Background
Portal vein tumor thrombosis (PVTT) is a common complication of hepatocellular carcinoma and is one of the most negative prognostic factors. The management of patients with PVTT is challenging. The aim of the study was to develop a score predictive of tumor thrombosis.
Methods
Data from a large cohort of 2243 hepatocellular carcinoma patients (all stages) recorded in the Progetto Epatocarcinoma Campania (January 2013–April 2021) database were analyzed. To construct the score, univariate generalized estimated equation models, the bootstrap approach for internal validation, and a regression coefficient-based scoring system were used.
Results
PVTT (any location) was found in 14.4% of cases and was related to shorter survival. Males, younger patients, and symptomatic cases were more prevalent among the PVTT group. At multivariate analysis, size ≥5 cm, massive or infiltrative hepatocellular carcinoma growth, and alpha-fetoprotein ≥400 ng/mL were significantly associated with PVTT. A risk prediction score of PVTT based on eight variables was developed. Using a continuous score, the risk was associated with an odds ratio (OR) of 1.30 (1.27–1.34; P < 0.001). Considering a dichotomous score >8 versus a score ≤8 the OR for PVTT was 11.33 (8.55–15.00; P < 0.001).
Conclusion
The risk score for PVTT might be useful for clinicians to optimize hepatocellular carcinoma management by picking out patients with more aggressive cancers and higher mortality rates. Prospective validation of the score is needed before its application in daily clinical practice.
Keywords: hepatocellular carcinoma, portal vein tumor thrombosis, risk score, survival
Introduction
The incidence of hepatocellular carcinoma has been increasing steadily over the past decades and is diagnosed in well over 500 000 people worldwide [1,2]. In Europe, according to the most recent data from GLOBOCAN, in 2020 the age-standardized incidence rate of liver cancer was 87.6 per 100 000 with 78.4 deaths per 100 000 [3]. This highlights that prognosis of hepatocellular carcinoma remains dismal, with one of the worst survival rates among all cancers, even though survival rates may vary somewhat among different countries and settings [4,5]. The primary risk factors for hepatocellular carcinoma include hepatitis B virus or hepatitis C virus (HCV) infection, alcoholic liver disease, and nonalcoholic fatty liver disease [1].
Hepatocellular carcinoma is characterized by a strong propensity to invade the surrounding liver vasculature [6]. Macrovascular invasion or gross tumor invasion into the main portal vein and branches, hepatic veins, and inferior vena cava, is common in hepatocellular carcinoma [6–8]. Portal vein tumor thrombosis (PVTT) is the most common form of macrovascular invasion and is present in 10–40% of all hepatocellular carcinoma patients at diagnosis [6–8]. At present, PVTT is associated with a very poor prognosis and overall survival (OS) of only 2–4 months with the best supportive care [7–10].
The poor prognosis in patients with PVTT results from a combination of multiple factors that include more aggressive behavior of the primary tumor, portal hypertension with reduced tolerance to treatment, and a decline in liver function reserve [11]. The obstruction of portal blood flow by the tumor increases portal pressure and consequently causes ascites, hepatic encephalopathy, liver failure, and esophageal varices [11]. PVTT can also facilitate tumor spread throughout the entire liver parenchyma. Given the hypothesized mechanisms and physiological consequences of PVTT, therapeutic options for hepatocellular carcinoma patients with PVTT are limited; however, the presence of PVTT, regardless of the extent of the primary hepatic tumor, is often considered a contraindication for traditional treatment options such as surgical resection or transarterial chemoembolization [11–13]. The available data on the use of anticoagulants has not yet provided definitive results that can guide daily practice [14].
An accurate scoring system to predict the risk of PVTT might help clinicians to optimize hepatocellular carcinoma management by selecting patients with more aggressive cancer types and higher risk of death. Herein, we present clinicopathological data and outcomes on a large cohort of hepatocellular carcinoma patients. We also use this data to develop a score to predict the risk of PVTT.
Materials and methods
Study population
This was a retrospective analysis of an online region-wide database (Progetto Epatocarcinoma Campania) containing clinicopathological information on 2243 consecutively observed patients with hepatocellular carcinoma. Data were prospectively collected at 18 community and academic centers located in the Campania region and updated every 6 months. The database covered the time period from January 2013 to April 2021. The consistency of the data set was checked by the group coordinator (G.G.D.C.) and discussed during periodic meetings. The electronic registry gathered data in compliance with the current Italian privacy laws. The present study conforms to the ethical guidelines of the Declaration of Helsinki and was approved by the respective Institutional Review Boards of the participating centers. All patients with hepatocellular carcinoma being treated at the participating centers were included in the database and informed consent was obtained.
When included in the database, all patients underwent abdominal computed tomography (CT) or MRI to provide information on size, number of tumor nodules, and presence or absence of PVTT and alpha-fetoprotein levels. Among the characteristics of hepatocellular carcinoma, monofocal, multifocal, infiltrative, and massive growth were evaluated. Hepatocellular carcinoma was classified with infiltrative growth when the margins of the nodules were poorly defined. The massive type was considered when the diameter of hepatocellular carcinoma was larger than 10 cm. A complete blood count, routine blood liver function tests (bilirubin, albumin, International Normalized Ratio [INR]), and patient demographics were also collected.
Diagnosis
Diagnosis was made by tumor biopsy in 12% of cases and according to noninvasive imaging criteria used by the European Association for the Study of the Liver in the remaining cases [12]. Liver function was evaluated by bilirubin, albumin, INR, and presence/absence of ascites at diagnosis. Hepatocellular carcinoma stage was classified according to the Barcelona Clinic Liver Cancer (BCLC) staging system [15]. The choice of the therapeutic strategy was performed according to the BCLC algorithm and the physician’s judgment, using a multidisciplinary approach.
Classification of portal vein tumor thrombosis
The most conventional classification for PVTT is the classification proposed by the Liver Cancer Study Group of Japan, which divides PVTT into four classes according to the extent of the thrombus: Vp1 is defined by the presence of a PVTT distal to, but not in, the second-order branches of the portal vein; Vp2 is defined by the presence of a PVTT in the second-order branches of the portal vein; Vp3 is defined by the presence of a PVTT in the first-order branches of the portal vein, and Vp4 is defined by the presence of a PVTT in the main trunk of the portal vein or a contralateral portal vein branch or both [16].
Characteristics considered diagnostic of neoplastic thrombosis included thrombus enhancement in the arterial phase by CT, MRI, or ultrasound with contrast, expansion of the portal vein, and clear relation between the thrombus and the neoplasm [17]. Thrombus was defined as vascular in cases not showing arterial enhancement and when it was noncontiguous with the tumor.
Outcomes and follow-up
Overall survival (OS) was defined as the interval between the first visit and either death or the last follow-up visit. Patients treated with surgical resection or locoregional treatments were followed with abdominal ultrasound every three months and CT or MRI every 6 months. During systemic treatment with a tyrosine-kinase inhibitor, CT or MRI was performed every 2–3 months. Any change to these schedules was made according to the physician’s judgment.
Statistical analysis
Descriptive characteristics of the patients and observed thromboses were reported as frequencies and percentages (%) for categorical variables and as mean (SD) for continuous variables. We used Pearson’s chi-squared test (categorical variables) and Wilcoxon test (continuous variables) to compare the characteristics of the patients without thrombosis to those who developed an episode of thrombosis. To investigate risk factors for thrombosis, we used univariate generalized estimated equation (GEE) models to account for the presence of thrombosis recurrences. We then performed multivariate analysis including only the variables showing a P value <0.10 in univariate analysis.
On the basis of multivariate analysis, we computed a score reflecting the risk of thrombosis. Specifically, we used the bootstrap approach (500 replications) for internal validation and then a regression coefficient-based (Schneeweiss) scoring system [18]. Subsequently, we used the GEE models to investigate the impact of the score on the risk of thrombosis, including the score as the independent variable, both as a continuous and as a dichotomous variable. Kaplan–Meier curves and Cox regression models were used to investigate differences in survival between patients who developed thromboses and those who did not. All statistical analyses were performed using Stata version 16.0 (Stata Corporation, College Station, Texas, USA).
Results
Patient characteristics
Baseline characteristics of the 2243 patients subdivided according to the presence of thrombosis are summarized in Table 1. Groups were comparable, although males (72%), younger age, and symptomatic cases (17%) were more prevalent in the thrombosis group. Hepatocellular carcinoma was diagnosed more frequently during surveillance in patients without thrombosis than with thrombosis, 52% versus 37%, respectively. In all, 398 (17.7%) patients had at least one thrombosis. Most patients were affected with cirrhosis and no difference in the cause of the liver disease was observed between patients with and without thrombosis. The majority of cases had a viral etiology, which was related to HCV in 65% of cases. Of note, in 7% of patients, the hepatic disease had a metabolic cause. The most frequent comorbidities were arterial hypertension (49%) and diabetes (30%).
Table 1.
Characteristics of patients with hepatocellular carcinoma
| All patients | No thrombosis | Thrombosis | P value | |
|---|---|---|---|---|
| Characteristic | N = 2243 | N = 1845 | N = 398 | |
| Male, N (%) | 1621 (72%) | 1308 (71%) | 313 (79%) | 0.002 |
| BMI, mean (SD) | 26.09 (3.98) | 26.12 (3.99) | 25.96 (3.95) | 0.455 |
| Age at diagnosis, mean (SD) | 68.45 (10.34) | 68.83 (10.37) | 66.69 (10.02) | <0.001 |
| Diagnosis type, N (%) | ||||
| Not known | 88 (4%) | 77 (4%) | 11 (3%) | <0.001 |
| Surveillance | 1109 (49%) | 962 (52%) | 147 (37%) | |
| Casual | 882 (39%) | 709 (38%) | 173 (43%) | |
| Symptomatic | 164 (7%) | 97 (5%) | 67 (17%) | |
| Liver disease, N (%) | ||||
| Not known or none | 140 (6%) | 116 (6%) | 24 (6%) | 0.922 |
| Chronic hepatitis | 153 (7%) | 127 (7%) | 26 (7%) | |
| Cirrhosis | 1933 (86%) | 1589 (86%) | 344 (86%) | |
| Steatosis | 17 (1%) | 13 (1%) | 4 (1%) | |
| Etiology, N (%) | ||||
| Alcohol | 166 (7%) | 129 (7%) | 37 (9%) | 0.267 |
| HBV | 317 (14%) | 257 (14%) | 60 (15%) | |
| HCV | 1453 (65%) | 1211 (66%) | 242 (61%) | |
| Metabolic | 168 (7%) | 133 (7%) | 35 (9%) | |
| Mixed viral | 38 (2%) | 29 (2%) | 9 (2%) | |
| Other/unknown | 101 (5%) | 86 (5%) | 15 (4%) | |
| Child–Pugh, N (%) | ||||
| A | 1153 (75%) | 1005 (79%) | 148 (55%) | <0.001 |
| B | 307 (20%) | 222 (17%) | 85 (32%) | |
| C | 80 (5%) | 46 (4%) | 34 (13%) | |
| Tumor growth, N (%) | ||||
| Monofocal | 1115 (53%) | 995 (58%) | 120 (32%) | <0.001 |
| Multifocal | 818 (39%) | 660 (39%) | 158 (41%) | |
| Infiltrative | 107 (5%) | 31 (2%) | 76 (20%) | |
| Massive | 50 (2%) | 23 (1%) | 27 (7%) | |
| Tumor size, cm | ||||
| <5 | 1907 (86%) | 1611 (89%) | 296 (75%) | <0.001 |
| ≥5 | 305 (14%) | 204 (11%) | 101 (25%) | |
| Alpha-fetoprotein, N (%) | ||||
| <400 ng/mL | 1646 (84%) | 1406 (88%) | 240 (66%) | <0.001 |
| ≥400 ng/mL | 318 (16%) | 192 (12%) | 126 (34%) | |
| Cardiopathy, N (%) | ||||
| Not known | 105 (5%) | 92 (5%) | 13 (3%) | 0.299 |
| Yes | 226 (10%) | 188 (10%) | 38 (10%) | |
| Hypertension, N (%) | ||||
| Not known | 95 (4%) | 85 (5%) | 10 (3%) | 0.002 |
| Yes | 1106 (49%) | 933 (51%) | 173 (43%) | |
| Type 2 diabetes, N (%) | ||||
| Not known | 103 (5%) | 91 (5%) | 12 (3%) | 0.253 |
| Yes | 668(30%) | 548 (30%) | 120 (30%) | |
| COPD, N (%) | ||||
| Not known | 109 (5%) | 94 (5%) | 15 (4%) | 0.465 |
| Yes | 228 (10%) | 190 (10%) | 38 (10%) | |
| Other tumors reported, N (%) | 132 (6%) | 114 (6%) | 18 (5%) | 0.203 |
Data are provided for patients with and without thrombosis.
HBV, hepatitis B virus; HCV, hepatitis C virus.
Wilcoxon test for continuous variables and Pearson’s chi-squared test for categorical variables.
Characteristics of thrombosis
Characteristics of thrombosis are shown in Table 2. PVTT was observed in 323 (14.4%) patients and vascular thrombosis in only 47 (2.1%) hepatocellular carcinoma patients. There were 448 thrombotic events: in 32% of cases the thrombus involved the main trunk, while it involved the right first-order branch and left first-order branch in 27% and 11% of cases, respectively. The thrombus was observed at other locations in smaller percentages of patients.
Table 2.
Characteristics of portal vein tumor thrombosis
| Type of thrombosis | N (%) |
|---|---|
| Neoplastic | 323 (81%) |
| Non-neoplastic | 47 (12%) |
| Doubtful/undefined | 28 (7%) |
| Site of thrombosis | |
| VP1-VP2 | 42 (8%) |
| VP3 | 172 (38%) |
| VP4 | 154 (35%) |
| Mesenteric vein | 17 (4%) |
| Splenic vein | 11 (2%) |
| Vena cava | 7 (2%) |
| Missing | 52 (12%) |
Data are provided for the total number of thromboses, and some patients had more than one thrombosis.
Risk factors for thrombosis
Patients were analyzed according to the presence of PVTT or vascular thrombosis. Among baseline variables, ascites, lesion size ≥5 cm, hepatocellular carcinoma growth (multifocal, infiltrative, massive), bilirubin >1.10 mg/dL, albumin <3.5 g/dL, INR > 1.0, alpha-fetoprotein ≥400 ng/mL, and moderate or severe congestive gastropathy were all related to the presence of PVTT at univariate analysis. The highest odds ratio (OR) was observed for the infiltrative growth pattern (Table 3).
Table 3.
Univariate analysis of factors related to portal vein tumor thrombosis
| OR (95% CI) | P value | |
|---|---|---|
| Ascites | ||
| Absence | 1.00 (ref) | – |
| Presence | 4.06 (3.28–5.04) | <0.001 |
| Diameter, cm | ||
| <5 | 1.00 (ref) | – |
| ≥5 | 2.62 (2.08–3.31) | <0.001 |
| Albumin, g/dL | 0.50 (0.41–0.60) | <0.001 |
| Albumin | ||
| Normal range: [3.5–5.3] | 1.00 (ref) | – |
| Low: <3.5 | 2.15 (1.74–2.65) | <0.001 |
| Bilirubin, mg/dL | 1.15 (1.10–1.21) | <0.001 |
| Bilirubin | ||
| Normal range [0.20–1.10] | 1.00 (ref) | – |
| High: >1.10 | 2.13 (1.73–2.63) | <0.001 |
| INR | 3.46 (2.39–5.01) | <0.001 |
| INR | ||
| Normal range | 1.00 (ref) | – |
| High: >1.0 | 1.99 (1.60–2.48) | <0.001 |
| Alpha-fetoprotein, ng/mL | ||
| <400 | 1.00 (ref) | – |
| ≥400 | 4.40 (3.43–5.66) | <0.001 |
| Tumor growth | ||
| Monofocal | 1.00 (ref) | – |
| Multifocal | 2.63 (2.05–3.36) | <0.001 |
| Infiltrative | 24.70 (16.84–36.22) | <0.001 |
| Massive | 11.14 (6.32–19.66) | <0.001 |
| Gastropathy | ||
| Absent/mild | 1.00 (ref) | – |
| Moderate/severe | 2.06 (1.53–2.76) | <0.001 |
| Etiology | ||
| HCV | 1.00 (ref) | – |
| HBV | 1.24 (0.91–1.70) | 0.166 |
| Alcohol | 1.42 (0.97–2.08) | 0.070 |
| Metabolic | 1.47 (0.99–2.17) | 0.055 |
| Mixed viral | 1.70 (0.82–3.53) | 0.156 |
| Other | 1.62 (0.60–4.38) | 0.344 |
P values in bold are significant.
CI, confidence interval; HBV, hepatitis B virus; HCV, hepatitis C virus; OR, odds ratio.
At multivariate analysis, ascites, tumor size ≥5 cm, massive or infiltrative hepatocellular carcinoma growth, and alpha-fetoprotein ≥400 ng/mL were all significantly associated with PVTT (Table 4). Among patients with vascular thrombosis, only bilirubin >1.10 mg/dL (OR = 3.70, 1.51–9.06; P = 0.004) and multifocal hepatocellular carcinoma (OR = 0.42, 0.18–0.95; P = 0.037) were the variables related to thrombosis.
Table 4.
Multivariate analysis of factors related to portal vein tumor thrombosis
| Characteristic | OR (95% CI) | P value |
|---|---|---|
| Ascites | ||
| Absence | 1.00 (ref) | – |
| Presence | 2.68 (1.81–3.95) | <0.001 |
| Diameter, cm | ||
| <5 | 1.00 (ref) | – |
| ≥5 | 2.15 (1.42–3.25) | <0.001 |
| Albumin, g/dL | ||
| Normal range: [3.5–5.3] | 1.00 (ref) | – |
| Low: <3.5 | 1.20 (0.80–1.78) | 0.375 |
| Bilirubin, mg/dL | ||
| Normal range [0.20–1.10] | 1.00 (ref) | – |
| High: >1.10 | 1.26 (0.85–1.87) | 0.250 |
| INR | ||
| Normal range | 1.00 (ref) | – |
| High: >1.0 | 1.17 (0.78–1.74) | 0.454 |
| Alpha-fetoprotein, ng/mL | ||
| <400 | 1.00 (ref) | – |
| ≥400 | 3.55 (2.37–5.33) | <0.001 |
| Tumor growth | ||
| Monofocal | 1.00 (ref) | – |
| Multifocal | 1.40 (0.96–2.04) | 0.081 |
| Infiltrative | 11.75 (6.01–22.95) | <0.001 |
| Massive | 3.38 (1.34–8.50) | 0.010 |
| Gastropathy | ||
| Absent/mild | 1.00 (ref) | – |
| Moderate/severe | 1.15 (0.77–1.71) | 0.505 |
P values in bold are significant.
CI, confidence interval; INR, International Normalized Ratio; OR, odds ratio.
Survival analysis
The median follow-up time in patients without PVTT was 15.55 (4.57–36.50) months and 11.13 (3.33–25.4) months in patients with PVTT. Patients with PVTT had a risk of death that was doubled compared to patients without thrombosis (Fig. 1). In patients with and without thrombosis the median survival time was 46.67 (42.60–50.77) and 20.3 (16.27–24.37) months, respectively. The probability of survival at 1 year in patients with and without thrombosis was 0.67 and 0.85, while at 2 years was 0.45 and 0.72, respectively [overall hazard ratio = 2.01 (1.71–2.36); P < 0.001]. There was no difference in OS in patients with PVTT at the main trunk, or first-order branches of the portal vein (log-rank test: P value = 0.3098). As shown in Fig. 2, the presence of diabetes was not related to OS but patients with PVTT suffering from diabetes had the highest risk of death [hazard ratio = 1.02 (0.88–1.18); P = 0.756].
Fig. 1.
OS according to presence or absence of PVTT (Kaplan–Meier method). OS, overall survival; PVTT, portal vein tumor thrombosis.
Fig. 2.
OS by presence or absence of PVTT and type 2 diabetes at baseline (Kaplan–Meier method). OS, overall survival; PVTT, portal vein tumor thrombosis.
Risk score for portal vein tumor thrombosis
Multivariate analysis was used to calculate a risk prediction score for PVTT (Table 5). The score was based on eight factors (Table 6 provided as a Supplementary Excel file, Supplemental digital content 1, http://links.lww.com/EJGH/A849 can be used to calculate the score for the individual patient). A total score of 22 was possible. Using a continuous score, the risk of PVTT was associated with an OR of 1.30 (1.27–1.34; P < 0.001); however, considering a dichotomous score >8 versus a score ≤8 the OR for the risk of PVTT was 11.33 (8.55–15.00; P < 0.001).
Table 5.
Risk factors used to calculate predictive score for portal vein tumor thrombosis
| Original β-coefficient ± SE | Bootstrap (500 replication) | Risk-scoring (Schneeweiss) | |
|---|---|---|---|
| Ascites | |||
| Absence | 0 | 0 | 0 |
| Presence | 0.98 ± 0.20 | 0.98 ± 0.20 | 3 |
| Diameter, cm | |||
| <5 | 0 | 0 | 0 |
| ≥5 | 0.77 ± 0.21 | 0.78 ± 0.25 | 3 |
| Albumin, g/dL | |||
| Normal range: [3.5–5.3] | 0 | 0 | 0 |
| Low: <3.5 | 0.18 ± 0.20 | 0.20 ± 0.22 | 1 |
| Bilirubin, mg/dL | |||
| Normal range [0.20–1.10] | 0 | 0 | 0 |
| High: >1.10 | 0.23 ± 0.20 | 0.21 ± 0.21 | 1 |
| INR | |||
| Normal range | 0 | 0 | 0 |
| High: >1.0 | 0.16 ± 0.20 | 0.16 ± 0.22 | 1 |
| Alpha-fetoprotein, ng/mL | |||
| <400 | 0 | 0 | 0 |
| ≥400 | 1.27 ± 0.21 | 1.27 ± 0.23 | 4 |
| Tumor aspect | |||
| Monofocal | 0 | 0 | 0 |
| Multifocal | 0.33 ± 0.19 | 0.33 ± 0.19 | 1 |
| Infiltrating | 2.47 ± 0.34 | 2.52 ± 0.39 | 8 |
| Massive | 1.22 ± 0.47 | 1.21 ± 0.65 | 4 |
| Gastropathy | |||
| No | 0 | 0 | 0 |
| Mild | 0.06 ± 0.22 | 0.04 ± 0.25 | 0 |
| Moderate | 0.20 ± 0.26 | 0.20 ± 0.29 | 1 |
| Severe | 0.07 ± 0.37 | 0.07 ± 0.40 | 0 |
INR, International Normalized Ratio; SE, standard error.
Discussion
PVTT is not an uncommon event in patients with hepatocellular carcinoma and represents one of the most important prognostic factors in the natural history of this neoplasm [19]. When portal thrombosis is present, the spectrum of treatments is narrowed and patient survival is poor [19,20]. The reason is that most of the thromboses observed in conjunction with hepatocellular carcinoma are caused by neoplastic invasion of the vein, which is therefore reflective of more aggressive tumor behavior. In a large cohort of consecutive hepatocellular carcinoma patients macrovascular invasion was associated with a worse performance status and liver function, and the extension of thrombosis was associated with greater impairment in liver function.
In a previous study, PVTT independently predicted a decreased long-term survival in patients undergoing both curative and noncurative treatments [20]. Cytological or histological characterization of the thrombus to obtain a definite diagnosis is not usually performed in routine practice due to technical difficulties as well as the risk of complications for which noninvasive criteria based on imaging techniques are used. The frequency of PVTT in our series is somewhat higher (14.4% vs. 11.1%) than that reported in a recent publication by the ITA.LI.CA group, which may be due to the lower number of patients with hepatocellular carcinoma under surveillance in our cohort (49% vs. 60%) and to the fact that we considered any anatomical location [21].
In daily clinical practice, the diagnosis of portal neoplastic thrombosis is challenging, and for this reason, we attempted to use readily available imaging, clinical, pathological, and laboratory data to create a risk score that is predictive of PVTT. The score found by multivariate analysis included eight factors: three related to liver function (INR, bilirubin, albumin), three to the pathological features of the tumor (alpha-fetoprotein, diameter, infiltration), and two to portal hypertension (ascites, portal hypertensive gastropathy). The proposed score may be useful not only for the diagnosis of PVTT but also to define patients with a more aggressive cancer. Actually, there are different treatments that can be used in patients with PVTT, in sequence or combination (radioembolization, systemic therapy). In line with previous studies, the presence of thrombosis was associated with the size of the tumor, multifocality, and high levels of alpha-fetoprotein [22,23]. In our analysis, two other features of hepatocellular carcinoma, namely infiltrative and massive growth, had a high predictive value for the presence of thrombosis. These characteristics are indicative of increased aggressiveness of the tumor but have not always been evaluated in previous studies on PVTT. This is especially true for infiltrative growth since this feature is not always been adequately described by radiologists [24]. An infiltrative growth pattern has been reported to be present in 7–20% of hepatocellular carcinoma in various studies [25–27]. Infiltrative growth has also been associated with poor response to therapy and worse prognosis [28]. Other factors related to the presence of PVTT include Child–Pugh classes B–C and severe portal hypertension [29].
The main biochemical tests of liver function, namely bilirubin, albumin, and INR, were all associated with a diagnosis of PVTT. Liver function impairment is mainly due to cirrhosis which is present in most patients with hepatocellular carcinoma. An alternative explanation is that thrombosis, especially when it affects the main branches, causes worsening of liver function due to the rapid growth and consequent rapid arrest of portal flow. Albumin and bilirubin are both associated with the prognosis of patients with hepatocellular carcinoma and are used in the formulation of the albumin-bilirubin score [30]. Indeed, liver failure is one of the primary determinants that have an impact on the natural history of hepatocellular carcinoma [30]. In addition, a low level of serum albumin is related to the presence of thrombosis in patients with cirrhosis, and a recent study has suggested that albumin can modulate the hemostatic process through interference with platelet activation [31]. While this suggests that anticoagulant therapies may be of benefit in patients with hepatocellular carcinoma and PVTT, their current clinical utility is still being debated [14]. Nonetheless, a low level of albumin also correlates with the activation of systemic inflammation, which may have a role in tumor growth through the modulation of alpha-fetoprotein or by affecting kinases that control growth [32]. Regarding bilirubin, an increase in bilirubin is also an index of liver function deterioration and is associated with a more aggressive phenotype as is multifocality, presence of PVTT, and higher levels of alpha-fetoprotein [33]. Lastly, considering parameters of liver function, alterations in INR may be a consequence not only of the impaired synthetic capacity of the liver but also of the activation of hemostatic processes. Indeed, the presence of hepatocellular carcinoma in a cirrhotic patient is responsible for a hypercoagulable state that may be associated with greater consumption of coagulation factors [34]. The moderate and severe portal hypertensive gastropathy observed in patients with hepatocellular carcinoma is a complication of portal hypertension, and the association between gastropathy and thrombosis may be explained by the sudden increase of portal hypertension caused by the neoplastic invasion of portal vein [35].
To our knowledge, this is one of the few studies that have attempted to construct a score to predict the presence of PVTT in patients with hepatocellular carcinoma. Using a cutoff value of 8, the score was highly accurate in predicting the presence of PVTT with an OR of 11.33. Akkiz et al. reported that three factors, tumor diameter >5.0 cm, tumor multifocality, and elevated alpha-fetoprotein, could be used together to predict the presence of PVTT with an OR of 17.9 [36]. In a study of 47 cirrhotic patients with hepatocellular carcinoma, Serag et al. reported that Annexin A5 and phosphatidylserine-bearing microparticles may be predictive of PVTT [37]. Herein, multivariate analysis found eight factors that were used to construct our score.
A major strength of our study is the large sample size of well characterized patients with a consecutive and prospective enrollment over a recent time period in a well defined geographical area; however, there are some limitations. Foremost among these is the need for prospective validation of the score on a larger cohort of patients in different settings; however, in the lack of additional studies, at present, the score may be of value to clinicians involved in the management of patients with hepatocellular carcinoma.
Acknowledgements
Editorial assistance was provided by Edra S.p.A. and unconditionally funded by AstraZeneca. The data from this study are available from the corresponding author upon reasonable request.
Conflicts of interest
There are no conflicts of interest.
Supplementary Material
Footnotes
Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website, www.eurojgh.com.
References
- 1.El-Serag HB. Hepatocellular carcinoma. N Engl J Med 2011; 365:1118–1127. [DOI] [PubMed] [Google Scholar]
- 2.El-Serag HB, Kanwal F. Epidemiology of hepatocellular carcinoma in the United States: where are we? Where do we go? Hepatology 2014; 60:1767–1775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Dyba T, Randi G, Bray F, Martos C, Giusti F, Nicholson N, et al. The European cancer burden in 2020: incidence and mortality estimates for 40 countries and 25 major cancers. Eur J Cancer 2021; 157:308–347. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Goutte N, Sogni P, Bendersky N, Barbare JC, Falissard B, Farges O. Geographical variations in incidence, management and survival of hepatocellular carcinoma in a Western country. J Hepatol 2017; 66:537–544. [DOI] [PubMed] [Google Scholar]
- 5.Llovet JM, Bustamante J, Castells A, Vilana R, Ayuso MC, Sala M, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic trials. Hepatology 1999; 29:62–67. [DOI] [PubMed] [Google Scholar]
- 6.Minagawa M, Makuuchi M. Treatment of hepatocellular carcinoma accompanied by portal vein tumor thrombus. World J Gastroenterol 2006; 12:7561–7567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Cheung TK, Lai CL, Wong BC, Fung J, Yuen M-F. Clinical features, biochemical parameters, and virological profiles of patients with hepatocellular carcinoma in Hong Kong. Aliment Pharmacol Ther 2006; 24:573–583. [DOI] [PubMed] [Google Scholar]
- 8.Stuart KE, Anand AJ, Jenkins RL. Hepatocellular carcinoma in the United States. Prognostic features, treatment outcome, and survival. Cancer 1996; 77:2217–2222. [DOI] [PubMed] [Google Scholar]
- 9.Chan SL, Chong CC, Chan AW, Poon DMC, Chok KSH. Management of hepatocellular carcinoma with portal vein tumor thrombosis: review and update at 2016. World J Gastroenterol 2016; 22:7289–7300. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fuster J, Garcia-Valdecasas JC, Grande L, Tabet J, Bruix J, Anglada T, et al. Hepatocellular carcinoma and cirrhosis. Results of surgical treatment in a European series. Ann Surg 1996; 223:297–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Liu PH, Huo TI, Miksad RA. Hepatocellular carcinoma with portal vein tumor involvement: best management strategies. Semin Liver Dis 2018; 38:242–251. [DOI] [PubMed] [Google Scholar]
- 12.European Association for the Study of the Liver. Electronic address eee, European Association for the Study of the L. EASL Clinical Practice Guidelines: management of hepatocellular carcinoma. J Hepatol 2018; 69:182–236. [DOI] [PubMed] [Google Scholar]
- 13.Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, et al. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology 2018; 67:358–380. [DOI] [PubMed] [Google Scholar]
- 14.Basili S, Pastori D, Raparelli V, Violi F. Anticoagulant therapy in patients with liver cirrhosis and portal vein thrombosis: insights for the clinician. Therap Adv Gastroenterol 2018; 11:1756284818793561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Semin Liver Dis 1999; 19:329–338. [DOI] [PubMed] [Google Scholar]
- 16.Ikai I, Kudo M, Arii S, Omata M, Kojiro M, Sakamoto M, et al. Report of the 18th follow-up survey of primary liver cancer in Japan. Hepatol Res 2010; 40:1043–1059. [DOI] [PubMed] [Google Scholar]
- 17.Sherman CB, Behr S, Dodge JL, Roberts JP, Yao FY, Mehta N. Distinguishing tumor from bland portal vein thrombus in liver transplant candidates with hepatocellular carcinoma: the A-VENA criteria. Liver Transpl 2019; 25:207–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mehta HB, Mehta V, Girman CJ, Adhikari D, Johnson ML. Regression coefficient-based scoring system should be used to assign weights to the risk index. J Clin Epidemiol 2016; 79:22–28. [DOI] [PubMed] [Google Scholar]
- 19.Chen JS, Wang Q, Chen XL, Huang X-H, Liang L-J, Lei J, et al. Clinicopathologic characteristics and surgical outcomes of hepatocellular carcinoma with portal vein tumor thrombosis. J Surg Res 2012; 175:243–250. [DOI] [PubMed] [Google Scholar]
- 20.Lee YH, Hsu CY, Huang YH, Hsia C-Y, Chiou Y-Y, Su C-W, et al. Vascular invasion in hepatocellular carcinoma: prevalence, determinants and prognostic impact. J Clin Gastroenterol 2014; 48:734–741. [DOI] [PubMed] [Google Scholar]
- 21.Guarino M, Cucchetti A, Pontillo G, Farinati F, Benevento F, Rapaccini GL, et al.; ITA.LI.CA Group. Pattern of macrovascular invasion in hepatocellular carcinoma. Eur J Clin Invest 2021; 51:e13542. [DOI] [PubMed] [Google Scholar]
- 22.Pawlik TM, Delman KA, Vauthey JN, Nagorney DM, Ng IO, Ikai I, et al. Tumor size predicts vascular invasion and histologic grade: implications for selection of surgical treatment for hepatocellular carcinoma. Liver Transpl 2005; 11:1086–1092. [DOI] [PubMed] [Google Scholar]
- 23.Sakata J, Shirai Y, Wakai T, Kaneko K, Nagahashi M, Hatakeyama K. Preoperative predictors of vascular invasion in hepatocellular carcinoma. Eur J Surg Oncol 2008; 34:900–905. [DOI] [PubMed] [Google Scholar]
- 24.Reynolds AR, Furlan A, Fetzer DT, Sasatomi E, Borhani AA, Heller MT, et al. Infiltrative hepatocellular carcinoma: what radiologists need to know. Radiographics 2015; 35:371–386. [DOI] [PubMed] [Google Scholar]
- 25.Kanematsu M, Semelka RC, Leonardou P, Mastropasqua M, Lee JKT. Hepatocellular carcinoma of diffuse type: MR imaging findings and clinical manifestations. J Magn Reson Imaging 2003; 18:189–195. [DOI] [PubMed] [Google Scholar]
- 26.Kneuertz PJ, Demirjian A, Firoozmand A, Corona-Villalobos C, Bhagat N, Herman J, et al. Diffuse infiltrative hepatocellular carcinoma: assessment of presentation, treatment, and outcomes. Ann Surg Oncol 2012; 19:2897–2907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Trevisani F, Caraceni P, Bernardi M, D'Intino PE, Arienti V, Amorati P, et al. Gross pathologic types of hepatocellular carcinoma in Italian patients. Relationship with demographic, environmental, and clinical factors. Cancer 1993; 72:1557–1563. [DOI] [PubMed] [Google Scholar]
- 28.Jang ES, Yoon JH, Chung JW, Cho EJ, Yu SJ, Lee J-H, et al. Survival of infiltrative hepatocellular carcinoma patients with preserved hepatic function after treatment with transarterial chemoembolization. J Cancer Res Clin Oncol 2013; 139:635–643. [DOI] [PubMed] [Google Scholar]
- 29.Nadinskaia MY, Kodzoeva KB, Ulyanova KA, Volkova AS, Rogacheva SI, Dekhanov AS, et al. Risk factors associated with portal vein thrombosis in liver cirrhosis: a case-control study. Ter Arkh 2019; 91:73–81. [DOI] [PubMed] [Google Scholar]
- 30.Pinato DJ, Sharma R, Allara E, Yen C, Arizumi T, Kubota K, et al. The ALBI grade provides objective hepatic reserve estimation across each BCLC stage of hepatocellular carcinoma. J Hepatol 2017; 66:338–346. [DOI] [PubMed] [Google Scholar]
- 31.Basili S, Carnevale R, Nocella C, Bartimoccia S, Raparelli V, Talerico G, et al.; PRO-LIVER Collaborators. Serum albumin is inversely associated with portal vein thrombosis in cirrhosis. Hepatol Commun 2019; 3:504–512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Bagirsakci E, Sahin E, Atabey N, Erdal E, Guerra V, Carr BI. Role of albumin in growth inhibition in hepatocellular carcinoma. Oncology (Huntingt) 2017; 93:136–142. [DOI] [PubMed] [Google Scholar]
- 33.Carr BI, Guerra V, Giannini EG, Farinati F, Ciccarese F, Ludovico Rapaccini G, et al.; Italian Liver Cancer (ITA.LI.CA) Group. Association of abnormal plasma bilirubin with aggressive hepatocellular carcinoma phenotype. Semin Oncol 2014; 41:252–258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Zanetto A, Campello E, Spiezia L, Burra P, Simioni P, Russo FP. Cancer-associated thrombosis in cirrhotic patients with hepatocellular carcinoma. Cancers (Basel) 2018; 10:450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Cubillas R, Rockey DC. Portal hypertensive gastropathy: a review. Liver Int 2010; 30:1094–1102. [DOI] [PubMed] [Google Scholar]
- 36.Akkiz H, Carr BI, Kuran S, Karaoğullarindan U, Üsküdar O, Tokmak S, et al. Macroscopic portal vein thrombosis in HCC patients. Can J Gastroenterol Hepatol 2018; 2018:3120185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Serag WM, Mohammed BSE, Mohamed MM, Elsayed BE. Predicting the risk of portal vein thrombosis in patients with liver cirrhosis and hepatocellular carcinoma. Heliyon 2020; 6:e04677. [DOI] [PMC free article] [PubMed] [Google Scholar]