Expression and prognostic value of soluble CD97 and its ligand CD55 in intrahepatic cholangiocarcinoma

Abstract

The incidence rate of intrahepatic cholangiocarcinoma is rising, and treatment options are limited. Therefore, new biological markers of intrahepatic cholangiocarcinoma are needed. Immunohistochemistry and enzyme-linked immunosorbent assay were applied to analyze the expressions of CD97, CD55, and soluble CD97 in 71 patients with intrahepatic cholangiocarcinoma and 10 patients with hepatolithiasis. CD97 and CD55 were not expressed in hepatolithiatic tissues, but positive expression was observed in 76.1% (54/71) and 70.4% (50/71) of intrahepatic cholangiocarcinoma patients. The univariate analyses indicated that the positive expressions of CD97 and CD55 were related to short intrahepatic cholangiocarcinoma survival of patients (both p = 0.001). Furthermore, CD97 and CD55 expressions and biliary soluble CD97 levels were significantly associated with histological grade (p = 0.004, 0.002, and 0.012, respectively), lymph node metastases (p = 0.020, 0.038, and 0.001, respectively), and venous invasion (p = 0.003, 0.002, and 0.001, respectively). The multivariate analyses indicated that lymph node metastases (hazard ratio: 2.407, p = 0.003), positive CD55 expression (hazard ratio: 4.096, p = 0.003), and biliary soluble CD97 levels (hazard ratio: 2.434, p = 0.002) were independent risk factors for the intrahepatic cholangiocarcinoma survival. The receiver operating characteristic (ROC) curve analysis indicated that when the cutoff values of biliary soluble CD97 were 1.15 U/mL, the diagnostic value for predicting lymph node metastasis had a sensitivity of 87.5% and a specificity of 51.3%. For intrahepatic cholangiocarcinoma patient death within 60 months at a cutoff value of 0.940 U/mL, the diagnostic value sensitivity was 89.3% and the specificity was 93.3%. Biliary soluble CD97 may be a new biological marker for early diagnosis, prediction of lymph node metastasis and poor prognosis, and discovery of a therapeutic target.

Introduction

The incidence of intrahepatic cholangiocarcinoma (ICC) has increased in Europe and North America, and ICC is the second most popular cancer after hepatocellular carcinoma.¹ ICC is refractory to existing chemotherapy and radiotherapy regimes. Hepatectomy is the sole potentially medicable option for this solid tumor, but most patients are in high risk with local recurrence or metastatic disease after curative resection.² New insights into the biological process of ICC and identification of novel biomarkers are essential for cancer management and treatment because of the increasing number of affected patients and the restricted treatment options.

CD97 is an important member of the epidermal growth factor 7-transmembrane (EGF-TM7) protein family for adhesive properties³ that is related to aggressiveness and metastasis in oral squamous cell carcinomas,⁴ colorectal cancer,⁵ thyroid tumors,⁶ gastric carcinoma,⁷ pancreatic cancer,⁸ and primary gallbladder carcinoma.⁹ CD97 can bind to its ligand CD55 (a glycosylphosphatidylinositol-anchored protein) through its EGF domain region, which is characterized by a low attraction, off-rate and is dependent on calcium.¹⁰ The EGF domain can maintain conformational stability to interdomain linkages in combination with calcium, which can be highly resistant to complement activation pathway.¹¹ The activation of complement immune system pathways leads to the formation of a membrane attack complex that results in proteolytic cleavage.¹⁰ These properties of CD97 and CD55 are suitable for cell–extracellular matrix (ECM) or cell–cell interactions, especially those related to adhesion, migration, or invasion. Many researches have indicated that CD97 and CD55 are related to tumor dedifferentiation, invasiveness, migration, and metastasis.^{7–9,12–14} For example, CD97 and CD55 were positively expressed in 69.6% and 65.2%, respectively, of gallbladder carcinomas in 138 patients and were significantly correlated with venous/lymphatic invasion, histological grade, clinical stage, and pathological T stage.⁹ In a sample of 37 human pancreatic cancers, CD97 and CD55 were upregulated and were significantly correlated with vascular invasion and lymph node metastasis.⁸ In a study of rectal adenocarcinoma, patients with higher CD97 and CD55 expression had significantly more recurrent or metastatic disease than patients with lower CD97 and CD55 expression.¹⁵ However, the relevance between CD97 and CD55 expression in ICC with the clinical significance has not been investigated. Moreover, CD97 can exist in a soluble form for the deciduous EGF domain. Many research reported that soluble CD97 (sCD97) was detected in body fluids of inflammatory sites^16,17 but no research reported about sCD97 in body fluids of tumor sites. Therefore, CD97 and CD55 expressions in ICC tissues with sCD97 in ICC bile were researched for the relevance with tumor aggressiveness and prognosis.

Materials and methods

Patients and samples

ICC tissue specimens were got from 71 patients (40 male and 31 female) who underwent radical resection from January 2009 to January 2014 in the Department of Hepatobiliary Surgery of the Affiliated Union Hospital of Fujian Medical University in China. All of the radical resections (R0 resection) were performed by the same hepatobiliary surgical team. All cancer specimens were defined as primary tumors arising from intrahepatic duct and histologically diagnosed as adenocarcinoma. In accordance with the World Health Organization classification, ICC tissue specimens were classified as well (G1), moderately (G2), or poorly differentiated (G3) adenocarcinomas. All of ICC patients included in the study were not treated with radiotherapy or chemotherapy in the follow-up period. Patients were excluded from this study if the direct cause of death was not related to ICC or was due to unexpected events. The data were collected by reviewing a database of medical records. The mean age of the ICC patients was 58 years (range: 33–83 years). The 71 patients routinely underwent follow-up via telephone or outpatient clinic visit. The definition of the follow-up period was from the date of surgery to death from ICC or until 60 months (range: 11–60 months). A total of 15 patients (21.1%) were alive, and 56 patients (78.9%) died from ICC during the follow-up period. Additional surgical intrahepatic bile duct tissue and bile were obtained for comparative analysis from 10 patients who underwent live resection for hepatolithiasis. The bile samples were obtained from a common bile duct using a puncture needle in surgery. After centrifugation of the cells, the supernatant was stored at −70°C until use. This study was approved by the Ethics Committee of the Medical Faculty of Fujian Medical University in China. All specimens were handled and made anonymous according to ethical and legal standards. Written informed consent was obtained from all patients before surgery.

Immunohistochemistry assay for CD97 and CD55 expression

Formalin-fixed, paraffin-embedded, sectioned tissues (4 μm thick) were immunostained using the Labelled Streptavidin Biotin 2 System. The tissue sections underwent heat-induced antigen retrieval at 95°C for 15 min while immersed in citrate buffer (pH 6.0) after deparaffinization and rehydration. Following peroxidase blocking with 3% hydrogen peroxidase for 30 min, the specimens were blocked with phosphate-buffered saline (PBS) containing 5 g/L bovine serum albumin. All specimens were incubated with a 1:200 dilution of rabbit anti-CD97 (ab108368, Abcam, United Kingdom) and a 1:100 dilution of rabbit anti-CD55 (ab133684, Abcam) overnight at 4°C. Then, the specimens were briefly washed in PBS and incubated at room temperature with the anti-rabbit antibody and avidin–biotin peroxidase. The specimens were washed in PBS and developed using a diaminobenzidine solution. After washing with water, the specimens were counterstained with hematoxylin. Negative controls were performed by replacing the primary antibody with nonimmune immunoglobulin (IgG). All of the immunostained sections were evaluated blindly by two independent pathologists. The existence of yellow-brown cytoplasm or membrane staining of tumor cells was defined as the expression of CD97 and CD55. CD97 positive and CD55 positive cells were counted in 10 typical microscopic fields. The intensity of staining was estimated as grades 0 (negative), 1 (weak), 2 (moderate), and 3 (strong), whereas the percentage of positive cells was judged following Remmele and Stegner’s¹⁸ criteria, estimated as 1 (<10%), 2 (10%–50%), 3 (51%–80%), and 4 (>80%). It was counted from the intensity score × percentage score. The total score ranged from 0 to 12. If the total score was more than 4, the samples were defined positive.

Enzyme-linked immunosorbent assay for biliary sCD97

An enzyme-linked immunosorbent assay (ELISA) system was used to survey the level of biliary sCD97. ELISA plates were coated with 5 mg/mL of the antibody (MEM180, Biermann, Germany) that binds to the stalk region of CD97. After washing, undiluted bile fluid was applied to the plates and they were incubated at room temperature for 60 min. Subsequently, the plates were incubated with biotin-conjugated monoclonal antibody CLB-CD97/1,³ a monoclonal antibody that binds to the first EGF domain of CD97. Streptavidin–peroxidase was appended to the plates before tetramethylbenzidine application and optical density measurement at OD450. The results are expressed as arbitrary units (units/mL) of a standard amount of CD97 in duplicate measurements. The detection limit of the ELISA was 0.01 U/mL.

Statistical analysis

All computations were performed by SPSS version 13.0 for Windows (SPSS Inc., IL, USA). The quantitative data were expressed as the mean ± standard deviation (SD) and compared by independent samples t-test for two groups. When the quantitative data were classified for more than two groups, they were compared by one-way analysis of variance (ANOVA). The categorical data were compared by the chi-square test. The definition of the ICC survival was from the date of surgery to death from ICC or until 60 months. The Kaplan–Meier curves were plotted to compare the survival information using a log-rank test. The Cox regression model was used to evaluate the risk factors for survival. The cutoff values of sCD97 were indicated by receiver operating characteristic (ROC) curve for predicting the lymph node metastasis and poor prognosis of ICC. The diagnostic values included the sensitivity, specificity, and area under the curve (AUC). The p-value <0.05 was considered statistically significant.

Results

Relevance of CD97 and CD55 expression with biliary sCD97 levels in ICC

CD97 and CD55 expressions in hepatolithiasis (10) and ICC (71) were investigated. We found that CD97 and CD55 were not expressed in the bile duct of hepatolithiatic tissues but were expressed in 76.1% (54/71) and 70.4% (50/71), respectively, of ICC (Table 1). CD97 immunoreactivity was found in the cytoplasm of tumor cells, and CD55 immunoreactivity was found in the cell membrane of tumor cells, both with different percentages of tumor intensities and staining (Figure 1). CD97 positive expression in ICC was significantly related to that of CD55 (χ² = 26.650, p < 0.001). The sCD97 level in ICC bile was significantly correlated with positive expression of CD97 (t = 10.188, p < 0.001) (Table 2) in ICC tissues but could not be detected in the bile of hepatolithiatic tissues.

Table 1.

Relevance of CD97 and CD55 expressions in ICC tissues.

CD97 expression	CD55 expression		Total
	−	+
−	14	3	17
+	7	47	54
Total	21	50	71

ICC: intrahepatic cholangiocarcinoma.

Figure 1.

Immunohistochemical staining for CD97 and CD55 in tissues (original magnification 400×). CD97 (a) and CD55 (b) staining were negative in hepatolithiatic tissues; (c) weak CD97 positive expression was discovered in the cytoplasm of ICC tissues; (d) strong CD97 positive expression was discovered in the cytoplasm of ICC tissues; (e) weak CD55 positive expression was discovered in the cell membrane of ICC tissues; (f) strong CD55 positive expression was discovered in the cell membrane of ICC tissues.

[Figure omitted. See PDF]

Table 2.

Relevance of CD97 and CD55 expressions with clinicopathological factors.

Factor	Number	CD97 expression, n (%)		p	CD55 expression, n (%)		p
		Negative	Positive		Negative	Positive
Gender
Female	31	6 (19.4)	25 (80.6)	0.605	8 (25.8)	23 (74.2)	0.726
Male	40	11 (27.5)	29 (72.5)		13 (32.5)	27 (67.5)
Age at diagnosis (years)	71	61.59 ± 10.74	57.61 ± 10.94	0.726	60.52 ± 9.37	57.74 ± 11.54	0.333
Blood types
A	23	6 (26.1)	17 (73.9)	0.766	7 (30.4)	16 (69.6)	0.455
B	17	3 (17.6)	14 (82.4)		4 (23.5)	13 (76.5)
AB	5	2 (40.0)	3 (60.0)		3 (60.0)	2 (40.0)
O	26	6 (23.1)	20 (76.9)		7 (26.9)	19 (73.1)
Histological grade
G1	7	5 (71.4)	2 (28.6)	0.004*	5 (71.4)	2 (28.6)	0.002*
G2	44	10 (22.7)	34 (77.3)		15 (34.1)	29 (65.9)
G3	20	2 (10.0)	18 (90.0)		1 (5.0)	19 (95.0)
Lymph node metastases
Negative	39	14 (35.9)	25 (64.1)	0.020*	16 (41.0)	23 (59.0)	0.038*
Positive	32	3 (9.4)	29 (90.6)		5 (15.6)	27 (84.4)
Venous invasion
Negative	43	16 (37.2)	27 (62.8)	0.003*	19 (44.2)	24 (55.8)	0.002*
Positive	28	1 (3.6)	27 (96.4)		2 (7.1)	26 (92.9)
Perineural invasion
Negative	41	9 (22.0)	32 (78.0)	0.858	12 (29.3)	29 (70.7)	1.000
Positive	30	8 (26.7)	22 (73.3)		9 (30.0)	21 (70.0)
Tumor size
<3 cm	20	6 (30.0)	14 (70.0)	0.454	6 (30.0)	14 (70.0)	1.000
>3 cm	51	11 (21.6)	40 (78.4)		15 (29.4)	36 (70.6)
Tumor nodule
Solitary	48	12 (25.0)	36 (75.0)	0.997	14 (29.2)	34 (70.8)	1.000
Multiple	23	5 (21.7)	18 (78.3)		7 (30.4)	16 (69.6)
Biliary sCD97 levels	71	0.48 ± 0.19	1.65 ± 0.46	0.001*	0.69 ± 0.51	1.66 ± 0.46	0.728

sCD97: soluble CD97.

*

p < 0.05.

Relevance of CD97 and CD55 expression and biliary sCD97 level with clinicopathological factors

We analyzed the relevance between CD97 and CD55 expression and different clinicopathological factors. The results are summarized in Table 2. The expressions of CD97 and CD55 were significantly related to histological differentiation (χ² = 10.838 and 12.117, respectively; p = 0.004 and 0.002, respectively), lymph node metastases (χ² = 5.411 and 4.293, respectively; p = 0.020 and 0.038, respectively), and venous invasion (χ² = 8.770 and 9.464, respectively; p = 0.003 and 0.002, respectively), showing a relevance between their expression and the aggressiveness of ICC. Moreover, other clinicopathological factors such as gender, age, blood type, perineural invasion, tumor size, and number of tumor nodules were not significantly associated with CD97 and CD55 expression. We also analyzed the relevance between the mean (±SD) sCD97 level and various clinicopathological factors (Table 3). The sCD97 level was also significantly related to histological grade (F = 4.687, p = 0.012), lymph node metastases (t = 3.486, p = 0.001), and venous invasion (t = 3.622, p = 0.001).

Table 3.

Relevance of biliary sCD97 level with conventional clinicopathological factors.

Factor	Biliary sCD97 levels (U/mL)a	p
Gender
Female	1.43 ± 0.60	0.535
Male	1.33 ± 0.68
Age at diagnosis (years)
<50	1.45 ± 0.68	0.881
50–60	1.42 ± 0.69
60–70	1.33 ± 0.63
>70	1.26 ± 0.66
Blood types
A	1.43 ± 0.69	0.955
B	1.36 ± 0.64
AB	1.27 ± 0.90
O	1.35 ± 0.60
Histological grade
G1	0.88 ± 0.81	0.012*
G2	1.32 ± 0.61
G3	1.67 ± 0.55
Lymph node metastases
Negative	1.15 ± 0.66	0.001*
Positive	1.65 ± 0.52
Venous invasion
Negative	1.16 ± 0.66	0.001*
Positive	1.69 ± 0.48
Perineural invasion
Negative	1.40 ± 0.66
Positive	1.34 ± 0.64	0.686
Tumor size
<3 cm	1.20 ± 0.69	0.175
>3 cm	1.44 ± 0.63
Tumor nodule
Solitary	1.32 ± 0.64	0.333
Multiple	1.48 ± 0.66

a

Presented by mean ± standard deviation.

*

p < 0.05.

Relevance of CD97 and CD55 expression and biliary sCD97 level with survival

We performed a survival analysis of 71 patients with ICC using information available from clinical follow-up. The median survival time was 32 months (range: 11–60 months). The 5-year survival rate of the ICC patients was 17.4%. The 71 patients were divided into positive CD97 expression and positive CD55 expression groups. The 5-year survival rates of the CD97 negative and CD97 positive groups were 70.8% and 1.9%, respectively (p < 0.001, Figure 2(a)) and 59.1% and 2.0% for the CD55 negative and CD55 positive groups, respectively (p < 0.001, Figure 2(b)). The Kaplan–Meier analyses showed the statistical significance of different CD97 and CD55 expression groups on clinical outcome using a log-rank test. The survival curves are shown in Figure 2.

Figure 2.

Relevance of CD97 (a) and CD55 (b) expression with the OS rate in 71 patients with ICC (compared using a log-rank test).

*p < 0.05.

[Figure omitted. See PDF]

According to the univariate Cox regression model analysis shown in Table 4, histological grade (p = 0.001), lymph node metastases (p = 0.001), venous invasion (p = 0.001), positive CD97 expression (p = 0.001), CD55 expression (p = 0.001), and biliary sCD97 level (p = 0.001) were statistically significant risk factors affecting the ICC survival of patients. To indicate the independent prognostic factors, multivariate analyses were performed by using the Cox proportional hazard model with forward selection. The results indicated that lymph node metastases (hazard ratio (HR): 2.407, p = 0.003), positive CD55 expression (HR: 4.096, p = 0.003), and biliary sCD97 level (HR: 2.434, p = 0.002) were independent risk factors for the ICC survival (Table 4).

Table 4.

Univariate and multivariate analysis for the relevance of prognosis with various clinicopathological factors.

Factor	Univariate analysis p	Multivariate analysis (Cox with forward selection)
		Hazard ratio (HR)	95% confidence interval (CI)	p
Gender	0.077
Age at diagnosis	0.736
Blood types	0.962
Histological grade	0.001*
Lymph node metastases	0.001*	2.407	1.360–4.258	0.003*
Venous invasion	0.001*
Perineural invasion	0.215
Tumor size	0.571
Tumor nodule	0.672
Biliary sCD97 levels	0.001*	2.434	1.390–4.264	0.002*
CD97 expression	0.001*
CD55 expression	0.001*	4.096	1.639–10.237	0.003*

sCD97: soluble CD97.

*

p < 0.05.

ROC curve analysis of sCD97 in bile

We used the following definition to discover the diagnostic values: it was identified as positive if the sCD97 level was higher than the cutoff. When the sCD97 cutoff for predicting the lymph node metastasis of ICC patients was 1.15 U/mL, the diagnostic value had a sensitivity of 87.5%, specificity of 51.3%, and AUC of 0.722 (Figure 3(a)). For ICC patient death within 60 months, when the cutoff of 0.940 U/mL was selected, the sCD97 diagnostic value had a sensitivity of 89.3%, specificity of 93.3%, and AUC of 0.933 (Figure 3(b)).

Figure 3.

ROC curve analysis for predicting lymph node metastasis in ICC (cutoff value 1.15 U/mL) (a) and ICC patient death within 60 months (cutoff value: 0.940 U/mL) (b) using biliary sCD97.

[Figure omitted. See PDF]

Discussion

The prognosis of patients with ICC includes recurrence and metastasis even after curative resection.¹⁹ The lack of suitable medical treatments for ICC calls for deeper investigation of its tumor biology. Therefore, identification of biological markers related to tumor progression is useful for early diagnosis, metastasis monitoring, and discovery of a therapeutic target.

In this study, we observed that 76.1% and 70.4% of ICCs were immunopositive for CD97 and CD55, respectively (Figure 1), and the CD97 expression pattern was significantly related to CD55 (χ² = 26.650, p < 0.001). As the ligand of CD97, CD55 expression was obviously correlated with that of CD97.¹² The malignant tumor behavior promotion by CD97 may occur because of interactions between CD97 and CD55.³ Our results revealed that the intensity and percentage of CD97 and CD55 expression were higher in ICC tissues than hepatolithiatic tissues, which did not show expression of these markers. High expressions of CD97 and CD55 were significantly related to tumor progression and poor histological grade, lymph node metastases, and positive venous invasion (Table 2) and significantly influenced survival after surgical resection (Table 4). Many research indicated that CD97 and CD55 were upregulated in various types of solid tumors, were related to the aggressiveness of the tumor, and were independent risk factors of the overall survival (OS) in patients^4,7–9,20 except for those with ICC. Our study evaluated the effect of CD97 and CD55 expression on the prognosis of ICC using Kaplan–Meier analyses. The tumor tissues with positive CD97 and CD55 expression were more aggressive than those with negative CD97 and CD55 expression. The ICC survival rates of negative CD97 and CD55 groups were significantly higher than those of CD97 positive and CD55 positive groups (Figure 2). The univariate analyses indicated that positive CD97 and CD55 expressions were statistically significant risk factors affecting the ICC survival of patients (Table 4). In the multivariate analyses, we used the Cox proportional hazard model with forward selection because of the clinicopathological correlation between CD97 and sCD97. The results showed that lymph node metastases, positive CD55 expression, and biliary sCD97 level were independent risk factors for prognosis (Table 4). These results indicate that sCD97 (not CD97) and CD55 are ideal tumor marker candidates for ICC. Moreover, the possible new role of the CD97–CD55 interaction in invasion and metastasis promotion may lead to an immunotherapeutic method for cancer therapy.²¹

sCD97 is a soluble form of CD97 that may result from proteolytic cleavage of the extracellular part of the membrane-expressed receptor, and the coactions of CD97 with CD55 may be the triggering event for release of sCD97,²² which has been observed in body fluids of inflammatory sites^16,17,23 but not in body fluids surrounding tumor sites. In our study, we found that sCD97 can be detected in ICC bile (not in hepatolithiatic bile) and that the sCD97 level in ICC bile was significantly correlated with positive expression of CD97 (t = 10.188, p < 0.001) (Table 2) in ICC tissues. These findings suggest that sCD97 was nonexistent in the noncancerous intrahepatic duct when transitioning into cancerous cells and that tumor cells release high sCD97 levels, thereby inducing the recurrence and metastasis of tumors. sCD97 levels were elevated in bile of many patients with ICC compared with those with hepatolithiasis, indicating that sCD97 is a useful marker for ICC and may be an indicator for diagnosis of ICC.

This study also analyzed the associations between clinicopathological factors and the sCD97 level in ICC. We found that the sCD97 level was related to the histological grade (F = 4.687, p = 0.012), lymph node metastases (t = 3.486, p = 0.001), and positive venous invasion (t = 3.622, p = 0.001) (Table 3), similar to positive CD97 expression in ICC tissue. The univariate and multivariate analyses indicated that the biliary sCD97 level of ICC patients was an independent risk factor for prognosis (Table 4). Previous studies have demonstrated that sCD97 acts as a potent chemoattractant for invasion and migration of human umbilical vein endothelial cells (HUVECs) and can contribute to angiogenesis associated with inflammation and tumor progression.²⁴ More studies are required to validate whether sCD97 in ICC bile can act as an inducement for tumor angiogenesis, which can contribute to lymph node metastasis, positive venous invasion, and poor prognosis.

It has not been reported that the biliary sCD97 level can predict the presence of lymph node metastasis and poor prognosis in various malignancies. In our study, when the sCD97 level cutoff as a marker for lymph node metastasis was 1.15 U/mL, the predicted values had a sensitivity of 87.5% and specificity of 51.3% (Figure 3(a)). Therefore, we suggest that biliary sCD97 level might be a reliable biomarker for lymph node metastasis in ICC. When the sCD97 level cutoff as a marker for ICC patient death within 60 months was 0.94 U/mL, the predicted values had a sensitivity of 89.3% and specificity of 93.3% (Figure 3(b)). These results suggest that biliary sCD97 could be helpful to predict the prognosis of ICC patients, possibly because high biliary sCD97 level in ICC patients increases the risk of positive venous invasion and lymph node metastasis, thereby resulting in poor outcomes. These conjectures must be rigorously tested by further experiments.

To our knowledge, this study is the first to discover the CD97 and CD55 expression and biliary sCD97 level in patients with ICC and their value in diagnosis and prognosis. However, this study has some limitations that need further research. First, it is not easy to obtain bile from patients except through an invasive operation, meaning that this technology cannot be commonly used. Second, postoperative biliary sCD97 level after radical or palliative resection should be detected and compared with the preoperative level to identify the validity of surgical result assessment using sCD97. In addition, larger sample sizes are necessary in future research.

In conclusion, our study indicated that the CD97 and CD55 proteins might be reliable biomarkers to predict the metastasis status and prognosis of ICC patients. The functional role of CD97 and CD55 will be further research in ICC patients by treatment with blocking mAb. Furthermore, biliary sCD97 may be a new biological marker associated with early diagnosis, prediction of lymph node metastasis and poor prognosis, and discovery of a therapeutic target. Therefore, the research of targeted medicine for sCD97 in the bile cannot be ignored.

To the best of our knowledge, our study is the first to describe the CD97 and CD55 expression and the biliary sCD97 levels in patients with ICC and their values in the diagnosis and prognosis.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by the Ethics Committee of the Medical Faculty of Fujian Medical University in China. Written informed consent was obtained from all patients before surgery. All specimens were handled and made anonymous according to the ethical and legal standards. Ethical approval number: 2016KY014.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the National Clinical Key Specialty Construction Project (General Surgery) of China, the MiaoPu Youth Fund in Fujian Medical University (no.2015MP002), and the Youth Fund of Fujian Provincial Health and Family Planning Commission (no.2016-1-48).