Risk Factor of Microvascular Invasion in Non-metastasis Single Lesion Hepatocellular Carcinoma

Positron Emission Tomography; AFP: Alfa-Feto-protein ALT: Alanine Transaminase; CEA: Carcinoembryonic Contrast Ultrasonography; Computerized Tomography; Gd-EOB-DT-PA-MRI: Gadolinium-ethoxybenzyl-dieth-ylenetriamine penta-acetic acid HBV: Hepatitis B Virus; HCC: Hepatocellular Carcinoma; HCV: C INR: Prothrombin Operating 95%CI: 1.040-2.077, P= .03) and tumor size ≥3.5 cm (OR=2.205, 95%CI: 1.123-4.333, P= .02) were independent risk factor of MVI. Conclusion: Poorly differentiated tumor and tumor size ≥3.5 cm are independent risk factor of MVI and could help us to predict the possibility of MVI in non-metastasis single lesion HCC patients.


Risk Factor of Microvascular Invasion in Nonmetastasis Single Lesion Hepatocellular Carcinoma
DOI: 10.26717/BJSTR.2020. 27.004558 Jonathan Hartanto Tan 1-4 , Ze Qian 1-4 , Diyu Chen 1-4 , Abdul Lateef Shaikh 1-4 and Shusen Zheng 1-4 * 1 treatment of HCC has been improve, but the long-term survival rates remain unsatisfactory [5]. Previous study found that the prognosis of HCC is depend on multiple factor such as the age of the patient [6], tumor size, number of the tumor, presence of tumor vascular invasion [7,8], serum Alfa-Fetoprotein (AFP) level [9], and presence extrahepatic metastasis [10]. Tumor vascular invasion was classified as macrovascular invasion, which grossly recognizable (mostly large and medium vessels) and Microvascular Invasion (MVI), which could only found in microscopic examination (mainly in small vessels such as portal vein branches, central vein in noncancerous liver tissue, and venous vessels in tumor capsule) [11,12]. Microvascular invasion also known for negative influence outcome following surgical treatment HCC, it has been reported to be one of the most important risk factor related to postoperative early tumor recurrence [13][14][15][16]. Lim et al. [17] reported that MVI is more prominent tumor recurrence and overall survival predictor than Milan criteria for HCC after surgical resection. Patients who fulfill Milan criteria should have single tumor not more than 5 cm or three or fewer tumors with the largest tumor not exceeding 3 cm, and no evidence of macrovascular invasion or extrahepatic metastasis [7]. Lim et al. [17] also concluded that there was no significant difference in overall survival rates of patients without MVI and not exceeding the Milan criteria relative to patients within Milan criteria. However, the overall survival rate decrease significantly with MVI. Therefore, an accurate preoperative prediction of MVI is important to make a treatment planning for impeding recurrence and improving outcome of the patients. The aim of this study was to analyze the correlation between preoperative clinicopathological data of the patients with histological result in hepatic resection single lesion non-metastasis HCC patients and try to found the risk factor that could help to predict MVI.

Patient Selection
Two hundred and ten patients with non-metastatic single lesion HCC underwent hepatic resection surgery at Department of Hepatobiliary Pancreatic Surgery, First Affiliated Hospital, School of Medicine, Zhejiang University, between 2017 and 2018, based on the following criteria: (1) hepatectomy as a treatment of HCC; (2) no adjuvant treatment before hepatectomy; (3) no invasive diagnostic procedure before hepatectomy; (4) no macrovascular invasion found; (5) complete resection with incised margin was negative; (6) without lymph node and distant metastasis; (7) complete preoperative data. All the data before surgery and histological data after surgery of every patients were collected for analysis of MVI.

Data Collection
Data of the patient before and after surgery were retrospectively collected. Some general clinical data like gender, age, history of hepatitis B, history of cirrhosis and child pugh status were collected.
Serum index like total bilirubin, albumin, Alanine Transaminase

Statistics
The statistical significance of differences of selected clinicopathology features between those with MVI and without MVI, for those continues variable were expressed as means ± standard deviation (SD) and compared using Student's t-test or Mann-Whitney test for variables with an abnormal distribution. All data will be dichotomized with a reliable cut-off value and will be presented in percentage and assessed by chi-square test. Variable at P-value < .05 on univariate analysis were subjected to multivariate analysis by fitting logistic regression model to identify independent predictors for MVI in non-metastatic single lesion HCC. The SPSS 23 statistical software was used to perform the statistical analyses.
Variable with P-value < .05 was considered statistically significant.

Determination of the Cut-Off Value
According to the Receiver Operating Characteristic (ROC) curve ( Figure 1), the cut-off value of tumor size for microvascular invasion was set to 3.5 cm (AUC=0.626, sensitivity=0.623 and 1-specificity=0.396). Thus, the patients were dichotomized into group of tumor size larger than 3.5 cm and smaller than 3.5 cm.

Univariate and Multivariate Analyses of MVI-Related Factors
Comparisons of clinical characteristic in univariate analysis between MVI (-) and MVI (+) group are summarized in Table 2.
There were no significant difference between patient with MVI

Discussion
Previous study shown that presence of MVI is one of the most important risk factor related to postoperative tumor recurrence and a prognostic factor associated with lower survival rate [13][14][15][16].
This make prediction of MVI become important for surgeon to make a good planning treatment for HCC patients. Therefore, this study was designed to investigate the risk factor of the microvascular invasion in HCC patient to help predict the MVI. In our results, Some published studies found the presence of MVI is closely related to tumor histological differentiation [18]. Du et al. [19] found that tumor histological grade is a strong predictor of MVI and also an important prognostic factor for survival of Small Hepatocellular Carcinoma (SHCC) patients. Esnaola et al. [20] also said that patient with poorly differentiated or undifferentiated tumors were 6 times more likely to have MVI than those with well differentiated tumor. However, even tumor differentiation is an independent risk factor of MVI, the current data of tumor differentiation was taken from histology examination after surgical procedure. So it is hard to make this variable to be a predictor of MVI before the surgery.
Pawlik et al. [21] demonstrate that needle core biopsy of HCC to determine histologic grade was inaccurate due to heterogenous tumor differentiation resulting in sampling error. In addition, there is a risk of needle-tract implantation as a complication of biopsy [22]. Therefore, to found another way to differentiate tumor cell before the surgery could help it to become an important predictor variable for the MVI.
Most studies found that tumor size is an independent predictor of MVI in HCC patient, Esnaola et al. [20] found that patient with tumor larger than 4 cm were 3 times more likely to have MVI than those with tumor diameter 4 cm or less. Yamashita et al. [23] from their study of MVI in SHCC patients also found that tumor diameter larger than 2 cm could be a predictor marker of MVI. Our study also parallel with those previous studies, we found that tumor size larger than 3.5 cm is an independent risk factor of MVI in non-metastasis

Conclusion
According to the study, we found that it is still possible to predict MVI in HCC before the surgical treatment. We found that poorly differentiated tumor and tumor size ≥3.5 cm are independent risk factor of MVI and could help us to predict the possibility of MVI in non-metastasis single lesion HCC patients. Tumor size could be examine before the surgical procedure with some imaging modality like CT scan or MRI. However, the definitive way to find tumor differentiation is only from histological finding, it is hard to determine it before the surgical procedure. With this few possible factors from this study, we hope it could be a trigger for another research to found a new biomarker or another method to predict MVI in HCC accurately.

Conflict of Interest
All the authors declare that there are no conflicts of interest concerning this study.