Association of serum lactate dehydrogenase levels with the severity of microvascular invasion in hepatocellular carcinoma: a retrospective cross-sectional study

Introduction

Primary liver cancer is the sixth most common cancer and the fourth leading cause of cancer death all over the world (1). According to the global cancer statistics report, in 2020 there were more than 900,000 new patients with primary liver cancer worldwide, and the number of deaths was approximately 830,000 (2). It is estimated that after 2025, the number of primary liver cancer patients worldwide will exceed 1 million each year (3,4). Hepatocellular carcinoma (HCC) is the most common pathological type of primary liver cancer, accounting for 75–85% of primary liver cancers. China is a country with a high incidence of HCC, accounting for 47% of the world’s new cases (5). Currently, surgical resection of tumors is still the preferred method for HCC treatment. However, the recurrence rate of HC within 5 years after hepatectomy is as high as 70% (1,5). According to the time of recurrence, postoperative recurrence can be divided into early recurrence (within 2 years after surgery) and late recurrence (after 2 years). The main cause of death within 2 years after HCC surgery is early recurrence of the tumor. Early postoperative recurrence seriously affects the prognosis of patients with HCC. Studies have shown that microvascular invasion (MVI) is one of the important risk factors for early postoperative recurrence of HCC (6,7).

MVI mainly refers to nests of cancer cells seen under a microscope in the lumen of blood vessels lined with endothelial cells, with the number of cancer cells exceeding 50 (7,8). MVI is mostly found in small branches of the portal vein in adjacent tissues of tumors (7). The exact mechanism of MVI in HCC has not been fully elucidated. It is currently believed to be a biological process involving multiple steps and factors (9). Cancer cells in the primary tumor acquire the ability to infiltrate through specific oncogenes, break away from the tumor cell nest, infiltrate surrounding tissues, and enter the blood vessel wall (10). Studies have shown that the molecular pathways involved in liver cancer cells invasion into blood vessels include growth factors, Wnt/β-catenin, YAP and c-MET, as well as epigenetic changes (11-13). MVI makes liver cancer cells highly invasive and more likely to produce satellite nodules around the cancer (14). Studies have demonstrated that regardless of tumor size and number, MVI positivity may also shorten the progression-free survival and overall survival of HCC patients (15,16). The importance of MVI in the management of HCC has been confirmed by many studies (9,14,15). Currently, MVI can only be definitively diagnosed through postoperative histopathology. Preoperative prediction relies on clinical and imaging features, such as large tumor size (>5 cm), multinodularity, and specific radiologic signatures. However, these predictors are not fully sensitive or specific, and their interpretation can be subjective, highlighting the need for complementary, objective biomarkers (17,18).

Due to the particularity of tumor metabolism, tumor cells mainly rely on glycolysis to metabolize glucose to provide energy even under aerobic conditions (19). This metabolic pattern of aerobic glycolysis is called the “Warburg” effect (20). Glycolysis is significantly elevated in liver cancer cells. During the glycolysis process, four key rate-limiting enzymes are involved, including hexokinase 2, phosphofructokinase, pyruvate kinase, and lactate dehydrogenase (LDH) (21,22). LDH is the final rate-limiting enzyme in the glycolysis, catalyzing pyruvate to lactate, which is a necessary condition for tumors to maintain energy metabolism (23,24). Importantly, LDH activity is not merely a correlate of tumor bulk but is implicated in creating a pro-invasive niche. The resultant lactate accumulation leads to extracellular acidification, which can activate proteases such as matrix metalloproteinases, induce epithelial-mesenchymal transition, and stabilize hypoxia-inducible factor-1α, thereby upregulating pro-angiogenic factors like VEGF—all central mechanisms in vascular invasion and metastasis (24-27). Studies have displayed that patients with HCC have higher LDH levels in blood and tumor tissue, which are significantly related to prognosis (28,29). And HCC patients with higher serum LDH levels tend to have poor prognosis (28).

Currently, the presence and severity of MVI in HCC patients can only be confirmed and evaluated through pathological assessment of surgically removed specimens. Therefore, this makes it meaningful to explore a simple, convenient, and economical marker to evaluate the possibility and severity of MVI before surgery. The relationship between serum LDH levels and the severity of MVI in HCC has not been proven. We conducted a retrospective analysis of Chinese adults with HCC based on a large sample from the database system of our hospital. We aimed to investigate the correlation between serum LDH levels and different degrees of MVI in adult patients with HCC. We present this article in accordance with the STROBE reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-944/rc).

Methods

Data source and participants

The data used for this study were obtained from the Clinical Database of Wuhan Union Hospital, which recorded the data of more than 1.6 million people between January 1, 2014 and June 30, 2023. This Clinical Database has been used for clinical epidemiological studies previously (30). This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The Ethics Committee of Union Hospital of Tongji Medical College Huazhong University of Science and Technology approved our study. The requirement for individual informed consent was waived by the Ethics Committee because this was a retrospective observational study based on de-identified clinical data and involved no direct patient contact or intervention.

We conducted a retrospective cross-sectional study by analyzing the data retrieved from the Clinical Database of Wuhan Union Hospital. Adult (>18 years) person with liver cancer who underwent partial liver resection and had pathology reports were identified in this database. Among patients with liver cancer after hepatectomy identified from the database, we excluded those lacking of MVI evaluation, lacking of blood LDH test, diagnosed with cholangiocarcinoma, liver metastasis and other liver tumors.

Clinical database

We obtained the data of blood tests and image data exporting from the Clinical Database of Wuhan Union Hospital. The peripheral blood samples of the study participants were collected under fasting conditions on the next day after hospitalization, and then were sent for examination immediately. The antigen-antibody immune-agglutination reaction method was used to identify ABO blood types of participants. The enzyme-linked immunoassay, colloidal gold method or chemiluminescence method were performed to detect hepatitis B surface antigen (HBsAg) and hepatitis C antibody (HCVAb). White blood cell count, red blood cell count, hemoglobin and platelet were tested using an automated hematology analyzer. The electrochemiluminescence method was conducted to determine the levels of serum alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9). Blood biochemical variables (including total bilirubin, direct bilirubin, alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), γ-glutamyl transpeptidase (γ-GTP), total protein, albumin, fasting blood glucose (FBG), cholesterol and LDH) were measured by automatic biochemical analyzers. The participants were divided into two groups, using the upper limit of normal value (ULN, 245 U/L) of serum LDH level as reference (group A: serum LDH level ≤ ULN, group B: serum LDH level > ULN). The time interval between blood test and pathology report shall not exceed one week. The image data, including tumor size and number, were derived from preoperative computed tomography/magnetic resonance imaging (CT/MRI) scans. Tumor size and number were assessed according to the 2024 edition of the liver cancer evaluation criteria (31). Tumor size was determined by the maximum diameter, measured as the longest axis on the phase with the clearest tumor margins, with the tumor capsule included in the measurement. A tumor count of two or more, including the presence of satellite nodules, was defined as multiple tumors.

Pathological evaluation of HCC specimens

After the liver cancer was surgically removed from the participants, the specimens were sent to the pathology department as intact as possible within 30 minutes of removal from the body. The specimens were fixed with 4% neutral formaldehyde fixative (10% neutral formalin fixative) for 12–24 h. The area surrounding liver cancer was a representative area of tumor biological behavior. Therefore, the 7-point baseline sampling method was adopted, that is, sampling at 12, 3, 6, and 9 o’clock positions of the tumor at the junction of the cancer and adjacent liver tissue at a ratio of 1:1. At least one piece of tissue should be sampled inside the tumor. And one piece of sample was collected from the near-paratumor tissue (≤1 cm from the tumor edge) and far-paratumor tissue (>1 cm from the tumor edge), respectively. For small HCC with the maximum diameter of a single tumor ≤3 cm, all specimens should be collected for examination. The MVI assessment for all harvested tissues was performed under microscope. According to the Practice Guidelines for the Pathological Diagnosis of Primary Liver Cancer, MVI was graded using a three-tiered system as follows: M0 (no MVI), M1 (1–5 MVI sites located within 1 cm of adjacent liver tissue), and M2 (>5 MVI sites or any MVI located >1 cm from adjacent liver tissue) (32,33). The pathologists who evaluated and graded MVI were blinded to all patient clinical and laboratory data, including preoperative serum LDH levels.

Statistical analysis

All statistical analyses were performed using SPSS version 25.0 (IBM, Chicago, IL, USA). Continuous variables were tested for normality using the Kolmogorov-Smirnov test and are presented as mean ± standard deviation (SD) or median interquartile range (IQR), as appropriate. The Pearson’s Chi-squared (χ²) test was used to compare categorical variables. For continuous variables, the independent-samples t-test was used for normally distributed data and nonparametric tests were used for non-normally distributed data. The spearman rank correlation analysis was conducted to potential variables associated with degree of MVI among HCC patients. Variables with P<0.20 in univariate analysis were selected into ordinal logistic regression analyses to determine the independent factors associated with severity of MVI in HCC patients. Furthermore, we examine the association between different degree of MVI and the categories of serum LDH levels. Additionally, we assessed the collinearity among all variables included in the logistic regression model, thereby strengthening the robustness of the regression analysis by confirming that multicollinearity did not distort the estimates. Meanwhile, receiver operating characteristic (ROC) curve was drawn to evaluate the ability of serum LDH to discriminate the presence of MVI. The area under the curve (AUC), sensitivity, and specificity were calculated. For sensitivity analyses, we excluded patients with hepatitis C virus (HCV) infection, then further excluded those with negative HBsAg from the study sample. In addition, after removing the factor of AFP, the multivariate logistic regression analysis was conducted. The other potential confounding variables including sociodemographic characteristics (age, white blood cell, total bilirubin, direct bilirubin, ALT, AST, ALP, γ-GTP, albumin levels) were assessed. All tests were two-sided, and P<0.05 was considered statistically significant.

Results

General characteristics of the participants

A total of 2,706 patients with liver cancer who underwent partial liver resection were retrieved from the Clinical Database of our hospital from 24 July, 2016 to 30 June, 2023. Two thousand one hundred and five cases were excluded from this study: 367 people had no results of MVI evaluation, 805 people were lacking of results of blood LDH test, 913 people were diagnosed with non-HCC (449 people with cholangiocarcinoma, 324 people with liver metastasis and 140 people with other liver tumors). Therefore, 621 HCC participants were included in the final sample. The flowchart of population selection is showed in Figure 1.

Figure 1 Study flowchart for participants screening. HCC, hepatocellular carcinoma; LDH, lactate dehydrogenase; MVI, microvascular invasion.

Among all participants in this study, 535 (86.2%) were male and 86 (13.8%) were female. There were 610 (98.2%) patients of HCC and 11 (1.8%) patients of combined hepatocellular cholangiocarcinoma (cHCC-CC). The mean age of all individuals was 56.1 (SD: 11.4) years old. The proportions of A, B, O and AB blood types were 29.6% (n=176), 25.5% (n=152), 35.3% (n=210) and 9.6% (n=57), respectively. The patients with positive HBsAg accounted for 83.2% (n=510). There were 50 (8.2%) patients with positive serum HCVAb. Among all participants in this study, 535 (86.2%) were male and 86 (13.8%) were female. There were 610 (98.2%) patients of HCC and 11 (1.8%) patients of cHCC-CC. The mean age of all individuals was 56.1 (SD: 11.4) years old. The proportions of A, B, O and AB blood types were 29.6% (n=176), 25.5% (n=152), 35.3% (n=210) and 9.6% (n=57), respectively. The patients with positive HBsAg accounted for 83.2% (n=510). There were 50 (8.2%) patients with positive serum HCVAb. A single tumor was present in 540 patients, and 81 patients had two or more tumors, with a median tumor size of 5 cm. The percentage of patients with M0, M1 and M2 were 55.7% (n=346), 27.2% (n=169) and 17.1% (n=106), respectively. The median serum LDH level of all subjects was 205 U/L (IQR, 72 U/L). The serum LDH level of 461 (74.2%) people was less than or equal to the ULN of LDH (group A). One hundred sixty (25.8%) people had serum LDH levels greater than the ULN of LDH (group B). Other sociodemographic and biological characteristics of the study participants are displayed in Table 1. The percentage of patients with M0, M1 and M2 were 55.7% (n=346), 27.2% (n=169) and 17.1% (n=106), respectively. The median serum LDH level of all subjects was 205 U/L (IQR, 72 U/L). The serum LDH level of 461 (74.2%) people was less than or equal to the ULN of LDH (group A). One hundred sixty (25.8%) people had serum LDH levels greater than the ULN of LDH (group B). Other sociodemographic and biological characteristics of the study participants are displayed in Table 1.

Comparison of various variables in group A and group B

We compared the indicators of people with serum LDH levels ≤ ULN (group A) and those with serum LDH levels > ULN (group B). The results are presented in Table 2. The white blood cell counts of group B (median: 5.68×10⁹/L, IQR, 3.03×10⁹/L) was higher than those of group A (median: 4.85×10⁹/L, IQR, 1.88×10⁹/L) (P<0.001). There were no differences in red blood cell count, hemoglobin level and platelet count between group A and group B. As shown in Table 2, we found that the liver function indicators of group A and group B were significantly different. The serum levels of total bilirubin, direct bilirubin, ALT, AST, ALP, and γ-GTP in group B were higher than those in group A. At the same time, the serum albumin level of group B was lower than that of group A. In addition, the median value of serum AFP level in group B was 160.2 µg/L (IQR, 1986.1 µg/L) which was significantly higher than that in group A with a median value of 18.0 µg/L (IQR, 224.5 µg/L) (P<0.001). Similarly, the median value of serum tumor size level in group B was 6.5 cm (IQR, 6.775 cm) which was significantly higher than that in group A with a median value of 4.2 cm (IQR, 4.5 cm) (P<0.001). There was no difference in the proportion of gender, ABO blood types, HBsAg-positive, and HCVAb-positive patients between group A and group B. The mean age of the two groups were 56.6 (SD: 11.3) and 54.8 (SD: 11.7) years old with no statistical difference (P=0.09). Meanwhile, there was no statistical difference in serum CEA, tumor number and CA 19-9 levels between the two groups, with P values of 0.34 and 0.13, respectively.

Correlation between serum LDH levels and different degrees of MVI

We considered the MVI (M0, M1, and M2) of HCC as an orderly classification variable. Spearman rank correlation analysis was performed to clarify the potential factors associated with the severity of MVI from 25 sociodemographic and blood variables of HCC patients. As shown in Table 3, we found that the mean ages of HCC patients with M2, M1, and M0 were 54.1 (SD: 11.6), 56.0 (SD: 11.4) and 56.8 (SD: 11.4) years, respectively. There was significant difference in the age of these three groups of HCC patients (P=0.03). In addition, with the increased grades of MVI, the serum levels of 4 liver enzymes (including ALT, AST, ALP, γ-GTP), LDL cholesterol and tumor size showed the significantly increasing trends with P values of 0.03, <0.001, 0.001, <0.001, 0.01 and 0.001, respectively. There was statistical difference in serum AFP levels of different MVI groups (P<0.001). However, the serum AFP level of the M1 group was the highest with median value of 213.7 µg/L (IQR, 1,983 µg/L); and the median serum AFP levels of the M0 and M2 groups were 10.5 µg/L (IQR, 96.5 µg/L) and 76.6 µg/L (IQR, 1,992.7 µg/L), respectively.

In Table 3, univariate analysis displayed that the correlation coefficient between different MVI grades and LDH levels was 0.212 (P<0.001). As shown in Figure 2, the median value of LDH in M2 group was 232 U/L (IQR, 90 U/L), which was significantly higher than that (median: 212 U/L, IQR, 70 U/L) in M1 group (Z=−2.98, P=0.003). Compared with patients with M1, patients without MVI had significantly lower LDH with median value of 196 U/L (IQR, 58 U/L) (Z=−2.70, P=0.007). Moreover, the patients with serum LDH level > ULN in M0, M1 and M2 group accounted for 18.2% (63/346), 28.4% (48/169) and 46.2% (49/106), respectively (r=0.223, P<0.001) (Table 3 and Figure 3). The proportions of M0, M1 and M2 were 61.4% vs. 39.4%, 26.2% vs. 30.0% and 12.4% vs. 30.6% in patients with serum LDH level ≤ ULN (group A) and > ULN (group B), respectively.

Figure 2 Box-plot of serum LDH levels in HCC patients with different degrees of MVI. HCC, hepatocellular carcinoma; IQR, interquartile range; LDH, lactate dehydrogenase; MVI, microvascular invasion.

Figure 3 Distribution of different MVI degrees in HCC patients with serum LDH levels ≤ ULN and > ULN. HCC, hepatocellular carcinoma; LDH, lactate dehydrogenase; MVI, microvascular invasion; ULN, upper limit of normal.

We selected the eight variables with P values less than 0.05, seven variables with P values between 0.05 and 0.2 [namely ABO blood types (P=0.09), white blood cell count (P=0.06), red blood cell count (P=0.09), hemoglobin level (P=0.17), platelet count (P=0.12), FBG level (P=0.19) and total cholesterol level (P=0.05)] in the previous univariate analysis and three confounding variables (including total bilirubin, direct bilirubin and albumin) into ordinal logistic regression analysis. Surprisingly, we found that only serum LDH was the independent risk factor associated with the severity of MVI in HCC patients, with coefficient (B) value of 0.006 [95% confidence interval (CI): 0.002–0.009, P=0.003]. In addition, to ensure the robustness of the logistic regression analysis, we performed multicollinearity diagnostics on all included variables. The results indicated no significant multicollinearity among tumor size (1.405), tumor number (1.032), AFP (1.035), and LDH (1.589) and other variables. Table 4 showed that the other 17 variables (including serum AFP level) were not the independent risk factors related to the different grades of MVI in HCC patients. We further explored whether serum AFP level was related to MVI of HCC. The results were displayed in Tables S1,S2. It demonstrated that serum AFP level was not an independent risk factor of MVI in HCC patients.

In Figure 4, ROC curve displayed that the AUC of LDH was 0.610, which indicated that LDH had modest predictive ability for MVI. The sensitivity of it was 54.9%, and the specificity was 65.0%.

Figure 4 Patients ROC curve of serum LDH for predicting MVI. LDH, lactate dehydrogenase; MVI, microvascular invasion; ROC, receiver operating characteristic.

Sensitivity analysis

Exclusion of HCC patients with positive serum HCVAb did not alter the association between serum LDH level and severity of MVI. In Table S3, it displayed that serum LDH level was the independent risk factor associated with the severity of MVI in HCC patients, with coefficient (B) value of 0.004 (95% CI: 0.001–0.008, P=0.02). Then excluding HCC patients with negative serum HBsAg did not modify the association of serum LDH level and different grades of MVI (P=0.044) (Table S4). Progressively exclusion of patients with positive HCVAb or negative HBsAg did not change the relationship between serum LDH level and severity of MVI in HCC with coefficient (B) value of 0.004 (95% CI: 6.16×10⁻⁵–0.008, P=0.046) (Table S5). As shown in Table S6, the association of serum LDH level with severity of MVI persisted after removing the variable of AFP and tumor size in multivariate analysis with coefficient (B) value of 0.003 (95% CI: 0.001–0.005, P=0.004).

Discussion

The goal of our study was to evaluate the relationship between preoperative serum LDH levels and different grades of MVI, an important but unknown parameter before surgery that was key to estimating metastasis and recurrence of HCC. First, we analyzed the proportion of different degree of MVI and the status of the serum LDH level of the HCC participants. Our results showed that the proportion of MVI in HCC patients was about 44.3%, of which M1 accounted for 27.2% and M2 accounted for 17.1%. Previous studies have shown that the prevalence of MVI in HCC ranges from 12.4% to 67.6% (7,9,34-37). Meanwhile, a study which analyzed 16,144 resected HCC specimens reported that the proportions of M1 and M2 in HCC were 26.2% and 20.4%, respectively (32). In present study, the proportions of all MVI and the different grades of MVI in HCC patients are approximately consistent with the previous reports.

Zhang et al. have reported that the median value of serum LDH levels of HCC patients is 208 U/L (IQR =87 U/L). Our data showed that the median serum LDH level of all participants was 205 U/L (IQR =72 U/L), which was similar to the result of previous study. Among the participants of this study, about a quarter of HCC person had serum LDH levels higher than the ULN. We observed that the HCC patients with elevated serum LDH had significantly an increased white blood cell count. One potential explanation might be that LDH can enhance the inflammatory responses resulting in activation and proliferation of white blood cell (38,39). In addition, the current study indicated that the serum bilirubin and liver enzymes (ALT, AST, ALP, γ-GTP) levels of patients with LDH more than ULN were higher, and albumin level was lower when compared to person with normal LDH. As we all known, LDH, bilirubin, ALT, AST, ALP and γ-GTP are positively correlated with hepatocytes injury; meanwhile, albumin is negatively correlated with liver damage. This may can support our results. Also, we found a significantly higher serum AFP level in patients with LDH more than ULN. Several studies also showed that the serum AFP level of HCC patients with elevated LDH was higher (40-42).

Currently, there is a large body of literature investigating markers associated with the existence of MVI in HCC (9,43). However, there are few research exploring the markers related to MVI grade. Existing studies have reported the multiple roles of LDH in cancer (26). The research of Xu et al., in which the normal range of serum LDH is 100.0 to 240.0 U/L, showed that serum LDH ≥176.58 U/L was independently associated with the presence of MVI (7). In present study, we found that higher serum LDH levels were independently associated with greater MVI severity in patients with HCC, although the association was modest. We noted that the proportion of M1 in HCC patients with serum LDH level < ULN (group A) was higher than that in people with serum LDH level ≥ ULN (group B). Meanwhile, our data demonstrated that the proportion of M2 in group B was almost three times more than that in group A (Figure 3). The mechanisms of relationship between serum LDH level and MVI still remain unknown. There may be some underlying mechanisms of the association between high serum LDH levels and increased risk of MVI grade in HCC. It has been hypothesized that high serum LDH levels may reflect the elevated glycolytic activity of cancer cells and hypoxia condition in tumor microenvironment (44). Hypoxia can trigger cancer epithelial-mesenchymal transition, which plays a pivotal role in promoting cancer progression by increasing cancer cell invasion and metastatic potential (45-48). In addition, LDH can catalyzes the conversion of pyruvate to lactic acid under anaerobic conditions. Further lactate can promote angiogenesis by increasing acidify of the tumor microenvironment (49). Increasing of the glycolysis metabolite lactate due to elevated serum LDH can result in immune reprogramming and disruption of blood vessel normalization of the tumor microenvironment, which can promote cancer cells to infiltrate into blood vessels (48,50).

Previous studies have shown that association between of serum AFP levels and MVI is controversial. Hu et al. found that serum AFP level greater than 400 µg/L was independently related to the presence of MVI in HCC patients comparing to serum AFP level less than 20 µg/L (51). Chang et al. indicated that preoperative AFP levels was an independent risk factor for predicting MVI of HCC on multivariable analysis (36). However, some studies have the opposite results. Poté et al. showed that the serum AFP level had no relationship with MVI on univariate and multivariate analyses (52). The results of Wang’s research also implicated that serum AFP level was not an independent risk factor related to MVI (53). The study of Beaufrère and colleagues indicated that although serum AFP level was related to MVI on univariate analysis with P value of 0.04, it was not an independent risk factor associated with MVI of HCC on multivariate analysis (14). Indeed, in the current study we showed that although there was a statistical difference in serum AFP levels between the HCC persons with MVI and without MVI on univariate analysis (Table S1), surprisingly the serum AFP levels of the M0 and M2 groups were significantly lower than that of the M1 group (Table 3). In addition, multivariate analysis showed that serum AFP level was not an independent risk factor related to MVI (Table S2), nor it was independent risk factor related to the severity of MVI in HCC patients (Table 4).

Based on these findings, we believe that serum LDH is a valuable, easily accessible, and cost-effective biomarker that can be used to refine preoperative risk stratification and guide clinical decision-making. For instance, integrating serum LDH with imaging features suggestive of MVI and other biomarkers such as AFP and PIVKA-II may facilitate the development of a multiparameter evaluation strategy or a composite risk score, allowing more accurate prediction of MVI severity. However, the clinical utility of serum LDH should be evaluated within composite scoring systems in future prospective studies. Its routine application for direct surgical guidance is not warranted by the present data, but its role as an adjunctive, biologically plausible marker for enhanced risk assessment merits further investigation.

This study found that both tumor size and serum LDH level are independent predictors of MVI in the multifactor logistic regression model. However, this does not mean that LDH is merely a surrogate marker for large tumor burden. First, statistical evidence supports the independence of the two variables. Collinearity diagnostics indicated that in this study cohort, there was no significant multicollinearity between LDH level and tumor size. This means that the predictive information carried by the two is largely non‑redundant and can be included in the model together without mutual interference. More importantly, their biological and clinical implications are fundamentally different. Tumor size is a structural indicator, primarily reflecting the anatomical volume or diameter of the tumor, which is the result of tumor growth. In contrast, serum LDH is a functional indicator; as a key enzyme in glycolysis, its level directly reflects the metabolic activity and invasive biological phenotype of tumor cells. A high LDH level indicates an active Warburg effect, acidification of the tumor microenvironment, and the consequent pro‑invasive processes. Therefore, a tumor with a small volume may exhibit extremely high metabolic activity and invasive potential, manifested as high LDH, leading to MVI; conversely, a large tumor may have relatively indolent metabolic activity, manifested as low LDH, and thus may not develop MVI. LDH provides precisely this additional information regarding the “intrinsic invasiveness” of the tumor, which cannot be captured by simple radiological size measurements alone. In summary, our study suggests that preoperative serum LDH is not a simple substitute for tumor size but serves as a complementary biomarker that offers unique predictive value from a metabolic‑functional perspective. Combining LDH with traditional indicators such as tumor size may help construct a more comprehensive preoperative risk assessment model for MVI, facilitating the identification of different risk‑subgroup patients, such as those with “small yet aggressive” or “large yet indolent” tumors, thereby optimizing clinical decision‑making. Certainly, the practical value of this integrated model requires further validation in future prospective studies.

We acknowledge that there are some limitations in this study. First, it is a retrospective study based on clinical database of medical records, blood tests and pathologic characteristics from a single center. We look forward to multicenter and prospective studies with large sample to confirm the results. Second, among 1,793 HCC patients in this study, approximately 65.3% person were excluded for lacking of the results of serum LDH test or lacking of the results of MVI assessment in pathological reports. Third, in this study the UNL of serum LDH levels which were tested by our hospital was adopted to group participants for analyses. The ULN values of serum LDH levels may vary with different detection devices in other medical centers. Forth, the mechanisms proposed in this study have not been validated through laboratory experiments; the findings are based solely on statistical associations. Further validation using in vitro and/or in vivo models is required in the future. Finally, there is no a universally recognized consensus among researchers worldwide regarding the standards for MVI evaluation. The three-tiered MVI grading system which has been confirmed by a multicenter team was used in our study (32). This may improve comparability with studies adopting the same criteria.

Therefore, we propose the following directions for future research: first, to develop and validate predictive models or machine learning algorithms that integrate LDH, imaging features, and other serum biomarkers to achieve individualized quantification of MVI risk. Second, to conduct prospective multicenter studies to verify the predictive performance of LDH and establish its optimal cutoff value. Third, to further investigate the biological mechanisms underlying the role of LDH in MVI development—specifically, whether it serves merely as a marker of glycolytic activity or directly promotes vascular invasion through the creation of a lactic acid-rich microenvironment. Fourth, to explore whether targeting glycolysis-related pathways could be therapeutically relevant in the context of MVI, with initial emphasis on preclinical studies to clarify feasibility, safety, and mechanism before considering clinical translation.

Conclusions

In summary, our study shows that serum LDH is an independent risk factor associating with the severity of MVI in HCC patients, although the association was modest. HCC patients with higher serum LDH levels have higher probability and more severe of MVI. Therefore, preoperative serum LDH levels can be used to estimate the MVI status of HCC. More experimental researches are needed to reveal the potential mechanisms of relationship between serum LDH and the severity of MVI in HCC.

Acknowledgments

A preliminary version of this work was presented as a poster at the International Congress of Asian Oncology Society (AOS) &51^st Annual Meeting of Korean Cancer Association (KCA) (July 3–4, 2025 in COEX, Seoul, Korea. No. 25-545).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-944/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-944/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-944/prf

Funding: This study was supported by grants from the National Natural Science Foundation of China (Nos. 82570789, 82002453, 82373195, and 81600482), the Natural Science Foundation of Hubei Province (Nos. 2020CFB600 and 2019CFB501), and China Postdoctoral Science Foundation (Nos. 2018M632875 and 2019T120671).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-aw-944/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was approved by the Ethics Committee of Union Hospital of Tongji Medical College Huazhong University of Science and Technology. The requirement for individual informed consent was waived by the Ethics Committee because this was a retrospective study based on de-identified clinical data and involved no direct patient contact or intervention.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Vogel A, Meyer T, Sapisochin G, et al. Hepatocellular carcinoma. Lancet 2022;400:1345-62. [Crossref] [PubMed]
2. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
3. Singal AG, Lampertico P, Nahon P. Epidemiology and surveillance for hepatocellular carcinoma: New trends. J Hepatol 2020;72:250-61. [Crossref] [PubMed]
4. Siegel RL, Miller KD, Fuchs HE, et al. Cancer statistics, 2022. CA Cancer J Clin 2022;72:7-33. [Crossref] [PubMed]
5. Yang JD, Heimbach JK. New advances in the diagnosis and management of hepatocellular carcinoma. BMJ 2020;371:m3544. [Crossref] [PubMed]
6. Krishnan MS, Rajan Kd A, Park J, et al. Genomic Analysis of Vascular Invasion in HCC Reveals Molecular Drivers and Predictive Biomarkers. Hepatology 2021;73:2342-60. [Crossref] [PubMed]
7. Xu W, Wang Y, Yang Z, et al. New Insights Into a Classification-Based Microvascular Invasion Prediction Model in Hepatocellular Carcinoma: A Multicenter Study. Front Oncol 2022;12:796311. [Crossref] [PubMed]
8. Lee S, Kim SH, Lee JE, et al. Preoperative gadoxetic acid-enhanced MRI for predicting microvascular invasion in patients with single hepatocellular carcinoma. J Hepatol 2017;67:526-34. [Crossref] [PubMed]
9. Zhao X, Wang Y, Xia H, et al. Roles and Molecular Mechanisms of Biomarkers in Hepatocellular Carcinoma with Microvascular Invasion: A Review. J Clin Transl Hepatol 2023;11:1170-83. [Crossref] [PubMed]
10. Zhu WH, Chen J, Huang RK, et al. Erythroid-transdifferentiated myeloid cells promote portal vein tumor thrombus in hepatocellular carcinoma. Theranostics 2023;13:4316-32. [Crossref] [PubMed]
11. He Q, Lin Z, Wang Z, et al. SIX4 promotes hepatocellular carcinoma metastasis through upregulating YAP1 and c-MET. Oncogene 2020;39:7279-95. [Crossref] [PubMed]
12. Poté N, Cros J, Laouirem S, et al. The histone acetyltransferase hMOF promotes vascular invasion in hepatocellular carcinoma. Liver Int 2020;40:956-67. [Crossref] [PubMed]
13. Nishida N. Role of Oncogenic Pathways on the Cancer Immunosuppressive Microenvironment and Its Clinical Implications in Hepatocellular Carcinoma. Cancers (Basel) 2021;13:3666. [Crossref] [PubMed]
14. Beaufrère A, Caruso S, Calderaro J, et al. Gene expression signature as a surrogate marker of microvascular invasion on routine hepatocellular carcinoma biopsies. J Hepatol 2022;76:343-52. [Crossref] [PubMed]
15. Peng Z, Chen S, Xiao H, et al. Microvascular Invasion as a Predictor of Response to Treatment with Sorafenib and Transarterial Chemoembolization for Recurrent Intermediate-Stage Hepatocellular Carcinoma. Radiology 2019;292:237-47. [Crossref] [PubMed]
16. Hwang YJ, Bae JS, Lee Y, et al. Classification of microvascular invasion of hepatocellular carcinoma: correlation with prognosis and magnetic resonance imaging. Clin Mol Hepatol 2023;29:733-46. [Crossref] [PubMed]
17. Ye JZ, Chen JZ, Li ZH, et al. Efficacy of postoperative adjuvant transcatheter arterial chemoembolization in hepatocellular carcinoma patients with microvascular invasion. World J Gastroenterol 2017;23:7415-24. [Crossref] [PubMed]
18. Zhong JH, Li LQ, Ye XP, et al. Efficacy trend of hepatic resection for hepatocellular carcinoma with large multinodular tumor or macrovascular invasion. J Clin Oncol 2015;33:427. Available online: doi:10.1200/jco.2015.33.3_suppl.427
19. Liang B, Jiang Y, Song S, et al. ASPP2 suppresses tumour growth and stemness characteristics in HCC by inhibiting Warburg effect via WNT/β-catenin/HK2 axis. J Cell Mol Med 2023;27:659-71. [Crossref] [PubMed]
20. Liao M, Yao D, Wu L, et al. Targeting the Warburg effect: A revisited perspective from molecular mechanisms to traditional and innovative therapeutic strategies in cancer. Acta Pharm Sin B 2024;14:953-1008. [Crossref] [PubMed]
21. Hicks KG, Cluntun AA, Schubert HL, et al. Protein-metabolite interactomics of carbohydrate metabolism reveal regulation of lactate dehydrogenase. Science 2023;379:996-1003. [Crossref] [PubMed]
22. Zhang W, Wang G, Xu ZG, et al. Lactate Is a Natural Suppressor of RLR Signaling by Targeting MAVS. Cell 2019;178:176-89.e15.
23. Faubert B, Li KY, Cai L, et al. Lactate Metabolism in Human Lung Tumors. Cell 2017;171:358-71.e9.
24. Liu W, Wang Y, Bozi LHM, et al. Lactate regulates cell cycle by remodelling the anaphase promoting complex. Nature 2023;616:790-7. [Crossref] [PubMed]
25. Huimin W, Xin W, Shan Y, et al. Lactate promotes the epithelial-mesenchymal transition of liver cancer cells via TWIST1 lactylation. Exp Cell Res 2025;447:114474. [Crossref] [PubMed]
26. Claps G, Faouzi S, Quidville V, et al. The multiple roles of LDH in cancer. Nat Rev Clin Oncol 2022;19:749-62. [Crossref] [PubMed]
27. Huang R, Zhang L, Jin J, et al. Bruceine D inhibits HIF-1α-mediated glucose metabolism in hepatocellular carcinoma by blocking ICAT/β-catenin interaction. Acta Pharm Sin B 2021;11:3481-92. [Crossref] [PubMed]
28. Su K, Huang W, Li X, et al. Evaluation of Lactate Dehydrogenase and Alkaline Phosphatase as Predictive Biomarkers in the Prognosis of Hepatocellular Carcinoma and Development of a New Nomogram. J Hepatocell Carcinoma 2023;10:69-79. [Crossref] [PubMed]
29. Gan Y, Gao F, Du B, et al. Effects of preoperative serum lactate dehydrogenase levels on long-term prognosis in elderly patients with hepatocellular carcinoma undergoing transcatheter arterial chemoembolization. Front Surg 2022;9:982114. [Crossref] [PubMed]
30. Zhang S, Zong Y, Hu Y, et al. High HBV-DNA serum levels are associated with type 2 diabetes in adults with positive HBsAg: An observational study. Front Endocrinol (Lausanne) 2023;14:1146798. [Crossref] [PubMed]
31. Zhou J, Sun H, Wang Z, et al. China Liver Cancer Guidelines for the Diagnosis and Treatment of Hepatocellular Carcinoma (2024 Edition). Liver Cancer 2025;14:779-835. [Crossref] [PubMed]
32. Sheng X, Ji Y, Ren GP, et al. A standardized pathological proposal for evaluating microvascular invasion of hepatocellular carcinoma: a multicenter study by LCPGC. Hepatol Int 2020;14:1034-47. [Crossref] [PubMed]
33. Zhou J, Sun H, Wang Z, et al. Guidelines for the Diagnosis and Treatment of Primary Liver Cancer (2022 Edition). Liver Cancer 2023;12:405-44. [Crossref] [PubMed]
34. Giannini EG, Bucci L, Garuti F, et al. Patients with advanced hepatocellular carcinoma need a personalized management: A lesson from clinical practice. Hepatology 2018;67:1784-96. [Crossref] [PubMed]
35. Ding GY, Zhu XD, Ji Y, et al. Serum PON1 as a biomarker for the estimation of microvascular invasion in hepatocellular carcinoma. Ann Transl Med 2020;8:204. [Crossref] [PubMed]
36. Chang Y, Guo T, Zhu B, et al. A novel nomogram for predicting microvascular invasion in hepatocellular carcinoma. Ann Hepatol 2023;28:101136. [Crossref] [PubMed]
37. Huang J, Li L, Liu FC, et al. Prognostic Analysis of Single Large Hepatocellular Carcinoma Following Radical Resection: A Single-Center Study. J Hepatocell Carcinoma 2023;10:573-86. [Crossref] [PubMed]
38. Chowdhury CS, Wareham E, Xu J, et al. Rap1b-loss increases neutrophil lactate dehydrogenase activity to enhance neutrophil migration and acute inflammation in vivo. Front Immunol 2022;13:1061544. [Crossref] [PubMed]
39. Seledtsov VI, Darinskas A, Von Delwig A, et al. Inflammation Control and Immunotherapeutic Strategies in Comprehensive Cancer Treatment. Metabolites 2023;13:123. [Crossref] [PubMed]
40. Wu SJ, Lin YX, Ye H, et al. Prognostic value of alkaline phosphatase, gamma-glutamyl transpeptidase and lactate dehydrogenase in hepatocellular carcinoma patients treated with liver resection. Int J Surg 2016;36:143-51. [Crossref] [PubMed]
41. Li MX, Zhao H, Bi XY, et al. Lactate dehydrogenase is a prognostic indicator in patients with hepatocellular carcinoma treated by sorafenib: results from the real life practice in HBV endemic area. Oncotarget 2016;7:86630-47. [Crossref] [PubMed]
42. Kong W, Zuo X, Liang H, et al. Prognostic Value of Lactate Dehydrogenase in Patients with Hepatocellular Carcinoma: A Meta-Analysis. Biomed Res Int 2018;2018:1723184. [Crossref] [PubMed]
43. Zheng Z, Guan R, Jianxi W, et al. Microvascular Invasion in Hepatocellular Carcinoma: A Review of Its Definition, Clinical Significance, and Comprehensive Management. J Oncol 2022;2022:9567041. [Crossref] [PubMed]
44. Peng X, He Z, Yuan D, et al. Lactic acid: The culprit behind the immunosuppressive microenvironment in hepatocellular carcinoma. Biochim Biophys Acta Rev Cancer 2024;1879:189164. [Crossref] [PubMed]
45. Brisson L, Bański P, Sboarina M, et al. Lactate Dehydrogenase B Controls Lysosome Activity and Autophagy in Cancer. Cancer Cell 2016;30:418-31. [Crossref] [PubMed]
46. Vadde R, Vemula S, Jinka R, et al. Role of hypoxia-inducible factors (HIF) in the maintenance of stemness and malignancy of colorectal cancer. Crit Rev Oncol Hematol 2017;113:22-7. [Crossref] [PubMed]
47. Joseph JP, Harishankar MK, Pillai AA, et al. Hypoxia induced EMT: A review on the mechanism of tumor progression and metastasis in OSCC. Oral Oncol 2018;80:23-32. [Crossref] [PubMed]
48. Sharma D, Singh M, Rani R. Role of LDH in tumor glycolysis: Regulation of LDHA by small molecules for cancer therapeutics. Semin Cancer Biol 2022;87:184-95. [Crossref] [PubMed]
49. Brown TP, Ganapathy V. Lactate/GPR81 signaling and proton motive force in cancer: Role in angiogenesis, immune escape, nutrition, and Warburg phenomenon. Pharmacol Ther 2020;206:107451. [Crossref] [PubMed]
50. Jin H, Liu Q, Li J, et al. Multifaceted roles of lactate dehydrogenase in liver cancer Int J Oncol 2025;66:50. (Review). [Crossref] [PubMed]
51. Hu HT, Wang Z, Huang XW, et al. Ultrasound-based radiomics score: a potential biomarker for the prediction of microvascular invasion in hepatocellular carcinoma. Eur Radiol 2019;29:2890-901. [Crossref] [PubMed]
52. Poté N, Cauchy F, Albuquerque M, et al. Performance of PIVKA-II for early hepatocellular carcinoma diagnosis and prediction of microvascular invasion. J Hepatol 2015;62:848-54. [Crossref] [PubMed]
53. Wang D, Xu Y, Goldstein JB, et al. Preoperative evaluation of microvascular invasion with circulating tumour DNA in operable hepatocellular carcinoma. Liver Int 2020;40:1997-2007. [Crossref] [PubMed]