Early Recurrence of Hepatocellular Carcinoma in Patients without Microscopic Vascular Invasion: Clinicopathological Characteristics and Risk Factors

Introduction

Hepatocellular carcinoma (HCC) is one of the most common and lethal malignancies worldwide, accounting for approximately 80% of primary liver cancer and contributing to over 700,000 deaths annually.^1,2 The incidence and mortality rates of HCC are closely associated with the prevalence of chronic liver diseases, particularly viral hepatitis, cirrhosis, and non-alcoholic fatty liver disease.³ Due to its heterogeneity in etiology, molecular characteristics, and clinical behavior, HCC remains a challenging malignancy to manage.⁴

The primary treatment modalities for HCC include surgical resection, liver transplantation, local ablative therapies, and systemic therapies. However, the prognosis for HCC patients remains poor, primarily due to the high rate of tumor recurrence.⁵ Early recurrence, within two years after curative treatment, is common and associated with unfavorable outcomes. This early recurrence is often attributed to intrahepatic metastasis, but may also result from multicentric tumor development, underlying liver disease, and tumor invasion at the resection margin.⁶

Several clinicopathological factors have been identified as predictors of early recurrence in HCC patients, including tumor size, multifocality, histological grade, vascular invasion, serum alpha-fetoprotein levels, and liver function parameters. Among these, microscopic vascular invasion (MVI) is recognized as one of the most important independent risk factors leading to early recurrence and reduced overall survival.⁶ MVI refers to the microscopic presence of tumor cells within blood vessels in either the tumor or adjacent liver parenchyma. MVI has been reported in approximately 30–60% of HCC patients undergoing surgical resection or liver transplantation.^7,8

While MVI-positive patients are at high risk of recurrence, early recurrence can still occur in MVI-negative cases.⁹ Identifying risk factors for recurrence among HCC patients with negative MVI can facilitate the detection of high-risk patients who could benefit from closer surveillance and adjuvant therapies post-treatment. Therefore, this study aims to determine the clinical and pathological factors associated with early recurrence in HCC patients without MVI who have undergone surgical resection and to analyze the clinical and pathological determinants influencing this recurrence.

Materials and Methods

Patient Selection and Data Collection

This retrospective cohort study included all 69 hCC patients without MVI who underwent liver resection at the University Medical Center, Ho Chi Minh City between January 2018 and December 2018. Patients were excluded if tissue samples were inadequate for histopathological evaluation, or if they were lost to follow-up immediately after surgery, or if the final diagnosis was not HCC (Figure 1).

Figure 1 Flowchart of patient selection.

All clinical, paraclinical and follow-up data were extracted from electronic medical records. Patients were followed every 3–6 months for 24 months postoperatively, including abdominal ultrasound and serum AFP measurement at each visit. If any abnormalities were detected, such as elevated AFP levels or suspicious ultrasound findings, patients underwent a contrast-enhanced CT scan of the abdomen. When new lesions exhibited typical imaging features of HCC, the case was reviewed in a multidisciplinary tumor board composed of oncologists, radiologists, pathologists, and surgeons. The diagnosis of tumor recurrence was based on the consensus of this multidisciplinary discussion and was documented in the electronic medical records. In this study, early recurrence was defined as HCC recurrence confirmed by the tumor board and occurring within 24 months after surgery, consistent with definitions used in previous studies.^6,10,11

Pathological Assessment

Tumor characteristics, including size, Edmondson-Steiner grade, histological subtype (classified according to the 2019 WHO classification),¹² necrosis, mitotic index, and MVI were assessed. MVI was defined as the presence of clusters of tumor cells within a vascular lumen lined by endothelium, identified on hematoxylin and eosin (H&E) stained sections. For each case, all tumor-containing blocks (typically three to five per case) were examined. All cases were reviewed by the same pathologist to ensure consistency in MVI evaluation.

Non-tumor liver tissue, sampled at least 1 cm from the tumor margin, was evaluated for inflammation and fibrosis using the Ishak-Knodell scoring system,¹³ and steatosis was categorized as none (<5%), mild (5–33%), moderate (34–66%), or severe (>67%).

Statistical Analysis

Statistical analyses were performed using Stata software. Continuous and categorical variables were summarized as means ± standard deviations and frequencies/percentages, respectively. Univariate and multivariate Cox regression analyses were conducted to identify factors associated with early recurrence. Variables with a p-value < 0.1 in univariate analysis or those considered clinically relevant were included in the multivariate model. Due to the retrospective nature of the study, missing data were present for some variables. However, since the missingness was likely not completely at random and the overall sample size was relatively small, we did not apply any imputation methods. Instead, patients with complete data for all selected covariates were included in the multivariate analysis. We did not apply automated selection procedures such as stepwise, backward, or forward selection. Instead, we used a manual approach to construct a parsimonious multivariate model based on variables with univariate significance. Given that our study was exploratory in nature and not designed to test a specific hypothesis, the multivariate analysis was intended to describe potential associations rather than to formally test hypotheses. A p-value < 0.05 was considered statistically significant.

Results

Among the 69 patients included in the analysis, 22 (31.9%) experienced tumor recurrence within two years post-surgery, while 47 (68.1%) remained recurrence-free. Most recurrences occurred between 6 and 18 months. The recurrence rate over time is illustrated in Figure 2.

Figure 2 Probability of tumor recurrence in patients after HCC surgery.

Clinical and Paraclinical Characteristics

The clinical and laboratory characteristics of the cohort are summarized in Table 1. The majority of patients were male (78.3%), with a mean age of 56.9 ± 12.8 years. Most had ALBI grade 1 (66.2%) with normal AST (85.3%) and ALT (80.9%) levels. AFP levels exceeded 200 ng/mL in one-third of cases. Patients with advanced BCLC stage at surgery had a more than four-fold increased risk of early recurrence compared to those with early-stage disease (HR = 4.23, 95% CI = 1.82–9.87, p < 0.001). A marginally statistical difference was observed in HCV infection where a recurrence group had a higher proportion of HCV infection compared to the non-recurrence group (47.4% vs 25.6%, p = 0.080). No statistically significant differences were found in other clinical characteristics among patients with early recurrence and those without early recurrence.

Table 1 The Association Between Clinical Features and Early Tumor Recurrence in Surgery HCC Patients

Histopathological Features

The histopathological characteristics of the tumor and non-tumor liver tissue are presented in Table 2. Tumor size >5 cm was observed in approximately 25% of cases and were significantly associated with early recurrence (HR = 3.24, 95% CI = 1.38–8.00, p = 0.007). Conventional HCC histology was predominant, with over 70% of cases lacking adverse histological features such as necrosis and high mitotic rate. Most HCCs developed in cirrhotic livers, with a significant percentage of hepatitis. HCV-infected patients had a higher tendency to develop cirrhosis (89.5% vs 61.5%, p = 0.059). However, non-tumor liver characteristics were not associated with early recurrence.

Table 2 The Association Between Clinical Characteristics and Early Tumor Recurrence in Surgery HCC Patients

Multivariate Analysis

Based on findings from Tables 1 and 2, multivariate analysis was conducted to include patient age, HCV infection, BCLC stage, preoperative treatment, and tumor size. Preoperative treatment was the strongest predictor of early recurrence (HR = 8.63, 95% CI = 1.45–51.38, p = 0.018). Other independent risk factors included HCV infection (HR = 4.61, 95% CI = 1.41–15.1, p = 0.012) and tumor size >5 cm (HR = 3.82, 95% CI = 1.17–12.42, p = 0.026). Although BCLC stage was statistically significant in the univariate analysis (HR = 4.23, p < 0.001), it did not remain significant in the multivariate model (HR = 2.66, 95% CI = 0.96–7.39, p = 0.061) (Figure 3).

Figure 3 Multivariable Cox regression of factors independently associated with tumor recurrence.

Discussion

Although microscopic vascular invasion (MVI) is a well-established predictor of early HCC recurrence,^14–16 HCC patients without MVI may also have high risk of early recurrence. In this study, nearly one third of HCC patients without MVI had early recurrence, mostly between 6- and 18-month post-surgery. Independent prognostic factors were HCV infection, tumor size >5 cm, and preoperative local/regional treatment. Our findings align with prior studies. Sumie et al reported a 27.7% early recurrence rate in MVI-negative patients, with no recurrences in the first few months after the surgery, unlike MVI-positive patients who experienced rapid recurrence.¹⁷ Yu et al found a higher recurrence rate of 41.8% in MVI-negative patients, with over 20% recurring within three months.¹⁸ However, it should be noted that while Sumie’s and our study’s MVI status was assessed by a pathologist, Yu’s study relied solely on information from the pathology report form and excluded cases where the MVI status was not reported.

Previous studies have shown that HCV infection is one of the significant risk factors for HCC development.¹⁹ In our study, HCV infection was identified as an independent risk factor for early recurrence among MVI-negative patients. This suggests that HCV may contribute directly to tumor recurrence through mechanisms beyond liver cirrhosis alone. Indeed, HCV-positive patients in our cohort had a higher prevalence of cirrhosis, which may partially mediate this association. However, chronic HCV infection itself has been implicated in hepatocarcinogenesis through sustained inflammation, immune evasion, and genomic instability.^20,21 Previous studies have also raised concerns about recurrence risk following direct-acting antiviral (DAA) therapy, though this remains controversial.^22,23 As we did not collect treatment data, we were unable to assess the potential impact of antiviral therapy in our cohort. Further studies incorporating both virological and treatment-related variables are needed to clarify the role of HCV in post-surgical recurrence. Tumor size is a well-known predictor of recurrence and survival in HCC patients. Larger tumors are associated with increased recurrence risk due to higher rates of poor differentiation, genetic alterations, and positive resection margins.^24,25 In this study, tumor size >5 cm was identified as an independent risk factor for early recurrence (HR = 3.82, 95% CI = 1.17–12.42, p = 0.026). This result is consistent with Yu et al, who demonstrated a significant correlation between tumor size >5 cm and early recurrence (HR = 1.584, 95% CI = 1.295–1.937, p < 0.001).¹⁸ Conversely, Zhang et al found that while larger tumors were associated with recurrence, tumor size was not an independent prognostic factor.²⁴ Shehta et al also highlighted the increased likelihood of positive surgical margins in larger tumors, rising from 9.1% in tumors <5 cm to 23.3% in tumors >10 cm.²⁵ Due to the retrospective nature of our data and financial constraints, a comprehensive evaluation of tumor margin status was not feasible.

In our study, although only three patients received preoperative treatment, this was significantly increased recurrence risk. Preoperative interventions, such as trans-arterial chemoembolization (TACE) and radiofrequency ablation (RFA), may influence tumor biology by altering the microenvironment, increasing inflammatory cytokines (IL6, Th2), and promoting tumor progression through hypoxia-induced signaling pathways.^26,27 Additionally, TACE/RFA selectively targets well-differentiated tumor cells, potentially leading to the survival of more aggressive tumor clones and the vascular dissemination of cancer cells may occur before treatment.^28,29 Delays in surgery following preoperative treatment may also contribute to higher recurrence rates.³⁰ On the other hand, a meta-analysis by Zhou et al found no significant difference in recurrence or five-year survival between patients with and without preoperative TACE (32.1% vs 30.0%, P = 0.17; 40.2% vs 45.2%, p = 0.37, respectively).³¹ Although our analysis suggested an association between preoperative TACE/RFA and early recurrence, the number of patients who received this treatment in our cohort was limited. Therefore, the result should be interpreted with caution. Further studies with larger sample sizes are warranted to confirm this finding and clarify the role of preoperative locoregional therapy in recurrence risk. Our findings highlight the importance of close monitoring of MVI-negative HCC patients, particularly during the high-risk period of 6–18 months after surgery. Preoperative TACE/RFA in surgically resectable cases, may negatively impact outcomes and should be considered with caution. Additionally, HCV status and tumor size should be factored into individualized treatment strategies. However, this study has several limitations. First, as a retrospective analysis, comprehensive clinical data were not available for all patients. Second, MVI assessment is highly dependent on grossing and histological evaluation, which may introduce variability. Third, the relatively small sample size may limit the statistical power and generalizability of our results. Fourth, surgical margin status, a factor potentially associated with early recurrence, was not evaluated in our study due to limitations in histological documentation and retrospective data collection.

Conclusion

Early tumor recurrence in MVI-negative HCC was prevalent. This study identified HCV infection, tumor size, and preoperative treatment as independent factors influencing recurrence. Future prospective studies are needed to refine recurrence prediction models and optimize therapeutic approaches for MVI-negative HCC patients.