Vessels that encapsulate tumor clusters (VETC) pattern predicts the efficacy of adjuvant TACE in hepatocellular carcinoma

Postoperative adjuvant trans-catheter arterial chemoembolization (TACE) is regarded as a common strategy for hepatocellular carcinoma (HCC) patients at a high risk of recurrence. However, there are currently no clinically available biomarkers to predict adjuvant TACE response. Vessels that encapsulate tumor clusters (VETC) can be used as an independent predictor of HCC prognosis. In this study, we aimed to explore whether the VETC pattern could predict adjuvant TACE benefit. Vascular pattern and HIF-1α expression were detected in immunohistochemistry. The survival benefit of adjuvant TACE therapy for patients with or without VETC pattern (VETC+ /VETC−) was evaluated. The adjuvant TACE therapy obviously improved the TTR and OS in VETC+ patients, while adjuvant TACE therapy could not benefit from VETC− patients. Univariate and multivariate analysis revealed that adjuvant TACE therapy significantly improved the TTR and OS in VETC+ patients, but not in VETC- patients. In addition, the VETC+ , but not VETC− , patients could benefit from adjuvant TACE therapy in patients with high-risk factors of vascular invasion, larger tumor or multiple tumor. The mechanistic investigations revealed that the favorable efficacy of adjuvant TACE on VETC+ patients, but not VETC− ones, may be not due to the activation of HIF-1α pathway. The VETC pattern may represent a novel and reliable factor for selecting HCC patients who may benefit from adjuvant TACE therapy, and the combination of VETC pattern and tumor characteristics may help stratify patients’ outcomes and responses to adjuvant TACE therapy.


Introduction
Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related death, and its incidence is increasing year by year worldwide (Siegel et al. 2021;Kulik and El-Serag 2019). Although there are many effective and feasible treatments for HCC, such as surgical resection, liver transplantation, chemotherapy, ablation, interventional therapy, targeted therapy and immunotherapy, etc. (Hartke et al. 2017), the prognosis of HCC patients is still very poor. Studies have reported that the 5-year survival rate is only 50% (17-69%) even for HCC patients undergoing surgical resection (Takayama 2011). The main reason is that HCC is prone to hematogenous metastasis, especially early intrahepatic metastasis (Forner et al. 2018). It is well known that trans-catheter arterial chemoembolization (TACE) is the most commonly used method to prevent postoperative recurrence of HCC and improve survival rate. It also reported that only selected patients would benefit from adjuvant TACE . Previous studies have reported that adjuvant TACE therapy is recommended for 1-2 months after surgery for patients with large tumors or venous invasion Imamura et al. 2003;Poon et al. 2000). However, TACE damages liver function and may negatively affect patients' survival (Sieghart et al. 2013). In addition, the prognosis of HCC patients with the same clinicopathological characteristics may be significantly different, which may be related to the heterogeneity of biological behavior of tumor cells (Aravalli et al. 2008;Dragani 2010). Therefore, it is essential to identify new predictors that can predict the efficacy of adjuvant TACE therapy and to help select subgroups of patients most likely to benefit from adjuvant TACE therapy.
Several approaches have been implemented to identify potential factors associated with the clinical benefit of adjuvant TACE. Ren et al. (Ren et al. 2004) evaluated the effect of postoperative TACE on the prognosis of HCC patients and concluded that in patients with high-risk factors of residual tumor (vascular invasion, tumor diameter > 5 cm or multiple nodules), postoperative adjuvant TACE significantly prolonged the patients' survival. However, many scholars draw an opposite conclusion. It is suggested that TACE can damage remnant liver function and immunological function resulting in poor survival Yang et al. 2018). We and other investigators revealed that patients with high expression of cezanne (Wang et al. 2017), Cripto-1 (Li et al. 2019), Cbx4 (Jiao et al. 2015), RMP  or low COCH expression (Wang et al. 2021) had favorable responses to adjuvant TACE therapy. Whereas, there is no clinically applied biomarker for predicting adjuvant TACE response in HCC.
Vessels that encapsulate tumor cluster (VETC) is a special vascular structure in HCC angiogenesis (Fang et al. 2015;Sugino et al. 2004Sugino et al. , 2008, which can provide rich blood supply for HCC and play an important role in tumor growth, invasion and metastasis (Fang et al. 2015;Ding et al. 2011). VETC provides an important mechanism for HCC metastasis by which entire tumor clusters may be released into the bloodstream in a manner independent of epithelial-mesenchymal transformation (EMT) (Fang et al. 2015). In addition, the presence of VETC can significantly increase the incidence of portal vein HCC thrombosis and para-carcinoma tissue micro-embolus, and has a higher risk of recurrence, which plays a key role in HCC metastasis (Fang et al. 2015). These results reveal that the VETC pattern represents a more aggressive HCC subtype. However, it is not clear whether different vascular patterns of HCC can benefit from anti-HCC treatment differently.
In this study, we aimed to investigate whether vascular patterns can predict the efficacy of adjuvant TACE therapy. Our results showed that adjuvant TACE therapy was effective in prolonging patients' survival in the VETC model, but not in patients with capillaries, suggesting that VETC pattern may serve as a predictor of optimal choice for HCC patients who may benefit from adjuvant TACE therapy after hepatectomy.

Patients and specimens
The study was approved by the Institutional Review Board and Human Ethics Committee of Affiliated Cancer Hospital & Institute of Guangzhou Medical University.
A total of 231 HCC tissues were consecutively collected from Department of Hepatobiliary Oncology, Sun Yat-sen University (Guangzhou, China) between July 2007 and January 2009. Only cases stages R0 after surgical resection were included in the study. Among these patients, one hundred and five patients (45.5%) also accepted adjuvant TACE 1-2 months after surgery. Patients included in this study were as follows: (1) all tissues were pathologically diagnosed; (2) no chemotherapy or radiotherapy before operation; (3) no distant metastasis or other malignant disease (enhanced CT and/or MRI). Tumor staging was performed according to the American Joint Committee on Cancer Staging, Edition 7th TNM classification and the Barcelona Clinical Liver Cancer (BCLC) staging system.

TACE treatment
All patients accepted adjuvant TACE 1-2 months after operation. Hepatic arterial angiography was performed first, and prophylactic chemoembolization was performed for the remaining liver without tumor staining. Prophylactic adjuvant TACE regimen consisted of lobaplatin 50 mg, epirubicin (EPI) 50 mg, and lipiodol 5 ml. One month later, enhanced CT or MRI scans were performed to complete the program.

Immunohistochemical staining
A total of 231 HCC tissues were used for IHC analysis. After baking for 2 h at 60 °C, the tissues were dewaxed in xylene and rehydrated using a series of graded alcohols. Tissues sections were then treated with 3% hydrogen peroxide methanol for 10 min to deplete endogenous peroxidase activity. The antigen was recovered by microwave in 0.01 M sodium citrate buffer (pH 6.0) for 30 min and then pre-incubated in 10% normal goat serum for 30 min to prevent non-specific staining. Sections were incubated overnight with rabbit CD31 polyclonal antibody (working dilution 1:100, Abcam, #ab28364, UK) and mouse HIF-1α monoclonal antibody (working dilution 1:200, Abcam, #ab16066, UK) at 4 °C. Slices were treated with a non-biotin horseradish-peroxidase detection system according to the manufacturer's instructions (DAKO, Glostrup, Denmark). Staining was assessed by two experienced pathologists without knowledge of the patients' identity and clinical status. In discrepant cases, the pathologist reviews the cases and reaches consensus.
The intensity and extent of immunostaining were considered in the analysis of data. The staining intensity score was 0-3, and the staining extent score was 0-100%. The final quantity of each stain was obtained by multiplying the two fractions. The expression of HIF-1α was higher than 1.6 and lower than 1.6, respectively. The VETC pattern was detected by the presence of sinusoid-like vessels, which formed cobweb-like networks and encapsulated individual tumor clusters. Cases with VETC pattern visible in all or part of the HCC area were classified as VETC+ , and those without VETC pattern were classified as VETC− (Fang et al. 2019).

Follow-up
The last follow-up was in May 2018. All patients were on follow-up visits every 3 months for 2 years after TACE treatment and every 3-6 months afterward, with routine monitoring of serum α-fetoprotein (AFP) levels, liver function tests, abdominal ultrasonography, and computed tomography or magnetic resonance imaging. The main causes of death were HCC recurrence or complicated cirrhosis of the liver.

Statistical analysis
The SPSS software package (version 26.0; Chicago, IL, USA) was used for statistical analysis. The experimental data between two groups were tested by a two-tailed Student's t test. The Kaplan-Meier analysis with log-rank test was used to calculate the survival time of HCC patients. Cox proportional risk regression model was used to analyze independent prognostic factors of HCC patients. All P values were bilateral and P value < 0.05 was considered statistically significant.

Patients' characteristics
There were 262 HCC patients who received operation between 2007 and 2009 in Sun Yat-Sen University Cancer Center. After excluded incomplete data and incomplete follow-up data, a total of 231 HCC patients were included in our study, which consisted of 105 HCC patients receiving adjuvant TACE after hepatectomy (designated as the TACE group) and 126 HCC patients without adjuvant TACE therapy (the control group) (Fig. 1). The patients' baseline demographic and clinicopathological features were analyzed in this study, and showed that the TACE group patients had larger tumor size (P = 0.013), multiple tumor (P < 0.001), vascular invasion (P = 0.001), advanced TNM stage (P < 0.001) and advanced BCLC stage (P < 0.001) than those of the control group patients (Table 1), which suggested that patients in TACE group were more advanced than those in control group. According to immunohistochemical staining of CD31 in human HCC tissues, a patient with a VETC pattern in tumor tissue was defined as a VETC+ patient, while a patient Fig. 1 Flowchart showing the selection process for HCC patients without a VETC pattern was defined as a VETC− patient ( Fig. 2A). The proportions of VETC+ patients were 42.9% in control group and 45.7% in TACE group. In addition, VETC pattern was more frequently observed in advanced TNM stage and advanced BCLC stage (Table S1), and VETC+ patients had poorer prognosis than VETC− patients (Fig. 2B), indicating that VETC pattern may represent a more malignant HCC subtype, which had similar results with previous reports (Fang et al. 2015(Fang et al. , 2019Zhou et al. 2016).  (Li et al. 2019). The survival benefit of adjuvant TACE therapy in VETC+ and VETC-patients was analyzed in our study. In the control group, VETC+ patients showed significantly shorter TTR (median TTR: 10.0 vs. 43.5 months, P < 0.001) and poorer OS (median OS: 17.7 vs. 56.5 months, P < 0.001) than VETC-patients (Fig. 2C). However, the TTR and OS were not significantly different between VETC + and VETC-patients in the TACE group (both P > 0.05, Fig. 2D). As VETC pattern represents a more malignant HCC subtype and could be an independent prognostic factor for HCC patients, we further explored whether the response of patients to adjuvant TACE therapy was different according to the VETC pattern. Interestingly, adjuvant TACE had no influence on tumor relapse and OS time in VETC-patients (Fig. 3A). On the contrary, the TACE therapy obviously improved the TTR (median TTR: 10.0 vs. 36.0 months, P = 0.008) and OS (median OS: 17.7 vs. 53.2 months, P = 0.005) in VETC + patients (Fig. 3B). Furthermore, the 5-year TTR and OS rates of the adjuvant TACE group were 44.8% and 54.0%, which were significantly higher than that of the control group (23.9% and 35.0%) (Fig. 3B). Cox proportional hazards regression was used to analyze the value of adjuvant TACE therapy in VETC− patients and VETC+ patients, respectively. Our results showed that adjuvant TACE could not be an independent prognostic factor for TTR or OS in VETC− patients (Table 2). Interestingly, the data indicated that adjuvant TACE therapy (P = 0.008), smaller tumor (P = 0.013), single tumor (P = 0.002), no

VETC pattern predicts response of adjuvant TACE in clinical subgroups
The HCC patients with high-risk factors of vascular invasion, larger tumor (> 5 cm in diameter) or multiple tumor are recommended to receive TACE therapy 1-2 months after hepatectomy Imamura et al. 2003;Poon et al. 2000). Adjuvant TACE may improve survival outcome and reduce the recurrence rate (Mathurin et al. 2003;Xi et al. 2012), which was consistent with our previous reports (Wang et al. 2017;Li et al. 2019). We performed the discriminative power of VETC pattern on postoperative adjuvant TACE in patients with vascular invasion, large tumor and multiple tumor. In patients with vascular invasion subgroup, the survival curves of the TACE therapy and control groups could not be significantly separated in VETC-patients (Fig. 4A) Fig. 4F). These data indicate that only VETC + , but not VETC− , patients could benefit

The different adjuvant TACE benefit from VETC− and VETC+ patients is not owing to the activation of HIF-1α signaling
We further investigated the mechanism underlying the different survival benefit of adjuvant TACE therapy in VETC− and VETC+ HCC. The activated HIF-1α pathway is involved in HCC angiogenesis and metastasis (Chen and Lou 2017) and has been reported to affect the therapeutic effect of adjuvant TACE (Wei et al. 2021). We detected the HIF-1α expression in VETC-and VETC+ HCC samples, and found that the levels of HIF-1α expression had no difference in VETC− and VETC+ HCC samples (Fig. 5A). Analysis of the correlation between VETC pattern and adjuvant TACE therapy suggested that VETC+ patients had significant improvement of TTR after receiving adjuvant TACE therapy in both low HIF-1α (VETC− vs. VETC+ : HR = 0.707 vs. 0.266, P interaction = 0.030, Fig. 5B) and high HIF-1α (VETC − vs. VETC + : HR = 2.189 vs. 0.452, P interaction = 0.021, Fig. 5C) groups. Therefore, the favorable efficacy of adjuvant TACE on VETC + patients, but not VETC-ones, may be not due to the activation of HIF-1α pathway.

Discussion
Postoperative recurrence is a major obstacle to long-term survival after HCC resection, and intrahepatic metastasis is considered to be closely related to postoperative recurrence (Ouchi et al. 2000). However, small intrahepatic metastasis is difficult to detect before or during surgery, and their presence contributes to intrahepatic recurrence. Adjuvant TACE have been used to prolong survival and prevent the recurrence of HCC postoperatively (Mathurin et al. 2003;Xi et al. 2012;Fukuda et al. 2002). However, adjuvant TACE is not a standard treatment, and its associated survival benefit is still controversial. Some studies reported that TACE damaged residual liver, worsened liver function, and was even associated with more frequent extrahepatic recurrences and adverse outcome (Lai et al. 1998;Zhong and Li 2010). Therefore, it is necessary to select patients who may benefit from TACE after operation. Several research groups have attempted to identify biomarkers to predict patients susceptible to adjuvant TACE therapy. However, there is currently no optimal biomarker for clinical prediction of the efficacy of adjuvant TACE. To date, it is not clear whether vascular pattern could predict the benefit of adjuvant TACE. In the present study, we evaluated the clinical effect of VETC pattern in predicting the response of adjuvant TACE in patients after surgery. To our knowledge, this is the first attempt to investigate the value of vascular patterns in predicting the efficacy of adjuvant TACE. In patients without received adjuvant TACE, VETC+ patients had a significantly worse prognosis than VETC− patients. Interestingly, the prognosis was not significantly different between VETC+ and VETC− patients in patients with adjuvant TACE, which suggests that VETC pattern may benefit from adjuvant TACE therapy. In addition, we also found that VETC+ patients had significantly reduced relapse and improved OS after adjuvant TACE therapy, while adjuvant TACE therapy had no significant effect on the survival in VETC− patients, indicating that the VETC pattern can be used as an indicator to select HCC patients for adjuvant TACE therapy. The use of VETC pattern as a predictive marker has several advantages. First, VETC pattern is a clinically feasible biomarker. Vascular invasion, tumor diameter > 5 cm and multiple nodules are considered as high-risk factors for postoperative recurrence of HCC. However, grouping tumors by diameter and number of tumors is difficult to capture the biological characteristics of tumors, and some patients with microvascular invasion may be missed due to pathological findings. VETC pattern can better reflect the biological characteristics of HCC and is consistent at multiple sites in the same tumor tissue (Fang et al. 2015). Secondly, VETC pattern is a low-cost biomarker. The VETC pattern could be easily distinguished from capillaries in frozen or paraffin-embedded tissue masses by immunohistochemical staining using CD34 or CD31 (Ding et al. 2011). Finally, VETC is a common vascular pattern that presented in 40-50% of HCC patients (Fang et al. 2015;Ding et al. 2011). Therefore, most HCC patients will benefit from the VETC pattern as the selection criteria.
Previous studies showed that adjuvant TACE therapy was recommended for patients with large tumor volume, vascular invasion or multiple nodules at high risks of recurrence (Ren et al. 2004;Xi et al. 2012;Peng et al. 2009;Huang et al. 2013). However, even patients with similar clinicopathologic parameters who received adjuvant TACE after operation may have different outcomes (Lai et al. 1998;Ono et al. 1997Ono et al. , 2001Schwartz et al. 2002). Therefore, we combined VETC pattern with highrisk factors of recurrence to evaluate the efficacy of adjuvant TACE therapy in HCC patients. In patients with vascular invasion, larger tumor size or multiple nodules, VETC− patients did not benefit from adjuvant TACE therapy, while adjuvant TACE therapy could significantly improve the TTR and OS times in VETC+ patients. These The different TTR benefits from adjuvant TACE for VETC− and VETC+ patients are not correlated with the HIF-1α level. The level of HIF-1α had no difference in VETC− and VETC+ tissues (A). The interaction between VETC pattern and adjuvant TACE was stratified by HIF-1α level (B-C) results suggest that the VETC pattern can help to identify different outcomes of patients with high-risk of recurrence who received TACE after hepatectomy, thus providing a theoretical basis for the treatment combined with the VETC pattern and tumor characteristics. According to the VETC pattern and clinicopathological features, postoperative evaluation of the efficacy of adjuvant TACE therapy will be a necessary condition to determine whether adjuvant TACE therapy can be used as first-line treatment after hepatectomy.
Interestingly, we found that the superior efficacy of adjuvant TACE therapy in patients with VETC+ but not VETC− was not due to activation of HIF-1α signaling. HIF-1α overexpression predicts poor prognosis in HCC patients. HIF-1α is mainly involved in promoting tumor migration, invasion and metastasis, angiogenesis and other aspects, involving various signaling pathways (Chu et al. 2022). However, it has been reported that tumor clusters were wrapped in endothelium and released into the bloodstream in VETC+ patients. VETC− mediated metastasis was dependent on vascular pattern, while the metastasis of VETC− HCC mainly depended on the migration and invasion capacity of HCC cells (Fang et al. 2015;Zhou et al. 2016). It is difficult to detect the minimal intrahepatic metastasis before or during operation. Postoperative adjuvant TACE mainly blocks tumor blood supply by embolization of tumor supplying artery, leading to tumor ischemia and hypoxia, thus achieving the purpose of inhibiting tumor growth and recurrence. Therefore, adjuvant TACE therapy may directly block VETC− mediated metastasis by disrupting VETC formation, but only indirectly affect VETC− HCC metastasis by reducing the number of blood vessels.

Conclusion
Our results revealed that the VETC pattern may represent a novel and reliable factor for selecting HCC patients who may benefit from adjuvant TACE therapy, and that the combination of VETC pattern and tumor characteristics may contribute to stratified patients' outcomes and response to adjuvant TACE therapy. However, it also has a few limitations in this study: the study subjects were retrospectively collected, and randomized controlled studies should be carried out in future. This project is a single-center study, which needs to be verified by a multi-center study.