- Email: [email protected]
Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis
Biochemical and Biophysical Research Communications xxx (2017) 1e8
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis Hai Wen a, Hong Ma a, Pengzhi Li a, Jiaoyun Zheng b, Yipin Yu a, Guohua Lv a, * a b
Department of Spine Surgery, The Second Xiangya Hospital of Central South University, Changsha 410011, China Department of Pathology, The Second Xiangya Hospital of Central South University, Changsha 410011, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 July 2017 Accepted 1 August 2017 Available online xxx
The far upstream element (FUSE)-binding protein 1 (FUBP1), a well-known transcriptional regulator of the proto-oncogene c-Myc, has been demonstrated by previous work to be aberrantly expressed in a variety of tumors and plays a critical role in tumor progression; however, its expression and function in relatively rare and aggressive chordomas remains unclear. In this retrospective study, we reviewed clinicopathologic characteristics of 40 patients diagnosed with sacral chordoma, and analyzed 40 tumor and 20 distant normal tissues obtained from patients during the primary surgical tumor excision. Using immunohistochemistry, we observed an up-regulation in the expression of FUBP1 and c-Myc in sacral chordomas compared with the normal tissues (P ¼ 0.001 for both). Additionally, positive correlations of FUBP1 expression with c-Myc (g ¼ 0.651, P < 0.001) and the cell proliferation index Ki-67 expression (g ¼ 0.447, P ¼ 0.004) were indicated using Spearman's rank correlation coefficient. Increased expression of FUBP1 was significantly associated with tumor invasion into the surrounding muscles (P ¼ 0.002). Kaplan-Meier curves demonstrated the association between FUBP1 levels and the patients' local recurrence-free survival (LRFS) (P < 0.001) but not with the overall survival (OS) (P ¼ 0.070). The independent prognostic significance of FUBP1 levels for the LRFS was indicated by multivariate analysis (HR ¼ 4.272; 95% CI, 1.133e16.112; P ¼ 0.032). Our findings demonstrate an association between FUBP1 levels and chordoma progression and prognosis, suggesting that FUBP1 can be used as a biomarker and a potential therapeutic target. © 2017 Elsevier Inc. All rights reserved.
Keywords: Sacral chordomas FUBP1 c-Myc Biomarker Recurrence
1. Introduction Chordomas are rare bony neoplasms that account for 1%e4% of all primary malignant bone tumors originating from embryonic notochordal remnants along the cerebrospinal axis [1e3]. Slowgrowing sacral chordomas, constituting approximately 50%e60% of all chordomas, present indolent symptoms, which often results in delayed diagnosis and treatment [4,5]. Because of the poor sensitivity of chordomas to chemotherapy and conventional radiotherapy, surgical excision remains the mainstay of treatment [6e9]; however, in patients with the late stage disease, the presence of inadequate resection margins significantly increases the risk of local tumor recurrence, influencing the patients' quality of
* Corresponding author. E-mail address: [email protected] (G. Lv).
life after surgery [7,10]. Although it is often technically difficult, achieving wide resection margins is the most important factor determining tumor recurrence and long-term prognosis [5,11,12]. Because the molecular mechanisms underlying tumorigenesis and progression of chordomas are not well-understood, it is crucial to identify effective biomarkers that can predict clinical prognosis and provide potential therapeutic targets. The far upstream element (FUSE)-binding protein 1 (FUBP1), one of a three-member family, was identified as a DNA-binding protein from undifferentiated HL60 cells; using four K-homology domains, FUBP1 binds to single-stranded FUSE located upstream of the proto-oncogene c-Myc promoter and regulates its transcription [13,14]. In recent years, it has been shown that the aberrant expression of FUBP1 is often correlated with an abnormal expression of c-Myc, and it was demonstrated in a growing number of malignancies; increasing studies have focused on investigating the conceivable roles of FUBP1 in tumor initiation and progression
http://dx.doi.org/10.1016/j.bbrc.2017.08.008 0006-291X/© 2017 Elsevier Inc. All rights reserved.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
2
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
because of its effect on cellular activities including proliferation, migration and apoptosis of tumor cells [15e17]. For example, it was revealed that overexpression of FUBP1 in human hepatocellular carcinoma (HCC) is associated with poor patient survival and increased cell proliferation in vitro [15]. It was also reported that patients with glioma have unfavorable prognosis when the levels of FUBP1 and c-Myc are elevated, and down-regulation of FUBP1 reduces c-Myc expression and arrests cell growth [18]. These results suggest that FUBP1 can be a new cancer biomarker and a potential treatment target for many human tumors. Although a significant up-regulation in the expression of the mRNAs of FUBP1 and Myc in chordomas compared with notochord tissues has been identified using microarrays with hierarchical clustering analysis previously [19], the evidence of the expression and function of FUBP1 and c-Myc in chordomas is very limited. We carried out this study to investigate the expression of the proteins of FUBP1 and c-Myc in sacral chordomas and assess the association between their expression and the clinicopathologic characteristics and prognosis of patients with sacral chordoma. 2. Materials and methods 2.1. Patients and tissue samples A total of 40 patients diagnosed with sacral chordoma who had received surgical treatment in the Department of Spine Surgery, The Second Xiangya Hospital of Central South University (Changsha, China) were enrolled in this study between January 2005 and October 2015. There were 23 males and 17 females with a mean age of 55.1 years old (range, 29e74 years old). The patients had not received any tumor-specific therapy, including radiotherapy and chemotherapy, prior to surgical treatment. Clinicopathologic parameters including age and gender of the patients; tumor location, size, grade, and stage; tumor invasion of the surrounding muscles; and the type of surgical resection performed were obtained from patients' medical records, retrospectively. Tumor grade and stage were evaluated based on the Enneking staging system for the surgical staging of malignant bone and soft tissue tumors [20e22]. Resected specimens were assessed either as Enneking appropriate (EA) or Enneking inappropriate (EI) by anatomic pathologists according to the Enneking principles [23]. Surrounding muscle invasion was determined using preoperative magnetic resonance imaging (MRI) and histologic examination of tissues [24]. In addition to tumor tissues, we also collected 20 distant normal tissues at least 3 cm away from the surgical margins as a control [25]. The tissue samples were fixed immediately after surgery in 10% buffered formalin and embedded in paraffin. Subsequently, the paraffin blocks were cut into 4-mm thick slices for immunohistochemical analysis. Tumor diagnosis was confirmed through histologic examination of hematoxylin and eosin-stained sections by two pathologists. The study was approved by the Ethical Committee of The Second Xiangya Hospital, Central South University, and informed consent was obtained from each patient enrolled in the study. 2.2. Follow-up Patients were clinically followed-up with radiographic evaluations every 3 months during the first 2 years, every 6 months for the next 3 years, and annually thereafter until October 2016. Tumor recurrence was diagnosed by analyzing clinical manifestations and imaging results. The interval between the primary surgery and the first incidence of local tumor recurrence or death from any cause was defined as the local recurrence-free survival (LRFS) or the overall survival (OS), respectively.
2.3. Immunohistochemistry Tissue sections were deparaffinized in xylene and rehydrated by passage through a series of ethanol solutions; subsequently, the tissues were immersed in 0.01 M citrate buffer (pH 6.0) and heated in a microwave oven at 121 C for 15 min to retrieve antigen activity. Potential endogenous peroxidase activity was quenched by submerging the samples in 3% hydrogen peroxide solution for 10 min. After rinsing with phosphate-buffered saline (PBS), tissue sections were blocked with 10% normal goat serum at room temperature for 10 min and were then incubated overnight at 4 C with primary antibodies anti-FUBP1 (1:100; Abcam), anti-c-Myc (1: 50; Abcam) or anti-Ki-67 (1:100; Abcam). The negative controls were not incubated with primary antibodies. Subsequently, the sections were rinsed with PBS three times, followed by a 30 min incubation at 37 C with horseradish peroxidase-conjugated anti-rabbit IgG (Boster, Wuhan, China). Antibody binding was visualized using 3,30 diaminobenzidine (DAB) solution. The tissue sections were counterstained with hematoxylin and were then mounted for observation. 2.4. Immunohistochemical evaluation The stained sections were evaluated independently by two experienced pathologists who were blinded to the patients' clinicopathologic information and were scored using a semiquantitative assessment reported previously [26]. Five highpower fields were randomly selected, and at least 300 cells per field were evaluated in each section. Cells with immunostained brown nucleus were considered positive for FUBP1, c-Myc or Ki-67 expression. Based on the method described by Zou et al. [27], we obtained the final score for each section by multiplying the staining intensity, established as 0 (no staining), 1 (weak staining), 2 (moderate staining) or 3 (strong staining), by the ratio of immunostained cells to the total number of cells evaluated; subsequently, we evaluated the mean value of all final scores for FUBP1 expression in tumor tissues as the cut-off value to distinguish between the high and low levels of FUBP1 expression for survival analysis. The same method was used to distinguish between the high and low levels of c-Myc expression. As described previously [24], Ki-67 index was defined as the percentage of Ki-67-positive cells and categorized as a low (<10%) or high (10%) index. 2.5. Statistical analysis All statistical analyses were conducted with SPSS 19.0 (SPSS Inc., Chicago, IL, USA). The difference in FUBP1 and c-Myc expression between the tumor and normal tissues and the association between FUBP1 expression and the clinicopathologic characteristics were examined using the c2 or the Fisher's exact test, as appropriate. Spearman's rank correlation coefficient was used to determine the strength of the paired correlation between FUBP1 and cMyc levels and Ki-67 index. A log-rank test was used to analyze the LRFS and OS; the data are shown as Kaplan-Meier curves. Multivariate analysis was subsequently performed by Cox's proportional hazard model to determine whether a statistically significant factor could independently affect the patients' prognosis. A P-value < 0.05 was considered statistically significant. 3. Results 3.1. Up-regulation of FUBP1 expression and its positive correlation with c-Myc expression and Ki-67 index in sacral chordomas Immunohistochemical analysis was used to investigate the
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
3
Fig. 1. Representative expression of FUBP1, c-Myc and Ki-67 in sacral chordoma tissues. (A) Low expression of FUBP1; (B) low expression of c-Myc; (C) low expression of Ki-67; (D) high expression of FUBP1; (E) high expression of c-Myc; (F) high expression of Ki-67; (G) negative control of FUBP1; (H) negative control of c-Myc; (I) negative control of Ki-67. All magnification, 400.
differential expression of FUBP1 and c-Myc in 40 tumor and 20 distant normal tissues from patients with sacral chordoma. The results indicate that the proteins were predominantly expressed in the nucleus of positive cells (Fig. 1). The mean scores for FUBP1 and c-Myc expression in tumor tissues were 1.174 ± 0.097 and 1.189 ± 0.091, which were considered as the cut-off values, respectively. Higher expression of FUBP1 and c-Myc was observed in 60% (24/40; Fig. 1D) and 55% (22/40; Fig. 1E) of tumor tissues, respectively; whereas only 15% (3/20) and 10% (2/20) of distant normal tissues expressed higher levels of FUBP1 and cMyc, respectively. Expression of FUBP1 and c-Myc was significantly higher in tumor tissues compared with the distant normal tissues (P ¼ 0.001 for both; Table 1). Spearman's correlation test was used to assess the correlation between FUBP1 and c-Myc expression levels. As shown in Fig. 2A, FUBP1 expression is positively correlated with c-Myc expression (Spearman's g ¼ 0.651, P < 0.001). We also examined the expression of Ki-67, a cell proliferation indicator, in sacral chordomas. Based on Ki-67 expression levels, tissues were divided into low expressing (16 samples; Fig. 1C) or high expressing tissues (24 samples; Fig. 1F). Expression of FUBP1 was significantly higher in tumor tissues with a high Ki-67 index than those with a low Ki-67 index (P < 0.001; Table 2). Additionally, the Spearman's rank correlation coefficient for FUBP1 expression and Ki-67 index is 0.447 (P ¼ 0.004; Fig. 2B). A positive correlation between c-Myc expression and Ki-67 index was also observed (Spearman's g ¼ 0.574, P < 0.001; Fig. 2C).
3.2. Association between FUBP1 expression and the clinicopathologic characteristics of patients with sacral chordoma To assess the association between FUBP1 expression and patients' clinicopathologic parameters, the patients were divided into two groups based on the FUBP1 expression levels. Consistent with the results described before, positive correlations of FUBP1 expression with c-Myc expression and Ki-67 index were confirmed by the c2 test (P < 0.001 for both; Table 2). Furthermore, FUBP1 expression was significantly associated with tumor invasion into the surrounding muscles (P ¼ 0.002). No statistically significant association was found between FUBP1 expression and other clinicopathologic characteristics. 3.3. Prognostic significance of FUBP1 expression in sacral chordomas After the primary tumor resection, patients were followed-up until October 2016. Of the total, 27 (67.5%) had a local tumor relapse, and 16 (40%) died during the follow-up period, with a median LRFS of 18 months (range, 5e45 months) and a median OS of 47 months (range, 12e141 months). The results of the KaplanMeier analysis demonstrate that high expression of either FUBP1 or c-Myc is significantly associated with a shorter LRFS time (P < 0.001 for both; Table 3 and Fig. 3A and B). Patients with low FUBP1/c-Myc coexpression had an increased LRFS time (P < 0.001; Fig. 3C), whereas those with high FUBP1/c-Myc coexpression had a
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
4
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
Table 1 Expression of FUBP1 and c-Myc in sacral chordomas compared with distant normal tissues. Tissue samples
Sacral chordomas Distant normal tissues
Total
40 20
FUBP1 expression
P-value
High (%)
Low (%)
24 (60%) 3 (15%)
16 (40%) 17 (85%)
0.001
c-Myc expression
P-value
High (%)
Low (%)
22 (55%) 2 (10%)
18 (45%) 18 (90%)
0.001
Fig. 2. Graphic representation of paired correlation between FUBP1 and c-Myc levels and Ki-67 index in sacral chordomas using Spearman's rank correlation test. (A) Positive correlation between FUBP1 and c-Myc expression; (B) positive correlation between Ki-67 index and FUBP1 expression; (C) positive correlation between Ki-67 index and c-Myc expression.
reduced LRFS time (P < 0.001; Fig. 3D). Shorter LRFS time was also observed in patients with higher Ki-67 index (P ¼ 0.016; Table 3 and Fig. 3E), tumor invasion of the surrounding muscles (P < 0.001; Table 3 and Fig. 3F), higher tumor stage (P < 0.001; Table 3 and Fig. 3G) and Enneking inappropriate resection (P < 0.001; Table 3 and Fig. 3H). Furthermore, shorter OS time was related to factors including higher c-Myc expression (P ¼ 0.011; Table 4 and Fig. 4A), high FUBP1/c-Myc coexpression (P ¼ 0.002; Table 4 and Fig. 4B), tumor invasion of the surrounding muscles (P ¼ 0.032; Table 4 and Fig. 4C), higher tumor stage (P < 0.001; Table 4 and Fig. 4D), and Enneking inappropriate resection (P ¼ 0.015; Table 4 and Fig. 4E). Lastly, the variables associated with LRFS and OS obtained from the Kaplan-Meier analysis were included in the multivariate analysis
using Cox's proportional hazard model. The results indicate that FUBP1 expression (HR ¼ 4.272; 95% CI, 1.133e16.112; P ¼ 0.032; Table 3), tumor stage (HR ¼ 1.872; 95% CI, 1.168e3.000; P ¼ 0.009; Table 3), tumor invasion of the surrounding muscles (HR ¼ 18.398; 95% CI, 3.454e98.001; P ¼ 0.001; Table 3), and the type of tumor resection (HR ¼ 8.025; 95% CI, 2.251e28.607; P ¼ 0.001; Table 3) were the independent prognostic factors for the LRFS, whereas tumor stage was the only independent prognostic factor for the OS (HR ¼ 1.749; 95% CI, 1.043e2.933; P ¼ 0.034; Table 4). 4. Discussion In this study, we observed an up-regulation of FUBP1
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8 Table 2 Association between FUBP1 expression and the clinicopathologic characteristics of patients with sacral chordoma. Variables
Age (years) < 50 50 Gender Male Female Tumor location Above S3 S3 and below Tumor size (cm) <5 5 Tumor grade High Low Tumor stage ⅠA ⅠB ⅡA ⅡB Ⅲ Surrounding muscle invasion Yes No Type of resection EI EA c-Myc expression High Low Ki-67 index High Low
Total
FUBP1 expression
P-value
High
Low
15 25
10 14
5 11
23 17
13 11
10 6
24 16
14 10
10 6
14 26
8 16
6 10
27 13
18 6
9 7
9 4 4 20 3
3 3 1 14 3
6 1 3 6 0
25 15
20 4
5 11
18 22
14 10
4 12
22 18
19 5
3 13
24 16
20 4
4 12
0.740
0.747
1.000
1.000
0.305
0.095*
0.002
0.054
<0.001
<0.001
EI, Enneking inappropriate; EA, Enneking appropriate. *, using the Fisher's exact test.
expression, which was positively correlated with c-Myc expression in sacral chordoma tissues. Increased expression of FUBP1 was significantly associated with a relatively high Ki-67 index, tumor invasion of the surrounding muscles, and poor LRFS. Moreover, FUBP1 has an independent prognostic significance for patients' LRFS. These results suggest the involvement of FUBP1 in tumor progression and its usefulness as a prognostic indicator of patients with sacral chordoma. Because of histologic specificity, the exact function of FUBP1 in
5
the molecular mechanisms of different tumors is unclear; for example, FUBP1 is overexpressed in nasopharyngeal carcinoma and ccRCC, where its knockdown inhibits cell proliferation [17,28], whereas the loss of FUBP1 is an important oncogenic event in oligodendroglioma and basal-like breast cancer [29,30]. In this study, elevated FUBP1 expression was observed in sacral chordomas compared with the normal tissues, suggesting that FUBP1 may promote tumor development in sacral chordomas. FUBP1 was discovered in HL60 cells as a transcriptional regulator of the wellknown proto-oncogene c-Myc whose activation causes malignant transformation [13]. Consistent with the evidence from other cancers, a positive correlation was demonstrated between FUBP1 and c-Myc expression in sacral chordomas, supporting the involvement of the c-Myc-dependent pathway in FUBP1-mediated carcinogenesis. In this study, we determined that up-regulation of FUBP1 expression in sacral chordomas is related to high levels of Ki67, a widely used cell proliferation indicator, suggesting the correlation between FUBP1 and cell proliferation. Based on our findings, we propose that FUBP1 could regulate proliferation of chordoma cells through a c-Myc-dependent pathway; the mechanisms of regulation, however, need to be further explored. The association of FUBP1 with the clinicopathologic characteristics and prognosis of patients with cancer has been frequently reported in recent years. Malz et al. demonstrated that cooverexpression of FUBP1 and FUBP2 was significantly associated with the poor survival of patients with HCC [15]. Ding et al. showed a close connection between FUBP1 expression and tumor stage, and the accumulative survival of patients with human glioma [18]. We observed a significant association between high FUBP1 levels and invasion of the surrounding muscles by sacral chordomas, which supports the importance of FUBP1 in chordoma progression. However, we did not observe an association between FUBP1 expression and tumor size, grade or stage, which could be due to a small sample size, relatively low statistical power or due to the influence of other risk factors. Because patients with sacral chordoma have a high rate of local tumor recurrence, it is critical to identify the related risk factors for an effective prediction of tumor prognosis. Our data show that shorter LRFS time is closely associated with high expression levels of FUBP1 and c-Myc, and certain clinicopathologic factors, including tumor stage, tumor invasion of the surrounding muscles and the resection type. The independent prognostic significance of FUBP1 in sacral chordomas was verified using a multivariate analysis to eliminate the influence of confounding factors. Although patients with sacral chordoma who had high FUBP1 levels did not show reduced OS in our study, more
Table 3 Kaplan-Meier analysis and multivariate Cox's proportional hazard analysis of the prognostic factors for the local recurrence-free survival of patients with sacral chordoma. Variables
Categories
Kaplan-Meier analysis
Multivariate analysis
P-value
P-value
HR (95% CI)
Age (years) Gender Tumor location Tumor size (cm) Tumor grade Tumor stage Surrounding muscle invasion Type of resection Ki-67 index FUBP1 expression c-Myc expression FUBP1/c-Myc coexpression
<50/ 50 Male/Female Above S3/S3 and below <5/ 5 High/Low ⅠA/ⅠB/ⅡA/ⅡB/Ⅲ Yes/No EI/EA High/Low High/Low High/Low Low/Low High/High High/Low and Low/High
0.671 0.587 0.091 0.124 0.595 <0.001 <0.001 <0.001 0.016 <0.001 <0.001 <0.001* <0.001*
0.009 0.001 0.001 0.139 0.032 0.081
1.872 (1.168e3.000) 18.398 (3.454e98.001) 8.025 (2.251e28.607) 0.221 (0.030e1.631) 4.272 (1.133e16.112) 5.634 (0.808e39.275)
EI, Enneking inappropriate; EA, Enneking appropriate. *, not included in the multivariate analysis, because the variables depend on the expression of FUBP1 and c-Myc.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
6 H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
Fig. 3. Kaplan-Meier curves of the local recurrence-free survival (LRFS) of patients with sacral chordoma. (A) Stratified by FUBP1 expression; (B) stratified by c-Myc expression; (C) stratified by FUBP1/c-Myc coexpression (low/low and others); (D) stratified by FUBP1/c-Myc coexpression (high/high and others); (E) stratified by Ki-67 index; (F) stratified by surrounding muscle invasion; (G) stratified by tumor stage; (H) stratified by type of resection.
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
7
Table 4 Kaplan-Meier analysis and multivariate Cox's proportional hazard analysis of the prognostic factors for the overall survival of patients with sacral chordoma. Variables
Age (years) Gender Tumor location Tumor size (cm) Tumor grade Tumor stage Surrounding muscle invasion Type of resection Ki-67 index FUBP1 expression c-Myc expression FUBP1/c-Myc coexpression
Categories
<50/ 50 Male/Female Above S3/S3 and below <5/ 5 High/Low ⅠA/ⅠB/ⅡA/ⅡB/Ⅲ Yes/No EI/EA High/Low High/Low High/Low Low/Low High/High High/Low and Low/High
Kaplan-Meier analysis
Multivariate analysis
P-value
P-value
HR (95% CI)
0.362 0.319 0.075 0.065 0.468 <0.001 0.032 0.015 0.628 0.070 0.011 0.166 0.002*
0.034 0.099 0.511
1.749 (1.043e2.933) 5.866 (0.719e47.862) 1.503 (0.446e5.063)
0.110
3.617 (0.748e17.483)
EI, Enneking inappropriate; EA, Enneking appropriate. *, not included in the multivariate analysis because it depends on the expression of FUBP1 and c-Myc.
Fig. 4. Kaplan-Meier curves of the overall survival (OS) of patients with sacral chordoma. (A) Stratified by c-Myc expression; (B) stratified by FUBP1/c-Myc coexpression (high/high and others); (C) stratified by surrounding muscle invasion; (D) stratified by tumor stage; (E) stratified by type of resection.
evidence is needed to rule out a relationship between FUBP1 expression and the OS. Furthermore, despite that c-Myc is not an independent predictive factor for the LRFS and OS of patients with sacral chordoma indicated by this study, the significance of c-Myc in tumorigenesis and development of chordomas cannot be underestimated and needs to be further explored.
In conclusion, our findings demonstrate that up-regulation of FUBP1 expression plays a substantial role in progression and prognosis of sacral chordomas. Although more research is needed, FUBP1 is a strong candidate for use as a new biomarker and a potential therapeutic target.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
8
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
Acknowledgements This work was supported by the Fundamental Research Funds for the Central Universities of Central South University (No. 2016zzts149). Conflict of interest The authors declare that they have no conflict of interest. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.008. References [1] M.L. McMaster, A.M. Goldstein, C.M. Bromley, et al., Chordoma: incidence and survival patterns in the United States, 1973-1995, Cancer Causes Control 12 (2001) 1e11. [2] J. Llauger, J. Palmer, S. Amores, et al., Primary tumors of the sacrum: diagnostic imaging, AJR Am. J. Roentgenol. 174 (2000) 417e424. [3] S. Stacchiotti, J. Sommer, Chordoma Global Consensus Group, Building a global consensus approach to chordoma: a position paper from the medical and patient community, Lancet Oncol. 16 (2015) e71e83. [4] R. Chugh, H. Tawbi, D.R. Lucas, et al., Chordoma: the nonsarcoma primary bone tumor, Oncologist 12 (2007) 1344e1350. [5] C.A. Hulen, H.T. Temple, W.P. Fox, et al., Oncologic and functional outcome following sacrectomy for sacral chordoma, J. Bone Jt. Surg. Am. 88 (2006) 1532e1539. [6] C. Catton, B. O'Sullivan, R. Bell, et al., Chordoma: long-term follow-up after radical photon irradiation, Radiother. Oncol. 41 (1996) 67e72. [7] E.Y. Cheng, R.A. Ozerdemoglu, E.E. Transfeldt, et al., Lumbosacral chordoma. Prognostic factors and treatment, Spine (Phila Pa 1976) 24 (1999) 1639e1645. [8] A. Azzarelli, V. Quagliuolo, S. Cerasoli, et al., Chordoma: natural history and treatment results in 33 cases, J. Surg. Oncol. 37 (1988) 185e191. [9] J.E. York, A. Kaczaraj, D. Abi-Said, et al., Sacral chordoma: 40-year experience at a major cancer center, Neurosurgery 44 (1999) 74e79 discussion 79e80. [10] D.R. Fourney, Z.L. Gokaslan, Current management of sacral chordoma, Neurosurg. Focus 15 (2003) E9. [11] B. Fuchs, I.D. Dickey, M.J. Yaszemski, et al., Operative management of sacral chordoma, J. Bone Jt. Surg. Am. 87 (2005) 2211e2216. [12] B. Kayani, M.D. Sewell, K.A. Tan, et al., Prognostic factors in the operative management of sacral chordomas, World Neurosurg. 84 (2015) 1354e1361.
[13] M.I. Avigan, B. Strober, D. Levens, A far upstream element stimulates c-myc expression in undifferentiated leukemia cells, J. Biol. Chem. 265 (1990) 18538e18545. [14] R. Duncan, L. Bazar, G. Michelotti, et al., A sequence-specific, single-strand binding protein activates the far upstream element of c-myc and defines a new DNA-binding motif, Genes Dev. 8 (1994) 465e480. [15] M. Malz, A. Weber, S. Singer, et al., Overexpression of far upstream element binding proteins: a mechanism regulating proliferation and migration in liver cancer cells, Hepatology 50 (2009) 1130e1139. [16] F. Zhang, Q. Tian, Y. Wang, Far upstream element-binding protein 1 (FUBP1) is overexpressed in human gastric cancer tissue compared to non-cancerous tissue, Onkologie 36 (2013) 650e655. [17] J. Duan, X. Bao, X. Ma, et al., Upregulation of far upstream element-binding protein 1 (FUBP1) promotes tumor proliferation and tumorigenesis of clear cell renal cell carcinoma, PLoS One 12 (2017) e0169852. [18] Z. Ding, X. Liu, Y. Liu, et al., Expression of far upstream element (FUSE) binding protein 1 in human glioma is correlated with c-Myc and cell proliferation, Mol. Carcinog. 54 (2015) 405e415. [19] C. Long, L. Jiang, F. Wei, et al., Integrated miRNA-mRNA analysis revealing the potential roles of miRNAs in chordomas, PLoS One 8 (2013) e66676. [20] I.R. Samson, D.S. Springfield, H.D. Suit, et al., Operative treatment of sacrococcygeal chordoma. A review of twenty-one cases, J. Bone Jt. Surg. Am. 75 (1993) 1476e1484. [21] B. Kayani, S.A. Hanna, M.D. Sewell, et al., A review of the surgical management of sacral chordoma, Eur. J. Surg. Oncol. 40 (2014) 1412e1420. [22] A. Jemal, R. Siegel, E. Ward, et al., Cancer statistics, 2007, CA Cancer J. Clin. 57 (2007) 43e66. [23] C.G. Fisher, D.D. Saravanja, M.F. Dvorak, et al., Surgical management of primary bone tumors of the spine: validation of an approach to enhance cure and reduce local recurrence, Spine (Phila Pa 1976) 36 (2011) 830e836. [24] M. Zhou, K. Chen, H. Yang, et al., Expression of insulin-like growth factor II mRNA-binding protein 3 (IMP3) in sacral chordoma, J. Neurooncol 116 (2014) 77e82. [25] C. Chen, H.L. Yang, K.W. Chen, et al., High expression of survivin in sacral chordoma, Med. Oncol. 30 (2013) 529. [26] W. Han, M. Ming, T.C. He, et al., Immunosuppressive cyclosporin A activates AKT in keratinocytes through PTEN suppression: implications in skin carcinogenesis, J. Biol. Chem. 285 (2010) 11369e11377. [27] M.X. Zou, G.H. Lv, J. Li, et al., Upregulated human telomerase reverse transcriptase (hTERT) expression is associated with spinal chordoma growth, invasion and poor prognosis, Am. J. Transl. Res. 8 (2016) 516e529. [28] Z.H. Liu, J.L. Hu, J.Z. Liang, et al., Far upstream element-binding protein 1 is a prognostic biomarker and promotes nasopharyngeal carcinoma progression, Cell Death Dis. 6 (2015) e1920. [29] C. Dong, T. Yuan, Y. Wu, et al., Loss of FBP1 by Snail-mediated repression provides metabolic advantages in basal-like breast cancer, Cancer Cell 23 (2013) 316e331. [30] C. Bettegowda, N. Agrawal, Y. Jiao, et al., Mutations in CIC and FUBP1 contribute to human oligodendroglioma, Science 333 (2011) 1453e1455.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis Hai Wen a, Hong Ma a, Pengzhi Li a, Jiaoyun Zheng b, Yipin Yu a, Guohua Lv a, * a b
Department of Spine Surgery, The Second Xiangya Hospital of Central South University, Changsha 410011, China Department of Pathology, The Second Xiangya Hospital of Central South University, Changsha 410011, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 July 2017 Accepted 1 August 2017 Available online xxx
The far upstream element (FUSE)-binding protein 1 (FUBP1), a well-known transcriptional regulator of the proto-oncogene c-Myc, has been demonstrated by previous work to be aberrantly expressed in a variety of tumors and plays a critical role in tumor progression; however, its expression and function in relatively rare and aggressive chordomas remains unclear. In this retrospective study, we reviewed clinicopathologic characteristics of 40 patients diagnosed with sacral chordoma, and analyzed 40 tumor and 20 distant normal tissues obtained from patients during the primary surgical tumor excision. Using immunohistochemistry, we observed an up-regulation in the expression of FUBP1 and c-Myc in sacral chordomas compared with the normal tissues (P ¼ 0.001 for both). Additionally, positive correlations of FUBP1 expression with c-Myc (g ¼ 0.651, P < 0.001) and the cell proliferation index Ki-67 expression (g ¼ 0.447, P ¼ 0.004) were indicated using Spearman's rank correlation coefficient. Increased expression of FUBP1 was significantly associated with tumor invasion into the surrounding muscles (P ¼ 0.002). Kaplan-Meier curves demonstrated the association between FUBP1 levels and the patients' local recurrence-free survival (LRFS) (P < 0.001) but not with the overall survival (OS) (P ¼ 0.070). The independent prognostic significance of FUBP1 levels for the LRFS was indicated by multivariate analysis (HR ¼ 4.272; 95% CI, 1.133e16.112; P ¼ 0.032). Our findings demonstrate an association between FUBP1 levels and chordoma progression and prognosis, suggesting that FUBP1 can be used as a biomarker and a potential therapeutic target. © 2017 Elsevier Inc. All rights reserved.
Keywords: Sacral chordomas FUBP1 c-Myc Biomarker Recurrence
1. Introduction Chordomas are rare bony neoplasms that account for 1%e4% of all primary malignant bone tumors originating from embryonic notochordal remnants along the cerebrospinal axis [1e3]. Slowgrowing sacral chordomas, constituting approximately 50%e60% of all chordomas, present indolent symptoms, which often results in delayed diagnosis and treatment [4,5]. Because of the poor sensitivity of chordomas to chemotherapy and conventional radiotherapy, surgical excision remains the mainstay of treatment [6e9]; however, in patients with the late stage disease, the presence of inadequate resection margins significantly increases the risk of local tumor recurrence, influencing the patients' quality of
* Corresponding author. E-mail address: [email protected] (G. Lv).
life after surgery [7,10]. Although it is often technically difficult, achieving wide resection margins is the most important factor determining tumor recurrence and long-term prognosis [5,11,12]. Because the molecular mechanisms underlying tumorigenesis and progression of chordomas are not well-understood, it is crucial to identify effective biomarkers that can predict clinical prognosis and provide potential therapeutic targets. The far upstream element (FUSE)-binding protein 1 (FUBP1), one of a three-member family, was identified as a DNA-binding protein from undifferentiated HL60 cells; using four K-homology domains, FUBP1 binds to single-stranded FUSE located upstream of the proto-oncogene c-Myc promoter and regulates its transcription [13,14]. In recent years, it has been shown that the aberrant expression of FUBP1 is often correlated with an abnormal expression of c-Myc, and it was demonstrated in a growing number of malignancies; increasing studies have focused on investigating the conceivable roles of FUBP1 in tumor initiation and progression
http://dx.doi.org/10.1016/j.bbrc.2017.08.008 0006-291X/© 2017 Elsevier Inc. All rights reserved.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
2
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
because of its effect on cellular activities including proliferation, migration and apoptosis of tumor cells [15e17]. For example, it was revealed that overexpression of FUBP1 in human hepatocellular carcinoma (HCC) is associated with poor patient survival and increased cell proliferation in vitro [15]. It was also reported that patients with glioma have unfavorable prognosis when the levels of FUBP1 and c-Myc are elevated, and down-regulation of FUBP1 reduces c-Myc expression and arrests cell growth [18]. These results suggest that FUBP1 can be a new cancer biomarker and a potential treatment target for many human tumors. Although a significant up-regulation in the expression of the mRNAs of FUBP1 and Myc in chordomas compared with notochord tissues has been identified using microarrays with hierarchical clustering analysis previously [19], the evidence of the expression and function of FUBP1 and c-Myc in chordomas is very limited. We carried out this study to investigate the expression of the proteins of FUBP1 and c-Myc in sacral chordomas and assess the association between their expression and the clinicopathologic characteristics and prognosis of patients with sacral chordoma. 2. Materials and methods 2.1. Patients and tissue samples A total of 40 patients diagnosed with sacral chordoma who had received surgical treatment in the Department of Spine Surgery, The Second Xiangya Hospital of Central South University (Changsha, China) were enrolled in this study between January 2005 and October 2015. There were 23 males and 17 females with a mean age of 55.1 years old (range, 29e74 years old). The patients had not received any tumor-specific therapy, including radiotherapy and chemotherapy, prior to surgical treatment. Clinicopathologic parameters including age and gender of the patients; tumor location, size, grade, and stage; tumor invasion of the surrounding muscles; and the type of surgical resection performed were obtained from patients' medical records, retrospectively. Tumor grade and stage were evaluated based on the Enneking staging system for the surgical staging of malignant bone and soft tissue tumors [20e22]. Resected specimens were assessed either as Enneking appropriate (EA) or Enneking inappropriate (EI) by anatomic pathologists according to the Enneking principles [23]. Surrounding muscle invasion was determined using preoperative magnetic resonance imaging (MRI) and histologic examination of tissues [24]. In addition to tumor tissues, we also collected 20 distant normal tissues at least 3 cm away from the surgical margins as a control [25]. The tissue samples were fixed immediately after surgery in 10% buffered formalin and embedded in paraffin. Subsequently, the paraffin blocks were cut into 4-mm thick slices for immunohistochemical analysis. Tumor diagnosis was confirmed through histologic examination of hematoxylin and eosin-stained sections by two pathologists. The study was approved by the Ethical Committee of The Second Xiangya Hospital, Central South University, and informed consent was obtained from each patient enrolled in the study. 2.2. Follow-up Patients were clinically followed-up with radiographic evaluations every 3 months during the first 2 years, every 6 months for the next 3 years, and annually thereafter until October 2016. Tumor recurrence was diagnosed by analyzing clinical manifestations and imaging results. The interval between the primary surgery and the first incidence of local tumor recurrence or death from any cause was defined as the local recurrence-free survival (LRFS) or the overall survival (OS), respectively.
2.3. Immunohistochemistry Tissue sections were deparaffinized in xylene and rehydrated by passage through a series of ethanol solutions; subsequently, the tissues were immersed in 0.01 M citrate buffer (pH 6.0) and heated in a microwave oven at 121 C for 15 min to retrieve antigen activity. Potential endogenous peroxidase activity was quenched by submerging the samples in 3% hydrogen peroxide solution for 10 min. After rinsing with phosphate-buffered saline (PBS), tissue sections were blocked with 10% normal goat serum at room temperature for 10 min and were then incubated overnight at 4 C with primary antibodies anti-FUBP1 (1:100; Abcam), anti-c-Myc (1: 50; Abcam) or anti-Ki-67 (1:100; Abcam). The negative controls were not incubated with primary antibodies. Subsequently, the sections were rinsed with PBS three times, followed by a 30 min incubation at 37 C with horseradish peroxidase-conjugated anti-rabbit IgG (Boster, Wuhan, China). Antibody binding was visualized using 3,30 diaminobenzidine (DAB) solution. The tissue sections were counterstained with hematoxylin and were then mounted for observation. 2.4. Immunohistochemical evaluation The stained sections were evaluated independently by two experienced pathologists who were blinded to the patients' clinicopathologic information and were scored using a semiquantitative assessment reported previously [26]. Five highpower fields were randomly selected, and at least 300 cells per field were evaluated in each section. Cells with immunostained brown nucleus were considered positive for FUBP1, c-Myc or Ki-67 expression. Based on the method described by Zou et al. [27], we obtained the final score for each section by multiplying the staining intensity, established as 0 (no staining), 1 (weak staining), 2 (moderate staining) or 3 (strong staining), by the ratio of immunostained cells to the total number of cells evaluated; subsequently, we evaluated the mean value of all final scores for FUBP1 expression in tumor tissues as the cut-off value to distinguish between the high and low levels of FUBP1 expression for survival analysis. The same method was used to distinguish between the high and low levels of c-Myc expression. As described previously [24], Ki-67 index was defined as the percentage of Ki-67-positive cells and categorized as a low (<10%) or high (10%) index. 2.5. Statistical analysis All statistical analyses were conducted with SPSS 19.0 (SPSS Inc., Chicago, IL, USA). The difference in FUBP1 and c-Myc expression between the tumor and normal tissues and the association between FUBP1 expression and the clinicopathologic characteristics were examined using the c2 or the Fisher's exact test, as appropriate. Spearman's rank correlation coefficient was used to determine the strength of the paired correlation between FUBP1 and cMyc levels and Ki-67 index. A log-rank test was used to analyze the LRFS and OS; the data are shown as Kaplan-Meier curves. Multivariate analysis was subsequently performed by Cox's proportional hazard model to determine whether a statistically significant factor could independently affect the patients' prognosis. A P-value < 0.05 was considered statistically significant. 3. Results 3.1. Up-regulation of FUBP1 expression and its positive correlation with c-Myc expression and Ki-67 index in sacral chordomas Immunohistochemical analysis was used to investigate the
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
3
Fig. 1. Representative expression of FUBP1, c-Myc and Ki-67 in sacral chordoma tissues. (A) Low expression of FUBP1; (B) low expression of c-Myc; (C) low expression of Ki-67; (D) high expression of FUBP1; (E) high expression of c-Myc; (F) high expression of Ki-67; (G) negative control of FUBP1; (H) negative control of c-Myc; (I) negative control of Ki-67. All magnification, 400.
differential expression of FUBP1 and c-Myc in 40 tumor and 20 distant normal tissues from patients with sacral chordoma. The results indicate that the proteins were predominantly expressed in the nucleus of positive cells (Fig. 1). The mean scores for FUBP1 and c-Myc expression in tumor tissues were 1.174 ± 0.097 and 1.189 ± 0.091, which were considered as the cut-off values, respectively. Higher expression of FUBP1 and c-Myc was observed in 60% (24/40; Fig. 1D) and 55% (22/40; Fig. 1E) of tumor tissues, respectively; whereas only 15% (3/20) and 10% (2/20) of distant normal tissues expressed higher levels of FUBP1 and cMyc, respectively. Expression of FUBP1 and c-Myc was significantly higher in tumor tissues compared with the distant normal tissues (P ¼ 0.001 for both; Table 1). Spearman's correlation test was used to assess the correlation between FUBP1 and c-Myc expression levels. As shown in Fig. 2A, FUBP1 expression is positively correlated with c-Myc expression (Spearman's g ¼ 0.651, P < 0.001). We also examined the expression of Ki-67, a cell proliferation indicator, in sacral chordomas. Based on Ki-67 expression levels, tissues were divided into low expressing (16 samples; Fig. 1C) or high expressing tissues (24 samples; Fig. 1F). Expression of FUBP1 was significantly higher in tumor tissues with a high Ki-67 index than those with a low Ki-67 index (P < 0.001; Table 2). Additionally, the Spearman's rank correlation coefficient for FUBP1 expression and Ki-67 index is 0.447 (P ¼ 0.004; Fig. 2B). A positive correlation between c-Myc expression and Ki-67 index was also observed (Spearman's g ¼ 0.574, P < 0.001; Fig. 2C).
3.2. Association between FUBP1 expression and the clinicopathologic characteristics of patients with sacral chordoma To assess the association between FUBP1 expression and patients' clinicopathologic parameters, the patients were divided into two groups based on the FUBP1 expression levels. Consistent with the results described before, positive correlations of FUBP1 expression with c-Myc expression and Ki-67 index were confirmed by the c2 test (P < 0.001 for both; Table 2). Furthermore, FUBP1 expression was significantly associated with tumor invasion into the surrounding muscles (P ¼ 0.002). No statistically significant association was found between FUBP1 expression and other clinicopathologic characteristics. 3.3. Prognostic significance of FUBP1 expression in sacral chordomas After the primary tumor resection, patients were followed-up until October 2016. Of the total, 27 (67.5%) had a local tumor relapse, and 16 (40%) died during the follow-up period, with a median LRFS of 18 months (range, 5e45 months) and a median OS of 47 months (range, 12e141 months). The results of the KaplanMeier analysis demonstrate that high expression of either FUBP1 or c-Myc is significantly associated with a shorter LRFS time (P < 0.001 for both; Table 3 and Fig. 3A and B). Patients with low FUBP1/c-Myc coexpression had an increased LRFS time (P < 0.001; Fig. 3C), whereas those with high FUBP1/c-Myc coexpression had a
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
4
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
Table 1 Expression of FUBP1 and c-Myc in sacral chordomas compared with distant normal tissues. Tissue samples
Sacral chordomas Distant normal tissues
Total
40 20
FUBP1 expression
P-value
High (%)
Low (%)
24 (60%) 3 (15%)
16 (40%) 17 (85%)
0.001
c-Myc expression
P-value
High (%)
Low (%)
22 (55%) 2 (10%)
18 (45%) 18 (90%)
0.001
Fig. 2. Graphic representation of paired correlation between FUBP1 and c-Myc levels and Ki-67 index in sacral chordomas using Spearman's rank correlation test. (A) Positive correlation between FUBP1 and c-Myc expression; (B) positive correlation between Ki-67 index and FUBP1 expression; (C) positive correlation between Ki-67 index and c-Myc expression.
reduced LRFS time (P < 0.001; Fig. 3D). Shorter LRFS time was also observed in patients with higher Ki-67 index (P ¼ 0.016; Table 3 and Fig. 3E), tumor invasion of the surrounding muscles (P < 0.001; Table 3 and Fig. 3F), higher tumor stage (P < 0.001; Table 3 and Fig. 3G) and Enneking inappropriate resection (P < 0.001; Table 3 and Fig. 3H). Furthermore, shorter OS time was related to factors including higher c-Myc expression (P ¼ 0.011; Table 4 and Fig. 4A), high FUBP1/c-Myc coexpression (P ¼ 0.002; Table 4 and Fig. 4B), tumor invasion of the surrounding muscles (P ¼ 0.032; Table 4 and Fig. 4C), higher tumor stage (P < 0.001; Table 4 and Fig. 4D), and Enneking inappropriate resection (P ¼ 0.015; Table 4 and Fig. 4E). Lastly, the variables associated with LRFS and OS obtained from the Kaplan-Meier analysis were included in the multivariate analysis
using Cox's proportional hazard model. The results indicate that FUBP1 expression (HR ¼ 4.272; 95% CI, 1.133e16.112; P ¼ 0.032; Table 3), tumor stage (HR ¼ 1.872; 95% CI, 1.168e3.000; P ¼ 0.009; Table 3), tumor invasion of the surrounding muscles (HR ¼ 18.398; 95% CI, 3.454e98.001; P ¼ 0.001; Table 3), and the type of tumor resection (HR ¼ 8.025; 95% CI, 2.251e28.607; P ¼ 0.001; Table 3) were the independent prognostic factors for the LRFS, whereas tumor stage was the only independent prognostic factor for the OS (HR ¼ 1.749; 95% CI, 1.043e2.933; P ¼ 0.034; Table 4). 4. Discussion In this study, we observed an up-regulation of FUBP1
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8 Table 2 Association between FUBP1 expression and the clinicopathologic characteristics of patients with sacral chordoma. Variables
Age (years) < 50 50 Gender Male Female Tumor location Above S3 S3 and below Tumor size (cm) <5 5 Tumor grade High Low Tumor stage ⅠA ⅠB ⅡA ⅡB Ⅲ Surrounding muscle invasion Yes No Type of resection EI EA c-Myc expression High Low Ki-67 index High Low
Total
FUBP1 expression
P-value
High
Low
15 25
10 14
5 11
23 17
13 11
10 6
24 16
14 10
10 6
14 26
8 16
6 10
27 13
18 6
9 7
9 4 4 20 3
3 3 1 14 3
6 1 3 6 0
25 15
20 4
5 11
18 22
14 10
4 12
22 18
19 5
3 13
24 16
20 4
4 12
0.740
0.747
1.000
1.000
0.305
0.095*
0.002
0.054
<0.001
<0.001
EI, Enneking inappropriate; EA, Enneking appropriate. *, using the Fisher's exact test.
expression, which was positively correlated with c-Myc expression in sacral chordoma tissues. Increased expression of FUBP1 was significantly associated with a relatively high Ki-67 index, tumor invasion of the surrounding muscles, and poor LRFS. Moreover, FUBP1 has an independent prognostic significance for patients' LRFS. These results suggest the involvement of FUBP1 in tumor progression and its usefulness as a prognostic indicator of patients with sacral chordoma. Because of histologic specificity, the exact function of FUBP1 in
5
the molecular mechanisms of different tumors is unclear; for example, FUBP1 is overexpressed in nasopharyngeal carcinoma and ccRCC, where its knockdown inhibits cell proliferation [17,28], whereas the loss of FUBP1 is an important oncogenic event in oligodendroglioma and basal-like breast cancer [29,30]. In this study, elevated FUBP1 expression was observed in sacral chordomas compared with the normal tissues, suggesting that FUBP1 may promote tumor development in sacral chordomas. FUBP1 was discovered in HL60 cells as a transcriptional regulator of the wellknown proto-oncogene c-Myc whose activation causes malignant transformation [13]. Consistent with the evidence from other cancers, a positive correlation was demonstrated between FUBP1 and c-Myc expression in sacral chordomas, supporting the involvement of the c-Myc-dependent pathway in FUBP1-mediated carcinogenesis. In this study, we determined that up-regulation of FUBP1 expression in sacral chordomas is related to high levels of Ki67, a widely used cell proliferation indicator, suggesting the correlation between FUBP1 and cell proliferation. Based on our findings, we propose that FUBP1 could regulate proliferation of chordoma cells through a c-Myc-dependent pathway; the mechanisms of regulation, however, need to be further explored. The association of FUBP1 with the clinicopathologic characteristics and prognosis of patients with cancer has been frequently reported in recent years. Malz et al. demonstrated that cooverexpression of FUBP1 and FUBP2 was significantly associated with the poor survival of patients with HCC [15]. Ding et al. showed a close connection between FUBP1 expression and tumor stage, and the accumulative survival of patients with human glioma [18]. We observed a significant association between high FUBP1 levels and invasion of the surrounding muscles by sacral chordomas, which supports the importance of FUBP1 in chordoma progression. However, we did not observe an association between FUBP1 expression and tumor size, grade or stage, which could be due to a small sample size, relatively low statistical power or due to the influence of other risk factors. Because patients with sacral chordoma have a high rate of local tumor recurrence, it is critical to identify the related risk factors for an effective prediction of tumor prognosis. Our data show that shorter LRFS time is closely associated with high expression levels of FUBP1 and c-Myc, and certain clinicopathologic factors, including tumor stage, tumor invasion of the surrounding muscles and the resection type. The independent prognostic significance of FUBP1 in sacral chordomas was verified using a multivariate analysis to eliminate the influence of confounding factors. Although patients with sacral chordoma who had high FUBP1 levels did not show reduced OS in our study, more
Table 3 Kaplan-Meier analysis and multivariate Cox's proportional hazard analysis of the prognostic factors for the local recurrence-free survival of patients with sacral chordoma. Variables
Categories
Kaplan-Meier analysis
Multivariate analysis
P-value
P-value
HR (95% CI)
Age (years) Gender Tumor location Tumor size (cm) Tumor grade Tumor stage Surrounding muscle invasion Type of resection Ki-67 index FUBP1 expression c-Myc expression FUBP1/c-Myc coexpression
<50/ 50 Male/Female Above S3/S3 and below <5/ 5 High/Low ⅠA/ⅠB/ⅡA/ⅡB/Ⅲ Yes/No EI/EA High/Low High/Low High/Low Low/Low High/High High/Low and Low/High
0.671 0.587 0.091 0.124 0.595 <0.001 <0.001 <0.001 0.016 <0.001 <0.001 <0.001* <0.001*
0.009 0.001 0.001 0.139 0.032 0.081
1.872 (1.168e3.000) 18.398 (3.454e98.001) 8.025 (2.251e28.607) 0.221 (0.030e1.631) 4.272 (1.133e16.112) 5.634 (0.808e39.275)
EI, Enneking inappropriate; EA, Enneking appropriate. *, not included in the multivariate analysis, because the variables depend on the expression of FUBP1 and c-Myc.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
6 H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
Fig. 3. Kaplan-Meier curves of the local recurrence-free survival (LRFS) of patients with sacral chordoma. (A) Stratified by FUBP1 expression; (B) stratified by c-Myc expression; (C) stratified by FUBP1/c-Myc coexpression (low/low and others); (D) stratified by FUBP1/c-Myc coexpression (high/high and others); (E) stratified by Ki-67 index; (F) stratified by surrounding muscle invasion; (G) stratified by tumor stage; (H) stratified by type of resection.
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
7
Table 4 Kaplan-Meier analysis and multivariate Cox's proportional hazard analysis of the prognostic factors for the overall survival of patients with sacral chordoma. Variables
Age (years) Gender Tumor location Tumor size (cm) Tumor grade Tumor stage Surrounding muscle invasion Type of resection Ki-67 index FUBP1 expression c-Myc expression FUBP1/c-Myc coexpression
Categories
<50/ 50 Male/Female Above S3/S3 and below <5/ 5 High/Low ⅠA/ⅠB/ⅡA/ⅡB/Ⅲ Yes/No EI/EA High/Low High/Low High/Low Low/Low High/High High/Low and Low/High
Kaplan-Meier analysis
Multivariate analysis
P-value
P-value
HR (95% CI)
0.362 0.319 0.075 0.065 0.468 <0.001 0.032 0.015 0.628 0.070 0.011 0.166 0.002*
0.034 0.099 0.511
1.749 (1.043e2.933) 5.866 (0.719e47.862) 1.503 (0.446e5.063)
0.110
3.617 (0.748e17.483)
EI, Enneking inappropriate; EA, Enneking appropriate. *, not included in the multivariate analysis because it depends on the expression of FUBP1 and c-Myc.
Fig. 4. Kaplan-Meier curves of the overall survival (OS) of patients with sacral chordoma. (A) Stratified by c-Myc expression; (B) stratified by FUBP1/c-Myc coexpression (high/high and others); (C) stratified by surrounding muscle invasion; (D) stratified by tumor stage; (E) stratified by type of resection.
evidence is needed to rule out a relationship between FUBP1 expression and the OS. Furthermore, despite that c-Myc is not an independent predictive factor for the LRFS and OS of patients with sacral chordoma indicated by this study, the significance of c-Myc in tumorigenesis and development of chordomas cannot be underestimated and needs to be further explored.
In conclusion, our findings demonstrate that up-regulation of FUBP1 expression plays a substantial role in progression and prognosis of sacral chordomas. Although more research is needed, FUBP1 is a strong candidate for use as a new biomarker and a potential therapeutic target.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008
8
H. Wen et al. / Biochemical and Biophysical Research Communications xxx (2017) 1e8
Acknowledgements This work was supported by the Fundamental Research Funds for the Central Universities of Central South University (No. 2016zzts149). Conflict of interest The authors declare that they have no conflict of interest. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2017.08.008. References [1] M.L. McMaster, A.M. Goldstein, C.M. Bromley, et al., Chordoma: incidence and survival patterns in the United States, 1973-1995, Cancer Causes Control 12 (2001) 1e11. [2] J. Llauger, J. Palmer, S. Amores, et al., Primary tumors of the sacrum: diagnostic imaging, AJR Am. J. Roentgenol. 174 (2000) 417e424. [3] S. Stacchiotti, J. Sommer, Chordoma Global Consensus Group, Building a global consensus approach to chordoma: a position paper from the medical and patient community, Lancet Oncol. 16 (2015) e71e83. [4] R. Chugh, H. Tawbi, D.R. Lucas, et al., Chordoma: the nonsarcoma primary bone tumor, Oncologist 12 (2007) 1344e1350. [5] C.A. Hulen, H.T. Temple, W.P. Fox, et al., Oncologic and functional outcome following sacrectomy for sacral chordoma, J. Bone Jt. Surg. Am. 88 (2006) 1532e1539. [6] C. Catton, B. O'Sullivan, R. Bell, et al., Chordoma: long-term follow-up after radical photon irradiation, Radiother. Oncol. 41 (1996) 67e72. [7] E.Y. Cheng, R.A. Ozerdemoglu, E.E. Transfeldt, et al., Lumbosacral chordoma. Prognostic factors and treatment, Spine (Phila Pa 1976) 24 (1999) 1639e1645. [8] A. Azzarelli, V. Quagliuolo, S. Cerasoli, et al., Chordoma: natural history and treatment results in 33 cases, J. Surg. Oncol. 37 (1988) 185e191. [9] J.E. York, A. Kaczaraj, D. Abi-Said, et al., Sacral chordoma: 40-year experience at a major cancer center, Neurosurgery 44 (1999) 74e79 discussion 79e80. [10] D.R. Fourney, Z.L. Gokaslan, Current management of sacral chordoma, Neurosurg. Focus 15 (2003) E9. [11] B. Fuchs, I.D. Dickey, M.J. Yaszemski, et al., Operative management of sacral chordoma, J. Bone Jt. Surg. Am. 87 (2005) 2211e2216. [12] B. Kayani, M.D. Sewell, K.A. Tan, et al., Prognostic factors in the operative management of sacral chordomas, World Neurosurg. 84 (2015) 1354e1361.
[13] M.I. Avigan, B. Strober, D. Levens, A far upstream element stimulates c-myc expression in undifferentiated leukemia cells, J. Biol. Chem. 265 (1990) 18538e18545. [14] R. Duncan, L. Bazar, G. Michelotti, et al., A sequence-specific, single-strand binding protein activates the far upstream element of c-myc and defines a new DNA-binding motif, Genes Dev. 8 (1994) 465e480. [15] M. Malz, A. Weber, S. Singer, et al., Overexpression of far upstream element binding proteins: a mechanism regulating proliferation and migration in liver cancer cells, Hepatology 50 (2009) 1130e1139. [16] F. Zhang, Q. Tian, Y. Wang, Far upstream element-binding protein 1 (FUBP1) is overexpressed in human gastric cancer tissue compared to non-cancerous tissue, Onkologie 36 (2013) 650e655. [17] J. Duan, X. Bao, X. Ma, et al., Upregulation of far upstream element-binding protein 1 (FUBP1) promotes tumor proliferation and tumorigenesis of clear cell renal cell carcinoma, PLoS One 12 (2017) e0169852. [18] Z. Ding, X. Liu, Y. Liu, et al., Expression of far upstream element (FUSE) binding protein 1 in human glioma is correlated with c-Myc and cell proliferation, Mol. Carcinog. 54 (2015) 405e415. [19] C. Long, L. Jiang, F. Wei, et al., Integrated miRNA-mRNA analysis revealing the potential roles of miRNAs in chordomas, PLoS One 8 (2013) e66676. [20] I.R. Samson, D.S. Springfield, H.D. Suit, et al., Operative treatment of sacrococcygeal chordoma. A review of twenty-one cases, J. Bone Jt. Surg. Am. 75 (1993) 1476e1484. [21] B. Kayani, S.A. Hanna, M.D. Sewell, et al., A review of the surgical management of sacral chordoma, Eur. J. Surg. Oncol. 40 (2014) 1412e1420. [22] A. Jemal, R. Siegel, E. Ward, et al., Cancer statistics, 2007, CA Cancer J. Clin. 57 (2007) 43e66. [23] C.G. Fisher, D.D. Saravanja, M.F. Dvorak, et al., Surgical management of primary bone tumors of the spine: validation of an approach to enhance cure and reduce local recurrence, Spine (Phila Pa 1976) 36 (2011) 830e836. [24] M. Zhou, K. Chen, H. Yang, et al., Expression of insulin-like growth factor II mRNA-binding protein 3 (IMP3) in sacral chordoma, J. Neurooncol 116 (2014) 77e82. [25] C. Chen, H.L. Yang, K.W. Chen, et al., High expression of survivin in sacral chordoma, Med. Oncol. 30 (2013) 529. [26] W. Han, M. Ming, T.C. He, et al., Immunosuppressive cyclosporin A activates AKT in keratinocytes through PTEN suppression: implications in skin carcinogenesis, J. Biol. Chem. 285 (2010) 11369e11377. [27] M.X. Zou, G.H. Lv, J. Li, et al., Upregulated human telomerase reverse transcriptase (hTERT) expression is associated with spinal chordoma growth, invasion and poor prognosis, Am. J. Transl. Res. 8 (2016) 516e529. [28] Z.H. Liu, J.L. Hu, J.Z. Liang, et al., Far upstream element-binding protein 1 is a prognostic biomarker and promotes nasopharyngeal carcinoma progression, Cell Death Dis. 6 (2015) e1920. [29] C. Dong, T. Yuan, Y. Wu, et al., Loss of FBP1 by Snail-mediated repression provides metabolic advantages in basal-like breast cancer, Cancer Cell 23 (2013) 316e331. [30] C. Bettegowda, N. Agrawal, Y. Jiao, et al., Mutations in CIC and FUBP1 contribute to human oligodendroglioma, Science 333 (2011) 1453e1455.
Please cite this article in press as: H. Wen, et al., Expression of far upstream element-binding protein 1 correlates with c-Myc expression in sacral chordomas and is associated with tumor progression and poor prognosis, Biochemical and Biophysical Research Communications (2017), http:// dx.doi.org/10.1016/j.bbrc.2017.08.008