.45 µm PVDF filters following concentration with a Lenti-X Concentrator (TaKaRa). For stable cell line construction, WERI-Rb1 or Y79 cells were infected with the concentrated lentivirus. Stable cell lines were selected with 2 mg/mL puromycin (InvivoGen) for 2 weeks and then cultured in 1 mg/mL puromycin.

SiRNA Assay

AURKA siRNA oligonucleotides (siAURKA-1: 5′‐CCUGUCUUACUGUCAUUCGAATT‐3′, siAURKA-2: 5′‐AGGCCACUGAAUAACACCCAATT‐3′, and siAURKA-3: 5′‐GAGUCUACCUAAUUCUGGAAUTT‐3′), and control siRNA (siNC: 5′‐TCCACTACCGTTGT TATAGGTG‐3′) were designed based on published synthetic siRNA sequences (Biotend Co., Ltd.). We transfected RB cells with the plasmid and siRNAs using Lipofectamine 3000 (Invitrogen) according to the manufacturer's instructions.

ShRNA Assay

The shRNA sequences targeting MYCN were cloned and inserted into the pLKO.1-puro vector (Addgene). The sequences used were as follows: MYCN, shRNA-1: CCGGGCCAGTATTAGACTGGAAGTTCTCGAGAACTTCCAGTCTAATACTGGCTTTTTGAATTC and shRNA-2: CCGGCTGAGCGATTCAGATGATGAACTCGAGTTCATCATCTGAATCGCTCAGTTTTTGAATTC.²³

Chromatin Immunoprecipitation–Quantitative RT-PCR

The chromatin immunoprecipitation (ChIP) assay was performed using the EZ-Magna ChIP A/G kit (Millipore) according to the manufacturer's instructions. Y79 and WERI-Rb1 cells (1 × 10⁷) were cross-linked and pelleted before resuspension in ChIP lysis buffer. The cellular DNA was then sonicated to lengths between 500 and 1000 bp and subjected to immunoprecipitation. The DNA was analyzed with quantitative RT-PCR after incubation with an antibody against MYCN (CST, 84406S) or normal rabbit IgG (CST, 2729S) overnight and with purification. Accordingly, the primers for the AURKA promoter region were as follows: AURKA-1 forward, 5′-TGGAAAAATGGAGGCGCGAA-3′; reverse, 5′-ATGGCGAGAAAAGCAAGAGAGA-3′; AURKA-2 forward, 5′-ACCGATAAATGGCCGACCG-3′; and reverse, 5′-CCCTAAGAACCCGGAAGTGG-3′.

Evaluation of Drug Toxicity

To evaluate the toxicity of compounds, we used zebrafish embryos 3 days post fertilization (3 dpf) in 1 mL egg water containing Alisetib and TCS7010. Thirty zebrafish embryos were placed in 24-well plates per well. The maximum tolerable dose was determined to be 10 µM in 2-fold dilution of each compound. The embryos were incubated at 28°C. If the survival rate of zebrafish embryos is higher than 80% after 6 days, the compound is considered non-toxic. All experimental procedures were approved by the Institutional Animal Care and Use Committee of the Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine.

In Vivo Animal Model Experiments

All experimental procedures were approved by the Institutional Animal Care and Use Committee of the Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. Four- to 5-week-old male BALB/c nude mice were obtained from the Model Animal Research Center of Shanghai Jiao Tong University School of Medicine. For the patient-derived xenograft PDX models, 4 × 10⁵ cells were injected into the right retinas of the mice to establish stable orthotopic tumors.

High Throughput Cleavage Under Targets and Tagmentation

The cleavage under targets and tagmentation (CUT & Tag) assay was performed as described previously²⁴ with modifications by Jiayin Biotechnology Ltd. (Shanghai, China). Y79 cells were harvested, counted, and centrifuged at 800 × g for 5 minutes at room temperature (RT). Briefly, 1 × 10⁵ cells were washed twice gently with wash buffer. Then, 10 µL concanavalin A coated magnetic beads (Bangs Laboratories) were added per sample and incubated at RT for 10 minutes. The removed unbound supernatant and resuspended bead-bound cells with dig wash buffer and a 1:50 dilution of primary antibody (MYCN, CST, and 84406S) or control IgG were incubated on a rotating platform overnight at 4°C. The primary antibody was removed using a magnet stand. Secondary antibody (Goat Anti-Rabbit IgG H&L, Abcam, ab6702) was diluted 1:100 in dig wash buffer and cells were incubated at RT for 1 hour. The cells were washed using the magnet stand three times in dig wash buffer. A 1:100 dilution of pA-Tn5 adapter complex was prepared in dig-med buffer and incubated with cells at RT for 1 hour. Finally, they were resuspended in tagmentation buffer, and DNA fragments were purified by phenol-chloroform-isoamyl alcohol extraction and ethanol precipitation. Sequencing was performed in the Illumina Novaseq 6000 using 150 bp paired-end following the manufacturer's instructions. Pathway analysis was used to find out the significant pathway of the genes according to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. We turned to the Fisher's exact test to select the significant pathway, and the threshold of significance was defined by P value and false discovery rate (FDR).

Bioinformatic and Statistical Analysis

We queried The Cancer Genome Atlas (TCGA) database from R2 (http://hgserver1.amc.nl), which provided tumor transcriptional data and follow-up information, to validate the prognostic value of MYCN and AURKA and the expression correlation of AURKA with MYCN and AURKA with RB1 in 76 patients with RB. All the experimental data were presented as the mean ± SEM. For statistical analysis, we used GraphPad Prism version 9.0 software. Differences between two groups were analyzed by 2-tailed Student's t-test, whereas differences among multiple groups were analyzed by 2-way analysis of variance (ANOVA). A P value < 0.05 was considered to indicate statistical significance, and asterisks denote statistical significance (*P < 0.05, **P < 0.01, and ***P < 0.001).

Data Availability

Single-cell RNA-seq data supporting the findings of this study was deposited in Gene Expression Omnibus (GEO; PRJNA737188).²⁵ MYCN cleavage under targets and tagmentation (CUT & Tag) data were obtained from the GEO database (PRJNA1125487). All relevant data are available from the authors upon request.

Results

AURKA Inhibitors Selectively Inhibit RB

To discover RB-specific compounds, the proliferation ratio of RB cell lines (Y79 and WERI-Rb1) was examined by screening after treatment with a chemical library composed of 3075 known target compounds (Fig. 1A). By overlapping the compounds with tumor cell killing efficiency >85% in both Y79 cells and WERI-Rb1 cells, 32 candidate compounds were identified, among which HDAC inhibitors, ATPase inhibitors, and AURKA inhibitors were the top 3 drug types (Figs. 1B, 1C). Notably, AURKA inhibitors were on the top list of drugs that efficiently kill RB cells, with a log2 fold change (log2 FC; compound/DMSO) below −2 (Fig. 1D). Moreover, in the TCGA-RB cohort, we detected a significant negative correlation between AURKA expression and RB1 expression (R = −0.266, P = 0.020; Fig. 1E). These results suggested that AURKA inhibitors may be a promising approach for effectively inhibiting RB progression.

Figure 1.

Identification of AURKA inhibitors. (A) Schematic diagram of high-throughput drug screening. (B) Venn diagram showing 32 candidate compounds. (C) Bar chart showing the top three drug types: HDAC inhibitors, ATPase inhibitors, and AURKA inhibitors. (D) Overview of the potency of all compounds represented by cell viability. The red dots indicate the positions of the AURKA inhibitors as effective in killing RB cells. (E) Correlation analysis of the expression of AURKA and RB1 (R = −0.266, P = 0.020) in human RB using RNA sequencing data from Dorsman-76-RMA-u133p2 (GSE59983).

AURKA Contributes to the Progression of RB

To explore the potential role of AURKA in the regulation of tumor progression, we performed scRNA-seq, which showed that upregulated genes in RB tumor cells were markedly enriched in the cell cycle pathway (Fig. 2A). Intriguingly, AURKA was the unique gene among the functional target genes of 32 candidate compounds by taking the intersection of the genes related to the cell cycle pathway through scRNA-seq (Fig. 2B). As expected, the bulk RNA-seq chromatograms of clinical samples (GSE111168) showed that AURKA were markedly high expression in RB than in the normal retina (Figs. 2C, 2D). Similarly, in the TCGA-RB cohort, AURKA was more highly expressed in RB tissues than in normal retina tissues (Supplementary Fig. S1). Moreover, AURKA exhibited significant upregulation in RB tumor cells versus cone precursors (Figs. 2E–I), as our previous research has already confirmed the cone precursors were the cells of origin of RB that were required for initiating the differentiation and malignancy process by single-cell RNA sequencing.^25,26 Taken together, these results underscored that AURKA might play a predominant role in the malignant transformation of RB.

Figure 2.

AURKA is upregulated in RB. (A) The violin plot indicated that the cell cycle genes upregulated in RB cells. (B) Venn diagram and scRNA-seq analysis showed AURKA was the unique gene among the functional target genes of 32 candidate compounds. (C) IGV tracks for AURKA based on RNA-seq data in healthy retina and RB tissues. (D) Volcano plot showing 8575 upregulated genes and 5064 downregulated genes in RB tissues compared with healthy retinal tissues. (E) Dot plot showing 180 upregulated genes and 27 downregulated genes in RB cells. (F–H) The t-SNE plot of single cells and trajectory analysis revealing that AURKA is upregulated in RB cells. (I) Violin plot for the expression of AURKA in RB cells compared with cone precursors. ***P < 0.001.

AURKA Inhibitors Suppress Tumor Growth of RB

We first examined the expression of AURKA in RB cells. As expected, both Y79 and WERI-Rb1 cells exhibited significantly increased AURKA expression (Fig. 3A). To assess the effect of AURKA on RB cell proliferation, we first silenced AURKA by transfecting cells with three individual siRNAs (Fig. 3B) and performed CCK8 assay. Notably, the cell growth assays revealed a significantly decreased proliferation rate compared to that of the control group (Fig. 3C).

Figure 3.

AURKA inhibition modulates tumor growth in vitro and in vivo. (A) Western blot assays of AURKA expression in the human retina cell line ARPE-19 and RB cell lines (Y79 and WERI-Rb1). (B) AURKA knockdown by three siRNAs in Y79 cell line. (C) CCK-8 assay showing tumor cell growth after AURKA silenced. Cell growth was obviously restrained at day 3 in AURKA-silenced RB cells. The absorbance values were detected at 24 hours, 48 hours, 72 hours, and 96 hours. (D) The statistical results of AURKA inhibitors’ safety test in zebrafish embryos (n = 30 for each group). (E) Western blot assays of MYCN and AURKA expression in the three samples used in RB PDX models establishment. (F) Representative photographs of eyeballs from untreated and alisertib-treated mice in PDX models. (G) Representative eyeballs and tumor weights reflecting the effect of alisertib on the growth of RB tumors in each PDX cohort.

We then choose Aurora A Inhibitor I (TCS7010) and alisertib to explore the role of AURKA inhibition in RB progression in vivo. AURKA inhibitors were validated as safe agents using zebrafish embryos (Fig. 3D). To further evaluate the efficacy of AURKA inhibitors in vivo, we established an RB PDX model in which the enucleated tumor tissues were cut into pieces and transplanted, after which the BALB/c mice were subjected to intraocular injection (Supplementary Fig. S2A). Between November 2022 and April 2023, there were 3 patients with RB who were diagnosed with group E advanced RB with endophytic tumor growth at our center. The detailed clinical characteristics are described in the Table. MYCN and AURKA were significantly upregulated in the three clinical samples utilized for establishing RB PDX models, compared to the normal retina tissue (Supplementary Fig. S2B). Intriguingly, the representative photographs of eyeballs revealed that intraocular xenograft-bearing mice treated with alisertib showed significant tumor regression (Fig. 3E). Concordantly, the images and statistical analysis showed that the tumor weight following alisertib treatment was more than twice as small as those of the control group (Fig. 3F). These findings suggested that AURKA inhibitors have exciting pharmacological efficacy against RB cells both in vitro and in vivo.

Table.

Clinical Details of Patient-Derived RB PDX Models

Sample	PDX Models	Age at Diagnosis, Mo	Age at Enucleation, Mo	Laterality	Chemotherapy
Patient 01	PDX01	3	3	Bilateral	No
Patient 02	PDX02	25	29	Unilateral	Yes
Patient 03	PDX03	14	14	Unilateral	No

AURKA Inhibition Induces Synthetic Lethality in MYCN-Amplification RB

To explore the mechanism by which AURKA inhibitors inhibit RB tumorigenesis, we performed the MYCN CUT & Tag assay in RB cells. Notably, the KEGG analysis of these direct target genes of MYCN revealed enrichment of pathways in the cell cycle (Fig. 4A). Additionally, we found that MYCN binding peaks were clearly enriched at the promoter region of AURKA (Fig. 4B). Similarly, MYCN was found significantly enriched to AURKA promoters in RB, as demonstrated by ChIP‒qPCR (Fig. 4C). Moreover, we found a significant positive correlation between AURKA and MYCN expression in a cohort of 76 RB samples (R = 0.390, P = 5.00e-04; Fig. 4D). Instead, once silenced MYCN expression by transfecting cells with two individual shRNAs, AURKA expression was significantly decreased at the protein level (Fig. 4E). Following this route, we then used HDAC inhibitor (quisinostat), PI3K inhibitor (fimepinostat), and selective AURKA inhibitors (alisertib and TCS7010) to evaluate the cytotoxic effects on MYCN-deficient RB cell lines (see Fig. 4F). These inhibitors exhibited exciting pharmacological efficacy in RB cell lines (see Supplementary Fig. S2C). However, when MYCN was depleted by two shRNAs in RB cells, the antiproliferative effect of AURKA inhibitors was significantly attenuated (see Figs. 4G, 4H). To further examine the significant function of MYCN amplification in synthetic lethality triggered by AURKA inhibition, we exogenously overexpressed MYCN. The overexpression of MYCN in Y79 cells was confirmed by Western blot assays (Supplementary Fig. S2D). This finding aligned perfectly with previous results indicating that MYCN overexpressing Y79 cells were more susceptible to AURKA inhibitors (see Figs. 4G, 4H). In contrast, the sensitivity of both MYCN-deficient and MYCN-overexpressing RB cells to HDAC inhibitors and PI3K inhibitors remained unaffected (Supplementary Fig. S2E). These results obviously demonstrated that the pharmacological efficacy of AURKA inhibitors might depend on MYCN amplification. In conclusion, we validated that MYCN-amplified RB cells exhibited more sensitive to AURKA inhibition, verifying the screening results.

Figure 4.

AURKA inhibition induces synthetic lethality in retinoblastoma. (A) KEGG pathway analysis of genes regulated by the transcription factor MYCN in RB cells. (B) CUT & Tag analysis of MYCN binding sites in the AURKA gene in RB cells. (C) ChIP-qPCR showed MYCN was enriched in the AURKA promoter in RB cell lines. The data are presented as the mean ± SEM of triplicate experiments. ***P < 0.001. (D) Correlation analysis of the expression of AURKA and MYCN (R = 0.390, P = 5.00e–04) in retinoblastoma using RNA sequencing data from Dorsman-76-RMA-u133p2 (GSE59983). (E) Western blot assays of AURKA and MYCN expression in Y79 cells after silencing MYCN with shRNAs. (F) Stepwise schematic of the pharmaceutically effective compounds and potential target screening. (G, H) Representative dose–response curves of MYCN-deficient and MYCN overexpressed Y79 cells to AURKA inhibitors (alisertib and TCS7010). The data are presented as the mean ± SEM of triplicate experiments.

Discussion

In this study, by combining high-throughput drug screening and single-cell RNA sequencing, we revealed potential therapeutic targets for RB. Further mechanistic studies revealed that the AURKA inhibitors induced synthetic lethality via RB1 inactivation and MYCN activation. Notably, we constructed a promising therapeutic strategy using AURKA as a target and conducted effectiveness and safety assessments in vivo and in vitro (Fig. 5).

Figure 5.

Schematic diagram of this study. (A) AURKA is specifically upregulated in RB and promotes RB cell proliferation and tumor progression. (B) AURKA inhibition demonstrated significant antitumor efficacy in RB cells with both RB1 mutation and MYCN amplification owing to synthetic lethality.

In recent decades, RB1 and MYCN have been commonly referred to as undruggable targets, which is a key challenge in RB treatment.^16,17 Currently, there is still a lack of effective pharmaceutical drugs for targeting classical genetic mutations in RB. Synthetic lethality, a phenomenon in which simultaneous alterations in two genes induce cell death, is a novel therapeutic strategy.^27,28 Moreover, poly (ADP-ribose) polymerase inhibitors (PARPis), the first clinically approved drugs designed to exploit synthetic lethality, have shown strong efficacy in treating tumors with either BRCA1 or BRCA2 mutation.^29,30 The discovery of PARPis provided interesting guidance for the development of other targets and therapies.^29,31 In addition, synthetic lethality approaches involving targeting AURKA and other cancer therapeutic targets and pharmacological combination strategies have shown varying degrees of success.³² Notably, AURKA inhibitors provided a promising therapeutic strategy for targeting RB, as AURKA inhibition was previously reported to be synthetically lethal in RB-mutant cancer cells.^18,19 Intriguingly, we further reveal that MYCN amplification also exhibits synthetic lethality with AURKA inhibition.

AURKA is a member of the aurora kinase family and plays a pivotal role in mitotic processes, DNA damage checkpoint recovery, and centrosome and spindle maturation.³³ The overexpression of AURKA has been observed in a variety of tumors, including small cell lung cancer, breast cancer, and glioblastoma.^34–36 Alterations in AURKA plays an integral role in multiple cancer hallmarks, such as proliferation, invasion and migration, metabolism, immune escape, and drug resistance, which are involved in cancer progression and metastasis.²⁷ Alisertib (MLN8237), a potent and selective AURKA inhibitor, has been exploited as a potential treatment for patients with advanced solid tumors. A series of randomized clinical trials demonstrated that alisertib showed considerable clinical activity and a tolerable safety profile in patients with resistant metastatic breast cancer and small cell lung cancer.^18,37,38 In addition, alisertib is active in neuroblastoma due to disruption of the AURKA/MYCN complex, resulting in the inhibition of MYCN-dependent transcription.^39,40 Based on the preclinical antitumor activity in models and adults, a phase I/II trial of alisertib in children with recurrent and refractory solid tumors evaluated its toxicity and efficacy.^41,42 However, there are no clinical trials of AURKA inhibitors in patients with RB. Our next task is to expand the RB PDX cohort and improve preclinical trials. It is also necessary to accelerate the clinical assessment of AURKA inhibitors to benefit patients.

Considering that AURKA inhibitors have exhibited promising efficacy in preclinical and clinical studies of RB1-deficient tumors, more attention should be given to the clinical application of high-risk RB treatment. It is meaningful to investigate the application of AURKA inhibitors alone or in combination with chemotherapy as a potential treatment for patients with RB. Further studies are required to validate the profile of AURKA occupancy along the whole genome for exploring the role of more AURKA-targeting marks in tumorigenesis.

In summary, for the first time, we revealed a promising synthetic lethality strategy involving AURKA inhibitors in RB. This study reveals a promising drug target for high-risk RB and a novel clinical application for AURKA inhibitors.