Based on our extensive literature search, this is the first comprehensive and quantitative analysis to provide a systematic research data on the literature related to HCC-MVI. This research provides a wide and visual analysis of publications in this field, which may help scholars gain a basic understanding, develop areas of focus, identify trends and pursue further practise in this field.
The top 10 countries that published articles in this field included three Asian countries, five European countries and two American countries, accounting for 62.5%,14.81%,11.67% of all documents, respectively .Furthermore, our investigation revealed that more than 40 percent of the countries had more than 100 publications. China, the only developing country in the top 10,also contributed more than 40% of documents,indicating its high academic impact on the HCC-MVI field.However, although they cooperated extensively with other countries, especially the US, their global cooperation was not strong enough (e.g. with Europe).Besides this,another remarkable phenomenon was that nine Chinese universities were among the top 10 institutions that contributed to publications, showing the dominant position of Chinese researchers in this field. Additionally, the incidence of HCC in Asia was higher than in Europe and America, which may be why China had the most prolific research on this topic.
Notably (Table 2), all of the top 10 active authors published at least 17 articles. In the list, however, although eight of the ten leading contributors were Chinese, which was consistent with the results of institutional research, none of them were reflected in the top 10 co-cited authors who generated high-quality articles and were more influential.As the highest co-cited author,Bruix J(the Barcelona Clinic Liver Cancer Group, University of Barcelona,Spain) contributed extensively to developing HCC management guidelines,and completed numerous of high-quality studies[7–9].Meanwhile,Llovet JM and Mazzaferro V made notable contributions to the HCC field[10–12].
Based on academic journals(Table 3),although the retrieved ones in this field were not high-impact factor journals, Frontiers in Oncology had the highest IF and published the highest number of articles.Nevertheless,they received the lowest number of citations in the top 10.Our study also revealed that although Annals of Surgical Oncology ,published original and educational manuscripts on oncology for surgeons from all specialities in academic and community settings,they merely ranked 4th,and then got the highest number of citations.The clear difference between the two journals might be related to the lower open-access articles.Remarkably, however, since 2020,the number of articles published by Frontiers in Oncology and European Radiology increased substantially.Based on co-cited references,as the head of the international agency for research on cancer,Bray F,who was not a highly co-cited author(Table 2),comprehensively described the epidemiological characteristics of malignancy,and had the highest number of co-citations in 2018.On the other hand, we observed that the Chinese published only three documents (Table 4).As a Chinese author with the most cited references,Lei Z,published a paper on ‘nomogram for preoperative estimation of microvascular invasion risk in hepatitis b virus-related hepatocellular carcinoma within the Milan criteria’,which elaborated how to construct a nomogram using clinical and imaging information and achieved an optimal preoperative prediction of MVI in HBV-related HCC within the Milan criteria.What is noteworthy is that the document,only ranked 6th in the top 10 co-cited references,written by an accomplished medical epidemiologist,Torre LA,and published in the journal of Ca-A Cancer Journal for Clinicians,with the highest IF 508 in academia.Simultaneously, while this study was the most cited from a Q1 journal, it was mainly a review based on a WoS ranking, which may be why it has been so widely cited over the past decade.
Research hotspots and frontiers
Keywords and burst keywords(Figs. 10 and 11) reflect the research’s focus.A decade ago,several scholars described how to predict MVI and interpret its effects based on prognosis.For example,Shirabe et al.revealed tumour size, serum des-gamma-carboxy prothrombin (DCP) levels and the maximum standardised uptake value (SUVmax) on 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography were independent clinical predictors of MVI after multivariate analyses.The doctors and radiologists at the centre for imaging science of Sungkyunkwan university,also combined two or more of the following, arterial peritumoral enhancement, a non-smooth tumour margin, and peritumoral hypointensity at the hepatobiliary phase,as preoperative imaging biomarkers to predict MVI through gadoxetic acid-enhanced magnetic resonance imaging. These biomarkers which are associated with early recurrence after curative resection of single HCC.Zhao et al. constructed a scoring system based on AFP/GGT level, tumour diameters and tumour number.Their results indicated when the score was greater than 3,the incidence of MVI increased by five times, whereas the overall survival rate decreased significantly.Meanwhile, the“Milan criteria”,which had the strongest citation burst(5.39), was the focus of the field from 2013 to 2018.Furthermore,several clinicians deemed MVI was one of the independent risk factors for tumour recurrence in HCC, classified within the Milan criteria.Moreover,esearch on similar articles showed a rapid growth trend, which may have accounted for the frequent keywords ‘survival’, ‘recurrence’, ‘risk factor’ and ‘Milan criteria’.We also observed that ‘gamma carboxy prothrombin’, also termed Protein Induced by Vitamin K Absence or Antagonist-II(PIVKA-II), was associated with early recurrence after curative resection of a single HCC and has been extensively applied in the MVI field.
Burst keywords refer to those widely cited in articles within a particular period, which are considered another significant indicator of research hot spots or emerging trends. For example, as shown in Fig. 11, research hotspots and frontiers in the MVI included ‘image’ and ‘tace’ within the last three years.
1. Image: An important issue affecting HCC prognosis is the absence of a highly reliable factor to predict MVI preoperatively.Thus,several image radiomics studies have been incorporated into the MVI report.For example,Jiang et al.developed predictive models using Xtreme Gradient Boosting (XGBoost) and deep learning based on computerised tomography images to predict MVI preoperatively.However,Li et al.verified through the promising technique of magnetic resonance elastography-based shear strain mapping that the low interface shear strain identified at tumour-liver boundaries was highly correlated with the presence of MVI.Except that,scholars have evaluated the maximum standardised uptake value (SUVmax) of preoperative fluorine 18 fluorodeoxyglucose positron emission tomography/computed tomography (F-FDG PET-CT),and summarised the combination of the two risk factors(SUVmax and AFP-L3),providing a reliable assessment for predicting MVI,with the sensitivity and specificity of 88.9% and 82.4%,respectively.In addition to this finding,based on a retrospective study,Xu et al. developed a computational approach integrated 15 clinical factors, and 12 imaging features from contrast-enhanced CT.Their study demonstrated good performance in predicting MVI and clinical outcomes,using multivariate logistic regression.
2. transarterial chemoembolization(TACE): As one of the important non-surgical treatment methods for HCC, TACE can reduce tumour recurrence rates and increase the overall survival rates,especially in patient in advanced stages[2, 24].The controversial focus of these studies was to investigate whether postoperative-TACE (p-TACE) improved the prognosis of patients with MVI.For example,Yang et al.included 13 studies in which 824 patients received p-TACE,which significantly improved overall survival(OS) and recurrence-free survivals(RFS) compared with postoperative conservative treatment in patients with HCC accompanied by MVI after curative resection.Similarly,Li et al.considered p-TACE as a safe intervention which reduced tumour recurrence rates and improved the OS for MVI using high-quality meta-analysis.Besides,the recent results showed that only patients with BCLC stage A, CNLC stage Ib, and AJCC stage II both were found to be benefited more from p-TACE, but not patients with BCLC stage 0/B, CNLC stage IIb, and AJCC stage IIIa.Contrastingly, they declared that p-TACE was not inferior to more cycles when improving prognosis, strongly recommending only one cycle following R0 resection[27, 28].
Research strengths and limitations
To our knowledge, this study is the first to systematically research the literature related to HCC-MVI.Our study also provided a comprehensive and quantitative analysis of the most significant MVI articles, acknowledging the key contributions made to the evolution and advancement of this specialised field.However,we encountered some limitations.First,the data analysed in the study were only from WoSCC-SCIE,which excluded other databases,such as PubMed,Scopus and Embase.Moreover,we partially selected documents (articles and reviews) and defined the focus language as English, resulting in some linguistic bias.Therefore, since data are typically prone to frequent changes, our results only reflected the current state of HCC-MVI research. Hence, data generated from studies published in the current year (2022) were incomplete in our analysis. Finally, self-citations are a significant problem that can influence these results.