Serum gamma-glutamyl transferase to alanine aminotransferase ratio predict the vascular invasion and outcome in hepatitis B virus related hepatocellular carcinoma

Zhifeng Zhao Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Jiayun Lin Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Xiaochun Ni Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Hongjie Li Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Lei Zheng Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Chihao Zhang Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Xiaoliang Qi Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Haizhong Huo Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Xiaolou Lou Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Qiang Fan Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Yongyang Bao Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital Meng Luo (  luosh9hospital@sina.com ) Shanghai Jiao Tong University School of Medicine A liated Ninth People's Hospital https://orcid.org/0000-0003-3975-7476


Introduction
In 2018, liver cancer was ranked as sixth most common cancer and fourth leading cause of cancer-related death in the world, with 841,000 new cases and 782,000 deaths each year [1], among which 85 ~ 90% cases belongs to hepatocellular carcinoma (HCC). Hepatitis B virus (HBV) is one of the major risk factors associated with HCC worldwide especially in China [2].
At present, the treatment of HCC mainly involves chemotherapy, surgical resection and liver transplantation [3]. However, high rates of metastasis and recurrence after operation deteriorate the prognosis severely, in this process vascular invasion accounts for the leading cause [4,5]. Kunutsor et al.[6] demonstrated that the intrahepatic vascular invasion occupied the prevalent cause of post-operative recurrence and cancer-related death in HCC patients. Therefore, accurate and effective assessment of vascular invasion before surgery is strongly needed to guide the treatment options.
Currently, preoperative assessment of vascular invasion is mainly performed by Computed Tomography (CT), while laboratory tests are equally indispensable for adjuvant assessment. Risk factors for HCC vascular invasion include HBV infection, tumor size, multifocal localization, α-fetoprotein (AFP), γ-glutamyl transferase (GGT), alanine transaminase (ALT) etc. [4,7,8]. As a major etiological factor, HBV infection changes the hepatic microenvironment, induces in ammatory response, promote the angiogenesis and vascular invasion. Several studies have con rmed the correlation between HBV infection and vascular invasion in HCC [9,10]. Therefore, it is possible to determining vascular invasion and poor prognosis by assessing HBV severity, in which GGT and ALT have been widely investigated with considerable potential.
High GGT/ALT ratio was initially found to be prognostically associated with worse condition and treatment response in viral hepatitis [11,12]. Further studies also indicated GGT/ALT as a positive predictor of HCC [13,14]. However, those previous studies emphasized more on the prediction of hepatitis or HCC, while the prognostic effect of GGT/ALT on vascular invasion in HBV-related HCC is still unknown. Considering the vital role of hepatitis B in vascular invasion of HCC combining with the diagnostic value of GGT/ALT in hepatitis, the aim of our study is to con rm GGT/ALT as the risk factor for vascular invasion, cancer severity and outcomes of HBV-related HCC patients. After the discharge, patients were followed up regularly every three months for the rst three years and every six months thereafter. Serum biomarkers of tumor and hepatitis B were examined periodically, and abdominal contrast-enhanced CT, ultrasound, hepatic arteriogram or invasive examination were performed as needed. Overall survival (OS) was de ned as the interval between the date of surgery and death or the last follow-up. Disease-free survival (DFS) was de ned as the time interval between the date of surgery and the date of con rmed HCC recurrence or the date of last follow-up. The last follow-up ended on July 31, 2020.

Method
The study was approved by the Ethical Committees of Zhongshan Hospital of Fudan University. Data collection The following clinical data of HBV infected HCC patients were collected in our study: 1) demographic data, including age and gender; 2) preoperative laboratory examination, including GGT, ALT, albumin, AFP and total bilirubin (TB); 3) Tumor information evaluated by imaging test, pathological examination and scoring system, including vascular invasion, tumor size, tumor multifocality, tumor capsule, pathological differentiation, cirrhosis, Barcelona Clinic Liver Cancer (BCLC) classi cation, Albumin-Bilirubin (ALBI) Grade; (4) Postoperative follow-up data including OS and DFS.

De nition
Laboratory examination data were manipulated by the laboratory Department of Zhongshan Hospital. The infection of HBV was de ned as positive test of hepatitis B virus surface antigen (HBsAg). The normal ranges of laboratory test include ALT ≤ 56 U/L, GGT ≤ 50 U/L, albumin ≤ 40 g/L, TB ≤ 20 µmol/L, and AFP ≤ 400 ng/mL, respectively. Tumor size, tumor number, tumor capsule, pathological differentiation, cirrhosis and vascular invasion were con rmed by CT imaging combined with postoperative pathological examination. Each specimen was reviewed independently by two liver pathologists. Microvascular invasion (MVI) is de ned as microscopically con rmed tumor clusters into the vascular cavity [4]. The assessment of BCLC classi cation was performed as described previously [15]. Pathological differentiation was conducted using the Edmondson grading system [16]. ALBI grade was employed as a powerful tool for survival analysis in HCC patients, with speci c calculation process as described previously [17].  Table 2.
Subsequently, we plotted the ROC curve of GGT/ALT ratio in the diagnosis of vascular invasion (Fig. 1). The optimal cut-off value of GGT/ALT in the diagnosis of vascular invasion was 2.95 by calculating the Youden index of the ROC curve. Based on this cut-off value of GGT/ALT, we classi ed patients into low group (n = 365) and high group (n = 193).
Next, by using subgroup analysis, we explored the heterogenetic prediction value of GGT/ALT in different subgroups (  Table 5). The DFS curve of GGT/ALT yielded that the DFS rate of higher GGT/ALT group was signi cantly lower similarly with the OS rate (P < 0.0001), as shown in Fig. 3. The 5-year DFS survival rates of the high and low GGT/ALT group were 43.35% and 58.70%, respectively.

PSM of GGT/ALT for vascular invasion prediction
To avoid the interference of other clinical parameters, PSM was used to match the data of high and low GGT/ALT groups. After the matching, there were 93 patients equally in the low and high GGT/ALT group. In   Table 7. The results con rmed the GGT/ALT as a good predictive value for vascular invasion.

Discussion
Our study demonstrated the GGT/ALT ratio as an independent predictive biomarker for vascular invasion in HBV-related HCC. After grouping patients with cut-off value of 2.95, high GGT/ALT group showed positive predictive value for vascular invasion, higher tumor severity, lower DFS and OS in HCC patients. HBV infection leads to multiple pathophysiological alteration including DNA oxidative damage, liver cell necrosis, in ammation responses, cytokine synthesis and release, brosis and tumor tumorigenesis [19]. In this process, HBV X protein (HBx) has been found related with the MVI development involved in postoperative recurrence [20,21]. Yang et al. [10] found that HBV-positive patients were more prone to develop vascular invasion in HCC. In this study, HBV was found to change the intrahepatic microenvironment through the transformer growth factor beta(TGF-β) -miR34a -C-C motif chemokine ligand 22 (CCL22) pathway, thereby promoting the angiogenesis and the development of HCC. In further clinical studies, Lei et al. [9] discovered that preoperative HBV DNA load larger than 104 IU/mL was an independent risk factor for vascular invasion in HBV-related HCC. Wei et al. [22] further summarized that both infection and active replication of HBV were associated with the in ammatory injuries, occurrence of vascular invasion and cancer metastasis in liver cancer. Therefore, the aim of our study is to predict the vascular invasion and outcomes of HCC via the assessment of HBV infection.
Among the examinations, GGT and ALT are adopted most broadly in hepatitis evaluation [23][24][25][26][27]. GGT could mediate the production of ROS, promote cell growth and proliferation [28], and is often elevated in liver diseases such as hepatitis, brosis and cancer [24,29]. ALT is also known as a marker of liver dysfunction and in ammation, which was found associated with recurrence and poor survival of HBV-related HCC [30,31]. However, the changes of different biomarkers were not equally accompanied with the uctuation of in ammatory responses, therefore the combination of GGT/ALT exerts better predictive accuracy in hepatitis compared with using alone [32]. Ebiling et al. [11] disclosed that higher GGT/ALT ratio could predict the worse prognosis in chronic hepatitis C. The research of Tarantino et al. [12] further supported this idea by the discoveries that a lower GGT/ALT ratio is an independent predictor of anti-virus therapy response. Consequently, GGT/ALT could re ect the hepatitis level both in primary disease and therapy responses, which made it possible in the prediction of hepatitis-induced HCC.
Yang et al. [14] found the incidence of primary hepatic carcinoma increased markedly when GGT/ALT elevated in chronic hepatitis B patients. Ju et al. [13] furtherly discovered that increased GGT/ALT ratio was closely related to tumor burden in HBV-infected HCC with Child-Pugh A class, including tumor size, vascular invasion, tumor capsule, and poor prognosis. Those studies proved the predictive potential of GGT/ALT not only in hepatitis but also in hepatic carcinoma. Combining the pathological role of HBV infection in vascular invasion thus in HCC progression, and the prediction value of GGT/ALT in both hepatitis B and HCC, we conducted this study to con rm GGT/ALT as an independent predictive factor of vascular invasion and outcomes in HBV-infected HCC.
As mentioned above, GGT/ALT have the ability to predict vascular invasion in HCC. However, different distribution of characteristics among HCC patients increased the heterology and might lead to false positive results. Therefore, sensitivity analysis including subgroup analysis and PSM was conducted. In subgroup analysis, the prediction value of GGT/ALT was examined effectively in most subgroups. The insigni cant predictive value in other subgroups might be attributed to relatively small amount of cases and this could be con rmed through larger database in future. Moreover, PSM was employed to balance the distributions of variables between high and low GGT/ALT groups. After the matching, GGT/ALT was still listed as an independently risk factor of vascular invasion, which proved its prediction value of the ratio solidly.
By now, there is no authoritative explanation for the prediction value of GGT/ALT. One possible hypothesis is proposed as follows: GGT is affected by both in ammation and tumors, while ALT is more responsive to the hepatic in ammation and liver functions relatively. Although the hepatitis in HCC patients has often been controlled before the resection, GGT and ALT might still be interfered by residual HBV infection or surgical stimulus. Therefore, the ratio of GGT/ALT could re ect the stages of primary tumor progression more precisely by minimizing the interference of in ammation. Besides, decreased ALT also indicated increment mortality rate in people aged over 60 years because of the depletion of liver function reserve, which could explain the positive correlation between GGT/ALT ratio and worse prognosis in another aspect [33].
Some scoring systems, such as BCLC classi cation or ALBI grade, have also exhibited good predicting values in HCC. However, those scoring system is relatively di cult to address, which make it limited in clinical use. As a powerful complement to imaging examination and scoring systems, our study provided a simple and feasible monitoring tool to assess the vascular invasion and outcome of HCC, which make it easier in the patients' management.
This study still had several limitations. Firstly, this study only involves a single center. More cases from different centers and regions are required in the future. Secondly, this study is a retrospective study, which may lead to selection bias. It may be further veri ed by prospective cohort studies in the future.

Conclusion
Our study proposed a relatively reliable evidence in proving the prognostic value of GGT/ALT. As a cheap and Committees, in which the patient data used were kept strictly con dential.

Consent for publication
Not Applicable Availability of data and material The datasets generated and analysed during the current study are not publicly available as the data are being used in next study, but are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study is nancially supported by the National Natural Science Fund of China (Project No. 81970526, 81900550 and 81770599).
The funding body had no role in the design of the study and collection, analysis, interpretation of data or preparation of the manuscript.
Authors' contributions ZZ, YB and ML was involved in the plan of program and drafted the manuscript. JL and XN participated the data collection and analysis. HL and LZ participated the data collection. CZ and XQ made substantial contributions to the manuscript and the analysis of data. HH, XL, QF and ZZ participated in drafting or revising the work. All authors have given nal approval of the version to be published. All authors agreed to be accountable for all aspects of the manuscript.       The analysis indicated that the high GGT/ALT resulted in poor overall survival rates.