Abstract
Purpose: Hepatocellular carcinoma (HCC) with portal vein tumor thrombus (PVTT) is associated with a poor prognosis for HCC patients. Herein we aimed to establish a scoring system to predict the risk of PVTT formation in hepatitis B virus (HBV)-associated HCC. Methods: A total of 848 patients from the Henan Province Traditional Chinese Medicine (TCM) Hospital with HCC were included in the study. Among them, 403 with and 445 without PVTT were retrospectively analyzed to identify the risk factors for PVTT formation, using a novel scoring system to predict the occurrence of PVTT in HBV-associated HCC patients. The scoring system was validated using clinical data from the First Affiliated Hospital of Henan University of TCM. Significant findings: The Cox proportional-hazard regression model revealed that gender, tumor size, the neutrophil-lymphocyte ratio, and alpha-fetoprotein and C-reactive protein concentrations were dependent clinical prognostic factors for PVTT, which were included in the final scoring model for PVTT prediction (AUC, 0.858; 95% CI: 0.832 to 0.881). The scoring model ranked HCC patients into 3 risk grades. A sensitivity analysis for validation of the scoring system was performed on 489 patients with HBV-related HCC. The proportion of patients in each grade was not significantly different. Conclusions: The study established a risk warning system for PVTT prediction in HCC patients. More substantial clinical data will be necessary to confirm these findings.
Keywords: Carcinoma, hepatitis B virus, hepatocellular, portal vein, thrombosis
Introduction
Hepatocellular carcinoma (HCC) is the third most commonly occurring cancer worldwide [1] and is most prevalent in Asia and Africa. Hepatitis B virus (HBV)-induced chronic hepatocyte damage contributes to HCC [1]. Due to widespread chronic HBV infection, HBV is responsible for half of the cases of HCC. When HCC cells invade the portal venous system, this can lead to the formation of a portal vein tumor thrombus (PVTT). PVTT is a sign of advanced-stage cancer and the median survival time of HCC patients with PVTT is circa 3 months without treatment [2].
According to the Barcelona Clinic Liver Cancer (BCLC) classification, which are widely adopted guidelines in Europe and America, HCC with PVTT is ranked as BCLC Stage C, with patients being recommended to receive treatment with the vascular endothelial growth factor (VEGF) inhibitor sorafenib, as well as other non-surgical treatments as first-line therapy [3]. However, a randomized controlled trial (RCT) reported that the median survival time after sorafenib treatment was only 6.5 months [4]. Thus, expert Chinese consensus recommended a more aggressive therapeutic approach such as hepatectomy for HCC patients with PVTT, which was found to prolong the median survival time to between 8 months and 22 months [4,5].
It has been reported that the type of PVTT is one of the risk factors for mortality after surgical resection in HCC patients [6], and that aggressive treatment could benefit selected HCC patients with PVTT. At present, only two classifications of PVTT are recognized, namely Cheng’s and the Japanese VP classifications [7]. However, no consensus has been reached on whether or not surgery should be conducted according to either classification standard. It should be noted that PVTT is asymptomatic and is always detected using imaging techniques [8]. Thus, more attention should be paid to potential early warning signs of PVTT.
The present study aimed to identify the risk factors that contribute to the formation of PVTT, using clinical data of PVTT and non-PVTT HCC cohorts of patients and to establish an early risk and warning scoring system for PVTT using risk factors detected non-invasively to facilitate clinical diagnosis and treatment. As PVTT is a common complication of liver cancer, earlier diagnosis of PVTT is likely to improve the prognoses of HBV-related HCC patients.
Materials and methods
Study population
This retrospective study was conducted on the clinical data of HCC patients with or without PVTT who had been admitted to Henan Province TCM Hospital. The requirement of informed consent was waived because all personal identifiers were removed before data collection. The study was approved by the ethics committee of Henan TCM Hospital (approval number/date: 2021-12-08).
The flow chart of the study is shown in Figure 1. A total of 2,758 patients with HBV-related primary HCC were admitted to the Henan Province TCM Hospital from January 2010 to December 2019. HCC was diagnosed according to the diagnostic criteria for HCC of the American Association for the Study of Liver Diseases (AASLD) [9], and the diagnosis of PVTT in patients with HCC was based on combined imaging studies, including B ultrasound, color Doppler, enhanced computer tomography (CT) scans, magnetic resonance imaging (MRI) and Digital Subtraction Angiography (DSA) [10]. Further stratification showed that of the 2,758 patients, 613 underwent conservative medical therapy, 699 received transcatheter arterial chemoembolization (TACE), 152 received radiofrequency ablation (RFA), 951 received TACE combined with RFA, 128 underwent hepatectomy, 85 underwent hepatectomy combined with TACE + RFA, 89 underwent hepatectomy combined with TACE and 41 underwent hepatectomy combined with RFA.
Figure 1.
Flow chart of the study cohort. HCC, hepatocellular carcinoma; PVTT, portal vein tumor thrombus.
The inclusion criteria were: (1) HBV-related HCC patients between the ages of 18 and 85 years; (2) male or female patients.
The exclusion criteria were: (1) patients with infection from another hepatotropic virus (hepatitis A, C, D, E) or non-hepatotropic virus co-infection; (2) patients with severe baseline diseases of the heart, lung, kidney, brain, blood and other important organs; (3) patients with mental illness; (4) metastatic liver cancer; (5) incomplete clinical data; and (6) patients not having complete 5-year follow-up data.
Based upon the inclusion and exclusion criteria, of 2,758 patients, 621 had incomplete clinical data, 177 had secondary hepatic carcinoma, 258 had severe comorbidities, 94 had HCV-related HCC, 109 had alcoholic hepatitis, 257 were lost to follow-up and 496 had follow-up periods < 5 years. Thus, a total number of 848 patients met the inclusion criteria, among whom 403 patients suffered with PVTT during the previous 5 years and 445 did not.
Population validation
To validate further the proposed scoring system, 489 patients with HBV-related HCC, diagnosed in the First Affiliated Hospital of Henan University of TCM from January 2012 to December 2019, were randomly selected for the prediction of the occurrence of PVTT.
R software was used to construct the nomogram model. The Hosmer-Lemeshow goodness-of-fit test, and the calibration curve were employed to evaluate the performance of the nomogram model.
Data collection
The clinical data at the time of diagnosis for all patients with HBV-related primary HCC were collected as follows: (1) baseline information including age, gender, smoker, alcohol drinker, past and family histories; (2) data from imaging studies of CT, ultrasonography and MRI; (3) laboratory indicators from routine blood tests (white blood cell [WBC], hemoglobin [HGB], the neutrophil-to-lymphocyte ratio [NLR], platelet [PLT]), liver and kidney functions (alanine aminotransferase [ALT], aspartate aminotransferase [AST], gamma-glutamyl transferase [GGT], total bilirubin [TBIL], albumin [ALB], albumin/globulin [A/G], creatinine [Cr]), fasting blood glucose (Glu), triglyceride (TC), coagulation function, inflammation index (C-reactive protein [CRP]), prothrombin activation (PTA), virological index (HBV-DNA) and tumor marker (alpha fetoprotein [AFP]).
Statistical analysis
Quantitative variables are presented as the median and 25% percentile and 75% percentile ([Q1, Q3] 25th, 75th percentile) and were compared using the Wilcoxon rank sum test. For comparisons between data presented as frequencies, a χ2 test was employed.
Cox regression analysis was used to analyze multiple risk factors for survival. Multiple factors with a single factor P < 0.05 were included in the multifactorial regression model. Hazard ratio (HR) values were converted into integers to score and accumulate. Excel was used to plot a heat map and risk groupings. MedCalc was used to draw the receiver operating characteristic (ROC) curve of the model and to evaluate it, and Graphpad was used to construct the Kaplan-Meier curve of tumor thrombus in the different groups of patients. The survival times were followed-up and the survival curves of different risk groups were plotted and compared with the log-rank test results. SPSS ver. 22.0 was used for all statistical analyses. A P-value < 0.05 was considered to be a significant finding.
Results
Baseline
The baseline characteristics of the model cohort are shown in Table 1. Of note, the PVTT group had a significantly higher proportion of males (83.87 vs. 68.76%, P < 0.001) and also exhibited significantly higher levels of NLR (3.08 vs. 1.83, P < 0.001), CRP (13.00 vs. 3.20 mg/L, P < 0.001) and GGT (70.90 vs. 42.70 U/L, P < 0.001) than the non-PVTT group. The PVTT group had lower levels of ALB (34.30 vs. 37.00 g/L, P < 0.001), TC (0.77 vs. 0.80 mmol/L, P = 0.002), the proportion of patients with AFP ≤ 350 ng/mL (68.98 vs. 90.11%, P < 0.001) and a lower proportion of patients with tumor size < 5 cm (56.58 vs. 84.94%, P < 0.001) than the non-PVTT group.
Table 1.
Baseline of characteristics of patients with and without PVTT in HBV related HCC
| Variable | None-PVTT (n = 445) | PVTT (n = 403) | P-value | |
|---|---|---|---|---|
| Gender | n | 0.001* | ||
| Male | 306 | 338 | ||
| Female | 139 | 65 | ||
| Age (years) | Median [Q1, Q3] | 56 [50-61] | 55 [49-61] | 0.393 |
| Alcohol history | n | 0.018* | ||
| Yes | 120 | 139 | ||
| No | 325 | 264 | ||
| Smoking history | n | 0.006* | ||
| Yes | 142 | 165 | ||
| No | 303 | 238 | ||
| WBC (× 109/L) | Median [Q1, Q3] | 4.07 [2.92-5.22] | 4.54 [3.13-6.39] | < 0.001 |
| HGB (g/L) | Median [Q1, Q3] | 129.90 [111.70-142.40] | 122.50 [106.95-139.05] | 0.002* |
| NLR | Median [Q1, Q3] | 1.83 [1.36-2.66] | 3.08 [2.07-5.12] | < 0.001* |
| PLT (× 109/L) | Median [Q1, Q3] | 79.00 [54.00-124.30] | 90.40 [56.20-146.70] | 0.020* |
| Cr (μmol/L) | Median [Q1, Q3] | 66.00 [55.00-76.70] | 65.10 [58.00-76.10] | 0.313 |
| Glu (mmol/L) | Median [Q1, Q3] | 5.42 [4.91-6.51] | 5.81 [5.07-7.49] | < 0.001 |
| ALT (U/L) | Median [Q1, Q3] | 31.10 [21.60-45.90] | 34.50 [24.05-59.10] | 0.002* |
| AST (U/L) | Median [Q1, Q3] | 36.00 [26.80-55.30] | 46.00 [31.30-73.35] | 0.001* |
| TBIL (μmol/L) | Median [Q1, Q3] | 17.80 [12.70-29.40] | 20.70 [14.20-36.20] | 0.001* |
| ALB (g/L) | Median [Q1, Q3] | 37.00 [31.50-40.90] | 34.30 [30.30-38.80] | < 0.001* |
| A/G | Median [Q1, Q3] | 1.20 [1.00-1.40] | 1.10 [0.80-1.30] | < 0.001 |
| GGT (U/L) | Median [Q1, Q3] | 42.70 [25.10-74.30] | 70.90 [39.00-138.25] | < 0.001* |
| TC (mmol/L) | Median [Q1, Q3] | 0.80 [0.64-1.05] | 0.77 [0.59-0.92] | 0.002* |
| PTA (%) | Median [Q1, Q3] | 76.00 [63.50-91.00] | 72.00 [59.50-84.60] | < 0.001* |
| CRP (mg/L) | Median [Q1, Q3] | 3.20 [3.20-8.10] | 13.00 [3.20-17.70] | < 0.001* |
| AFP (ng/mL) | n | < 0.001* | ||
| ≤ 350 | 401 | 278 | ||
| > 350 | 44 | 125 | ||
| Single/multiple tumor | n | 0.001* | ||
| Single | 287 | 217 | ||
| Multiple | 158 | 186 | ||
| Child-Pugh score | n | < 0.001* | ||
| A | 247 | 169 | ||
| B | 146 | 162 | ||
| C | 52 | 72 | ||
| Tumor size (cm) | n | < 0.001* | ||
| < 5 | 378 | 228 | ||
| ≥ 5 | 67 | 175 | ||
| HBV-DNA (IU/mL) | n | 0.059 | ||
| < 500 | 187 | 181 | ||
| ≥ 500 | 258 | 222 |
AFP, alpha fetoprotein; A/G, albumin/globulin; ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; Cr, creatinine; CRP, C-reactive protein; GGT, gamma-glutamyl transferase; Glu, fasting blood glucose; HBV, hepatitis B virus; HGB, hemoglobin; [Q1, Q3], [25% percentile, 75% percentile]; NLR, neutrophil-to-lymphocyte ratio; PLT, platelet; PTA, prothrombin activation; PVTT, portal vein tumor thrombus; TBIL, total bilirubin; TC, triglyceride; WBC, white blood cell.
P-value < 0.05.
Univariate and multivariate analysis of PVTT predictors in HBV-related primary hepatic carcinoma
Predictors of PVTT were assessed using forward stepwise Cox proportional hazards regression analysis. The baseline variables with significance between two groups, included gender, NLR, tumor size, AFP and CRP, which were screened as prognostic factors using univariate analysis, which identified gender (P < 0.001), alcohol history (P = 0.006), smoking history (P = 0.007), the Child-Pugh score (P < 0.001), single/multiple tumors (P = 0.001), tumor size (P < 0.001), treatment (P = 0.001), WBC (P = 0.005), HGB (P = 0.010), PLT (P = 0.005), NLR (P < 0.001), Glu (P = 0.001), ALT (P = 0.009), AST (P < 0.001), ALB (P < 0.001), A/G (P = 0.001), GGT (P < 0.001), PTA (P = 0.024), TC (P < 0.001), AFP (P < 0.001), HBV-DNA (P = 0.008) and CRP (P < 0.001) as significant prognostic factors (Table 2).
Table 2.
Univariate and multivariate Cox regression analysis of PVTT in HBV-related HCC
| Variable | Univariate analysis | Multivariate analysis | |||||
|---|---|---|---|---|---|---|---|
|
|
|
||||||
| HR | 95% CI | P-value | Adjusted HR | 95% CI | P-value | ||
| Gender | Male/female | 1.901 | [1.458-2.481] | < 0.001* | 2.151 | [1.592-2.909] | < 0.001* |
| Age (years) | ≥ 50 | 0.995 | [0.985-1.005] | 0.321 | |||
| Alcohol history | Yes/no | 1.333 | [1.086-1.638] | 0.006* | |||
| Smoking history | Yes/no | 1.316 | [1.079-1.605] | 0.007* | |||
| Hypertension | Yes/no | 0.868 | [0.687-1.096] | 0.235 | |||
| Diabetes | Yes/no | 1.036 | [0.826-1.300] | 0.760 | |||
| Child-Pugh score | A/B/C | 1.279 | [1.124-1.455] | < 0.001* | |||
| Single/multiple tumor | Single/multiple | 1.381 | [1.134-1.682] | 0.001* | |||
| Tumor size (cm) | ≥ 5 | 2.865 | [2.349-3.493] | < 0.001* | 2.520 | [2.013-3.153] | < 0.001* |
| Treatment | Conservative treatment/minimally invasive surgery/resection/minimally invasive surgery + resection | 0.735 | [0.639-0.847] | 0.001* | |||
| WBC (× 109/L) | 1.331 | [1.091-1.624] | 0.005* | ||||
| NLR | ≥ 1.91 | 3.410 | [2.648-4.391] | < 0.001* | 2.400 | [1.831-3.145] | < 0.001* |
| HGB (g/L) | > 120 | 0.773 | [0.635-0.940] | 0.010* | |||
| PLT (× 109/L) | > 130 | 1.329 | [1.091-1.617] | 0.005* | |||
| Cr (μmol/L) | > 88.2 | 1.190 | [0.885-1.600] | 0.250 | |||
| Glu (mmol/L) | > 6.1 | 1.396 | [1.145-1.702] | 0.001* | |||
| ALT (U/L) | > 40 | 1.301 | [1.067-1.585] | 0.009* | |||
| AST (U/L) | > 40 | 1.747 | [1.431-2.133] | < 0.001* | |||
| TBIL (μmol/L) | > 17.1 | 1.201 | [0.983-1.468] | 0.073 | |||
| ALB (g/L) | > 35 | 0.686 | [0.564-0.835] | < 0.001* | |||
| A/G | > 1 | 0.714 | [0.586-0.869] | 0.001* | |||
| GGT (U/L) | > 45 | 1.947 | [1.583-2.394] | < 0.001* | |||
| TC (mmol/L) | > 0.78 | 0.619 | [0.476-0.806] | < 0.001* | |||
| PTA (%) | > 70 | 0.797 | [0.655-0.970] | 0.024* | |||
| AFP (ng/mL) | > 350 | 2.875 | [2.324-3.555] | < 0.001* | 2.304 | [1.810-2.934] | < 0.001* |
| CRP (mg/L) | > 5 | 5.024 | [4.101-6.155] | < 0.001* | 4.136 | [3.278-5.219] | < 0.001* |
| HBV-DNA (IU/mL) | > 500 | 1.340 | [1.079-1.644] | 0.008* | |||
AFP, alpha fetoprotein; A/G, albumin/globulin; ALB, albumin; ALT, alanine aminotransferase; AST, aspartate aminotransferase; Cr, creatinine; CRP, C-reactive protein; GGT, gamma-glutamyl transferase; Glu, fasting blood glucose; HBV, hepatitis B virus; HGB, hemoglobin; HR, hazard ratio; NLR, neutrophil-to-lymphocyte ratio; PLT, platelet; PTA, prothrombin activation; PVTT, portal vein tumor thrombus; TBIL, total bilirubin; TC, triglyceride; WBC, white blood cell.
P-value < 0.05.
Next, multivariate analysis was conducted, which confirmed that only gender (P < 0.001, HR: 2.151, 95% CI: [1.592-2.909]), tumor size (P < 0.001, HR: 2.520, 95% CI: [2.013-3.153]), NLR (P < 0.001, HR: 2.400, 95% CI: [1.831-3.145]), AFP (P < 0.001, HR: 2.304, 95% CI: [1.810-2.934]) and CRP (P < 0.001, HR: 4.136, 95% CI: [3.278-5.219]) remained as predictors of PVTT.
Establishment of an early risk and warning scoring system for PVTT in HBV-related primary HCC
The scoring system proposed for PVTT was based on the results of multivariate regression shown in Figure 2, which are detailed as male and NLR ≥ 1.91. Tumor sizes ≥ 5 cm were each assigned a value of 2; AFP > 350 ng/mL was given the value 3; CRP > 5 mg/L 4 and other variables including female, tumors < 5 cm, NLR < 1.91, AFP < 350 ng/mL and CRP < 5 mg/L values of 0. Based on this scoring system, the PVTT risk was stratified as low-risk: total score 0-4, medium-risk: total score 5-8 and high-risk: total score 9-13. The AUC of the model was 0.858 (95% CI: 0.832 to 0.881, Figure 3).
Figure 2.
Definition of scoring system of PVTT prediction. AFP, alpha fetoprotein; CRP, C-reactive protein; NLR, neutrophil-to-lymphocyte ratio; PVTT, portal vein tumor thrombus.
Figure 3.
Receiver-operating characteristic curve of the prediction model.
Prediction of survival in HCC patients with PVTT using the scoring system
K-M survival analysis with a log-rank test was performed for the cohort of patients with PVTT. As shown in Figure 4, the high-risk group exhibited a significantly higher incidence of PVTT formation, with the low-risk group tended to have the lowest incidence of PVTT. The total rank score was significantly associated with the incidence of PVTT formation in HCC patients (log-rank P-value < 0.001), and a higher total score indicated a higher probability of PVTT formation. Taken together, these results indicated that the scoring system could potentially serve as a prediction method for PVTT formation.
Figure 4.
Kaplan-Meier analyses for the incidence of PVTT in patients with different risk. PVTT, portal vein tumor thrombus.
Since all included cohorts were followed up for at least 5-years, the 1-year progression-free survival (PFS) curve was first plotted using the Kaplan-Meier method. Patients in the high-risk group had shorter 1-year PFS times (Figure 5A) and 1-year all-cause survival (Figure 5B) than the medium-risk and low-risk groups (P < 0.001). On the other hand, patients in the low-risk group had higher 1-year PFS times and a 1-year all-cause survival times than those in the medium-risk and high-risk groups (P ≤ 0.001). The 1-year PFS and the 1-year all-cause survival were significantly different among patients in the three groups (P < 0.001).
Figure 5.
Kaplan-Meier analyses for the 1-year PFS rate (A), 1-year all-cause survival rate (B), 5-year PFS rate (C) and 5-year all-cause survival rate (D) in patients with different risk factors. PFS, progression-free survival.
Similarly, patients in the high-risk group had shorter 5-year PFS times (Figure 5C) and 5-year all-cause survival (Figure 5D) than those in the medium-risk and low-risk groups (P < 0.001). In contrast, patients in the low-risk group had longer 5-year PFS times and a 5-year all-cause survival than those in the medium-risk and high-risk groups (P < 0.001). The 5-year PFS and the 5-year all-cause survival times were significantly different among patients in the three groups (P < 0.001). These results indicated good discrimination of survival times among the different risk groups of patients.
Performance of the scoring system for PVTT prediction
The effectiveness of the scoring system was tested in a validation cohort of 489 patients with HBV-related primary HCC diagnosed in the First Affiliated Hospital of Henan University of TCM from January 2012 to December 2019. The calibration curves for the probability of occurrence of PVTT displayed high consistency between the predicted values and the actual observations in the validation set (Figure 6). As expected, the prediction efficiency of the scoring system delivered a better performance in the HCC cohort enrolled in the present study.
Figure 6.
Calibration curves for the PVTT prediction model. PVTT, portal vein tumor thrombus.
Discussion
Portal metastasis has been reported to occur in 30-100% of HCC cases [11]. The present retrospective study first employed univariate combined with multivariate analyses to identify 5 different variables as independent risk factors for PVTT formation and then established a novel PVTT prediction scoring system with further validation. First, an HBV-related HCC sex disparity was found, evidenced by HCC being more frequently and more aggressive in males than in females, suggesting a correlation between gender and PVTT. It has been shown that estrogen can not only can protect females infected with HBV from having a high-risk of developing HCC [12] but can also slow down HCC-related HCC progression in female HCC. Conversely, androgens may promote HCC-related hepatocarcinogenesis as the androgen-androgen receptor complex can activate transcription of the HBV genome, and the male gender has been reported to be an independent risk factor for poor HCC outcomes. Thus, the gender of HCC patients can be taken as a potential parameter for predicting the occurrence of PVTT.
Tumor size has long been recognized as an important predictor of HCC prognosis and also an independent risk factor for the recurrence of HCC. A recent study suggested that tumor size was a risk factor for the formation of PVTT [13], and other research has confirmed that a tumor size > 8 cm is an independent predictor for the occurrence of PVTT in HCC [14]. More importantly, HCC cells in tumors > 5 cm exhibited a higher probability of migrating into the portal vein [15]. Therefore, we adapted a cutoff tumor size > 5 cm to evaluate the probability of PVTT formation in the scoring system.
AFP, a glycoprotein, is a diagnostic tool for HCC as it is secreted from HCC cells into the serum [16], and has been accepted as a prognosis marker, with a cut-off concentration ≥ 400 ng/mL indicating a poor prognosis. A serum AFP concentration between 20 and 400 ng/mL has been reported to be a feasible cutoff for long-term outcome prediction in unselected HCC patients [17]. According to previous studies, the concentration of AFP in the serum is positively correlated with tumor size [17,18]. Although the use of AFP as a prognosis marker has been challenged, a retrospective study revealed that AFP is still a significant diagnosis and prognosis marker of HBV-related HCC, rather than non-HBV-related HCC [19].
CRP, an inflammation marker synthesized by hepatocytes under the control of cytokines IL-6, is associated with the poor prognosis of several types of cancer and can be considered as a useful marker when AFP presents negative in HBV-related HCC patients. Thus, CRP carries diagnostic and prognosis potential in HBV-related HCC.
NLR was identified as another independent predictor of PVTT. A previous study reported that an increased NLR in HCC patients was associated with poor prognosis in primary HCC patients [20] and could predict the surgical outcomes of HCC patients [21]. More importantly, the preoperative NLR has been reported to be a prognostic factor after hepatectomy for HCC patients with PVTT [22]. Thus, we included NLR in the predictive scoring system for PVTT formation.
Previously, two retrospective studies revealed that the presence of cirrhosis is one of the risk factors for the development of PVTT [13,14]. Cirrhosis has been proven to be important in the fundamental pathogenesis of HCC, which can be graded using the Child-Pugh score for prognostically assessing the overall survival of HCC patients. However, in the present study, a relationship between the grade of cirrhosis and PVTT prognosis was not found, with the proportion of each cirrhosis grade being similar between PVTT and non-PVTT patients. Thus, the cirrhosis grade was not used as a prognosis marker for PVTT formation in the present study.
The serum HBV load, also, was not significantly related to PVTT formation in HBV-associated HCC in the present study. A previous study reported that HBV infection could induce the alteration of the TGF-β-miR-34a-CCL22 axis, and thus abnormality of the liver microenvironment, which may induce the formation of PVTT [23]. A retrospective study also revealed that active HBV replication might contribute to vascular invasion in HCC patients [24]. That is, positive HBsAg or a certain level of HBsAg might be a more accurate clinical marker to predict the occurrence of PVTT.
The limitation of the present study is rooted in its retrospective case-control single-center nature, which not only carries recall and selection bias but also can only be considered as level III evidence. Thus, further investigations with randomized control trials and long-term, large-scale analyses of clinical data are required to confirm this scoring system and its prediction capability.
In conclusion, gender, tumor size, NLR, AFP and CRP concentrations are reliable factors that can predict the likelihood of PVTT formation in HBV-related HCC patients, and indirectly indicate the prognosis and long-term consequences for HCC patients.
Acknowledgements
This work was supported by the Major Scientific and Technological Project for the Prevention and Treatment of AIDS, Viral Hepatitis, and Other Major Infectious Diseases (grant number: 2018ZX10303502), and scientific research special item of National TCM Clinical Research Base (grant number: 2021JDZX2070).
Disclosure of conflict of interest
None.
References
- 1.Chen W, Zheng R, Baade PD, Zhang S, Zeng H, Bray F, Jemal A, Yu XQ, He J. Cancer statistics in China, 2015. CA Cancer J Clin. 2016;66:115–132. doi: 10.3322/caac.21338. [DOI] [PubMed] [Google Scholar]
- 2.Pawarode A, Voravud N, Sriuranpong V, Kullavanijaya P, Patt YZ. Natural history of untreated primary hepatocellular carcinoma: a retrospective study of 157 patients. Am J Clin Oncol. 1998;21:386–391. doi: 10.1097/00000421-199808000-00014. [DOI] [PubMed] [Google Scholar]
- 3.Cerrito L, Annicchiarico BE, Iezzi R, Gasbarrini A, Pompili M, Ponziani FR. Treatment of hepatocellular carcinoma in patients with portal vein tumor thrombosis: beyond the known frontiers. World J Gastroenterol. 2019;25:4360–4382. doi: 10.3748/wjg.v25.i31.4360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Shi J, Lai EC, Li N, Guo WX, Xue J, Lau WY, Wu MC, Cheng SQ. Surgical treatment of hepatocellular carcinoma with portal vein tumor thrombus. Ann Surg Oncol. 2010;17:2073–2080. doi: 10.1245/s10434-010-0940-4. [DOI] [PubMed] [Google Scholar]
- 5.Ye JZ, Zhang YQ, Ye HH, Bai T, Ma L, Xiang BD, Li LQ. Appropriate treatment strategies improve survival of hepatocellular carcinoma patients with portal vein tumor thrombus. World J Gastroenterol. 2014;20:17141–17147. doi: 10.3748/wjg.v20.i45.17141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhang XP, Gao YZ, Chen ZH, Wang K, Cheng YQ, Guo WX, Shi J, Zhong CQ, Zhang F, Cheng SQ. In-hospital mortality after surgical resection in hepatocellular carcinoma patients with portal vein tumor thrombus. J Cancer. 2019;10:72–80. doi: 10.7150/jca.27102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Shuqun C, Mengchao W, Han C, Feng S, Jiahe Y, Guanghui D, Wenming C, Peijun W, Yuxiang Z. Tumor thrombus types influence the prognosis of hepatocellular carcinoma with the tumor thrombi in the portal vein. Hepatogastroenterology. 2007;54:499–502. [PubMed] [Google Scholar]
- 8.Wu B, Zhang Y, Tan H, Shi H. Value of (18)F-FDG PET/CT in the diagnosis of portal vein tumor thrombus in patients with hepatocellular carcinoma. Abdom Radiol (NY) 2019;44:2430–2435. doi: 10.1007/s00261-019-01997-2. [DOI] [PubMed] [Google Scholar]
- 9.Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, Zhu AX, Murad MH, Marrero JA. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology. 2018;67:358–380. doi: 10.1002/hep.29086. [DOI] [PubMed] [Google Scholar]
- 10.Catalano OA, Choy G, Zhu A, Hahn PF, Sahani DV. Differentiation of malignant thrombus from bland thrombus of the portal vein in patients with hepatocellular carcinoma: application of diffusion-weighted MR imaging. Radiology. 2010;254:154–162. doi: 10.1148/radiol.09090304. [DOI] [PubMed] [Google Scholar]
- 11.Subbotin VM. A hypothesis on paradoxical privileged portal vein metastasis of hepatocellular carcinoma. Can organ evolution shed light on patterns of human pathology, and vice versa? Med Hypotheses. 2019;126:109–128. doi: 10.1016/j.mehy.2019.03.019. [DOI] [PubMed] [Google Scholar]
- 12.Yu MW, Chang HC, Chang SC, Liaw YF, Lin SM, Liu CJ, Lee SD, Lin CL, Chen PJ, Lin SC, Chen CJ. Role of reproductive factors in hepatocellular carcinoma: impact on hepatitis B- and C-related risk. Hepatology. 2003;38:1393–1400. doi: 10.1016/j.hep.2003.09.041. [DOI] [PubMed] [Google Scholar]
- 13.Chen J, Shi X, Luo T, Zhao Y, Ye J, Bai T, Li L. The correlations between hepatitis B virus infection and hepatocellular carcinoma with portal vein tumor thrombus or extrahepatic metastasis. Eur J Gastroenterol Hepatol. 2020;32:373–377. doi: 10.1097/MEG.0000000000001514. [DOI] [PubMed] [Google Scholar]
- 14.Chen JS, Wang Q, Chen XL, Huang XH, Liang LJ, Lei J, Huang JQ, Li DM, Cheng ZX. Clinicopathologic characteristics and surgical outcomes of hepatocellular carcinoma with portal vein tumor thrombosis. J Surg Res. 2012;175:243–250. doi: 10.1016/j.jss.2011.03.072. [DOI] [PubMed] [Google Scholar]
- 15.Zhong Y, Deng M, Xu R. Reappraisal of evidence of microscopic portal vein involvement by hepatocellular carcinoma cells with stratification of tumor size. World J Surg. 2015;39:1142–1149. doi: 10.1007/s00268-014-2807-5. [DOI] [PubMed] [Google Scholar]
- 16.Nomura F, Ohnishi K, Tanabe Y. Clinical features and prognosis of hepatocellular carcinoma with reference to serum alpha-fetoprotein levels. analysis of 606 patients. Cancer. 1989;64:1700–1707. doi: 10.1002/1097-0142(19891015)64:8<1700::aid-cncr2820640824>3.0.co;2-z. [DOI] [PubMed] [Google Scholar]
- 17.Hsu CY, Liu PH, Lee YH, Hsia CY, Huang YH, Lin HC, Chiou YY, Lee FY, Huo TI. Using serum alpha-fetoprotein for prognostic prediction in patients with hepatocellular carcinoma: what is the most optimal cutoff? PLoS One. 2015;10:e0118825. doi: 10.1371/journal.pone.0118825. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.An SL, Xiao T, Wang LM, Rong WQ, Wu F, Feng L, Liu FQ, Tian F, Wu JX. Prognostic significance of preoperative serum alpha-fetoprotein in hepatocellular carcinoma and correlation with clinicopathological factors: a single-center experience from China. Asian Pac J Cancer Prev. 2015;16:4421–4427. doi: 10.7314/apjcp.2015.16.10.4421. [DOI] [PubMed] [Google Scholar]
- 19.Yao M, Zhao J, Lu F. Alpha-fetoprotein still is a valuable diagnostic and prognosis predicting biomarker in hepatitis B virus infection-related hepatocellular carcinoma. Oncotarget. 2016;7:3702–3708. doi: 10.18632/oncotarget.6913. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Shiraki T, Ishizuka M, Kubota K, Kato M, Matsumoto T, Mori S, Shimizu T, Aoki T. An elevated neutrophil-to-lymphocyte ratio predicts a poor postoperative survival in primary hepatocellular carcinoma patients with a normal preoperative serum level of alpha-fetoprotein. Surg Today. 2019;49:661–669. doi: 10.1007/s00595-019-01781-1. [DOI] [PubMed] [Google Scholar]
- 21.Mano Y, Shirabe K, Yamashita Y, Harimoto N, Tsujita E, Takeishi K, Aishima S, Ikegami T, Yoshizumi T, Yamanaka T, Maehara Y. Preoperative neutrophil-to-lymphocyte ratio is a predictor of survival after hepatectomy for hepatocellular carcinoma: a retrospective analysis. Ann Surg. 2013;258:301–305. doi: 10.1097/SLA.0b013e318297ad6b. [DOI] [PubMed] [Google Scholar]
- 22.Li SH, Wang QX, Yang ZY, Jiang W, Li C, Sun P, Wei W, Shi M, Guo RP. Prognostic value of the neutrophil-to-lymphocyte ratio for hepatocellular carcinoma patients with portal/hepatic vein tumor thrombosis. World J Gastroenterol. 2017;23:3122–3132. doi: 10.3748/wjg.v23.i17.3122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Yang P, Li QJ, Feng Y, Zhang Y, Markowitz GJ, Ning S, Deng Y, Zhao J, Jiang S, Yuan Y, Wang HY, Cheng SQ, Xie D, Wang XF. TGF-beta-miR-34a-CCL22 signaling-induced Treg cell recruitment promotes venous metastases of HBV-positive hepatocellular carcinoma. Cancer Cell. 2012;22:291–303. doi: 10.1016/j.ccr.2012.07.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wei X, Li N, Li S, Shi J, Guo W, Zheng Y, Cheng S. Hepatitis B virus infection and active replication promote the formation of vascular invasion in hepatocellular carcinoma. BMC Cancer. 2017;17:304. doi: 10.1186/s12885-017-3293-6. [DOI] [PMC free article] [PubMed] [Google Scholar]