Liver cancer is expected to be the sixth most commonly diagnosed cancer, and it ranks as the fourth leading cause of cancer death. With the absence of early diagnosis and limited treatment methods, Bray et al. estimates the global incidence of liver cancer to be 841,000 new cases and 782,000 deaths in 2018, with greater prevalence in Northern and Western Africa and Eastern and South-Eastern Asia (Bray et al., 2018). Hepatocellular carcinoma (HCC) is the most prevalent form of primary liver cancer, which accounts for >90% of primary liver cancer cases (Llovet et al., 2021). The primary risk factors for HCC include chronic viral hepatitis infection (hepatitis B or C), aflatoxin B1 intake, obesity, and excessive alcohol consumption (Chhonker et al., 2021; Choi et al., 2021; Guan et al., 2021; Hwang et al., 2021; Lockart et al., 2021). These factors may be responsible for cancer-related mutations, DNA damage, and epigenetic alterations, resulting in inactivated oncogenes or inactivated tumor suppressor genes, ultimately leading to HCC development (Rovida et al., 2015; Ally et al., 2017; Dang et al., 2017). Current treatment strategies for HCC include specific kinase inhibitors, anti-hepatitis vaccine, and liver resection and transplantation (Dageforde et al., 2021; Pattyn et al., 2021; Sbenati et al., 2021; Yamamoto et al., 2021). These therapeutic approaches in combination with biomarker screening reduce HCC-related death. However, its treatment efficiency is far from satisfactory. Thus, novel treatment approaches with higher efficacy for HCC are an urgently needed.

Earlier studies related to HCC primarily concentrated on the coding genes in identified regions with their pivotal roles in many central biological processes (Ladero et al., 1996; Thorgeirsson et al., 2006; Park et al., 2016). However, only 2% of human genome transcripts comprise protein coding sequences, and nearly 98% of sequences have no coding for proteins; the transcripts that cannot be ultimately translated into proteins have been characterized as non-coding RNAs (ncRNAs) (Rosenbloom et al., 2012; Beermann et al., 2016). There is a growing body of evidences that support that these evolutionarily conserved ncRNAs, particularly in the long non-coding RNAs (lncRNAs), which are more than 200 nucleotides in length, play a regulatory role in various developmental or physiological processes of HCC (Huang et al., 2020; Qin et al., 2020; Zuo et al., 2020). The number of transcribed lncRNAs continues to grow rapidly (Xie et al., 2021). Several studies have proved that these lncRNAs are differentially expressed in a variety of tissues and HCC cells (Hongfeng et al., 2020; Topel et al., 2020; Gamaev et al., 2021). Currently, the function of a small fraction of lncRNAs in HCC has been clearly defined, such as DNA binding (Zhou et al., 2020), associating with proteins (Gou et al., 2018), RNA interaction (Zhou and Xia, 2020), and producing small peptides (Wu et al., 2020). Intriguingly, some HCC-related lncRNAs have been founded in bodily fluids, making them attractive alternative biomarkers for HCC (Burenina et al., 2021). Studies have shown that these lncRNAs serve as potential cancer biomarkers and therapeutic targets for treating HCC (Klingenberg et al., 2017; Ghafouri-Fard et al., 2020; Matboli et al., 2020; Sheng et al., 2021).

Recently, there has been increased interest in bibliometric studies. For example, Li et al. explored the research hotspots of external beam radiotherapy in prostate cancer (Li R. et al., 2021); the hotspots of the role of anesthesia on tumor prognosis was performed by Luo et al. (2021); Martynov et al. conducted a bibliometric study to find the hotspot trend related to neuroblastoma research (Martynov et al., 2020). However, few studies are available for the research hotspots of lncRNAs in HCC development by using bibliometric methods. Bibliometric analysis regarding the application of lncRNAs in HCC research can provide more detailed insights into how lncRNAs play a critical role in HCC development. In recent years, bibliometric approach has been used as the method for quantitative and qualitative analysis (Li Q. et al., 2021). Researchers may have a better understanding of research trends and research hotspots in particular areas from bibliometric analysis (Carollo et al., 2021; Kilicaslan et al., 2021). Thus, we preferentially used CiteSpace to explore the frontiers of the roles of lncRNAs in HCC development. The results would provide scientists with a valuable reference for future studies.

All relevant publications were retrieved from the Web of Science Core Collection (WoSCC) database with the following search strategies: TS = (Long non coding RNA OR LncRNA OR Lnc RNA OR Long noncoding RNA) AND TS = (Hepatocellular carcinoma OR Hepatocarcinoma OR HCC patient OR HCC cancer OR Hepatic carcinoma) AND Document types = (Article OR Review) AND Language = English, with a limited time frame from 2010 to 2020, index = Science Citation Index Expanded (SCI-EXPANDED). The data extraction was completed for all included studies within the day on August 20, 2021. In total, 2,667 publications were finally included. The data were retrieved by one author (ZL) and double checked by another author (NT). We then cleaned the data, such as merging “TAIWAN” into “China” in country cooperative analysis, unifying “long noncoding rna” and “long non-coding rna” as “LncRNA” in keyword co-occurrence analysis, and so on.

The CiteSpace version 5.6.R4 was used to conduct bibliometric analyses (Chen, 2006). Some valuable parameters were included, such as publication number, total number of citations, impact factor (IF), citation burst, and centrality. Productivity was calculated for individuals, countries, and institutions on the basis of the total number of publications. The total number of citations was used to evaluate the international impact on both authors and articles (Chen et al., 2020). IF was used for assessing the international impact of journals, and it was obtained from the 2020 Journal Citation Reports (Pacey, 2020). Burst keywords referred to keywords which were cited frequently over a period of time, and they were considered indicators of research frontiers (Zhu et al., 2020). Centrality was a critical parameter for measuring the importance of nodes in a network. The nodes with a higher centrality (≥0.1) in the network were highlighted using purple rings and were usually regarded as high influence in a network (Liang et al., 2017).

The retrieved data were used for bibliometric analysis, and analyses were performed with the software CiteSpace version 5.6.R4, such as cooperative analysis, document co-citation analysis, keyword co-occurrence, timeline view of keywords, and citation burst analysis of keywords. The top 50 most cited articles in each time splicing with 1 year were selected to create a network. In order to highlight the key network, minimum spanning tree was used to explore the co-author citation. Minimum duration was set to 1 in the burst analysis of keywords.

To the best of our knowledge, this was the first study to explore the mapping intellectual structure for lncRNAs and HCC development by using CiteSpace. The number of published papers related to the field of lncRNAs in HCC development increased sharply in recent years. We found the knowledge base for the field of lncRNAs and HCC development by combining co-citation analysis with co-word analysis.

Regarding the top 10 prolific countries, four were developing countries, and six were developed countries. In terms of absolute publication numbers, China was the leading country, followed by the United States; this indicated that China had made significant progress in this area. However, China had a lower centrality value compared to the United States; it suggested that although China was the most prolific country in producing research publications, the United States played a leading role in collaborative network in this research field. There were extensive collaborations between western countries in the collaboration network. Transatlantic collaborations were the strongest among the United States, China, Germany, and Italy. The top 10 institutions were all from China. Among them, the papers published by Nanjing Medical University were cited with the highest frequency. For example, a study showed that Lnc-EGFR could suppress cytotoxic T lymphocyte (CTL) activity, stimulates Treg differentiation, and promotes HCC development through an EGFR-dependent signaling pathway (Jiang et al., 2017). Moreover, Zhuo et al. suggested that MEG3 serves as a prognostic biomarker for HCC (Zhuo et al., 2016). In addition to that, a previous study suggested that lncRNA ANRIL, as a cell growth modulator, may act as a novel potential therapeutic target for HCC (Huang et al., 2015).

In terms of the top 10 co-cited journals, six of the journals were from the United States. It was noteworthy that none of the top 10 co-cited journals were from China. However, Oncotarget was dropped by Science Citation Index Expanded. This demonstrated that these papers published in the journal need to be further improved. Moreover, impact factor ranged from 3.24 to 508.702. Additionally, this field had been more widely studied, and it covered more scientific disciplines in recent years, such as oncology (Lai et al., 2019; Lim et al., 2019; Wei et al., 2019), cell biology (Li et al., 2019; Song et al., 2019), and biochemistry molecular biology (Du et al., 2019; Guo et al., 2019). In the dual map, papers published in the molecular/biology/genetics journals were mostly cited by the studies published in molecular/biology/immunology and medicine/medical/clinical journals.

As for the top 10 prolific authors, each of them contributed to at least 11 publications. However, only one of the top 10 prolific authors was included in the top 10 co-cited authors, indicating that the authors should consider improving the quality of their articles in future. Co-cited authors include Yang F, who suggested that lncRNA-HEIH served as an oncogenic lncRNA that promotes tumor progression in HCC (Yang et al., 2011); Gupta RA, who indicated that lncRNAs played a positive role in regulating the cancer epigenome and could be the important targets in the diagnosis and therapy of HCC (Gupta et al., 2010); Torre LA, who reported that a large proportion of HCC cases and deaths could be effectively avoided by averting with appropriate preventive measures, such as vaccination, the use of early detection tests, and tobacco control (Torre et al., 2015); and Ponting CP, who explored the evolution of lncRNAs contributions to epigenetic gene regulation and transcriptional regulation (Ponting et al., 2009). Even though these authors were not prolific authors, they had already made significant contributions in this field.

Burst keywords were considered to discover emerging trends or research frontiers during a period of time (Chen, 2006). In this study, CiteSpace 5.6.R4 was used to capture the strongest citation bursts. At exploration stage (2010–2013), since the lncRNA HOTAIR had been proven to promote breast cancer metastasis by participating in chromatin remodeling (Gupta et al., 2010), researchers paid a lot of rising attention on lncRNAs. More and more functions of lncRNAs had been found. For example, lncRNAs could influence the development or progression of cancer (Ling et al., 2018; Chen and Xia, 2019). Subsequently, both the primary and secondary structures of lncRNAs had been discovered by enzymatic probing (Ilik et al., 2013) or chemical probing (Novikova et al., 2013; Smola et al., 2016). Researchers went one step further to divide lncRNAs into intronic, intergenic, sense, antisense, and bidirectional according to their genomic location relative to protein coding genes (Batagov et al., 2013).

At the rapid development stage (2014–2018), the mechanism, function, and application of lncRNA had been further studied. Based on the different mechanisms of lncRNAs acting on cellular processes, they are divided into five categories, such as signal (Pandey et al., 2008; Wang and Chang, 2011), decoy (Hung et al., 2011), guide (Wang and Chang, 2011), scaffold (Wang and Chang, 2011), and enhancer (Goyal et al., 2018; Liao et al., 2018). Moreover, some existing studies have suggested that the various functions of lncRNAs serve as critical roles in diverse biological processes, including epigenetic modification (Bhan and Mandal, 2014), transcriptional regulation (Wei et al., 2019), post-transcriptional regulation (Cesana et al., 2011), translational regulation (Yoon et al., 2012), and post-translational regulation (Wang et al., 2014). In addition, there had been more and more applications of lncRNAs discovered in cancer research. Initially, lncRNAs were reported to be associated with breast cancer (Shore et al., 2012). Some subsequent research were also reported in HCC (Shi et al., 2019), ovarian cancer (Oncul et al., 2020), and colorectal cancer (Huang et al., 2017). These lncRNAs could be used to work as novel prognostic biomarkers and intervention strategies for HCC.

At the stable-growth stage (2019–2020), with the rapid development of network public databases, there were many lncRNAs databases established, such as LNCipedia, LNCBook, TargetScan, Starbase (ENCORI), etc. Some of them were used to explore interactions (RNA–RNA interactions, RNA–DNA interactions, and RNA–Protein interactions) (Li et al., 2014) and pathway analyses (Qian et al., 2020), which were considered as a research frontier in the field of lncRNAs and HCC development. Furthermore, The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases could allow researchers to obtain raw data for free. These two databases were combined with some predictive or prognostic databases to perform bioinformatics analysis, such as UALCAN (Chandrashekar et al., 2017) and Kaplan–Meier Plotter (Hou et al., 2017), etc.

This analysis urges other researchers to discover the research hotspots and frontiers in the field of lncRNAs and HCC development from year 2010 to year 2020. A total of 2,667 records on this topic were initially found worldwide. The annual related publications output had increased dramatically over this time. Although China was the most prolific country in terms of research publication, the United States played a leading role in collaborative network. The Nanjing Medical University was the most productive institute in this field. Gang Chen was the most prolific researcher, while Yang F was the most frequently co-cited author. Oncotarget, Cell, and Oncogene were the most highly co-cited journals. The most recent burst keywords were interaction, database, and pathway.