Elevated CircYthdc2 expression is correlated with aggressive features and poor progression-free survival in hepatocellular carcinoma

Abstract

Objective

Liver hepatocellular carcinoma (LIHC) is a major contributor to cancer-related mortality. While early detection is crucial for improving outcomes, current diagnostic methods lack optimal sensitivity and specificity. Circular RNAs (circRNAs) have emerged as promising biomarkers due to their stability and tissue-specific expression.

Methods

Bioinformatic analysis was performed using UALCAN database to examine YTHDC2 methylation and expression patterns. Tissue and serum samples were collected from 72 LIHC patients and matched controls. CircYTHDC2 expression was assessed in tissues, serum, and cell lines via qRT-PCR. CircYTHDC2’s diagnostic potential was evaluated through ROC analysis and stability testing. Associations between circYTHDC2 levels, clinicopathological features, and survival were analyzed.

Results

Bioinformatic analysis revealed reduced YTHDC2 promoter methylation and elevated expression in LIHC tissues. CircYTHDC2 showed significant upregulation in LIHC tissues, serum, and cell lines compared to controls. ROC analysis demonstrated high diagnostic accuracy for circYTHDC2 in tissues (AUC = 0.846) and serum (AUC = 0.788). CircYTHDC2 exhibited remarkable stability against RNase degradation. Elevated circYTHDC2 levels significantly correlated with advanced BCLC stage, larger tumor size, intrahepatic metastasis, and portal invasion. High circYTHDC2 expression was associated with shorter progression-free survival (P = 0.025). Additionally, we found circYTHDC2 bound to YTHDC2 and was positively regulated by YTHDC2 in an m6A-dependent manner.

Conclusion

CircYTHDC2 represents a stable, clinically viable biomarker for LIHC, demonstrating significant diagnostic accuracy and prognostic value. Its strong correlation with aggressive tumor features and survival outcomes suggests potential utility in clinical management of LIHC patients.

Introduction

Liver hepatocellular carcinoma (LIHC) is one of the most prevalent and aggressive malignancies and has the fourth highest cancer-related mortality rate worldwide. Fewer than 16% of LIHC patients survive for more than 5 years [1]. In Asia, the burden of LIHC is especially heavy, accounting for 72.5% of the global incidence and 73.3% of the global deaths associated with LIHC in 2020 [2]. Early detection and accurate prognostic evaluation remain crucial challenges in improving patient outcomes. While conventional diagnostic methods like computed tomography and magnetic resonance imaging are commonly used, their sensitivity and specificity are suboptimal, despite the improvement by using deep learning methods [3], highlighting the need for reliable diagnostic biomarkers [4].

Circular RNAs (circRNAs) represent a unique class of RNA molecules formed by backsplicing of exons, characterized by their stable circular structure and tissue-specific expression patterns [5]. Their high stability and resistance to degradation make them particularly attractive as potential biomarkers. In LIHC specifically, recent studies have demonstrated increased expression of various circRNAs in tumor tissues compared with adjacent nontumor tissues [6]. These expression changes have been associated with tumor occurrence, pathological grade, and treatment response, suggesting a significant role in LIHC pathogenesis.

The YT521-B homology domain containing 2 (YTHDC2) protein belongs to the YT521-B nuclear reader family and plays a crucial role in RNA metabolism through N6-methyladenosine (m6A) modification interpretation [7]. This post-transcriptional modification has emerged as a critical regulator of various cellular processes, including cell growth, development, and differentiation [8].

Previous studies have implicated YTHDC2 in various cancers, including pancreatic adenocarcinoma [9] and liver cancer [10], suggesting its potential involvement in tumor development and progression.

CircYTHDC2, derived from exons 13–18 of the YTHDC2 gene, represents a highly conserved circular RNA with a 634-nucleotide sequence [11]. Its expression has been documented across various species, from fish to mammals, indicating its evolutionary significance. Recent research has shown that circYTHDC2 can influence cellular processes through m6A modification [11], suggesting a potential regulatory role in cancer development. Notably, the interaction between circYTHDC2 and m6A modification machinery may represent a novel mechanism in cancer progression [12].

Despite these advances in understanding circRNA biology, there remains a significant gap in our knowledge regarding the role of circYTHDC2 in LIHC pathogenesis and its potential utility as a biomarker. The stability of circular RNAs, combined with preliminary evidence suggesting YTHDC2’s involvement in liver cancer, makes circYTHDC2 a promising candidate for investigation. Recent studies have highlighted the importance of m6A-related molecules in cancer development [13], yet the specific contribution of circYTHDC2 to LIHC progression remains unexplored.

Therefore, this study aimed to evaluate the expression patterns of circYTHDC2 in LIHC tissues, serum, and cell lines, and to assess its potential as a diagnostic and prognostic biomarker. Furthermore, we investigated the relationship between circYTHDC2 expression and various clinicopathological features to better understand its role in LIHC progression.

Materials and methods

Bioinformatic analysis

The promoter methylation level of YTHDC2 in LIHC samples was analyzed using the UALCAN database (https://ualcan.path.uab.edu), comparing 377 LIHC tissues with 50 normal tissues. The expression level of YTHDC2 in LIHC was also analyzed using the UALCAN database, comparing 371 LIHC tissues with 50 normal tissues.

Sample collection

Tissue and serum samples were collected from 72 patients with LIHC who underwent surgery between November 2020 and April 2023 at The First Affiliated Hospital of Nanjing Medical University and The Affiliated Jiangning Hospital of Nanjing Medical University. From each LIHC patient, tumor tissue samples and matched adjacent normal tissues were collected during surgery and immediately frozen at −80 °C until RNA extraction. For the serum analysis component, blood samples were collected from these same LIHC patients, 40 cirrhosis patients, as well as from 72 healthy volunteers who served as controls. The LIHC patient group consisted of 40 males and 32 females, with a median age of 56.47 ± 4.73 years (range 37–72 years), the cirrhosis patients included 24 males and 16 females, with a median age of 59.35 ± 8.64 years (range 41–74 years), while the healthy control group was age-matched with a median age of 58.14 ± 4.74 years (range 36–70 years). Patients were included if they had pathologically confirmed LIHC diagnosis [14] and normal neurovolitional function, while those with other malignant tumors or severe blood system lesions were excluded. All liver biopsy tissues were confirmed by two pathologists. Written informed consent was obtained from all participants, and the study was conducted in accordance with the Declaration of Helsinki and approved by the institutional Ethics Committees.

Cell lines and culture

Human liver hepatocellular carcinoma cell lines (Bel-7402 and PLC/PRF/5) and normal human liver cell line L02 (American Type Culture Collection) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (Gibco) at 37 °C with 5% CO₂ in a humidified incubator.

Cell transfection

Bel-7402 and PLC/PRF/5 cells were seeded in 6-well plates and incubated to reach 80% confluence. To overexpress YTHDC2, pcDNA3.1/YTHDC2 vector (2 µg) and empty pcDNA3.1 vector (GenePharma, Shanghai, China) were transfected into cells using Lipofectamine 3000 (Invitrogen, USA). After 48 h, cells were harvested and the expression of YTHDC2 was measured using qRT-PCR analysis.

RNA extraction and expression analysis

Total RNA was extracted from tissues, serum, and cells using TRIzol reagent (Yeasen) according to manufacturer’s instructions. RNA quality was assessed by measuring D260/D280 ratios using a Nanodrop 2000 spectrophotometer (Thermo Fisher), with ratios between 1.8 and 2.0 considered acceptable.

For qRT-PCR analysis, one microgram of RNA was reverse transcribed using the RT‒PCR Transcriptor First Strand cDNA Synthesis Kit (Roche). CircYTHDC2 expression was quantified using the LightCycle PCR system (Roche) with miScript SYBR Green PCR kit (Qiagen). The following primers were used: CircYTHDC2: Forward 5’-GATGCCACTGCAGGCCAT-3’, Reverse 3’-GTGGAGCAGATGTTGCATAGCA-5’; GAPDH: Forward 5’-ACCTTCACTCCTCCATCTT-3’, Reverse 3’-AGGTCACAGACACGGTTG-5’.

Relative expression was calculated using the 2^-ΔΔCt^ method with GAPDH as internal control.

RNase R Treatment The stability of circYTHDC2 was assessed via RNase R digestion assays. RNA extracted from Bel-7402 and PLC/PRF/5 cells was divided into RNase treatment and control groups. For RNase treatment, 1 µg RNA was incubated with 1U RNase at 37 °C for 15 min. RNA was then extracted using the benzene chloroform precipitation method and analyzed by qRT-PCR.

Actinomycin D treatment

Bel-7402 and PLC/PRF/5 cells were seeded in 6-well plates and incubated to reach 70% confluence. Then cells were treated with 5 µg/ml actinomycin D (MedChemExpress, USA) and harvested after incubation for 0, 8, 16, 24, 32, 40, and 48 h. Next, TRIzol was used to extract the total RNA, and the RNA expression of circYTHDC2 was measured using qRT-PCR analysis.

RNA Pull-down assay

The biotinylated circYTHDC2 probe (bio-circYTHDC2) and negative control probe (bio-NC) were provided by Ribobio (Guangzhou, China). After incubating the probes with cell lysates of Bel-7402 and PLC/PRF/5 cells at 4 °C for 4 h, 50 µL of streptavidin magnetic beads (Thermo Fisher, USA) were added to incubate with the probes for another 2 h. Subsequently, beads were washed twice and RNA was purified using TRIzol reagent. The enrichment of YTHDC2 was detected by western blot analysis and quantified using ImageJ software.

RNA Immunoprecipitation (RIP)

The RIP assay was performed using the Magna RIP Kit (EMD Millipore, MA, USA) according to the manufacturer’s protocol. Briefly, Bel-7402 and PLC/PRF/5 cells were lysed with RIP lysis buffer and centrifuged. Next, the supernatant was collected and incubated with Protein-A/G magnetic beads (Thermo Fisher, USA) conjugated to human antibodies against YTHDC2 (Proteintech, China), m6A (Proteintech, China) or IgG (Proteintech, China) overnight at 4 °C. The precipitated RNAs were detected by qRT-PCR analysis.

Clinical correlation and statistical analysis

Patients were stratified based on circYTHDC2 expression levels to analyze associations with clinicopathological features including BCLC stage, tumor size, intrahepatic metastasis, and portal invasion. Progression-free and overall survival were analyzed using Kaplan-Meier methods. Statistical analyses were performed using GraphPad Prism 8.0 and SPSS 12.0. Data are presented as means ± SD. Comparisons between groups were performed using unpaired t-tests. ROC curves were generated to assess diagnostic potential. P values < 0.05 were considered statistically significant.

Results

Bioinformatic analysis reveals the potential role of YTHDC2 in LIHC

To establish a foundation for investigating circYTHDC2 in LIHC, we first examined the characteristics of its host gene, YTHDC2. Analysis of the UALCAN database, comprising 377 LIHC tissues and 50 normal tissues, revealed significant hypomethylation of the YTHDC2 promoter in LIHC tissues compared to normal liver tissues (Fig. 1A). Consistently, we also demonstrated that YTHDC2 was highly expressed in LIHC samples compared to normal liver samples (Fig. 1B). This finding suggested potential dysregulation of YTHDC2 in LIHC. These initial findings provided a strong rationale for investigating the expression patterns of circYTHDC2 in LIHC.

Fig. 1.

YTHDC2 methylation and survival analysis in LIHC. A Analysis of YTHDC2 promoter methylation levels in 377 LIHC tissues compared to 50 normal liver tissues using the UALCAN database, showing significant hypomethylation in LIHC tissues. B Analysis of YTHDC2 expression levels in 371 LIHC tissues compared to 50 normal liver tissues using the UALCAN database, showing significant upregulation in LIHC tissues

CircYTHDC2 expression is elevated across multiple LIHC models

Given the significant associations observed with YTHDC2, we next investigated its circular RNA derivative, circYTHDC2, across multiple experimental models. Initial RT-qPCR analysis demonstrated markedly elevated circYTHDC2 expression in LIHC tissues compared to adjacent normal tissues (P < 0.001; Fig. 2A). To validate these findings in a more clinically accessible context, we examined serum samples, where circYTHDC2 levels were significantly higher in LIHC patients compared to cirrhosis patients or healthy controls (P < 0.001; Fig. 2B).

Fig. 2.

CircYTHDC2 expression analysis across multiple LIHC models. A RT-qPCR analysis of circYTHDC2 expression in 72 paired LIHC tissues and adjacent normal tissues (P < 0.001). B Serum circYTHDC2 levels measured by RT-qPCR in 72 LIHC patients, 40 cirrhosis patients and 72 healthy controls. C CircYTHDC2 expression levels in LIHC cell lines (Bel-7402 and PLC/PRF/5) compared to normal human liver cell line L02. ***P < 0.001

To further understand the biological basis of these clinical observations, we examined circYTHDC2 expression in cellular models. Notably, both LIHC cell lines (Bel-7402 and PLC/PRF/5) exhibited significantly elevated circYTHDC2 expression compared to the normal hepatic cell line L02 (P < 0.001; Fig. 2C). This consistent upregulation across tissues, serum, and cell lines strongly suggested a fundamental role for circYTHDC2 in LIHC pathogenesis.

CircYTHDC2 demonstrates strong diagnostic potential

The robust expression differences observed prompted us to evaluate circYTHDC2’s potential as a diagnostic biomarker. ROC curve analysis revealed strong diagnostic performance for tissue circYTHDC2 (AUC: 0.846, 95% CI: 0.775–0.914) (Fig. 3A). Importantly, serum circYTHDC2 also demonstrated promising diagnostic capability (AUC: 0.788, 95% CI: 0.703–0.873; Fig. 3B), suggesting its potential as a non-invasive diagnostic tool.

Fig. 3.

Diagnostic potential and stability analysis of circYTHDC2. A ROC curve analysis of tissue circYTHDC2 expression for LIHC diagnosis (AUC: 0.846, 95% CI: 0.775–0.914; sensitivity: 82.4%, specificity: 76.8%). B ROC curve analysis of serum circYTHDC2 levels (AUC: 0.788, 95% CI: 0.703–0.873). C RNase R resistance assay in Bel-7402 cells showing maintained circYTHDC2 levels compared to linear YTHDC2 mRNA after RNase R treatment. D RNase R resistance assay in L02 cells demonstrating similar stability characteristics of circYTHDC2

For a biomarker to be clinically viable, stability is crucial. We therefore assessed circYTHDC2’s resistance to degradation through RNase R treatment. Remarkably, circYTHDC2 maintained most of its baseline levels in both LIHC cell lines, while linear YTHDC2 mRNA showed significant degradation (P < 0.05; Fig. 3C and D). This enhanced stability further supported circYTHDC2’s potential as a reliable clinical biomarker.

Clinical significance of CircYTHDC2 in LIHC progression

To understand the clinical implications of circYTHDC2 upregulation, we analyzed its relationship with key pathological features. Based on the median circYTHDC2 expression value (13.41), patients were divided into low (N = 36) and high (N = 36) expression groups. While no significant differences were observed in age or gender between the groups (P > 0.05), significant associations were found with multiple indicators of disease progression: advanced BCLC stage (57.7% vs. 42.3%; P = 0.004), larger tumor size ≥ 5 cm (75.0% vs. 25.0%; P = 0.004), presence of intrahepatic metastasis (66.7% vs. 33.3%; P = 0.001), and portal invasion (64.4% vs. 35.6%; P = 0.001) (Table 1). These associations suggested that circYTHDC2 might serve as an indicator of disease advancement and aggression.

Table 1.

The correlation of circYthdc2 with clinicopathological features of LIHC

Clinicopathological features	n	High circYthdc2 group	Low circYthdc2 group	χ2	P
Sample number	72	36(50%)	36(50%)
Age				0.239	0.536
> 50	48	26(54.2%)	22(45.8%)
< 50	24	10(41.7%)	14(58.3%)
Gender				1.341	0.263
Man	40	18(45.0%)	22(55.0%)
Woman	32	18(56.3%)	14(43.7%)
BCLC stage				7.366	0.004
O/A	20	6(30.0%)	14(70.0%)
B/C	52	30(57.7%)	22(42.3%)
Tumor size				8.941	0.004
< 5 cm	32	6(18.7%)	26(81.3%)
≥ 5 cm	40	30(75.0%)	10(25.0%)
Intrahepatic metastasis				9.134	0.001
Without	30	8(26.7%)	22(73.3%)
With	42	28(66.7%)	14(33.3%)
Portal invasion				9.003	0.001
Without	27	7(25.9%)	20(74.1%)
With	45	29(64.4%)	16(35.6%)

Abbreviattions: BCLC Stage; Barcelona Clinic Liver Cancer staging system, LIHC; Liver hepatocellular carcinoma, circYthdc2; Circular RNA derived from the YTHDC2 gene, χ2; Chi-squared test, P; P-value

Association between circYTHDC2 expression and patient survival

Given the strong correlation with aggressive disease features, we examined whether circYTHDC2 expression could predict patient outcomes. Kaplan-Meier analysis revealed significantly shorter progression-free survival in patients with high circYTHDC2 expression compared to those with low expression (P = 0.025; Fig. 4A). While overall survival analysis showed a trend toward poorer outcomes in the high-expression group, this difference did not reach statistical significance (P = 0.265; Fig. 4B). These findings establish circYTHDC2 not only as a stable and reliable biomarker but also as a clinically relevant indicator of disease progression and patient outcomes in LIHC.

Fig. 4.

Survival analysis based on circYTHDC2 expression levels. A Kaplan-Meier analysis of progression-free survival in LIHC patients stratified by circYTHDC2 expression levels. B Overall survival analysis comparing high and low circYTHDC2 expression groups. Patients were stratified based on median circYTHDC2 expression level (13.41). Risk tables below each curve indicate the number of patients at risk at 0, 10, 20, and 30 months of follow-up

CircYTHDC2 is regulated by YTHDC2 via m6A modification

Next, we explored whether YTHDC2 regulates the expression of circYTHDC2 in LIHC considering the role of YTHDC2 as an m6A reader and regulate mRNA stability via m6A modification [7]. Pearson correlation analysis indicated that the expression levels of circYTHDC2 and YTHDC2 were positively correlated in LIHC samples (R² = 0.506, P < 0.001; Fig. 5A). Then we overexpressed YTHDC2 in Bel-7402 and PLC/PRF/5 cells, and the overexpression efficiency was validated by qRT-PCR analysis (P < 0.001; Fig. 5B). The impact of YTHDC2 overexpression on circYTHDC2 expression was examined, and results from qRT-PCR analysis showed that the circYTHDC2 level was increased in response to YTHDC2 overexpression in Bel-7402 and PLC/PRF/5 cells (P < 0.001, Fig. 5C), which indicated that YTHDC2 positively modulated circYTHDC2 expression. Meanwhile, we also found that the YTHDC2 overexpression increased the half-life of circYTHDC2 in Bel-7402 (P = 0.012) and PLC/PRF/5 cells (P = 0.002) in the presence of ActD treatment (Fig. 5D). Furthermore, YTHDC2 overexpression was demonstrated to promote the binding between circYTHDC2 and anti-m6A antibody, suggesting that YTHDC2 modulated circYTHDC2 expression via m6A modification (P < 0.001, Fig. 5E). Results of RIP assays further revealed that circYTHDC2 was precipitated by the anti-YTHDC2 antibody in Bel-7402 and PLC/PRF/5 cells (P < 0.001, Fig. 5F). Consistently, RNA-Pull down assays showed that the YTHDC2 was enriched in the complexes pulled down by bio-circYTHDC2, suggesting the binding between circYTHDC2 and YTHDC2 in Bel-7402 and PLC/PRF/5 cells (P < 0.001, Fig. 5G). Overall, these results indicate that YTHDC2 enhances the stability of circYTHDC2 via m6A modification.

Fig. 5.

YTHDC2 stabilizes circYTHDC2 via m6A modification. A Pearson correlation analysis was used to assess the correlation between the expression of YTHDC2 and circYTHDC2 in the tumor samples from LIHC patients (n = 72). B qRT-PCR analysis was conducted to determine the overexpression efficiency of YTHDC2 in Bel-7402 and PLC/PRF/5 cells. C qRT-PCR analysis was used to measure the expression of circYTHDC2 in Bel-7402 and PLC/PRF/5 cells after YTHDC2 overexpression. D The impact of YTHDC2 overexpression on the RNA stability of circYTHDC2 was measured in Bel-7402 and PLC/PRF/5 cells with ActD treatment for indicated time. E RIP assays were used to detect the binding between anti-m6A with circYTHDC2 in Bel-7402 and PLC/PRF/5 cells after YTHDC2 overexpression. F RIP asssays were performed to explore the interaction between circYTHDC2 and YTHDC2 in Bel-7402 and PLC/PRF/5 cells. G RNA-Pull down assays were used to determine the binding between circYTHDC2 and YTHDC2 in Bel-7402 and PLC/PRF/5 cells. *P < 0.05, **P < 0.01, ***P < 0.001

Discussion

This comprehensive study demonstrates that circYTHDC2 serves as a clinically relevant molecular marker in LIHC, with significant implications for both diagnosis and prognosis. Our findings establish elevated circYTHDC2 expression across multiple experimental models and reveal its strong associations with aggressive disease features and poor clinical outcomes.

Primary liver cancer remains one of the most lethal malignancies worldwide, ranking third in cancer-related mortality after lung and colorectal cancer, with approximately 830,000 deaths annually [15]. Among its subtypes, LIHC presents particular challenges due to its aggressive nature and poor prognosis. The absence of early symptoms often leads to delayed diagnosis, by which time patients have missed the optimal window for surgical intervention [16].

The emergence of circRNAs as regulatory factors in cancer has opened new avenues for understanding and predicting LIHC progression [17]. Through our multi-level analysis approach, we have uncovered several key insights into circYTHDC2’s role in LIHC pathogenesis. The consistent upregulation of circYTHDC2 across tissues, serum, and cell lines suggests its fundamental involvement in disease development. Moreover, its remarkable stability, demonstrated through RNase resistance testing, reinforces its potential as a reliable clinical biomarker.

The molecular architecture of circYTHDC2, comprising a highly conserved 634-nt sequence derived from exons 13–18 of the YTHDC2 gene [11], provides a stable foundation for its biological functions. Of particular interest is the relationship between circYTHDC2 and m6A modification, especially through its interaction with the YTHDC2 protein [7]. Consistently, in our study, we also demonstrated that YTHDC2 bound to circYTHDC2 and enhances the RNA stability of circYTHDC2 via m6A modification in Bel-7402 and PLC/PRF/5 cells. Recent advances in understanding m6A’s influence on RNA processing and cellular regulation [8] suggest that circYTHDC2 might serve as a crucial mediator in LIHC progression through epigenetic mechanisms.

The clinical significance of our findings is underscored by the strong associations between circYTHDC2 expression and multiple indicators of disease advancement. The correlation with advanced BCLC stage, increased tumor size, and metastatic features suggests that circYTHDC2 could serve as a valuable tool for risk stratification in LIHC patients. Furthermore, its prognostic value, demonstrated through progression-free survival analysis, indicates potential utility in treatment planning and monitoring. Although the overall survival analysis did not exhibit statistical difference between circYTHDC2 high and low groups, the results were potentially confounded considering the different types and durations of treatment regimens applied post progression.

Our observations regarding circYTHDC2’s enhanced stability and presence in serum samples suggest promising applications in liquid biopsy-based diagnostics. The ability to detect and quantify circYTHDC2 in serum could facilitate non-invasive monitoring of disease progression and treatment response. The significant diagnostic accuracy demonstrated by ROC analysis further supports its potential as a clinical biomarker.

Despite these promising findings, several aspects warrant further investigation. First, the modest sample size might restrain the generalization of the results. Second, the function and molecular mechanisms through which circYTHDC2 influences LIHC progression remain to be fully elucidated. Future research should focus on deciphering its interactions with the cellular machinery, particularly its potential roles in regulating gene expression. The validation of circYTHDC2’s predictive utility in larger, diverse patient populations would strengthen its clinical applicability. Additionally, investigating its potential as a therapeutic target could open new avenues for LIHC treatment.

The integration of circYTHDC2 assessment into clinical practice could potentially enhance current diagnostic and prognostic approaches in LIHC. Its stability, accessibility in serum, and strong correlation with disease features make it an attractive candidate for routine clinical testing. These characteristics, combined with its significant prognostic value, suggest that circYTHDC2 could become an integral component of LIHC patient management strategies.

Authors’ contributions

YG and XL performed the experiments and analyzed the data. YG drafted the manuscript. MW conceived and designed the study, supervised the research, and revised the manuscript. All the authors read and approved the final manuscript.

Funding

This work was supported by a study on the role and mechanism of ASPM alternative splicing mediated by the RBM10 C761Y mutation in hilar cholangiocarcinoma (No. 82273066).

Data availability

All the datasets generated and analyzed during the current study are available in this published article. Bioinformatic data were obtained from publicly available databases: GEPIA2 (http://gepia2.cancer-pku.cn/#index) and Kaplan‒Meier Plotter (http://kmplot.com/analysis/).

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committees of The First Affiliated Hospital of Nanjing Medical University and The Affiliated Jiangning Hospital of Nanjing Medical University having ethical approval number (2023YJK87ZH). Written informed consent was obtained from all participants prior to sample collection.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.