Med Oncol (2015) 32:131 DOI 10.1007/s12032-015-0574-2
ORIGINAL PAPER
Overexpression of AGGF1 is correlated with angiogenesis and poor prognosis of hepatocellular carcinoma Wei Wang1 • Guang-Yao Li2 • Jian-Yu Zhu2 • Da-Bing Huang1 Hang-Cheng Zhou3 • Wen Zhong3 • Chu-Shu Ji1
•
Received: 9 October 2014 / Accepted: 13 March 2015 / Published online: 22 March 2015 Ó Springer Science+Business Media New York 2015
Abstract Angiogenic factor with G-patch and FHA domains 1 (AGGF1) is a factor implicating in vascular differentiation and angiogenesis. Several lines of evidence indicate that aberrant expression of AGGF1 is associated with tumor initiation and progression. The aim of this study was to investigate the expression and prognostic value of AGGF1 in hepatocellular carcinoma (HCC), as well as its relationship with clinicopathological factors and tumor angiogenesis. Immunohistochemistry was performed to evaluate the expression of AGGF1 in HCC and paracarcinomatous tissues collected from 70 patients. Vascular endothelial growth factor (VEGF) and CD34 expression levels were examined in the 70 HCC tissues. Prognostic significance of tumoral AGGF1 expression was determined. Notably, AGGF1 expression was significantly
Wei Wang, Guang-Yao Li, and Jian-Yu Zhu have contributed equally to this work.
Electronic supplementary material The online version of this article (doi:10.1007/s12032-015-0574-2) contains supplementary material, which is available to authorized users. & Chu-Shu Ji [email protected]; [email protected] 1
Department of Medical Oncology, Anhui Provincial Hospital, Anhui Medical University, 17# Lujiang Road, Hefei 230001, People’s Republic of China
2
Department of Hepatic Surgery, Anhui Provincial Hospital, Anhui Medical University, Hefei 230001, People’s Republic of China
3
Department of Pathology, Anhui Provincial Hospital, Anhui Medical University, Hefei 230001, People’s Republic of China
higher in HCC than in surrounding non-tumor tissues (65.7 vs. 25.7 %; P \ 0.001). AGGF1 expression was significantly correlated with tumoral VEGF expression and CD34-positive microvessel density. Moreover, AGGF1 expression was significantly associated with tumor size, tumor capsule, vascular invasion, Edmondson grade, alphafetoprotein level, and TNM stage. Kaplan–Meier survival analysis showed that high AGGF1 was correlated with reduced overall survival (OS) rate (P = 0.001) and disease-free survival (DFS) rate (P \ 0.001). Multivariate analysis identified AGGF1 as an independent poor prognostic factor of OS and DFS in HCC patients (P = 0.043 and P = 0.010, respectively). Taken together, increased AGGF1 expression is associated with tumor angiogenesis and serves as an independent unfavorable prognostic factor for OS and DFS in HCC. AGGF1 may represent a potential therapeutic target for HCC. Keywords AGGF1 Hepatocellular carcinoma Angiogenesis Prognosis Abbreviations AFP Alpha-fetoprotein AGGF1 Angiogenic factor with G-patch and forkheadassociated domain 1 DAB 3,3-diaminobenzidine tetrahydrochloride DFS Disease-free survival HBsAg Hepatitis B s antigen HCC Hepatocellular carcinoma HE Hematoxylin and eosin KTS Klippel–Trenaunay syndrome MVD Microvessel density OS Overall survival PBS Phosphate-buffered saline TNM Tumor-node-metastasis
123
131 Page 2 of 9
UICC VEGF
International Union Against Cancer Vascular endothelial growth factor
Med Oncol (2015) 32:131
Materials and methods Patients and samples
Introduction Hepatocellular carcinoma (HCC), a predominant histological subtype of primary liver cancer, is the sixth most common neoplasm and the third most frequent cause of cancer-related mortality [1]. An estimated 748,300 new liver cancer cases and 695,900 cancer deaths occurred worldwide in 2008 [2]. Its prognosis is gloomy with a 5-year survival of 11 %. Surgery (e.g., hepatectomy and liver transplantation) provides a potentially curative treatment option for HCC patients. However, the prognosis remains poor due to high potential metastasis and recurrence. Besides, the vast majority of HCC patients are diagnosed at an advanced stage at which surgical treatments are unfeasible [3, 4]. Till now, the molecular mechanism(s) underlying HCC pathogenesis is still poorly understood. Identification of novel therapeutic targets and prognostic biomarkers is of significance to improve the survival of HCC patients. Angiogenesis, a fundamental event in numerous pathologic and physiologic conditions, is essential for tumor growth, invasion, and metastasis [5, 6]. HCC, the most common primary liver cancer, is one of the highly vascularized tumors characterized by active neoangiogenesis [7]. Tumor angiogenesis contributes to the aggressiveness and poor prognosis of HCC [7]. Therefore, targeting angiogenesis is a promising therapeutic strategy for controlling HCC development and progression. AGGF1, an angiogenic gene that is located on human chromosome 5q13.3, encodes an angiogenic factor with G-patch and forkhead-associated (FHA) domain [8, 9]. Upregulation of AGGF1 is associated with congenital vascular disease Klippel–Trenaunay syndrome (KTS) [9, 10]. It has been documented that increased AGGF1 expression in zebrafish embryos activates the AKT signaling pathway and leads to increased vein differentiation and angiogenesis [11]. Similar to vascular endothelial growth factor (VEGF), AGGF1 can promote angiogenesis in a chicken chorioallantoic membrane assay and in a mouse hindlimb ischemia model [9, 12]. Despite the pro-angiogenic activity of AGGF1 in several physiologic and pathologic models, its role in tumor angiogenesis is still unclear. In the current study, immunohistochemistry was used to evaluate the expression of AGGF1, VEGF, and CD34-labeled microvessel density (MVD) in HCC tumor tissues and paracarcinomatous tissues. The expression of AGGF1 in HCC and its associations with clinicopathological features, angiogenesis, and prognosis were examined.
123
Paired samples of cancerous liver tissue and paracarcinomatous for immunohistochemistry were obtained from 70 patients with HCC who underwent curative hepatectomy from 2006 to 2011 at the Department of Hepatic Surgery, Anhui Provincial Hospital (Hefei, China). None of the patients had undergone preoperative intervention therapy or chemotherapy. Detailed demographic and clinicopathological data such as age, gender, tumor size, tumor nodule, tumor capsula, vascular invasion, Edmondson grade, cirrhosis, HBsAg status, Child-Pugh grade, alpha-fetoprotein (AFP), and TNM stage were gathered and summarized. The group 70 patients included 59 males and 11 females with a mean age of 53.5 years (range 30–74 years). Preoperative liver function was assessed by using the Child-Pugh scoring system. Tumor differentiation was defined according to the Edmondson grading system [13]. The tumor stage was performed according to the sixth edition of the tumor-nodemetastasis (TNM) classification of the International Union Against Cancer (UICC). Tumor size was measured as the largest dimension of the tumor by gross examination. Follow-up data were obtained from all 70 patients. The mean follow-up period for patients in the present investigation was 28.6 months (range 2–68 months). After hepatectomy, all patients were monitored using CT and/or MRI as preferential, AFP, chest X-ray, and abdominal ultrasonography every 3–6 months after surgery. Overall survival (OS) was calculated from the date of surgery to the date of death. Diseasefree survival (DFS) time was defined as the interval between the date of surgery and the date of recurrence. If recurrence was not diagnosed, the survivors were censored on the date of death or the last date of follow-up. This study was approved by the Research Ethics Committee of Anhui Provincial Hospital, as stipulated by the Helsinki Declaration. All patients provided written informed consent.
Histology and immunohistochemistry Immunohistochemistry was performed as described by using a two-step protocol according to the manufacturer’s instructions [14]. Briefly, tumor tissues were fixed in 10 % formalin, paraffin-embedded, and sectioned. Sections (4 lm) were deparaffinized in xylene (dimethyl benzene), rehydrated in graded alcohol, and washed with phosphatebuffered saline (PBS). After heated in citrate buffer (pH 6.0) for 20 min in a microwave oven, slides were cooled at room temperature. Endogenous peroxidase activity was then blocked with 3 % hydrogen peroxide (H2O2). Then, sections were, respectively, incubated with primary rabbit
Med Oncol (2015) 32:131
anti-AGGF1 antibody (Beijing Biosynthesis Biotechnology Co., Beijing, China), anti-VEGF antibody, or anti-CD34 antibody (Zhongshan Jinqiao Co., Beijing, China) at 4 °C overnight in a moist chamber. The specificity of these antibodies has been validated by the manufacturers. Following incubation, slides were rinsed with PBS. After incubating with biotin-labeled secondary antibody (mouse anti-rabbit IgG; Zhongshan Jinqiao Co., Beijing, China) for 20 min, color was developed by incubating sections with 3,3-diaminobenzidine (DAB; Zhongshan Jinqiao Co., Beijing, China). Finally, slides were counterstained with hematoxylin followed by dehydration and coverslip mounting. Replacement of the primary antibody with PBS served as a negative control. Immunohistochemical analysis The immunohistochemical results were blindly assessed by two independent pathologists. The percentage of immunoreactive cells to the total cells in 3 random 200 9 microscopic fields was determined for each section. Five sections were randomly selected for each tumor sample. AGGF1 and VEGF expression levels were graded as follows: -:\10 % of tumor cells with positive staining; ?: 10–30 % of tumor cells with positive staining; ??:[30 % of tumor cells with positive staining. The cutoff values used in this study are consistent with a previous report by Kudo et al. [15]. MVD assessment CD34 antibody was used to stain vascular endothelial cells, and CD34-positive MVD was then calculated. Five areas with the greatest number of distinctly highlighted microvessels (hot spots) were selected, and the number of vessels was counted at a high-power magnification (4009). An average count was calculated by average measurements. All sections were independently assessed by two pathologists who had no knowledge of the corresponding clinicopathologic characteristics, and discrepancies were resolved by consensus. Statistical analysis Quantitative data were presented as mean ± standard deviation (SD). The association analysis between the expression of AGGF1 and clinicopathological features was performed by Pearson’s Chi-square test or Fisher´s exact test. Spearman correlation coefficients (r) were employed to analyze the correlation between expression of AGGF1 and VEGF. The Kaplan–Meier method and the log-rank test were used for survival analysis. Cox proportional hazards regression model was used for multivariate survival analysis to
Page 3 of 9 131
identify prognostic factors that were significant in the univariate analysis. All statistical analyses were performed using SPSS 19.0 for Windows (SPSS, Inc., Chicago, IL, USA). A P value\0.05 was considered statistically significant.
Results AGGF1 expression in HCC tissues and its correlation with clinicopathological features To further investigate the clinicopathological and prognostic roles of AGGF1 expression in HCC, immunohistochemical analysis was performed in 70 HCC tissue samples and matched corresponding paracarcinomatous. The staining of AGGF1 protein was mainly located at cytoplasm, and various staining intensities could be observed in different tissues shown in Figs. 1, 2a, d. To verify the antibody specificity, Western blot analysis was performed to detect AGGF1 protein in a panel of human HCC cell lines (Huh7, SMMC7721, HepG2, and Hep3B) using the same anti-AGGF1 antibody. The results showed a single 84-kDa band that corresponds to the expected size of AGGF1 protein (Supplementary Fig. S1). Therefore, the antibody was specific for AGGF1. The positive rate of AGGF1 expression was 65.7 % (46/ 70). In adjacent non-cancerous tissues, most cases were negative for AGGF1 expression. Only 18 of 70 corresponding paracarcinomatous (25.7 %) showed AGGF1positive staining (Table 1). The AGGF1-positive frequency was significantly different between HCC tissues and adjacent non-cancerous liver tissues (P \ 0.001). The associations of AGGF1 expression with clinicopathological parameters are listed in Table 2. Increased expression of AGGF1 was significantly correlated with tumor size (P = 0.033), tumor capsule (P = 0.004), vascular invasion (P = 0.001), Edmondson grade (P = 0.001), AFP level (P = 0.016), and TNM stage (P \ 0.001). However, AGGF1 expression was not significantly correlated with age, gender, tumor nodule number, cirrhosis, HBsAg status, or Child-Pugh grade (P [ 0.05). Correlation between AGGF1 and VEGF protein expression in HCC tissues To determine whether AGGF1 is indeed associated with angiogenesis in HCC tissues, the expression of VEGF and CD34 was examined in HCC patients using immunohistochemistry. The staining of VEGF was mainly localized in the cytoplasm. Forty-eight of the 70 HCC samples showed VEGF positivity (68.6 %; Fig. 2b, e). The positive rate of both AGGF1 and VEGF was 57.1 % (40/70), and their negative expression rate was 22.9 % (16/70). The consistency of both AGGF1 and
123
131 Page 4 of 9
Med Oncol (2015) 32:131
Fig. 1 Representative immunohistochemical staining of AGGF1 in hepatocellular carcinoma (HCC) tissues. AGGF1 expression showed diffuse cytoplasmic staining in tumor cells. a ?? AGGF1 staining
(bar = 100 lm); b ?AGGF1 staining (bar = 100 lm); c Negative expression of AGGF1 (bar = 100 lm)
Fig. 2 Immunohistochemical staining of AGGF1, VEGF, and CD34 in one representative HCC case. a Positive for AGGF1 staining (bar = 100 lm). AGGF1 expression showed diffuse cytoplasmic staining in tumor cells (shown by black arrows); b Positive for VEGF staining (bar = 100 lm); c Positive for CD34 staining (bar = 100 lm);
d Partial enlargement of AGGF1 staining with the magnifying power of 4009 (bar = 50 lm); e Partial enlargement of VEGF staining with the magnifying power of 4009 (bar = 50 lm); f Partial enlargement of CD34 staining with the magnifying power of 4009 (bar = 50 lm)
Table 1 Differential expression of AGGF1 in HCC tissues and corresponding paracarcinomatous tissues
Tissues
Case number
-
?*??
Positive rate (%)
P
\0.001
HCC tissues
70
24
46
65.7
Paracarcinomatous tissues
70
52
18
25.7
VEGF expression was 80.0 % (56/70). The discrepancy between AGGF1 and VEGF expression was 20.0 % (14/70). Moreover, Spearman’s rank correlation test proved a significant correlation between tumoral expression of AGGF1 and VEGF (r = 0.548, P \ 0.001; Table 3).
123
AGGF1 expression
Relationship between AGGF1 expression and MVD in HCC tissues To further evaluate the association between AGGF1 and angiogenesis, MVD was calculated by measuring the CD34
Med Oncol (2015) 32:131
Page 5 of 9 131
Table 2 AGGF1 expression in relation to clinicopathologic features Clinicopathologic data
n
AGGF1 expression n (%)
?*?? n (%)
P value
Table 3 The expression correlation between AGGF1 and VEGF (cases) Stain
AGGF1 –
Age (years)
R
P value
0.548
\0.001
?*??
VEGF
\65
60
21 (35)
39 (65)
C65 Gender
10
3 (30)
7 (70)
Male
59
22 (37.3)
37 (62.7)
Female
11
2 (18.2)
9 (81.8)
1.000
– ?*??
16
6
8
40
VEGF vascular endothelial growth factor 0.379
Tumor size (cm) B5
26
13 (50)
13 (50)
[5
44
11 (25)
33 (75)
Single
58
23 (39.7)
35 (60.3)
Multiple
12
1 (8.3)
11 (91.7)
0.033
Association between AGGF1 expression and prognosis
0.081
AGGF1 expression in HCC patients was found to inversely correlate with patient survival, which was revealed by Kaplan–Meier survival analysis and the log-rank test. The median OS time in HCC patients with positive expression of AGGF1 was significantly reduced, compared to those with negative expression of AGGF1 (26.1 vs. 47.0 months, P = 0.001; Fig. 4a). Likewise, patients with AGGF1positive expression had a shorter DFS time compared to those with AGGF1-negative expression (22.6 vs. 46.6 months, P \ 0.001; Fig. 4b). To further analyze the relationship between AGGF1 expression level and survival, Kaplan– Meier method was used to compare the survival of patients with weak and strong AGGF1 expression (Fig. 5). Notably, HCC patients with strong AGGF1 expression had a significantly shorter OS (P = 0.002) and DFS (P \ 0.001) than those with weak AGGF1 expression.
Tumor nodule number
Tumor capsule Absent
28
4 (14.3)
24 (85.7)
Present
42
20 (47.6)
22 (52.4)
No
50
23 (46)
27 (54)
Yes Edmondson grade
20
1 (5)
19 (95)
I–II
34
18 (52.9)
16 (47.1)
III–IV
36
6 (16.7)
30 (83.3)
0.004
Vascular invasion 0.001
0.001
Cirrhosis Absent
6
4 (66.7)
2 (33.3)
Present
64
20 (31.2)
44 (68.8)
Positive
59
21 (35.6)
38 (64.4)
Negative
11
3 (27.3)
8 (72.7)
A
66
22 (33.3)
44 (66.7)
B
4
2 (50)
2 (50)
B20
22
12 (54.5)
10 (45.5)
[20
48
12 (25)
36 (75)
TNM stage I/II
38
22 (57.9)
16 (42.1)
III/IV
32
2 (6.2)
30 (93.8)
0.194
HBsAg 0.851
Child–Pugh grade 1.000
AFP (ng/ml) 0.016
\0.001
HBsAg hepatitis B s antigen, AFP alpha-fetoprotein, TNM tumornode-metastasis
immunohistochemical staining of endothelial cells (Fig. 2c, f). The MVD in HCC tissues ranged from 0 to 138/200 per field (median, 64/200 per field). AGGF1-positive HCC specimens had a significantly greater MVD than AGGF1negative ones (70.9 ± 38.3 vs. 51.3 ± 30.2, P = 0.034; Fig. 3).
Fig. 3 Immunohistochemical staining of CD34 for microvessel density (MVD) in HCC tissues. Tumors with positive AGGF1 expression had a significantly higher MVD compared to tumors with negative AGGF1 expression (P = 0.034). Data are expressed as the number of CD34-positive microvessels per field
123
131 Page 6 of 9
Med Oncol (2015) 32:131
Fig. 4 Kaplan–Meier analysis of overall survival (OS) and diseasefree survival (DFS) curves of patients with HCC based on AGGF1 expression as positive or negative. a OS curve of patients with HCC based on AGGF1 expression; b DFS curve of patients with HCC
based on AGGF1 expression. The HCC patients with AGGF1-positive showed notably worse OS and DFS rates than those with AGGF1negative
Univariate and multivariate survival analyses of prognostic variables in HCC patients
expression of AGGF1 enhances angiogenesis and increases the blood flow in ischemic hindlimbs in a mouse model [12]. A recent study has reported that AGGF1 protects from myocardial ischemia/reperfusion injury by reducing myocardial apoptosis and enhancing angiogenesis [17]. In addition to endothelial cells, some types of cancer cells (e.g., bladder urothelial cancer cells [18]) can express AGGF1. Our present data showed that AGGF1 expression was upregulated in HCC compared with adjacent noncancerous tissues. Similarly, malignant pleural mesothelioma displays overexpression of AGGF1 [19]. Our results further demonstrated that AGGF1 overexpression was significantly associated with increased tumor size, absence of tumor capsule, vascular invasion, increased AFP levels, and advanced Edmondson grade and TNM stage. These findings suggest an important role for AGGF1 in tumor development and progression. HCC is one of the most hypervascular tumors characterized by active tumor angiogenesis. Tumor angiogenesis plays a pivotal role in HCC development and progression [7, 20]. Increased expression of VEGF is correlated with aggressive behaviors and poor prognosis of tumors [21]. Preclinical studies have shown that overexpression of VEGF promotes tumor growth [7, 22]. To check the involvement of AGGF1 in tumor angiogenesis in HCC, we examined the associations of AGGF1 expression with VEGF expression and MVD in HCC specimens. We found that there was a significant positive correlation between
Univariate analysis showed that AGGF1 expression, tumor size, tumor nodule, tumor capsule, vascular invasion, Edmondson grade, VEGF expression, and MVD had significantly prognostic influences on OS and DFS (Table 4). Moreover, multivariate survival analysis using Cox proportional hazard analyses of factors that were significant in the univariate analyses indicated that AGGF1 expression remained an independent prognostic factors for OS [hazard ratio (HR) 2.328; 95 % CI 1.151–4.710; P = 0.019] and DFS [HR 2.336; 95 % CI 1.144–4.770; P = 0.020] (Table 5). Additionally, vascular invasion and MVD were also independently associated with OS and DFS in HCC patients. However, VEGF expression had no independent impact on their survival.
Discussion AGGF1, previously known as VG5Q, is a potent angiogenic factor critical to vasculogenesis and angiogenesis. AGGF1 shows strong expression in blood vessels and the ability to promote endothelial cell proliferation [9]. Deregulation of AGGF1 has been shown to affect endothelial cell survival and function, contributing to a vascular malformation consistent with KTS [9, 16]. Enforced
123
Med Oncol (2015) 32:131
Page 7 of 9 131
Fig. 5 Kaplan–Meier analysis of overall survival (OS) and diseasefree survival (DFS) curves of patients with HCC based on AGGF1 expression as strongly positive, weakly positive or negative. a OS curve of patients with HCC based on AGGF1 expression; b DFS
curve of patients with HCC based on AGGF1 expression. The HCC patients with AGGF1-positive had a much worse OS and DFS rates than those with AGGF1-negative. The survival of patients with strongly positive AGGF1 expression was poorest
Table 4 Univariate analysis of factors associated with OS and DFS Characteristics
OS Hazard ratio
DFS 95 % CI
P
Hazard ratio
95 % CI
P \0.001
AGGF1 (negative/positive)
2.904
1.481–5.694
0.002
3.363
1.711–6.608
Tumor size (B5/[5)
1.959
1.065–3.605
0.031
1.994
1.080–3.682
0.027
Age (\65/C65 years)
1.162
0.522–2.587
0.713
1.056
0.474–2.353
0.893
Gender (male/female) Tumor nodule (single/multiple)
1.631 1.948
0.784–3.390 0.990–3.835
0.191 0.053
1.648 2.429
0.797–3.408 1.233–4.783
0.178 0.010
Tumor capsule (present/absent)
2.154
1.223–3.795
0.008
2.148
1.219–3.783
0.008
Vascular invasion (no/yes)
3.169
1.700–5.909
\0.001
3.138
1.699–5.796
\0.001
Edmondson grade (I–II/III–IV)
1.386
1.042–1.844
0.025
1.395
1.051–1.852
0.021
Cirrhosis (absent/present)
1.318
0.521–3.334
0.559
1.469
0.578–3.732
0.419
HBsAg (positive/negative)
1.378
0.669–2.838
0.385
1.439
0.697–2.973
0.325
Child–Pugh grade (A/B)
0.871
0.211–3.604
0.849
0.845
0.204–3.491
0.816
AFP (B20/[20 ng/ml)
0.907
0.503–1.635
0.745
0.966
0.537–1.737
0.907
VEGF (negative/positive)
2.646
1.345–5.206
0.005
2.806
1.423–5.533
0.003
MVD (low/high)
2.435
1.342–4.416
0.003
2.430
1.330–4.441
0.004
OS overall survival, DFS disease-free survival, HBsAg hepatitis B s antigen, AFP alpha-fetoprotein, TNM tumor-node-metastasis
AGGF1 and VEGF expression. Moreover, HCC cases with high AGGF1 expression had a significantly greater MVD than those with low AGGF1 expression. These results suggest that AGGF1 is implicated in tumor angiogenesis in HCC, which may be linked to cooperation with VEGF.
Major et al. [23] characterized AGGF1 as a nuclear chromatin-associated protein that participates in b-catenin– mediated transcription in human colon cancer cells. bCatenin has been found to regulate in colon cancer the transcription of VEGF gene, which contains seven
123
131 Page 8 of 9
Med Oncol (2015) 32:131
Table 5 Multivariate analysis of prognostic parameters associated with OS and DFS Characteristics
OS
DFS
Hazard ratio
95 % CI
AGGF1 (negative/positive)
2.101
1.023–4.316
Tumor size (B5/[5)
–
–
Tumor nodule (single/multiple)
–
–
Tumor capsule (present/absent)
–
–
Vascular invasion (no/yes)
2.241
Edmondson grade (I–II/III–IV)
–
VEGF (negative/positive) MVD (low/high)
– 2.012
1.153–4.355 – – 1.095–3.696
P
Hazard ratio
95 % CI
0.043
2.576
1.258–5.273
NA
–
–
NA
NA
–
–
NA
NA
–
–
NA
0.017
2.018
NA
–
NA 0.024
– 1.970
1.049–3.881 – – 1.059–3.664
P 0.010
0.035 NA NA 0.032
HR hazard ratio, 95 % CI 95 % confidence interval, OS overall survival, DFS disease-free survival, TNM tumor-node-metastasis, NA not applicable
consensus binding sites for beta-catenin/TCF [24]. This raises a possibility that AGGF1 might be involved in the induction of VEGF expression via b-catenin-dependent signaling. Nevertheless, additional studies are required to address the exact roles of AGGF1 in tumor angiogenesis in HCC. Although our data suggest a tumor-promoting role for AGGF1 in HCC, this gene may exert opposing effects on some other cancers. Xu et al. [18] reported that downregulation of AGGF1 promotes the survival of high-grade urothelial cancer cells under hypoxia, suggesting its suppressive role in this malignancy. We next explored the prognostic significance of AGGF1 in HCC. Kaplan–Meier survival analysis demonstrated that HCC patients with high tumoral AGGF1 tended to have reduced OS and DFS, compared to those with low tumoral AGGF1. Both univariate and multivariate analyses confirmed that AGGF1 was an independent poor predictor of OS and DFS in HCC patients following curative resection. To the best of our knowledge, this is the first report of the prognostic value of tumoral AGGF1 expression in resectable HCC patients. Some limitations of this study should be noted. First, this is a retrospective study with relatively small sample size, which is thus associated with a potential selection bias. Second, immunohistochemistry is considered to be an arbitrary technique, yielding semiquantitative data. Many factors such as antigen retrieval protocol, antibody quality, and detection systems may influence the final signal intensity and distribution. The choice of cutoff values for immunohistochemical evaluation is based on the previous study that reports the expression of periostin in head and neck cancer [15]. Therefore, these limitations associated with immunohistochemical studies should be taken into consideration when interpreting the prognostic significance of AGGF1 in HCC. Finally, there is no direct evidence for the function of AGGF1 in HCC development and progression.
123
In conclusion, we report for the first time that upregulation of AGGF1 is significantly associated with miscellaneous clinicopathologic characteristics and prognosis in patients with resectable HCC. There is a significant correlation of AGGF1 expression with VEGF expression and CD34-positive MVD in HCC specimens. Survival analysis demonstrated that overexpression of AGGF1 is an independent poor prognostic factor for OS and DFS of HCC. These results should be confirmed by others. Nevertheless, our data suggest that AGGF1 may contribute to tumor angiogenesis of HCC and represent a potential therapeutic target for this disease. Acknowledgments This research was partly supported by the National Natural Science Foundation of China (No. 81201906). Conflict of interest The authors declare that they have no competing financial interests.
References 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer. 2010;127:2893–917. 2. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin. 2011;61:69–90. 3. Blechacz B, Mishra L. Hepatocellular carcinoma biology. Recent Results Cancer Res. 2013;190:1–20. 4. Semela D, Dufour JF. Angiogenesis and hepatocellular carcinoma. J Hepatol. 2004;41:864–80. 5. Bishayee A, Darvesh AS. Angiogenesis in hepatocellular carcinoma: a potential target for chemoprevention and therapy. Curr Cancer Drug Targets. 2012;12:1095–118. 6. Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell. 1996;86:353–64. 7. Muto J, Shirabe K, Sugimachi K, Maehara Y. Review of angiogenesis in hepatocellular carcinoma. Hepatol Res. 2015;45:1–9. 8. Finsterbusch T, Steinfeldt T, Doberstein K, Rodner C, Mankertz A. Interaction of the replication proteins and the capsid protein of
Med Oncol (2015) 32:131
9.
10.
11.
12.
13. 14.
15.
16.
porcine circovirus type 1 and 2 with host proteins. Virology. 2009;386:122–31. Tian XL, Kadaba R, You SA, Liu M, Timur AA, Yang L, et al. Identification of an angiogenic factor that when mutated causes susceptibility to Klippel–Trenaunay syndrome. Nature. 2004;427:640–5. Hu Y, Li L, Seidelmann SB, Timur AA, Shen PH, Driscoll DJ, et al. Identification of association of common AGGF1 variants with susceptibility for Klippel–Trenaunay syndrome using the structure association program. Ann Hum Genet. 2008;72:636–43. Chen D, Li L, Tu X, Yin Z, Wang Q. Functional characterization of Klippel–Trenaunay syndrome gene AGGF1 identifies a novel angiogenic signaling pathway for specification of vein differentiation and angiogenesis during embryogenesis. Hum Mol Genet. 2013;22:963–76. Lu Q, Yao Y, Yao Y, Liu S, Huang Y, Lu S, et al. Angiogenic factor AGGF1 promotes therapeutic angiogenesis in a mouse limb ischemia model. PLoS One. 2012;7:e46998. Edmondson HA, Steiner PE. Primary carcinoma of the liver: a study of 100 cases among 48,900 necropsies. Cancer. 1954;7:462–503. Sun QK, Zhu JY, Wang W, Lv Y, Zhou HC, Yu JH, et al. Diagnostic and prognostic significance of peroxiredoxin 1 expression in human hepatocellular carcinoma. Med Oncol. 2014;31:786. Kudo Y, Ogawa I, Kitajima S, Kitagawa M, Kawai H, Gaffney PM, et al. Periostin promotes invasion and anchorage-independent growth in the metastatic process of head and neck cancer. Cancer Res. 2006;66:6928–35. Fan C, Ouyang P, Timur AA, He P, You SA, Hu Y, et al. Novel roles of GATA1 in regulation of angiogenic factor AGGF1 and endothelial cell function. J Biol Chem. 2009;284:23331–43.
Page 9 of 9 131 17. Liu Y, Yang H, Song L, Li N, Han QY, Tian C, et al. AGGF1 protects from myocardial ischemia/reperfusion injury by regulating myocardial apoptosis and angiogenesis. Apoptosis. 2014;19:1254–68. 18. Xu Y, Zhou M, Wang J, Zhao Y, Li S, Zhou B, et al. Role of microRNA-27a in down-regulation of angiogenic factor AGGF1 under hypoxia associated with high-grade bladder urothelial carcinoma. Biochim Biophys Acta. 2014;1842:712–25. 19. Roe OD, Anderssen E, Sandeck H, Christensen T, Larsson E, Lundgren S. Malignant pleural mesothelioma: genome-wide expression patterns reflecting general resistance mechanisms and a proposal of novel targets. Lung Cancer. 2010;67:57–68. 20. Tanigawa N, Lu C, Mitsui T, Miura S. Quantitation of sinusoidlike vessels in hepatocellular carcinoma: its clinical and prognostic significance. Hepatology. 1997;26:1216–23. 21. Kaya M, Wada T, Akatsuka T, Kawaguchi S, Nagoya S, Shindoh M, et al. Vascular endothelial growth factor expression in untreated osteosarcoma is predictive of pulmonary metastasis and poor prognosis. Clin Cancer Res. 2000;6:572–7. 22. Yoshiji H, Kuriyama S, Noguchi R, Yoshii J, Ikenaka Y, Yanase K, et al. Angiopoietin 2 displays a vascular endothelial growth factor dependent synergistic effect in hepatocellular carcinoma development in mice. Gut. 2005;54:1768–75. 23. Major MB, Roberts BS, Berndt JD, Marine S, Anastas J, Chung N, et al. New regulators of Wnt/beta-catenin signaling revealed by integrative molecular screening. Sci Signal. 2008;1:ra12. 24. Easwaran V, Lee SH, Inge L, Guo L, Goldbeck C, Garrett E, et al. beta-Catenin regulates vascular endothelial growth factor expression in colon cancer. Cancer Res. 2003;63:3145–53.
123
ORIGINAL PAPER
Overexpression of AGGF1 is correlated with angiogenesis and poor prognosis of hepatocellular carcinoma Wei Wang1 • Guang-Yao Li2 • Jian-Yu Zhu2 • Da-Bing Huang1 Hang-Cheng Zhou3 • Wen Zhong3 • Chu-Shu Ji1
•
Received: 9 October 2014 / Accepted: 13 March 2015 / Published online: 22 March 2015 Ó Springer Science+Business Media New York 2015
Abstract Angiogenic factor with G-patch and FHA domains 1 (AGGF1) is a factor implicating in vascular differentiation and angiogenesis. Several lines of evidence indicate that aberrant expression of AGGF1 is associated with tumor initiation and progression. The aim of this study was to investigate the expression and prognostic value of AGGF1 in hepatocellular carcinoma (HCC), as well as its relationship with clinicopathological factors and tumor angiogenesis. Immunohistochemistry was performed to evaluate the expression of AGGF1 in HCC and paracarcinomatous tissues collected from 70 patients. Vascular endothelial growth factor (VEGF) and CD34 expression levels were examined in the 70 HCC tissues. Prognostic significance of tumoral AGGF1 expression was determined. Notably, AGGF1 expression was significantly
Wei Wang, Guang-Yao Li, and Jian-Yu Zhu have contributed equally to this work.
Electronic supplementary material The online version of this article (doi:10.1007/s12032-015-0574-2) contains supplementary material, which is available to authorized users. & Chu-Shu Ji [email protected]; [email protected] 1
Department of Medical Oncology, Anhui Provincial Hospital, Anhui Medical University, 17# Lujiang Road, Hefei 230001, People’s Republic of China
2
Department of Hepatic Surgery, Anhui Provincial Hospital, Anhui Medical University, Hefei 230001, People’s Republic of China
3
Department of Pathology, Anhui Provincial Hospital, Anhui Medical University, Hefei 230001, People’s Republic of China
higher in HCC than in surrounding non-tumor tissues (65.7 vs. 25.7 %; P \ 0.001). AGGF1 expression was significantly correlated with tumoral VEGF expression and CD34-positive microvessel density. Moreover, AGGF1 expression was significantly associated with tumor size, tumor capsule, vascular invasion, Edmondson grade, alphafetoprotein level, and TNM stage. Kaplan–Meier survival analysis showed that high AGGF1 was correlated with reduced overall survival (OS) rate (P = 0.001) and disease-free survival (DFS) rate (P \ 0.001). Multivariate analysis identified AGGF1 as an independent poor prognostic factor of OS and DFS in HCC patients (P = 0.043 and P = 0.010, respectively). Taken together, increased AGGF1 expression is associated with tumor angiogenesis and serves as an independent unfavorable prognostic factor for OS and DFS in HCC. AGGF1 may represent a potential therapeutic target for HCC. Keywords AGGF1 Hepatocellular carcinoma Angiogenesis Prognosis Abbreviations AFP Alpha-fetoprotein AGGF1 Angiogenic factor with G-patch and forkheadassociated domain 1 DAB 3,3-diaminobenzidine tetrahydrochloride DFS Disease-free survival HBsAg Hepatitis B s antigen HCC Hepatocellular carcinoma HE Hematoxylin and eosin KTS Klippel–Trenaunay syndrome MVD Microvessel density OS Overall survival PBS Phosphate-buffered saline TNM Tumor-node-metastasis
123
131 Page 2 of 9
UICC VEGF
International Union Against Cancer Vascular endothelial growth factor
Med Oncol (2015) 32:131
Materials and methods Patients and samples
Introduction Hepatocellular carcinoma (HCC), a predominant histological subtype of primary liver cancer, is the sixth most common neoplasm and the third most frequent cause of cancer-related mortality [1]. An estimated 748,300 new liver cancer cases and 695,900 cancer deaths occurred worldwide in 2008 [2]. Its prognosis is gloomy with a 5-year survival of 11 %. Surgery (e.g., hepatectomy and liver transplantation) provides a potentially curative treatment option for HCC patients. However, the prognosis remains poor due to high potential metastasis and recurrence. Besides, the vast majority of HCC patients are diagnosed at an advanced stage at which surgical treatments are unfeasible [3, 4]. Till now, the molecular mechanism(s) underlying HCC pathogenesis is still poorly understood. Identification of novel therapeutic targets and prognostic biomarkers is of significance to improve the survival of HCC patients. Angiogenesis, a fundamental event in numerous pathologic and physiologic conditions, is essential for tumor growth, invasion, and metastasis [5, 6]. HCC, the most common primary liver cancer, is one of the highly vascularized tumors characterized by active neoangiogenesis [7]. Tumor angiogenesis contributes to the aggressiveness and poor prognosis of HCC [7]. Therefore, targeting angiogenesis is a promising therapeutic strategy for controlling HCC development and progression. AGGF1, an angiogenic gene that is located on human chromosome 5q13.3, encodes an angiogenic factor with G-patch and forkhead-associated (FHA) domain [8, 9]. Upregulation of AGGF1 is associated with congenital vascular disease Klippel–Trenaunay syndrome (KTS) [9, 10]. It has been documented that increased AGGF1 expression in zebrafish embryos activates the AKT signaling pathway and leads to increased vein differentiation and angiogenesis [11]. Similar to vascular endothelial growth factor (VEGF), AGGF1 can promote angiogenesis in a chicken chorioallantoic membrane assay and in a mouse hindlimb ischemia model [9, 12]. Despite the pro-angiogenic activity of AGGF1 in several physiologic and pathologic models, its role in tumor angiogenesis is still unclear. In the current study, immunohistochemistry was used to evaluate the expression of AGGF1, VEGF, and CD34-labeled microvessel density (MVD) in HCC tumor tissues and paracarcinomatous tissues. The expression of AGGF1 in HCC and its associations with clinicopathological features, angiogenesis, and prognosis were examined.
123
Paired samples of cancerous liver tissue and paracarcinomatous for immunohistochemistry were obtained from 70 patients with HCC who underwent curative hepatectomy from 2006 to 2011 at the Department of Hepatic Surgery, Anhui Provincial Hospital (Hefei, China). None of the patients had undergone preoperative intervention therapy or chemotherapy. Detailed demographic and clinicopathological data such as age, gender, tumor size, tumor nodule, tumor capsula, vascular invasion, Edmondson grade, cirrhosis, HBsAg status, Child-Pugh grade, alpha-fetoprotein (AFP), and TNM stage were gathered and summarized. The group 70 patients included 59 males and 11 females with a mean age of 53.5 years (range 30–74 years). Preoperative liver function was assessed by using the Child-Pugh scoring system. Tumor differentiation was defined according to the Edmondson grading system [13]. The tumor stage was performed according to the sixth edition of the tumor-nodemetastasis (TNM) classification of the International Union Against Cancer (UICC). Tumor size was measured as the largest dimension of the tumor by gross examination. Follow-up data were obtained from all 70 patients. The mean follow-up period for patients in the present investigation was 28.6 months (range 2–68 months). After hepatectomy, all patients were monitored using CT and/or MRI as preferential, AFP, chest X-ray, and abdominal ultrasonography every 3–6 months after surgery. Overall survival (OS) was calculated from the date of surgery to the date of death. Diseasefree survival (DFS) time was defined as the interval between the date of surgery and the date of recurrence. If recurrence was not diagnosed, the survivors were censored on the date of death or the last date of follow-up. This study was approved by the Research Ethics Committee of Anhui Provincial Hospital, as stipulated by the Helsinki Declaration. All patients provided written informed consent.
Histology and immunohistochemistry Immunohistochemistry was performed as described by using a two-step protocol according to the manufacturer’s instructions [14]. Briefly, tumor tissues were fixed in 10 % formalin, paraffin-embedded, and sectioned. Sections (4 lm) were deparaffinized in xylene (dimethyl benzene), rehydrated in graded alcohol, and washed with phosphatebuffered saline (PBS). After heated in citrate buffer (pH 6.0) for 20 min in a microwave oven, slides were cooled at room temperature. Endogenous peroxidase activity was then blocked with 3 % hydrogen peroxide (H2O2). Then, sections were, respectively, incubated with primary rabbit
Med Oncol (2015) 32:131
anti-AGGF1 antibody (Beijing Biosynthesis Biotechnology Co., Beijing, China), anti-VEGF antibody, or anti-CD34 antibody (Zhongshan Jinqiao Co., Beijing, China) at 4 °C overnight in a moist chamber. The specificity of these antibodies has been validated by the manufacturers. Following incubation, slides were rinsed with PBS. After incubating with biotin-labeled secondary antibody (mouse anti-rabbit IgG; Zhongshan Jinqiao Co., Beijing, China) for 20 min, color was developed by incubating sections with 3,3-diaminobenzidine (DAB; Zhongshan Jinqiao Co., Beijing, China). Finally, slides were counterstained with hematoxylin followed by dehydration and coverslip mounting. Replacement of the primary antibody with PBS served as a negative control. Immunohistochemical analysis The immunohistochemical results were blindly assessed by two independent pathologists. The percentage of immunoreactive cells to the total cells in 3 random 200 9 microscopic fields was determined for each section. Five sections were randomly selected for each tumor sample. AGGF1 and VEGF expression levels were graded as follows: -:\10 % of tumor cells with positive staining; ?: 10–30 % of tumor cells with positive staining; ??:[30 % of tumor cells with positive staining. The cutoff values used in this study are consistent with a previous report by Kudo et al. [15]. MVD assessment CD34 antibody was used to stain vascular endothelial cells, and CD34-positive MVD was then calculated. Five areas with the greatest number of distinctly highlighted microvessels (hot spots) were selected, and the number of vessels was counted at a high-power magnification (4009). An average count was calculated by average measurements. All sections were independently assessed by two pathologists who had no knowledge of the corresponding clinicopathologic characteristics, and discrepancies were resolved by consensus. Statistical analysis Quantitative data were presented as mean ± standard deviation (SD). The association analysis between the expression of AGGF1 and clinicopathological features was performed by Pearson’s Chi-square test or Fisher´s exact test. Spearman correlation coefficients (r) were employed to analyze the correlation between expression of AGGF1 and VEGF. The Kaplan–Meier method and the log-rank test were used for survival analysis. Cox proportional hazards regression model was used for multivariate survival analysis to
Page 3 of 9 131
identify prognostic factors that were significant in the univariate analysis. All statistical analyses were performed using SPSS 19.0 for Windows (SPSS, Inc., Chicago, IL, USA). A P value\0.05 was considered statistically significant.
Results AGGF1 expression in HCC tissues and its correlation with clinicopathological features To further investigate the clinicopathological and prognostic roles of AGGF1 expression in HCC, immunohistochemical analysis was performed in 70 HCC tissue samples and matched corresponding paracarcinomatous. The staining of AGGF1 protein was mainly located at cytoplasm, and various staining intensities could be observed in different tissues shown in Figs. 1, 2a, d. To verify the antibody specificity, Western blot analysis was performed to detect AGGF1 protein in a panel of human HCC cell lines (Huh7, SMMC7721, HepG2, and Hep3B) using the same anti-AGGF1 antibody. The results showed a single 84-kDa band that corresponds to the expected size of AGGF1 protein (Supplementary Fig. S1). Therefore, the antibody was specific for AGGF1. The positive rate of AGGF1 expression was 65.7 % (46/ 70). In adjacent non-cancerous tissues, most cases were negative for AGGF1 expression. Only 18 of 70 corresponding paracarcinomatous (25.7 %) showed AGGF1positive staining (Table 1). The AGGF1-positive frequency was significantly different between HCC tissues and adjacent non-cancerous liver tissues (P \ 0.001). The associations of AGGF1 expression with clinicopathological parameters are listed in Table 2. Increased expression of AGGF1 was significantly correlated with tumor size (P = 0.033), tumor capsule (P = 0.004), vascular invasion (P = 0.001), Edmondson grade (P = 0.001), AFP level (P = 0.016), and TNM stage (P \ 0.001). However, AGGF1 expression was not significantly correlated with age, gender, tumor nodule number, cirrhosis, HBsAg status, or Child-Pugh grade (P [ 0.05). Correlation between AGGF1 and VEGF protein expression in HCC tissues To determine whether AGGF1 is indeed associated with angiogenesis in HCC tissues, the expression of VEGF and CD34 was examined in HCC patients using immunohistochemistry. The staining of VEGF was mainly localized in the cytoplasm. Forty-eight of the 70 HCC samples showed VEGF positivity (68.6 %; Fig. 2b, e). The positive rate of both AGGF1 and VEGF was 57.1 % (40/70), and their negative expression rate was 22.9 % (16/70). The consistency of both AGGF1 and
123
131 Page 4 of 9
Med Oncol (2015) 32:131
Fig. 1 Representative immunohistochemical staining of AGGF1 in hepatocellular carcinoma (HCC) tissues. AGGF1 expression showed diffuse cytoplasmic staining in tumor cells. a ?? AGGF1 staining
(bar = 100 lm); b ?AGGF1 staining (bar = 100 lm); c Negative expression of AGGF1 (bar = 100 lm)
Fig. 2 Immunohistochemical staining of AGGF1, VEGF, and CD34 in one representative HCC case. a Positive for AGGF1 staining (bar = 100 lm). AGGF1 expression showed diffuse cytoplasmic staining in tumor cells (shown by black arrows); b Positive for VEGF staining (bar = 100 lm); c Positive for CD34 staining (bar = 100 lm);
d Partial enlargement of AGGF1 staining with the magnifying power of 4009 (bar = 50 lm); e Partial enlargement of VEGF staining with the magnifying power of 4009 (bar = 50 lm); f Partial enlargement of CD34 staining with the magnifying power of 4009 (bar = 50 lm)
Table 1 Differential expression of AGGF1 in HCC tissues and corresponding paracarcinomatous tissues
Tissues
Case number
-
?*??
Positive rate (%)
P
\0.001
HCC tissues
70
24
46
65.7
Paracarcinomatous tissues
70
52
18
25.7
VEGF expression was 80.0 % (56/70). The discrepancy between AGGF1 and VEGF expression was 20.0 % (14/70). Moreover, Spearman’s rank correlation test proved a significant correlation between tumoral expression of AGGF1 and VEGF (r = 0.548, P \ 0.001; Table 3).
123
AGGF1 expression
Relationship between AGGF1 expression and MVD in HCC tissues To further evaluate the association between AGGF1 and angiogenesis, MVD was calculated by measuring the CD34
Med Oncol (2015) 32:131
Page 5 of 9 131
Table 2 AGGF1 expression in relation to clinicopathologic features Clinicopathologic data
n
AGGF1 expression n (%)
?*?? n (%)
P value
Table 3 The expression correlation between AGGF1 and VEGF (cases) Stain
AGGF1 –
Age (years)
R
P value
0.548
\0.001
?*??
VEGF
\65
60
21 (35)
39 (65)
C65 Gender
10
3 (30)
7 (70)
Male
59
22 (37.3)
37 (62.7)
Female
11
2 (18.2)
9 (81.8)
1.000
– ?*??
16
6
8
40
VEGF vascular endothelial growth factor 0.379
Tumor size (cm) B5
26
13 (50)
13 (50)
[5
44
11 (25)
33 (75)
Single
58
23 (39.7)
35 (60.3)
Multiple
12
1 (8.3)
11 (91.7)
0.033
Association between AGGF1 expression and prognosis
0.081
AGGF1 expression in HCC patients was found to inversely correlate with patient survival, which was revealed by Kaplan–Meier survival analysis and the log-rank test. The median OS time in HCC patients with positive expression of AGGF1 was significantly reduced, compared to those with negative expression of AGGF1 (26.1 vs. 47.0 months, P = 0.001; Fig. 4a). Likewise, patients with AGGF1positive expression had a shorter DFS time compared to those with AGGF1-negative expression (22.6 vs. 46.6 months, P \ 0.001; Fig. 4b). To further analyze the relationship between AGGF1 expression level and survival, Kaplan– Meier method was used to compare the survival of patients with weak and strong AGGF1 expression (Fig. 5). Notably, HCC patients with strong AGGF1 expression had a significantly shorter OS (P = 0.002) and DFS (P \ 0.001) than those with weak AGGF1 expression.
Tumor nodule number
Tumor capsule Absent
28
4 (14.3)
24 (85.7)
Present
42
20 (47.6)
22 (52.4)
No
50
23 (46)
27 (54)
Yes Edmondson grade
20
1 (5)
19 (95)
I–II
34
18 (52.9)
16 (47.1)
III–IV
36
6 (16.7)
30 (83.3)
0.004
Vascular invasion 0.001
0.001
Cirrhosis Absent
6
4 (66.7)
2 (33.3)
Present
64
20 (31.2)
44 (68.8)
Positive
59
21 (35.6)
38 (64.4)
Negative
11
3 (27.3)
8 (72.7)
A
66
22 (33.3)
44 (66.7)
B
4
2 (50)
2 (50)
B20
22
12 (54.5)
10 (45.5)
[20
48
12 (25)
36 (75)
TNM stage I/II
38
22 (57.9)
16 (42.1)
III/IV
32
2 (6.2)
30 (93.8)
0.194
HBsAg 0.851
Child–Pugh grade 1.000
AFP (ng/ml) 0.016
\0.001
HBsAg hepatitis B s antigen, AFP alpha-fetoprotein, TNM tumornode-metastasis
immunohistochemical staining of endothelial cells (Fig. 2c, f). The MVD in HCC tissues ranged from 0 to 138/200 per field (median, 64/200 per field). AGGF1-positive HCC specimens had a significantly greater MVD than AGGF1negative ones (70.9 ± 38.3 vs. 51.3 ± 30.2, P = 0.034; Fig. 3).
Fig. 3 Immunohistochemical staining of CD34 for microvessel density (MVD) in HCC tissues. Tumors with positive AGGF1 expression had a significantly higher MVD compared to tumors with negative AGGF1 expression (P = 0.034). Data are expressed as the number of CD34-positive microvessels per field
123
131 Page 6 of 9
Med Oncol (2015) 32:131
Fig. 4 Kaplan–Meier analysis of overall survival (OS) and diseasefree survival (DFS) curves of patients with HCC based on AGGF1 expression as positive or negative. a OS curve of patients with HCC based on AGGF1 expression; b DFS curve of patients with HCC
based on AGGF1 expression. The HCC patients with AGGF1-positive showed notably worse OS and DFS rates than those with AGGF1negative
Univariate and multivariate survival analyses of prognostic variables in HCC patients
expression of AGGF1 enhances angiogenesis and increases the blood flow in ischemic hindlimbs in a mouse model [12]. A recent study has reported that AGGF1 protects from myocardial ischemia/reperfusion injury by reducing myocardial apoptosis and enhancing angiogenesis [17]. In addition to endothelial cells, some types of cancer cells (e.g., bladder urothelial cancer cells [18]) can express AGGF1. Our present data showed that AGGF1 expression was upregulated in HCC compared with adjacent noncancerous tissues. Similarly, malignant pleural mesothelioma displays overexpression of AGGF1 [19]. Our results further demonstrated that AGGF1 overexpression was significantly associated with increased tumor size, absence of tumor capsule, vascular invasion, increased AFP levels, and advanced Edmondson grade and TNM stage. These findings suggest an important role for AGGF1 in tumor development and progression. HCC is one of the most hypervascular tumors characterized by active tumor angiogenesis. Tumor angiogenesis plays a pivotal role in HCC development and progression [7, 20]. Increased expression of VEGF is correlated with aggressive behaviors and poor prognosis of tumors [21]. Preclinical studies have shown that overexpression of VEGF promotes tumor growth [7, 22]. To check the involvement of AGGF1 in tumor angiogenesis in HCC, we examined the associations of AGGF1 expression with VEGF expression and MVD in HCC specimens. We found that there was a significant positive correlation between
Univariate analysis showed that AGGF1 expression, tumor size, tumor nodule, tumor capsule, vascular invasion, Edmondson grade, VEGF expression, and MVD had significantly prognostic influences on OS and DFS (Table 4). Moreover, multivariate survival analysis using Cox proportional hazard analyses of factors that were significant in the univariate analyses indicated that AGGF1 expression remained an independent prognostic factors for OS [hazard ratio (HR) 2.328; 95 % CI 1.151–4.710; P = 0.019] and DFS [HR 2.336; 95 % CI 1.144–4.770; P = 0.020] (Table 5). Additionally, vascular invasion and MVD were also independently associated with OS and DFS in HCC patients. However, VEGF expression had no independent impact on their survival.
Discussion AGGF1, previously known as VG5Q, is a potent angiogenic factor critical to vasculogenesis and angiogenesis. AGGF1 shows strong expression in blood vessels and the ability to promote endothelial cell proliferation [9]. Deregulation of AGGF1 has been shown to affect endothelial cell survival and function, contributing to a vascular malformation consistent with KTS [9, 16]. Enforced
123
Med Oncol (2015) 32:131
Page 7 of 9 131
Fig. 5 Kaplan–Meier analysis of overall survival (OS) and diseasefree survival (DFS) curves of patients with HCC based on AGGF1 expression as strongly positive, weakly positive or negative. a OS curve of patients with HCC based on AGGF1 expression; b DFS
curve of patients with HCC based on AGGF1 expression. The HCC patients with AGGF1-positive had a much worse OS and DFS rates than those with AGGF1-negative. The survival of patients with strongly positive AGGF1 expression was poorest
Table 4 Univariate analysis of factors associated with OS and DFS Characteristics
OS Hazard ratio
DFS 95 % CI
P
Hazard ratio
95 % CI
P \0.001
AGGF1 (negative/positive)
2.904
1.481–5.694
0.002
3.363
1.711–6.608
Tumor size (B5/[5)
1.959
1.065–3.605
0.031
1.994
1.080–3.682
0.027
Age (\65/C65 years)
1.162
0.522–2.587
0.713
1.056
0.474–2.353
0.893
Gender (male/female) Tumor nodule (single/multiple)
1.631 1.948
0.784–3.390 0.990–3.835
0.191 0.053
1.648 2.429
0.797–3.408 1.233–4.783
0.178 0.010
Tumor capsule (present/absent)
2.154
1.223–3.795
0.008
2.148
1.219–3.783
0.008
Vascular invasion (no/yes)
3.169
1.700–5.909
\0.001
3.138
1.699–5.796
\0.001
Edmondson grade (I–II/III–IV)
1.386
1.042–1.844
0.025
1.395
1.051–1.852
0.021
Cirrhosis (absent/present)
1.318
0.521–3.334
0.559
1.469
0.578–3.732
0.419
HBsAg (positive/negative)
1.378
0.669–2.838
0.385
1.439
0.697–2.973
0.325
Child–Pugh grade (A/B)
0.871
0.211–3.604
0.849
0.845
0.204–3.491
0.816
AFP (B20/[20 ng/ml)
0.907
0.503–1.635
0.745
0.966
0.537–1.737
0.907
VEGF (negative/positive)
2.646
1.345–5.206
0.005
2.806
1.423–5.533
0.003
MVD (low/high)
2.435
1.342–4.416
0.003
2.430
1.330–4.441
0.004
OS overall survival, DFS disease-free survival, HBsAg hepatitis B s antigen, AFP alpha-fetoprotein, TNM tumor-node-metastasis
AGGF1 and VEGF expression. Moreover, HCC cases with high AGGF1 expression had a significantly greater MVD than those with low AGGF1 expression. These results suggest that AGGF1 is implicated in tumor angiogenesis in HCC, which may be linked to cooperation with VEGF.
Major et al. [23] characterized AGGF1 as a nuclear chromatin-associated protein that participates in b-catenin– mediated transcription in human colon cancer cells. bCatenin has been found to regulate in colon cancer the transcription of VEGF gene, which contains seven
123
131 Page 8 of 9
Med Oncol (2015) 32:131
Table 5 Multivariate analysis of prognostic parameters associated with OS and DFS Characteristics
OS
DFS
Hazard ratio
95 % CI
AGGF1 (negative/positive)
2.101
1.023–4.316
Tumor size (B5/[5)
–
–
Tumor nodule (single/multiple)
–
–
Tumor capsule (present/absent)
–
–
Vascular invasion (no/yes)
2.241
Edmondson grade (I–II/III–IV)
–
VEGF (negative/positive) MVD (low/high)
– 2.012
1.153–4.355 – – 1.095–3.696
P
Hazard ratio
95 % CI
0.043
2.576
1.258–5.273
NA
–
–
NA
NA
–
–
NA
NA
–
–
NA
0.017
2.018
NA
–
NA 0.024
– 1.970
1.049–3.881 – – 1.059–3.664
P 0.010
0.035 NA NA 0.032
HR hazard ratio, 95 % CI 95 % confidence interval, OS overall survival, DFS disease-free survival, TNM tumor-node-metastasis, NA not applicable
consensus binding sites for beta-catenin/TCF [24]. This raises a possibility that AGGF1 might be involved in the induction of VEGF expression via b-catenin-dependent signaling. Nevertheless, additional studies are required to address the exact roles of AGGF1 in tumor angiogenesis in HCC. Although our data suggest a tumor-promoting role for AGGF1 in HCC, this gene may exert opposing effects on some other cancers. Xu et al. [18] reported that downregulation of AGGF1 promotes the survival of high-grade urothelial cancer cells under hypoxia, suggesting its suppressive role in this malignancy. We next explored the prognostic significance of AGGF1 in HCC. Kaplan–Meier survival analysis demonstrated that HCC patients with high tumoral AGGF1 tended to have reduced OS and DFS, compared to those with low tumoral AGGF1. Both univariate and multivariate analyses confirmed that AGGF1 was an independent poor predictor of OS and DFS in HCC patients following curative resection. To the best of our knowledge, this is the first report of the prognostic value of tumoral AGGF1 expression in resectable HCC patients. Some limitations of this study should be noted. First, this is a retrospective study with relatively small sample size, which is thus associated with a potential selection bias. Second, immunohistochemistry is considered to be an arbitrary technique, yielding semiquantitative data. Many factors such as antigen retrieval protocol, antibody quality, and detection systems may influence the final signal intensity and distribution. The choice of cutoff values for immunohistochemical evaluation is based on the previous study that reports the expression of periostin in head and neck cancer [15]. Therefore, these limitations associated with immunohistochemical studies should be taken into consideration when interpreting the prognostic significance of AGGF1 in HCC. Finally, there is no direct evidence for the function of AGGF1 in HCC development and progression.
123
In conclusion, we report for the first time that upregulation of AGGF1 is significantly associated with miscellaneous clinicopathologic characteristics and prognosis in patients with resectable HCC. There is a significant correlation of AGGF1 expression with VEGF expression and CD34-positive MVD in HCC specimens. Survival analysis demonstrated that overexpression of AGGF1 is an independent poor prognostic factor for OS and DFS of HCC. These results should be confirmed by others. Nevertheless, our data suggest that AGGF1 may contribute to tumor angiogenesis of HCC and represent a potential therapeutic target for this disease. Acknowledgments This research was partly supported by the National Natural Science Foundation of China (No. 81201906). Conflict of interest The authors declare that they have no competing financial interests.
References 1. Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer. 2010;127:2893–917. 2. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin. 2011;61:69–90. 3. Blechacz B, Mishra L. Hepatocellular carcinoma biology. Recent Results Cancer Res. 2013;190:1–20. 4. Semela D, Dufour JF. Angiogenesis and hepatocellular carcinoma. J Hepatol. 2004;41:864–80. 5. Bishayee A, Darvesh AS. Angiogenesis in hepatocellular carcinoma: a potential target for chemoprevention and therapy. Curr Cancer Drug Targets. 2012;12:1095–118. 6. Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell. 1996;86:353–64. 7. Muto J, Shirabe K, Sugimachi K, Maehara Y. Review of angiogenesis in hepatocellular carcinoma. Hepatol Res. 2015;45:1–9. 8. Finsterbusch T, Steinfeldt T, Doberstein K, Rodner C, Mankertz A. Interaction of the replication proteins and the capsid protein of
Med Oncol (2015) 32:131
9.
10.
11.
12.
13. 14.
15.
16.
porcine circovirus type 1 and 2 with host proteins. Virology. 2009;386:122–31. Tian XL, Kadaba R, You SA, Liu M, Timur AA, Yang L, et al. Identification of an angiogenic factor that when mutated causes susceptibility to Klippel–Trenaunay syndrome. Nature. 2004;427:640–5. Hu Y, Li L, Seidelmann SB, Timur AA, Shen PH, Driscoll DJ, et al. Identification of association of common AGGF1 variants with susceptibility for Klippel–Trenaunay syndrome using the structure association program. Ann Hum Genet. 2008;72:636–43. Chen D, Li L, Tu X, Yin Z, Wang Q. Functional characterization of Klippel–Trenaunay syndrome gene AGGF1 identifies a novel angiogenic signaling pathway for specification of vein differentiation and angiogenesis during embryogenesis. Hum Mol Genet. 2013;22:963–76. Lu Q, Yao Y, Yao Y, Liu S, Huang Y, Lu S, et al. Angiogenic factor AGGF1 promotes therapeutic angiogenesis in a mouse limb ischemia model. PLoS One. 2012;7:e46998. Edmondson HA, Steiner PE. Primary carcinoma of the liver: a study of 100 cases among 48,900 necropsies. Cancer. 1954;7:462–503. Sun QK, Zhu JY, Wang W, Lv Y, Zhou HC, Yu JH, et al. Diagnostic and prognostic significance of peroxiredoxin 1 expression in human hepatocellular carcinoma. Med Oncol. 2014;31:786. Kudo Y, Ogawa I, Kitajima S, Kitagawa M, Kawai H, Gaffney PM, et al. Periostin promotes invasion and anchorage-independent growth in the metastatic process of head and neck cancer. Cancer Res. 2006;66:6928–35. Fan C, Ouyang P, Timur AA, He P, You SA, Hu Y, et al. Novel roles of GATA1 in regulation of angiogenic factor AGGF1 and endothelial cell function. J Biol Chem. 2009;284:23331–43.
Page 9 of 9 131 17. Liu Y, Yang H, Song L, Li N, Han QY, Tian C, et al. AGGF1 protects from myocardial ischemia/reperfusion injury by regulating myocardial apoptosis and angiogenesis. Apoptosis. 2014;19:1254–68. 18. Xu Y, Zhou M, Wang J, Zhao Y, Li S, Zhou B, et al. Role of microRNA-27a in down-regulation of angiogenic factor AGGF1 under hypoxia associated with high-grade bladder urothelial carcinoma. Biochim Biophys Acta. 2014;1842:712–25. 19. Roe OD, Anderssen E, Sandeck H, Christensen T, Larsson E, Lundgren S. Malignant pleural mesothelioma: genome-wide expression patterns reflecting general resistance mechanisms and a proposal of novel targets. Lung Cancer. 2010;67:57–68. 20. Tanigawa N, Lu C, Mitsui T, Miura S. Quantitation of sinusoidlike vessels in hepatocellular carcinoma: its clinical and prognostic significance. Hepatology. 1997;26:1216–23. 21. Kaya M, Wada T, Akatsuka T, Kawaguchi S, Nagoya S, Shindoh M, et al. Vascular endothelial growth factor expression in untreated osteosarcoma is predictive of pulmonary metastasis and poor prognosis. Clin Cancer Res. 2000;6:572–7. 22. Yoshiji H, Kuriyama S, Noguchi R, Yoshii J, Ikenaka Y, Yanase K, et al. Angiopoietin 2 displays a vascular endothelial growth factor dependent synergistic effect in hepatocellular carcinoma development in mice. Gut. 2005;54:1768–75. 23. Major MB, Roberts BS, Berndt JD, Marine S, Anastas J, Chung N, et al. New regulators of Wnt/beta-catenin signaling revealed by integrative molecular screening. Sci Signal. 2008;1:ra12. 24. Easwaran V, Lee SH, Inge L, Guo L, Goldbeck C, Garrett E, et al. beta-Catenin regulates vascular endothelial growth factor expression in colon cancer. Cancer Res. 2003;63:3145–53.
123
