The present study explored the correlation of ecotropic viral integration site 5 (EVI5) expression with clinicopathological features and prognosis in hepatocellular carcinoma (HCC). A total of 205 HCC patients were included retrospectively. Quantitative real-time polymerase chain reaction (RT-qPCR) and western blotting were performed to detect the profile of EVI5 expression in HCC cell lines and fresh tissues. Archived paraffin-embedded specimens were investigated for EVI5 expression by immunohistochemistry (IHC). Both the mRNA and protein levels of EVI5 were obviously upregulated in HCC cell lines and tumor tissues. EVI5 protein level was closely associated with the clinicopathological characteristics, including liver function (P=0.013), venous invasion (P=0.015) and TNM stage (P=0.014). Furthermore, univariate analysis showed that the patients with high EVI5 expression indicated shorter overall survival (OS, P<0.001) and recurrence-free survival (RFS, P=0.001) than those with low EVI5 expression. Importantly, high EVI5 expression also exerts predictive power for higher postoperative recurrence rate by stratified analysis. Multivariate Cox regression analysis demonstrated that OS was correlated with both tumor number (P=0.046) and EVI5 expression (P<0.001) and that RFS was correlated with serum AFP (P=0.023), tumor number (P=0.036) and EVI5 expression (P<0.001). Taken together, EVI5 is an useful independent prognostic marker of survival and recurrence in hepatocellular carcinoma.
Hepatocellular carcinoma (HCC) is one of the deadliest malignancies worldwide. According to the data from the International Agency for Research on Cancer, more than half of global incidence (782000 estimated new cases) and mortality (746000 estimated deaths) were in China in 2012 (
EVI5 (ecotropic viral integration site 5) is an 810 amino acid protein, which belongs to a small subfamily of the Tre-2/Bub2/Cdc16 (TBC) domain-containing proteins. Earlier studies showed that EVI5 was expressed in various tissues, including in liver, brain, thymus and adrenal tissues (
During the past two decades, EVI5 has been recognized as a potential oncogene and a cell cycle regulator (
Human normal liver cell line LO2 and seven hepatocellular carcinoma cell lines were cultured in this study. H2M and H2P were gifts from Professor Tiebang Kang laboratory (State Key Laboratory of Oncology in South China). HuH7, SMMC7721, BEL7402 and QGY7703 were purchased from the Shanghai Cell Bank Chinese Academy Sciences. HepG2 was purchased from the American Type Culture Collection (ATCC; Manassas, VA, USA). Cultured cells were maintained in Dulbeccos modified Eagles medium ((DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen), and incubated in a humidified atmosphere of 5% CO2 at 37°C.
A total of 205 consecutive patients who underwent radical hepatectomy between January 1, 2002 and December 31, 2004, at the Affiliated Tumor Hospital of Xinjiang Medical University were enrolled in the present study. Hepatectomy was carried out under general anesthesia using a right subcostal incision with a midline extension. Intraoperative ultrasound was routinely used. Radical hepatectomy was performed aiming at a surgical margin of at least 2 cm. The clinicopathological information of the patients is listed in
The detailed protocol of this experiment was described in a previous study (
Western blotting was performed as previously described (
After hematoxylin and eosin (H&E) staining, the degree of tumor differentiation was evaluated in terms of Edmondson-Steiner histological grading system. IHC was performed on hepatocellular carcinoma and paratumor tissues as previously described (
The SPSS statistical software package (version 16.0; SPSS, Inc., Chicago, IL, USA) was used. The Pearsons Chi-square test was utilized to analyze the relationship between the EVI5 expression and the clinicopathological parameters. Survival curves were generated by the Kaplan-Meier method and compared by the log-rank test. Univariate analysis identified variables associated with survival, and than significant variables associated with survival in univariate analysis were entered into a multivariate Cox proportional hazard regression model. Finally, independent prognostic factors were determined. A two-sided P<0.05 was considered to be statistically significant.
To detect EVI5 expression profile, initially, multiple HCC cell lines were used to measure mRNA and protein expression levels of EVI5 by RT-qPCR and western blotting assay, respectively. Both the mRNA and protein levels of EVI5 were elevated in the 7 HCC cell lines, including H2M, H2P, HuH7, SMMC7721, HepG2, BEL7402 and QGY7703, than those in LO2 (
To further evaluate the relationship between the EVI5 protein expression and the clinicopathological characteristics of HCC, 205 paraffin-embedded HCC samples were detected by IHC. According to the cut-off value of 6, all of patients could be divided into high-expression group (45.4%, 93/205) and low-expression group (54.6%, 112/205). Chi-square test indicated that the EVI5 protein level was closely associated with some of the clinicopathological variables of HCC, including liver function (P=0.013), venous invasion (P=0.015) and TNM stage (P=0.014). However, EVI5 protein expression is unrelated to age, sex, hepatitis B surface antigen (HBs Ag), serum AFP, tumor size, tumor number and histological grade (
Among this retrospective cohort of 205 HCC cases, including 148 cases of stage I (72.2%), 32 cases of stage II (15.6%), 25 cases of stage III (12.2%), IHC was conducted to explore the expression pattern of EVI5. The immunostaining of EVI5 was stronger in HCC than in matched adjacent non-tumor tissue (
The EVI5 protein level was negatively correlated with overall survival (OS, P<0.001;
Univariate analysis revealed there were 7 risk factors associated with OS and RFS, respectively (
Furthermore, based on the analysis of the 3 subgroups, the influence of EVI5 expression on postoperative recurrence was further revealed by stratified analysis. As shown in
Hepatocellular carcinoma (HCC) which is one of the most common malignant tumors worldwide, it is a serious threat to human health and survival. In recent years, the studies focusing on tumor molecular biomarkers seemed to provide new insight into diagnosis, treatment and prognostic assessment for precise medicine. Therefore, it is significant to identify sensitive and specific biomarkers for HCC. Here, we identified and validated EVI5 as a novel prognostic biomarker for HCC.
Previous studies show that EVI5 was located in the nucleus and cytoplasm (
In the initial studies, EVI5 participated in regulation of cell cycle in different compartments of cell, functioned by stabilizing the Emi1 protein and promoting cyclin-A accumulation, EVI5 misregulation led to abnormity of cell cycle and cell division (
Our results support the above previous findings. We showed that EVI5 protein overexpressed in HCC cell lines and tumor tissues compared with normal liver cell line (LO2) and corresponding paratumor tissues. Furthermore, we validated the differential expression by RT-qPCR and IHC. In particular, a pair of cell models, from heptic portal vein tumor thrombus (H2M) and primary tumor (H2P) (
Chi-square analysis demonstrated that EVI5 expression level was correlated with serum AFP, tumor number and TNM stage. Among TNM stages, the lower percentage of high EVI5 in advanced stage II/III (19.4%) than stage I (80.6%), may be due to fewer subjects with stage II/III (57/205, 27.8%). Interestingly, we still found the high EVI5 expression tended to occur in sufferers with more advanced stage (stage II, 25.0% vs. stage III, 40.0%), suggesting that EVI5 might be involved in tumor progression. Kaplan-Meier analysis showed that the OS and RFS curves were significantly separated between two cohorts with different EVI5 expression. The high expression of EVI5 was associated with shorter OS and RFS, low expression group showed contrary results. Tumor relapse is a key cause for therapeutic failure. Therefore, we investigated the prognostic factors that attributed to recurrence by stratified analysis. Cox regression analysis showed that serum AFP, tumor number and EVI5 expression were independent prognostic factors. TNM stage was failed to be selected, which may be due to fewer advanced patients (TNM stage II/III, 27.8%). Subgroup analysis demonstrated that among these patients with lower serum AFP, solitary or TNM stage I, high EVI5 expression was consistently associated with shorter RFS, which means that these patients could be distinguished for early intervention to prevent tumor recurrence. To those patients with EVI5 high expression, accompanying with high serum AFP, multiple or advanced stages, comprehensive treatment strategy should be considered as soon as possible.
In conclusion, our findings showed that the EVI5 was upregulated in HCC. The high expression of EVI5 indicated a worse prognosis in postoperative HCC subjects. EVI5 is a novel predictive biomarker that may help guide the clinical therapy for HCC patients.
The present study was supported by grants from the National Natural Science Foundation in China (grant no. 81460360 to B.W. and grant no. 81560403 to J.T.); the Natural Science Foundation of Xinjiang Uygur Autonomous Region, (grant no. 2015211C126 to B.W. and grant no. 2016D01C374 to J.T.). The authors thank professor Tiebang Kang of Sun Yat-sen University Cancer Center for providing experimental platform and expert opinions.
Expression of EVI5 in HCC cell lines and tissue samples. (A) The mRNA levels of EVI5 in immobilized liver cell line (LO2) and HCC cell lines were determined using RT-qPCR. (B) Representative pattern of EVI5 protein expressed in cell lines are shown. (C and D) Eight pairs of HCC T (tumor) tissues and their NT (non-tumor) counterparts were examined by RT-qPCR and western blotting. Relative mRNA (C) and protein (D) levels of EVI5 were upregulated in HCC clinical samples.
Expression of EVI5 in HCC and adjacent non-tumor tissues by IHC. EVI5 staining was observed mainly localized within the cytoplasm. (A) Positive EVI5 staining in tumor tissue. (B) Negative EVI5 staining in adjacent non-tumor liver tissue (ANLT). (C-F). The intensities of EVI5 staining in HCC were scored as four grades (0, 1, 2 and 3), respectively as indicated. Magnification, ×200 and ×400.
Correlation of EVI5 expression in 205 HCC patients with overall survival (A) and recurrence-free survival (B). Kaplan-Meier survival analysis (log-rank test) demonstrated that high EVI5 expression indicated unfavorable OS and RFS.
Association of EVI5 expression with recurrence-free survival in HCC subgroups. Survival analysis was performed according to the factors that were contributed to tumor recurrence, (A and B) serum AFP, (C and D) tumor number, (E and F) TNM stage. Kaplan-Meier stratified survival analysis was used (log-rank test).
Clinicopathological correlation of EVI5 expression in HCC patients.
EVI5 expression (n, %) | ||||
---|---|---|---|---|
Variables | N | Low | High | P-value |
All cases | 205 | 112 | 93 | |
Age (years) | 0.414 | |||
≤50 | 110 (53.7) | 63 (56.3) | 47 (50.5) | |
>50 | 95 (46.3) | 49 (43.7) | 46 (49.5) | |
Sex | 0.698 | |||
Female | 24 (11.7) | 14 (12.5) | 10 (10.8) | |
Male | 181 (88.3) | 98 (87.5) | 83 (89.2) | |
HBsAg | 0.737 | |||
Negative | 26 (12.7) | 15 (13.4) | 11 (11.8) | |
Positive | 179 (87.3) | 97 (86.6) | 82 (88.2) | |
Liver function | ||||
Child-Pugh A | 176 (85.9) | 90 (80.4) | 86 (92.5) | |
Child-Pugh B | 29 (14.1) | 22 (19.6) | 7 (7.5) | |
Serum AFP (ng/ml) | 0.942 | |||
≤25 | 70 (34.1) | 38 (33.9) | 32 (34.4) | |
>25 | 135 (65.9) | 74 (66.1) | 61 (65.6) | |
Tumor size (cm) | 0.632 | |||
≤5 | 160 (78.0) | 86 (76.8) | 74 (79.6) | |
>5 | 45 (22.0) | 26 (23.2) | 19 (20.4) | |
Tumor number | 0.384 | |||
Solitary | 188 (91.7) | 101 (90.2) | 87 (93.5) | |
Multiple | 17 (8.3) | 11 (9.8) | 6 (6.5) | |
Venous invasion | ||||
Absent | 180 (87.8) | 104 (92.9) | 76 (81.7) | |
Present | 25 (12.2) | 8 (7.1) | 17 (18.3) | |
Histological grade |
0.592 | |||
I/II | 110 (53.7) | 62 (55.4) | 48 (51.6) | |
III/IV | 95 (46.3) | 50 (44.6) | 45 (48.4) | |
TNM stage |
||||
Stage I | 148 (72.2) | 73 (65.2) | 75 (80.6) | |
Stage II/III | 57 (27.8) | 39 (34.8) | 18 (19.4) |
Edmondson-Steiner grade
according to the 7th edition AJCC/UICC TNM stage system. HCC, hepatocellular carcinoma; HBsAg, hepatitis B surface antigen; AFP, α-fetal protein; TNM, tumor-node-metastasis; EVI5, ecotropic viral integration site 5. Bold, P<0.05.
Univariate prognostic analysis of overall survival and recurrence-free survival rates for 205 surgical HCC patients.
Overall survival | Recurrence-free survival | |||
---|---|---|---|---|
Variables | HR (95% CI) | P-value | HR (95% CI) | P-value |
Age, years (≤50 vs. >50) | 1.359 (0.943–1.960) | 0.100 | 1.267 (0.908–1.767) | 0.163 |
Sex (female vs. male) | 1.122 (0.630–1.998) | 0.696 | 1.368 (0.787–2.378) | 0.267 |
HBsAg (negative vs. positive) | 1.095 (0.626–1.916) | 0.750 | 1.183 (0.703–1.994) | 0.527 |
Serum AFP, ng/ml (≤25 vs. >25) | 1.548 (1.032–2.320) | 1.771 (1.221–2.568) | ||
Liver function (Child-Pugh A vs. B) | 1.008 (0.699–1.454) | 0.965 | 0.891 (0.638–1.242) | 0.495 |
Histological grade |
1.511 (1.048–2.179) | 1.669 (1.195–2.331) | ||
Tumor number (solitary vs. multiple) | 2.190 (1.201–3.994) | 2.143 (1.223–3.755) | ||
Tumor size, cm (≤5 vs. >5) | 1.655 (1.090–2.513) | 1.740 (1.189–2.546) | ||
Venous invasion (absent vs. present) | 2.366 (1.467–3.814) | 2.341 (1.474–3.718) | ||
TNM stage |
1.550 (1.205–1.992) | 1.505 (1.040–2.170) | ||
EVI5 expression (low vs. high) | 1.903 (1.316–2.751) | 1.706 (1.221–2.383) |
Edmondson-Steiner grade.
According to the 7th edition AJCC/UICC TNM stage system. HCC, hepatocellular carcinoma; HBsAg, hepatitis B surface antigen; CI, confidence interval; HR, hazard ratio; AFP, α-fetal protein; TNM, tumor-node-metastasis; EVI5, ecotropic viral integration site 5. Bold, P<0.05.
Cox multivariate regression analysis on overall survival and recurrence-free survival in 205 HCC patients after hepatectomy.
Overall survival | Recurrence-free survival | |||
---|---|---|---|---|
Variables | HR (95% CI) | P-value | HR (95% CI) | P-value |
Serum AFP ng/ml (≤25 vs. >25) | 1.378 (0.892–2.130) | 0.149 | 1.577 (1.066–2.333) | |
Histological grade |
1.356 (0.917–2.005) | 0.127 | 1.570 (0.826–2.984) | 0.169 |
Tumor number (solitary vs. multiple) | 1.992 (1.011–3.925) | 1.467 (1.026–2.099) | ||
Tumor size, cm (≤5 vs. >5) | 1.324 (0.764–2.294) | 0.318 | 1.512 (0.921–2.483) | 0.102 |
Venous invasion (absent vs. present) | 1.255 (0.668–2.360) | 0.480 | 1.191 (0.671–2.111) | 0.551 |
TNM stage |
1.155 (0.673–1.983) | 0.600 | 1.236 (0.754–2.027) | 0.401 |
EVI5 expression (low vs. high) | 2.141 (1.437–3.190) | 1.903 (1.334–2.715) |
Edmondson-Steiner grade.
According to the 7th edition AJCC/UICC TNM stage system. HCC, hepatocellular carcinoma; CI, confidence interval; HR, hazard ratio; AFP, α-fetal protein; TNM, tumor-node-metastasis; EVI5, ecotropic viral integration site 5. Bold, P<0.05.