Introduction

Hepatocellular carcinoma (HCC) is a heterogeneous tumor with complicated genetic and epigenetic variations, therapeutic options for HCC have been limited and include surgery, local ablation, and liver transplantation in early disease stages.1–3 However, patients with advanced HCC have a poor overall survival (OS).1,2 Transarterial chemoembolization (TACE) blocks the arterial supply to malignant tissues by injecting small embolic particles containing certain cytotoxic chemotherapeutic agents into the hepatic artery and further promotes ischemic necrosis of the tumor. Multiple clinical trials have indicated that TACE significantly prolongs OS of unresectable HCC cases.3–5 Interestingly, it has been demonstrated that TACE promotes the release of angiogenic growth factors and elevates the levels of tumor antigens, suggesting that novel strategies combining TACE, PD-1/PD-L1 inhibitors and antiangiogenic therapy seem to have the potential to tip the balance of the tumor microenvironment and improve treatment response. Currently, TACE is the standard therapy for HCC cases with unresectable diseases.3–5 Clinically, not all HCC patients exhibit therapeutic responses and a small proportion of cases even show an uncontrolled increase in tumor burden after treatment.4,5 As a result, it is important to further elucidate how individual genetic background impacts the therapeutic efficacy of TACE and patients’ prognosis.

The transcriptome modifications are emerging as a crucial player during malignant transformation. As an RNA modification, RNA editing modifies specific nucleotides, such as adenosine, in RNA sequences.6,7 The adenosine-to-inosine (A-to-I) change is the most plentiful kind of editing in human RNA and known to diversify transcripts.6,7 Dysregulated levels of editing genes in cancerous cells lead to abnormal A-to-I editing, which contributes to the development of m HCC and other malignancies.6–12 Adenosine deaminase RNA specific B1 (ADARB1), adenosine deaminase RNA specific B2 (inactive) (ADARB2), and adenosine deaminase RNA specific (ADAR) are three main ADAR family enzymes catalyzing deamination of adenosine nucleotides in RNA.6,7 Likewise, as a key enzyme negatively regulating ADAR activities, aminoacyl tRNA synthetase complex interacting multifunctional protein 2 (AIMP2) is also involved in regulating RNA A-to-I editing.13 The main functions of the ADARB1 protein are recoding editing sites in mammals.14 A marked reduction in ADARB1 expression and editing capacity led to a malignant phenotype.15–18 Nevertheless, it remains largely unclear how genetic variations of the RNA editing genes confer to the prognosis of HCC cases treated with TACE.

In this study, we tested the hypothesis that single nucleotide polymorphisms (SNPs) in the four RNA A-to-I editing genes (ADAR, ADARB1, ADARB2 and AIMP2) may regulate expression of certain genes and prognosis of HCC cases treated with TACE. We verified a functional rs2253763 SNP which locates in the ADARB1 3’-untranslated region (3’-UTR) and leads to down-regulated ADARB1 expression in an allelic manner. The ADARB1 rs2253763 C-allele 3’-UTR exhibits higher binding affinity with miR-542-3p in comparison with the ADARB1 T-allele 3’-UTR in HCC cells. Interestingly, the HCC patients with rs2253763 C-allele and decreased ADARB1 expression showed shorter survival time than cases with of the T-allele after TACE treatment.

Materials and Methods Patient Cohorts

In the current study, there were 564 TACE-treated HCC patients who were recruited from two TACE patient cohorts (Jiangsu set and Shandong set) (Supplementary Table 1). The characteristics of all HCC cases have been reported previously.19–21 All cases are Han Chinese. The written informed consent was signed by each patient at recruitment. This study was approved by the Institutional Review Board of Shandong Cancer Hospital and Institute and complied with the Declaration of Helsinki.

Selection and Genotyping of Candidate SNPs

As previously described, we selected candidate genetic polymorphisms of ADARB1, ADARB2, ADAR, and AIMP2 which are potentially functional by analyzing multiple databases.19–21 We genotyped candidate genetic polymorphisms using the iPLEX MassARRAY (Sequenom) as described previously.22,23 In both cohorts, genotypes of twenty-eight candidate SNPs were successfully determined.

Reverse Transcription Quantitative PCR (RT-qPCR)

SYBR-Green RT-qPCR was performed to detect human ADARB1 and ß-actin mRNA expression (Supplementary Table 2) as well as human miR-542-3p, miR-4734 and U6 small RNA levels as described previously.23,24 Each test was performed in triplicate.

ADARB1 Dual Luciferase Reporter Gene Assays

Human HCC HepG2 or Li-7 cell lines were obtained from the cell bank of type culture, Chinese Academy of Sciences (Shanghai). Cells were cultured in DMEM medium or PRMI 1640 medium. All cells were used within six months of passaging from original stocks. The specific PCR primers were used to amplify the ADARB1 180bp DNA fragments (3121bp-3300bp) with either the rs2253763 TT human genomic DNA or the CC genotype genomic DNA (Supplementary Table 3). The PCR products were then cloned into pGL3-Control (Promega) and the plasmids were named as pGL3-ADARB1-T allele or pGL3-ADARB1-C allele. During the reporter gene assays, HepG2 and Li-7 HCC cells were co-transfected with the pGL3 luciferase reporter plasmids (pGL3-ADARB1-T or pGL3-ADARB1-C), miRNA mimics (miR-542-3p or miR-4734 mimics) and pRL-SV40 (Promega). At 48h after transfection, dual luciferase activities of the HCC cells were measured as previously described.23

Western Blot

Western blot was conducted as reported previously.23,24 The anti-ADARB1 antibody (Proteintech, 22248-1-AP) or anti-GAPDH antibody (Proteintech, 60004-1-Ig) was used to detect protein levels of ADARB1 or GAPDH in cells.

Oxaliplatin Drug Sensitivity Assays

Human HCC HepG2 and Li-7 cells were obtained from the cell bank of type culture, Chinese Academy of Sciences (Shanghai). As described previously, HepG2 or Li-7 HCC cells were firstly seeded in 96-well culture plates and then transfected with the pcDNA3.1 vector (as negative control) or pcDNA-ADARB1 plasmid (as the ADARB1-overexpression group).19 After 24h, cells into each well were treated with oxaliplatin (Selleck, S1224) or not. For HepG2 cells, a total of 1.5μmol/L, 3.0μmol/L and 4.5μmol/L oxaliplatin was used to treat cells. For Li-7 cells, a total of 0.5μmol/L, 1.0μmol/L, 1.5μmol/L and 2.0μmol/L oxaliplatin was utilized to treat cells. After 48 h, the MTT assays were performed as reported previously.19

Statistics

OS differences between each genotype were calculated using Log rank tests. The Kaplan–Meier curves of different ADARB1 genotypes were also analyzed. The correlations between the RNA editing genetic variants and death risk of HCC cases treated with TACE were evaluated using the multivariate Cox regression. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated after adjustment with patient characteristics. A P < 0.05 was considered as statistically significant and all statistical analyses were calculated using SPSS or GraphPad software.

Results Effects of Candidate SNPs of the RNA Editing Genes on OS of TACE-Treated HCC Cases

The characteristics of HCC cases are shown in Supplementary Table 1 and reported previously.19,22 The univariate Cox regression model elucidated that two ADARB1 genetic polymorphisms (rs1051367 and rs2253763) were markedly correlated with mortality risk of HCC cases who were treated with TACE (both P < 0.01) (Table 1). The ADARB1 rs1051367 genetic variant exhibited evidently correlation with 1.30-fold elevated death risk of HCC cases received TACE treatment (95% CI = 1.07–1.58, P = 0.008). On the contrary, the ADARB1 rs2253763 polymorphism markedly contributed to 0.64-fold reduced risk of mortality of HCC patients treated with TACE (95% CI = 0.53–0.77, P = 3.8×10−6) (Table 1).

Table 1 Cox-Regression Analyses of Twenty-Eight Candidate Genetic Variants in RNA Editing-Related Genes for OS in Jiangsu Cohort

Impacts of ADARB1 Genetic Variants on OS After TACE Treatment

Interestingly, HCC cases with the ADARB1 rs1051367 GG genotype showed markedly shorter OS time than HCC patients with the rs1051367 AA and AG genotypes after TACE treatment (Jiangsu set: log-rank P = 0.002; Shandong set: log-rank P < 0.001) (Figure 1A). For the carriers of the ADARB1 rs1051367 AA, AG or GG genotype, the median survival time (MST) was 19.5, 21 or 10 months in Jiangsu set. In Shandong set, the MST of HCC patients with the ADARB1 rs1051367 AA, AG and GG genotype was 18, 18 and 9 months. In contrast, carriers of the ADARB1 rs2253763 TT or CT genotype after TACE treatment had evidently prolonged survival time in comparison with subjects carrying the rs2253763 CC genotype (Jiangsu set: log-rank P < 0.001; Shandong set: log-rank P = 0.002) (Figure 1B). The MSTs for cases with rs2253763 TT and CT genotypes were 22 and 20 months in Jiangsu set or 23 and 17 months in Shandong set, which were significantly longer than 8 months of individuals carrying the CC genotype in Jiangsu set or 14 months in Shandong set.

Figure 1 Kaplan–Meier curves of OS for TACE-treated HCC patients with various ADARB1 genotypes. (A) rs1051367 in Jiangsu set (left panel) or Shandong set (right panel). (B) rs2253763 in Jiangsu set (left panel) or Shandong set (right panel).

The multivariate Cox regression model demonstrated that the ADARB1 rs1051367 GG genotype was markedly correlated with elevated death risk compared to the rs1051367 AA genotype (Jiangsu set: HR = 1.37, 95% CI = 1.12–1.68, P = 0.002; Shandong set: HR = 2.51, 95% CI = 1.73–3.64, P = 1.4×10−6) (Table 2). However, no such evident correlation between the rs1051367 AG genotype and death risk was observed (Jiangsu set: HR = 1.03, 95% CI = 0.79–1.35, P = 0.827; Shandong set: HR = 1.18, 95% CI = 0.90–1.56, P = 0.240) (Table 2). On the contrary, HCC patients carrying the ADARB1 rs2253763 TT or CT genotypes showed a 37% or 32% reduced death risk in comparison with cases carrying the rs2253763 CC genotype (95% CI = 0.53–0.88, P = 0.004; or 95% CI = 0.49–0.81, P = 2.8×10−4) after TACE therapy (Table 2). Similarly, the ADARB1 rs2253763 TT genotype was notably correlated with decreased death risk in comparison with the rs2253763 CC genotype in Shandong set (HR = 0.46, 95% CI = 0.28–0.74, P = 0.002) (Table 2). Nevertheless, there was no significant association between the rs2253763 CT genotype and OS in Shandong set (HR = 0.82, 95% CI = 0.64–1.06, P = 0.130).

Table 2 Multivariate Cox-Regression Analyses of ADARB1 rs1051367 and rs2253763 Genetic Variants for OS in Jiangsu Cohort and Shandong Cohort

Decreased ADARB1 Expression Levels in HCC Specimens

The ADARB1 levels in HCC specimens and normal tissues from twenty-six HCC patients were detected. It has been found that there was a significantly reduced ADARB1 expression level in cancerous specimens in comparison with its expression in normal tissues (P<0.001) (Figure 2A). Consistently, a significantly decreased ADARB1 expression level was found in malignant tissues in comparison with normal tissues in a cohort including 268 tumors and 243 adjacent normal tissues (GEO: GSE25097) (P = 0.001) (Figure 2B). Consistent with these data, ADARB1 expression was also lower in HCC tissues (GEO: GSE25097) in comparison with normal specimens in several patient cohorts from Koo Foundation SYS Cancer Center (Taipei, China) (GEO: GSE60502) (P = 0.006) (Figure 2C), Centre de Recherche des Cordeliers (Paris, France) (GEO: GSE62232) (P<0.001) (Figure 2D), and National Cancer Centre Singapore (Singapore) (GEO: GSE121248) (P < 0.001) (Figure 2E). Together, these data suggested that ADARB1 might function as a novel tumor suppressor gene in HCC.

Figure 2 ADARB1 expression in HCC specimens and normal tissues. (A) ADARB1 expression was detected in 26 tumor tissues and paired normal tissues. The GAPDH mRNA levels were used as the endogenous controls in these specimens. (B) ADARB1 expression in HCC and normal tissues from GEO: GSE25097. (C) ADARB1 expression in HCC and normal tissues from GEO: GSE60502. (D) ADARB1 expression in HCC and normal tissues from GEO: GSE62232. (E) ADARB1 expression in HCC and normal tissues from GEO: GSE121248. **P < 0.01, ***P < 0.001.

The rs2253763 SNP Caused Allelic Interactions Between miR-542-3p and ADARB1 mRNA

Multiple lines of evidence elucidated that SNPs in 3’-UTR of certain genes might disturb the interactions between miRNA(s) and target mRNAs. We then investigated the correlations between rs2253763 genotypes and ADARB1 expression levels in malignant specimens and normal specimens, as rs2253763 is in the gene 3’-UTR. Interestingly, we found that an evident difference of allele-specific expression existed between the ADARB1 rs2253763 C-allele HCC tissues and the T-allele HCC specimens (P<0.05) (Figure 3A). We also observed similar results in normal specimens (P < 0.05) (Figure 3A). An evident increased ADARB1 gene expression was observed in both HCC specimens (P = 0.028) and normal specimens (P = 0.013) of carriers with the ADARB1 rs2253763 TC and TT genotypes compared to the CC genotype (Figure 3A). Intriguingly, the MicroSNiPer algorithm demonstrated that the rs2253763 T-to-C genetic variant may lead to the gain of a miR-542-3p target binding site and the loss of a miR-4734 target site in ADARB1 3’-UTR. Compared to normal tissues, there was evidently increased miR-542-3p levels in HCC specimens (P<0.001) (Figure 3C). However, we did not observe such differences of miR-4734 expression levels between HCC and normal tissues for (P = 0.573) (Figure 3D). Next, we performed the luciferase reporter gene assays to examine how miR-542-3p and miR-4734 impact ADARB1 gene expression in both HCC cell lines (Figure 3E). After transfecting HepG2 or Li-7 cells with miR-542-3p or miR-4734 mimics as well as the pGL3-ADARB1-T allele construct or the pGL3-ADARB1-C allele construct, we then detected luciferase reporter gene activities in the HCC cell lines (Figure 3F and G). Importantly, the pGL3-ADARB1-C allele luciferase construct showed significantly lower reporter gene activity in HCC cells transfected with miR-542-3p mimics and the pGL3-ADARB1-C construct in comparison with cells with co-transfections of miR-542-3p mimics and the pGL3-ADARB1-T construct (HepG2 cells: P=0.002; Li-7 cells: P=0.002) (Figure 3F). In contrast, miR-4734 could not influence luciferase reporter gene activities of the pGL3-ADARB1-T or pGL3-ADARB1-C reporter construct in HCC cells (HepG2 cells: P=0.937; Li-7 cells: P=0.818) (Figure 3G). These data elucidated that the rs2253763 genetic variation caused allele-differential binding of miR-542-3p and, thus, allelic ADARB1 expression in HCC cells.

Figure 3 Allelic impacts of the SNP rs2253763 on ADARB1 expression. (A) ADARB1 expression in HCC and normal tissues grouped by rs2253763 genotypes. (B) The rs2253763 T-to-C change may cause the gain of target sites of miR-542-3p as well as the loss of target sites of miR-4734. (C) miR-542-3p expression in HCC and normal tissues. (D) miR-4734 expression in HCC and normal tissues. (E–G) Transient luciferase reporter gene assays in HepG2 or Li-7 cells transfected with both miR-542-3p/miR-4734 mimics and constructs containing different rs2253763 allele of ADARB1 3’-UTR region. pRL-SV40 were used to standardize transfection efficiency. Fold-changes were calculated by defining the luciferase activity of cells co-transfected with pGL3-Control as 1. All experiments were performed in triplicates in three independent transfection experiments and each value represents mean±SD. *P < 0.05, **P < 0.01, ***P < 0.001.

Abbreviation: ns, not significant.

ADARB1 Notably Enhances Oxaliplatin Sensitivity

Oxaliplatin is a commonly used chemotherapeutic agent in HCC patients during TACE. As a result, we investigated whether dysregulated ADARB1 expression in HCC cells impacts oxaliplatin sensitivities (Figure 4). Forced expression of ADARB1 in HepG2 cells significantly promoted the antineoplastic efficacy of oxaliplatin (P<0.01) (Figure 4A and B). Consistently, overexpression of ADARB1 in Li-7 cells also evidently promotes inhibition effects of oxaliplatin (P < 0.001) (Figure 4A and C).

Figure 4 ADARB1 profoundly enhanced oxaliplatin sensitivity. (A) Over-expression of mRNA and protein levels of ADARB1 in HepG2 or Li-7 cell lines. Forced expression of ADARB1 promoted antineoplastic efficiency of oxaliplatin in HepG2 cells (B) or Li-7 cells (C). All experiments were performed in triplicates in three independent transfection experiments and each value represents mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.

Discussion

Development of treatment resistance is a common cause of disease progression among HCC patients treated with TACE therapy. There are multiple possible explanations for the development of treatment resistance. Major reasons include proliferation of sub-clones with somatic evolution and intratumor genetic heterogeneity. Besides somatic mutations, aberrant A-to-I editing in RNA sequences induced non-genetic heterogeneity also facilitate drug resistance and cancer progression.8,25,26 For example, silencing of ADAR1 suppresses A-to-I editing of interferon-induced RNA species and tumor inflammation. As a result, ADAR1 loss in cancer cells overcomes cancer resistance to anti-PD-1 antibody.27 Multiple Phase 1, 2, or 3 clinical trials have shown that a combined PD-1 checkpoint blockade with molecular targeted agents or other locoregional therapy is an effective strategy for HCC therapy.28,29 Nevertheless, it is still largely unknown how the genes involved in RNA editing and their potential functional SNPs contribute to TACE resistance in HCC.

In the current study, we thoroughly explored correlations between genotypes of the ADAR, ADARB1, ADARB2 and AIMP2 genes involved in RNA editing and death risk of HCC cases received TACE therapy. We identified two novel ADARB1 rs1051367 and rs2253763 genetic polymorphisms which were significantly correlated with OS of HCC cases who received TACE therapy. Moreover, we revealed that the rs2253763 SNP in ADARB1 3’-UTR could disturb miR-542-3p binding and dysregulated ADARB1 expression in an allelic manner. Consistently, ADARB1 levels were increased in both HCC specimens and normal specimens of the rs2253763 TC and TT genotype carriers in comparison with the CC genotype. Our study presents new clues to involve the ADARB1 in disease progression of HCC.

There is growing evidence that HCC shows a severe disorder of A-to-I editing in RNA sequences compared to normal liver tissues.11,30 It has been reported that ADARB1 acts as a tumor suppressive gene in HCC.11 For instance, Kang et al found that ADARB1 expression levels were inhibited in HCC tumors and patients with reduced ADARB1 expression in cancerous tissues showed an elevated risk of recurrence and poor prognoses.11 Consistent with these findings, we also observed a marked reduce of ADARB1 expression in HCC specimens from our center and multiple other cancer centers around the world. Although functionally critical target genes for ADARB1 have rarely been reported, COPA (coatomer subunit α) is one of these editing targets.31 Binding of ADARB1 to COPA precursor mRNA results in the change of residue 164 isoleucine to residue 164 valine. After RNA editing, wide-type oncoprotein COPA is transformed into a tumor suppressor, possibly through the down-regulation of caveolin-1 to shut down the PI3K/AKT/mTOR pathway.31 Additionally, miR-214 and miR-122 RNA precursors have been identified as RNA editing targets by ADARB1 in HCC.32

Besides TACE, liver transplantation, radiofrequency ablation, and tyrosine protein kinase inhibitors are also common treatments for advanced HCC.33 It has been found that the tumor microenvironment of HCC is largely immunosuppressive. As a result, HCC immune escapes are commonly observed in immunotherapy of immune checkpoint inhibitors (ICIs). However, combinations of TACE, immunotherapy based on ICIs or anti-angiogenic drugs offer great promise during HCC treatment.33

In summary, we declared novel prognostic values of the RNA editing gene ADARB1 and its functional rs2253763 SNP in HCC patients after TACE therapy. TACE in combination with targeting ADARB1 for cases with unresectable HCC may be a promising therapeutic strategy. These insights will promote our understanding for genetic clarifications and outcome assessments of HCC cases, which may ultimately lead to patient-tailored TACE.

Data Sharing Statement

All data are available on reasonable request from the corresponding authors – Ming Yang; Ziqiang Li.

Ethics Statement

The Ethics Committee of Shandong Cancer Hospital and Institute approved the study (ID: SDCH-JYK-JL-123).

Acknowledgments

We acknowledge all participants enrolled in the study.

Funding

National Natural Science Foundation of China (82173070 and 82103291); Natural Science Foundation of Shandong Province (ZR2021LZL004 and ZR202102250889); Major Scientific and Technological Innovation Project of Shandong Province (2021ZDSYS04); Taishan Scholars Program of Shandong Province (tsqn202211340 and tstp20221141); Program of Science and Technology for the youth innovation team in universities of Shandong Province (2020KJL001 and 2022KJ316).

Disclosure

The authors report no conflicts of interest in this work.

References

1. Eisenberg E, Levanon EY. A-to-I RNA editing – immune protector and transcriptome diversifier. Nat Rev Genet. 2018;19(8):473–490. doi:10.1038/s41576-018-0006-1

2. Fritzell K, Xu LD, Lagergren J, Ohman M. ADARs and editing: the role of A-to-I RNA modification in cancer progression. Semin Cell Dev Biol. 2018;79123–79130. doi:10.1016/j.semcdb.2017.11.018

3. Tan MH, Li Q, Shanmugam R, et al. Dynamic landscape and regulation of RNA editing in mammals. Nature. 2017;550(7675):249–254. doi:10.1038/nature24041

4. Chen L, Li Y, Lin CH, et al. Recoding RNA editing of azin1 predisposes to hepatocellular carcinoma. Nat Med. 2013;19(2):209–216. doi:10.1038/nm.3043

5. Hu X, Wan S, Ou Y, et al. RNA over-editing of blcap contributes to hepatocarcinogenesis identified by whole-genome and transcriptome sequencing. Cancer Lett. 2015;357(2):510–519. doi:10.1016/j.canlet.2014.12.006

6. Kang L, Liu X, Gong Z, et al. Genome-wide identification of RNA editing in hepatocellular carcinoma. Genomics. 2015;105(2):76–82. doi:10.1016/j.ygeno.2014.11.005

7. Shi L, Yan P, Liang Y, et al. Circular RNA expression is suppressed by androgen receptor (ar)-regulated adenosine deaminase that acts on RNA (ADAR1) in human hepatocellular carcinoma. Cell Death Dis. 2017;8(11):e3171. doi:10.1038/cddis.2017.556

8. Clutterbuck DR, Leroy A, O’Connell MA, Semple CA. A bioinformatic screen for novel A-I RNA editing sites reveals recoding editing in bc10. Bioinformatics. 2005;21(11):2590–2595. doi:10.1093/bioinformatics/bti411

9. Maas S, Patt S, Schrey M, Rich A. Underediting of glutamate receptor GluR-b mRNA in malignant gliomas. Proc Natl Acad Sci U S A. 2001;98(25):14687–14692. doi:10.1073/pnas.251531398

10. Galeano F, Rossetti C, Tomaselli S, et al. ADAR2-editing activity inhibits glioblastoma growth through the modulation of the CDC14b/Skp2/p21/p27 axis. Oncogene. 2013;32(8):998–1009. doi:10.1038/onc.2012.125

11. Chen YB, Liao XY, Zhang JB, et al. ADAR2 functions as a tumor suppressor via editing IGFBP7 in esophageal squamous cell carcinoma. Int J Oncol. 2017;50(2):622–630. doi:10.3892/ijo.2016.3823

12. Cesarini V, Silvestris DA, Tassinari V, et al. ADAR2/miR-589-3p axis controls glioblastoma cell migration/invasion. Nucleic Acids Res. 2018;46(4):2045–2059. doi:10.1093/nar/gkx1257

13. Ferlay J, Colombet M, Soerjomataram I, et al. Estimating the global cancer incidence and mortality in 2018: globocan sources and methods. Int J Cancer. 2019;144(8):1941–1953. doi:10.1002/ijc.31937

14. Llovet JM, Zucman-Rossi J, Pikarsky E, et al. Hepatocellular carcinoma. Nat Rev Dis Primers. 2016;216018. doi:10.1038/nrdp.2016.18

15. Young S, Craig P, Golzarian J. Current trends in the treatment of hepatocellular carcinoma with transarterial embolization: a cross-sectional survey of techniques. Eur Radiol. 2019;29(6):3287–3295. doi:10.1007/s00330-018-5782-7

16. Llovet JM, Real MI, Montana X, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet. 2002;359(9319):1734–1739. doi:10.1016/S0140-6736(02)08649-X

17. Lo CM, Ngan H, Tso WK, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology. 2002;35(5):1164–1171. doi:10.1053/jhep.2002.33156

18. Liu J, Wang D, Zhou J, et al. N6-methyladenosine reader ythdc2 and eraser fto may determine hepatocellular carcinoma prognoses after transarterial chemoembolization. Arch Toxicol. 2021;95(5):1621–1629. doi:10.1007/s00204-021-03021-3

19. Song Y, Xing H, Zhou L, Zhang N, Yang M. Lncrna h19 modulated by mir-146b-3p/mir-1539-mediated allelic regulation in transarterial chemoembolization of hepatocellular carcinoma. Arch Toxicol. 2021;95(9):3063–3070. doi:10.1007/s00204-021-03119-8

20. Hua H, Zeng J, Xing H, et al. The RNA editing enzyme ADAR modulated by the rs1127317 genetic variant diminishes EGFR-TKIs efficiency in advanced lung adenocarcinoma. Life Sci. 2022;296120408. doi:10.1016/j.lfs.2022.120408

21. Li W, Wang T, Fu G, et al. The allelic regulation of tumor suppressor ADARB2 in papillary thyroid carcinoma. Endocr Relat Cancer. 2022;30. doi:10.1530/ERC-22-0189

22. Xu Y, Wang T, Zeng J, et al. Integrative functional genomics implicated the key t-/b-cell deficiency regulator rag1 in transarterial chemoembolization of hepatocellular carcinoma. Front Cell Dev Biol. 2021;9720791. doi:10.3389/fcell.2021.720791

23. Zhang N, Song Y, Xu Y, et al. Med13l integrates mediator-regulated epigenetic control into lung cancer radiosensitivity. Theranostics. 2020;10(20):9378–9394. doi:10.7150/thno.48247

24. Zheng Y, Lei T, Jin G, et al. Lncpsca in the 8q24.3 risk locus drives gastric cancer through destabilizing ddx5. EMBO Rep. 2021;22(11):e52707. doi:10.15252/embr.202152707

25. Wang C, Zou J, Ma X, Wang E, Peng G. Mechanisms and implications of ADAR-mediated RNA editing in cancer. Cancer Lett. 2017;41127–41134. doi:10.1016/j.canlet.2017.09.036

26. Walkley CR, Li JB. Rewriting the transcriptome: adenosine-to-inosine RNA editing by ADARs. Genome Biol. 2017;18(1):205. doi:10.1186/s13059-017-1347-3

27. Ishizuka JJ, Manguso RT, Cheruiyot CK, et al. Loss of ADAR1 in tumours overcomes resistance to immune checkpoint blockade. Nature. 2019;565(7737):43–48. doi:10.1038/s41586-018-0768-9

28. Xu F, Jin T, Zhu Y, Dai C. Immune checkpoint therapy in liver cancer. J Exp Clin Cancer Res. 2018;37(1):110. doi:10.1186/s13046-018-0777-4

29. Li Z, Wu X, Zhao Y, et al. Clinical benefit of neoadjuvant anti-pd-1/pd-l1 utilization among different tumors. MedComm. 2021;2(1):60–68. doi:10.1002/mco2.61

30. Chan TH, Lin CH, Qi L, et al. A disrupted RNA editing balance mediated by ADARs (adenosine deaminases that act on RNA) in human hepatocellular carcinoma. Gut. 2014;63(5):832–843. doi:10.1136/gutjnl-2012-304037

31. Song Y, An O, Ren X, et al. RNA editing mediates the functional switch of COPA in a novel mechanism of hepatocarcinogenesis. J Hepatol. 2021;74(1):135–147. doi:10.1016/j.jhep.2020.07.021

32. Liu WH, Chen CH, Yeh KH, et al. ADAR2-mediated editing of miR-214 and miR-122 precursor and antisense RNA transcripts in liver cancers. PLoS One. 2013;8(12):e81922. doi:10.1371/journal.pone.0081922

33. Leone P, Solimando AG, Fasano R, et al. The evolving role of immune checkpoint inhibitors in hepatocellular carcinoma treatment. Vaccines. 2021;9(5):532. doi:10.3390/vaccines9050532