## Abstract

Background: Esophagogastric junction tumor (EGJ) is a rare but fatal disease with a rapid rising incidence worldwide in the late 20 years, and it lacks a convenient and safe method for diagnosis. The present study aimed to evaluate the potential of serum CYR61 as a biomarker for the diagnosis of EGJ tumor.

Methods: Enzyme-linked immunosorbent assay (ELISA) was used to estimate CYR61 levels in sera of 152 EGJ tumor patients and 137 normal controls. Receiver operating characteristics (ROC) was carried out to evaluate the diagnostic accuracy. The Mann–Whitney’s U test was used to compare the difference of serum levels of CYR61 between groups. And chi-square tests were employed to estimate the correlation of the positive rate of serum CYR61 between/among subgroups.

Results: Serum CYR61 levels were statistically lower in EGJ tumor and early-stage EGJ tumor patients than those in normal controls (P<0.0001). The sensitivity, specificity and the area under the curve (AUC) of this biomarker in EGJ tumor were 88.2%, 43.8% and 0.691, respectively, and those for early stage of EGJ tumor were 80.0%, 66.4% and 0.722, respectively. Analyses showed that there was no correlation between the clinical data and the levels of CYR61 (P>0.05).

Conclusion: The present study showed that CYR61 might be a potential biomarker to assist the diagnosis of EGJ tumor.

## Introduction

Esophagogastric junction (EGJ) tumor is a rare but fatal disease with a rapid rising incidence worldwide in the late 20 years [1]. Studies have shown that incidence rate of EGJ tumor in China is higher than that in western countries [2–4]. Adenocarcinoma is the most common histology type, accounting for more than 90% of all EGJ tumors [5,6]. Due to the lack of epidemiological available data and public supervision, the diagnosis of EGJ tumor has always been complex. So far, the primary strategy in clinic of early detection for EGJ tumor is endoscopy that is invasive, unacceptable to some patients and proved to have side effects [7]. In western countries, patients with EGJ tumors are always diagnosed as advanced cancer with poor prognosis because of the nonspecific symptoms at early stage [1]. What’s more, despite a variety of treatment options, such as radical surgery, chemotherapy and radiotherapy, patients with EGJ tumor still appear extremely low survival rate [8–10]. Five-year overall survival (OS) rates with surgery alone are gloomy at approximately 25% [11]. Thus, a reliable and sensitive early detection method that has clinical value for effective treatment and improving the prognosis of patients is urgently needed for EGJ tumor patients.

CCN1/CYR61 is a protein from CCN family, which contains five parts: an N-terminal secretory signal peptide and four functional domains: an insulin-like growth factor-binding protein domain (IGFBP), a Von Willebrand factor domain (VWC), a thrombospondin type-1 repeat module (TSP-1) and a CT [12]. It can be induced rapidly by growth factors. As an angiogenic inducer that can promote tumor growth and vascularization, it plays an important part in promoting cell survival, proliferation, differentiation, angiogenesis and inducing apoptosis and senescence by binding directly to the integrins and heparin sulfate proteoglycans or activating multiple signaling transduction pathways [13–15]. This suggests that CYR61 might be useful as a biomarker or therapeutic target in certain diseases. Some studies have indicated that high expression of CYR61 was related to colorectal cancer [16], prostate cancer [17,18], ovarian cancer [19], glioma [20], osteosarcoma [21], gastric cancer [22] and breast cancer [23,24]. Meanwhile, it was proved that the expression of CYR61 reduced in high-grade chondrosarcomas [25], advanced gastric cancers [25], endometrial cancer [26] and lung cancer [27]. What’s more, multiple studies showed that CYR61 could be a metastatic biomarker for prediction of poor prognosis of EGJ tumors [28] and a potential diagnosis biomarker for colorectal cancer [29] and endometrial cancer [15]. However, the application of serum CYR61 as a clinical biomarker in the diagnosis of EGJ tumor patients has rarely been reported. The aim of our study is to examine the use of serum CYR61 as a potential biomarker for the diagnosis of EGJ tumor.

## Materials and methods

### Study sample

In the present study, we set the sample size required for the EGJ tumor group and the normal control group to be equal. In order to estimate the sample size, we calculate it according to the following formula [30,31].
$Sample size(n)based on Sensitivity=[Z1−α/32×SN×(1−SN)](L2×P)$
$Sample size(n)based on Specificity=[Z1−α/22×SP×(1−SP)][L2×(1−P)]$

Z1-α/2 is the value of Z when the cumulative probability in the normal distribution is equal to α/2, When α is 0.05, Z1-α/2 is 1.96, and when α is 0.01, Z1-α/2 is 2.58. L is the width of the 95% interval of sensitivity or specificity that we allow, which is artificially designated by the researcher, and is generally set at 0.03–0.1. Here, we set the allowable error (L) as 0.1 and α as 0.05. In the preliminary experiment, we concluded that the sensitivity (SN) of cyr61 for EGJ diagnosis is 0.4, the specificity (SP) is 0.9, and the disease prevalence (P) is 0.6.

When we use the sensitivity to estimate the sample size, according to the formula:
$N1=[1.962×0.4×(1−0.4)](0.12×0.6)≈154$
When we use the specificity to estimate the sample size, according to the formula:
$N2=[1.962×0.9×(1−0.9)][0.12×(1−0.6)]≈86$

Because N1 > N2, according to the principle of which general election, and taking N1 as a reference, it reminds us that we need to include at least 154 research subjects. For some other reasons, the actual sample size we collected was 152 cases in the EGJ tumor group and 137 cases in the normal control group. Among the 152 serum samples from EGJ tumor patients, 81 were diagnosed at the First Affiliated Hospital of Shantou University Medical College and 71 were diagnosed at Cancer Hospital of Shantou University Medical College from October 2017 to December 2019. And the 137 serum samples from normal controls were selected from the First Affiliated Hospital of Shantou University Medical College. All the normal samples were healthy subjects without cancer signs. After being coagulated at room temperature for 30 min and centrifuged at 1250g for 5 min, all the serum samples were stored at −80°C until use.

The diagnosis of EGJ tumor was confirmed by histopathology and the tumor stage was referred to the eight edition of the American Joint Committee on Cancer (AJCC) Cancer Staging Manual [32]. In the present study, EGJ tumor with TNM stage 0 + I + IIA was defined as early-stage EGJ tumor.

### Enzyme-Linked Immunosorbent Assay (ELISA) for CYR61

The Serum concentrations of CYR61 were tested by ELISA Kit based on manufacturer’s directions. Reagents, samples and standards were prepared as instructed. The CYR61 standard concentrations for standard curve were 0, 78, 156, 312, 625, 1250, 2500 and 5000 pg/ml, respectively. It was proved in our preliminary experiments that the most appropriate dilution ratio was 1:1. After adding 100 μl standards and serum samples (a 2-fold dilution) per well, the 96-well plate were incubated for 2 h at 37°C. Then the liquid was removed and 100 μl of biotin-antibody (1×) was added to each well and incubated for 1 h at 37°C followed by washing the plate for three times using microplate washer with water buffer. Before accomplishing the same washing procedure for another five times, 100 μl of HRP-avidin was added to each well and incubated for 1 h at 37°C. After adding 90 μl TMB substrate to each well, the plate was incubated for 20 min at 37°C protected from light. Color formation was stopped by 50 μl Stop Solution, and the optical density (OD) value was read at wavelength of 450 and 590 nm on a plate microplate reader within 5 min. Corresponding concentrations were converted from OD values using standard curve method (Table 1). Each serum sample was tested twice and the average value was taken for analysis.

Table 1
Participant information and clinicopathological characteristics
GroupEGJ tumor patients (n=152)Normal controls (n=137)
Age, years
Mean ± SD 64.34 ± 9.578 48.32 ± 12.529
Range 22-93 24-81
Gender
Male 125 65
Female 27 72
Smoke
Yes 25
No 72
Unknown 55
TNM Stage
0
I 13
II
III 29
IV 44
Unknown 58
Histological stage
Unknown 101
Depth of tumor invasion (T staging)
Tis
T1 12
T2
T3 22
T4 45
Unknown 62
Regional lymph nodes (N staging)
N0 27
N1 19
N2 16
N3 27
Unknown 63
Metasstasis
M0 83
M1
Unknown 60
GroupEGJ tumor patients (n=152)Normal controls (n=137)
Age, years
Mean ± SD 64.34 ± 9.578 48.32 ± 12.529
Range 22-93 24-81
Gender
Male 125 65
Female 27 72
Smoke
Yes 25
No 72
Unknown 55
TNM Stage
0
I 13
II
III 29
IV 44
Unknown 58
Histological stage
Unknown 101
Depth of tumor invasion (T staging)
Tis
T1 12
T2
T3 22
T4 45
Unknown 62
Regional lymph nodes (N staging)
N0 27
N1 19
N2 16
N3 27
Unknown 63
Metasstasis
M0 83
M1
Unknown 60

Abbreviation: EGJ tumor, esophagogastric junction tumor.

### Statistical analysis

With Microsoft Excel, SPSS (version 20.0), GraphPad Prism 8.0 and Sigma Plot 10.0 software, data analyses were performed statistically. The Mann–Whitney’s U test was used to compare the difference of serum levels of CYR61 between EGJ tumor group and normal group, early-stage EGJ tumor group and normal group. And chi-square tests were used to estimate the correlation between different clinical data and the positive rate, and the correlation between different groups. Plotting ROC curves and calculating the area under ROC curves (AUC) [33] with 95% confidence interval were used to analyze the accuracy of diagnostic value. The optimum cut-off values were obtained from the Youden’s indexes of the ROC curves and the maximum indexes were calculated by the sum of sensitivity and specificity minus 1. And sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV), false positive rate (FPR), false negative rate (FNR), positive likelihood ratio (PLR) and negative likelihood ratio (NLR) were calculated using the optimum cut-off values to further evaluate the diagnostic value. P<0.05 (two-sided) was considered as statistically significant in all the analyses.

## Results

### The levels of serum CYR61 in EGJ tumor patients and normal controls

In our study, 289 serum samples were tested, including EGJ tumor group (n=152) and normal control group (n=137), with the mean age of 64 years old and 48 years old respectively. (Table 1) The mean concentration of serum CYR61 in EGJ tumor group (n=152), early-stage EGJ tumor group (n=15) and normal group (n=137) was 258.515 ± 191.736 ng/ml, 225.146 ± 114.316 ng/ml, and 429.115 ± 273.432 ng/ml, respectively (Table 2). There was a difference between the distribution of EGJ tumor group and normal control group. The EGJ tumor group accounted for more histogram volume at low concentration while normal group accounted for more at high concentration (Figure 1). More intuitive distribution and dispersion could be seen in combined scatter plot and box plot (Figure 2). Confirmed by statistics, the level of serum CYR61 in EGJ tumor group was lower than that in normal controls (P<0.0001), and there was also a significant difference between early-stage EGJ tumor group and normal control group (P<0.0001). (Table 2)

#### Bar chart of concentration of CYR61 from EGJ tumor serum and normal serum

Figure 1
Bar chart of concentration of CYR61 from EGJ tumor serum and normal serum

The diagram of EGJ tumor (n=152) is in blue; the one of normal control (n=137) is in orange. The lowest concentration was 24.85 ng/ml in EGJ tumor and the highest one was 1280.64 ng/ml in normal control. The concentration was divided into 20 sections equally. EGJ tumor stands for more histogram volume on lower concentration while normal control accounts for more on higher concentration.

Figure 1
Bar chart of concentration of CYR61 from EGJ tumor serum and normal serum

The diagram of EGJ tumor (n=152) is in blue; the one of normal control (n=137) is in orange. The lowest concentration was 24.85 ng/ml in EGJ tumor and the highest one was 1280.64 ng/ml in normal control. The concentration was divided into 20 sections equally. EGJ tumor stands for more histogram volume on lower concentration while normal control accounts for more on higher concentration.

#### Scatter plots and box plots of concentration of CYR61 from EGJ tumor serum, early-stage EGJ tumor serum and normal serum

Figure 2
Scatter plots and box plots of concentration of CYR61 from EGJ tumor serum, early-stage EGJ tumor serum and normal serum

Every sample of the concentration of serum CYR61 in three groups were shown in scatter plots and box plots (P<0.0001). The central line is median. It showed the degree of dispersion. The lines up and down are the extremum; EGJ tumor, esophagogastric junction tumor. CYR61 is a protein from CCNs family.

Figure 2
Scatter plots and box plots of concentration of CYR61 from EGJ tumor serum, early-stage EGJ tumor serum and normal serum

Every sample of the concentration of serum CYR61 in three groups were shown in scatter plots and box plots (P<0.0001). The central line is median. It showed the degree of dispersion. The lines up and down are the extremum; EGJ tumor, esophagogastric junction tumor. CYR61 is a protein from CCNs family.

Table 2
Comparison between three groups
NMean ± SDP value95%CI
EGJ tumor 152 258.515 ± 191.736 *(<0.0001) 227.788–289.243
Early-stage EGJ tumor (0+I+IIA) 15 225.146 ± 114.316 *(<0.0001) 161.840–288.453
Normal controls 137 429.115±273.432  382.917–475.312
NMean ± SDP value95%CI
EGJ tumor 152 258.515 ± 191.736 *(<0.0001) 227.788–289.243
Early-stage EGJ tumor (0+I+IIA) 15 225.146 ± 114.316 *(<0.0001) 161.840–288.453
Normal controls 137 429.115±273.432  382.917–475.312

*compared with normal controls; EGJ tumor, esophagogastric junction tumor.

### The diagnostic value of serum CYR61 in EGJ tumor and early-stage EGJ tumor

ROC curve was established to evaluate the diagnostic value of CYR61 in EGJ tumor. According to the ROC curve of EGJ tumor group and normal group (Figure 3), AUC was 0.691, and the optimized cutoff value of 445.708 ng/ml was selected with specificity of 43.8% and sensitivity of 88.2%. For early-stage EGJ tumor group with AUC of 0.722 and the cutoff value of 281.947 ng/ml, specificity and the sensitivity were 66.4% and 80.0%, respectively. And the positive rates of EGJ tumor and early-stage EGJ tumor were much higher than that of the normal controls (Table 2). In order to better explain the clinical value, more relevant indicators were calculated and the results were displayed with 95% confidence interval, including false positive rate, false negative rate, positive predictive value, negative predictive value, positive likelihood ratio and negative likelihood ratio (Table 3).

#### ROC curve analysis in the diagnosis of EGJ tumor and early-stage EGJ tumor

Figure 3
ROC curve analysis in the diagnosis of EGJ tumor and early-stage EGJ tumor

Two groups versus normal controls group are in different colors. The area under the red line is 0.5 for reference; ROC curve, receiver operating characteristic curve; EGJ tumor, esophagogastric junction tumor.

Figure 3
ROC curve analysis in the diagnosis of EGJ tumor and early-stage EGJ tumor

Two groups versus normal controls group are in different colors. The area under the red line is 0.5 for reference; ROC curve, receiver operating characteristic curve; EGJ tumor, esophagogastric junction tumor.

Table 3
Evaluation of the detection value of CYR61 in the diagnosis of EGJ tumor
AUCSENSPEFPRFNRPPVNPVPLRNLR
EGJ tumor vs. NC 0.691 88.2% (81.7–92.6%) 43.8% (35.4–52.5%) 56.2% (47.5–64.6%) 11.8% (7.4–18.3%) 63.5% (56.6–69.9) 76.9% (65.8–85.5%) 1.57 (1.34–1.84) 0.27 (0.17–0.42)
Early-stage EGJ tumor vs. NC 0.722 80.0% (51.4–94.7%) 66.4% (57.8–74.1%) 33.6% (25.9–42.2%) 20.0% (5.3–48.6%) 20.7% (11.6–33.7%) 96.8% (90.3–99.2%) 2.38 (1.68–3.36) 0.30 (0.11–0.83)
AUCSENSPEFPRFNRPPVNPVPLRNLR
EGJ tumor vs. NC 0.691 88.2% (81.7–92.6%) 43.8% (35.4–52.5%) 56.2% (47.5–64.6%) 11.8% (7.4–18.3%) 63.5% (56.6–69.9) 76.9% (65.8–85.5%) 1.57 (1.34–1.84) 0.27 (0.17–0.42)
Early-stage EGJ tumor vs. NC 0.722 80.0% (51.4–94.7%) 66.4% (57.8–74.1%) 33.6% (25.9–42.2%) 20.0% (5.3–48.6%) 20.7% (11.6–33.7%) 96.8% (90.3–99.2%) 2.38 (1.68–3.36) 0.30 (0.11–0.83)

95% CI were given in brackets for each group. AUC, area under the ROC curve; EGJ tumor, esophagogastric junction tumor; FNR, false negative rate; FPR, false positive rate; NC, normal controls; NLR, negative likelihood ratio; NPV, negative predictive value; PLR, positive likelihood ratio; PPV, positive predictive value; SEN, sensitivity; SPE, specificity.

### Correlation between serum concentration of CYR61 and clinical data in patients with EGJ tumor

The association between serum CYR61 level of patients with EGJ tumor and clinical data variables was shown in Table 4. There was no statistically significant correlation between positive rate of serum CYR61 and clinical data, including age, gender, smoking status, depth of tumor invasion, lymph node status, metastasis, histological grade and early-stage or advanced-stage of EGJ tumor (all P>0.05).

Table 4
Correlation between CYR61 and clinical data in EGJ patients
NPositive%95%CIP
Age     0.823
≥60 115 101 87.8% 80.1–92.9
<60 37 33 89.2% 73.6–96.4
Gender     0.598
Male 125 111 88.8% 81.6–93.5
Female 27 23 85.2% 65.4–95.1
Smoke     0.780
Yes 25 21 84.0% 63.1–94.7
No 72 64 88.9% 78.7–94.7
Unknown 55 49 89.1% 77.1–95.5
T     0.819
Tis+T1+T2 23 21 91.3% 70.5–98.5
T3+T4 67 58 86.6% 75.5–93.3
Unknown 62 55 88.7% 77.5–95.0
N     0.360
N0 27 22 81.5% 61.3–93.0
N1+N2+N3 62 57 91.9% 81.5–97.0
Unknown 63 55 87.3% 76.0–94.0
M     0.119
M0 83 74 89.2% 79.9–94.6
M1 66.7% 30.9–91.0
Unknown 60 54 90.0% 78.8–95.9
G1 100.0% 46.3–100
G2 19 17 89.5% 65.5–98.2
G3 27 23 85.2% 65.4–95.1
Unknown 101 89 88.1% 79.8–93.4
TNM stage     0.658
Early 15 14 93.3% 66.0–99.7
Unknown 58 52 89.7% 78.2–95.7
NPositive%95%CIP
Age     0.823
≥60 115 101 87.8% 80.1–92.9
<60 37 33 89.2% 73.6–96.4
Gender     0.598
Male 125 111 88.8% 81.6–93.5
Female 27 23 85.2% 65.4–95.1
Smoke     0.780
Yes 25 21 84.0% 63.1–94.7
No 72 64 88.9% 78.7–94.7
Unknown 55 49 89.1% 77.1–95.5
T     0.819
Tis+T1+T2 23 21 91.3% 70.5–98.5
T3+T4 67 58 86.6% 75.5–93.3
Unknown 62 55 88.7% 77.5–95.0
N     0.360
N0 27 22 81.5% 61.3–93.0
N1+N2+N3 62 57 91.9% 81.5–97.0
Unknown 63 55 87.3% 76.0–94.0
M     0.119
M0 83 74 89.2% 79.9–94.6
M1 66.7% 30.9–91.0
Unknown 60 54 90.0% 78.8–95.9
G1 100.0% 46.3–100
G2 19 17 89.5% 65.5–98.2
G3 27 23 85.2% 65.4–95.1
Unknown 101 89 88.1% 79.8–93.4
TNM stage     0.658
Early 15 14 93.3% 66.0–99.7
Unknown 58 52 89.7% 78.2–95.7

Abbreviations: EGJ tumor, esophagogastric junction tumor.

## Discussion

At present, the diagnostic examination choice for EGJ tumor is upper esophagogastroduodenoscopy [1,5], an invasive method with serious side effects, which is not suitable for the screening and detection of asymptomatic population. With the development of clinical medicine, the detection of serum tumor biomarkers, as a painless, convenient, and most importantly, non-invasive detection method, has been widely developed in clinical diagnosis. Tumor biomarkers are a kind of substances reflecting the existence of tumors. When these substances reach a certain level in vivo, they can predict the existence of some tumors, which makes it possible to diagnose EGJ tumor early [34]. In this regard, our study found that CYR61 might be a potential biomarker for the diagnosis of EGJ tumor.

CYR61 plays an important role in tumor angiogenesis, tumor cell proliferation, apoptosis and tumor metastasis, which closely participates in the occurrence and development of tumors [13–15]. An increasing number of studies have proved CYR61 to be a metastatic biomarker for prediction of prognosis in osteosarcoma [21], gastric cancer [22], colorectal cancer [16], laryngeal tumor [35], ovarian carcinoma [36] and prostate cancer [37]. As for EGJ tumors, a study has suggested that CYR61 might serve as a metastatic predictor of poor prognosis and provide a potential molecular target for anti-metastatic therapy of EGJ tumor [28]. In addition, many other studies have shown that CYR61 could also act as a diagnosis predictor in patients with colorectal cancer [29] and endometrial cancer [15]. However, CYR61 as a potential biomarker for diagnosis of EGJ tumor has not yet been reported. In the present study, ROC results showed that AUC was 0.691, specificity was 43.8%, and sensitivity was 88.2%, suggesting the diagnostic value of serum CYR61 for EGJ tumor. Similar results could also be demonstrated in early EGJ tumor. Taking other diagnostic evaluation indices into consideration contributes to better understanding of the diagnostic value of serum CYR16 in EGJ tumor, including false positive rate (FPR) of 56.2% (95%CI: 47.5–64.6%), false negative rate (FNR) of 11.8% (95%CI: 7.4–18.3%), positive predictive value (PPV) of 63.5% (95%CI: 56.6–69.9%), negative predictive value (NPV) of 76.9% (95%CI: 65.8–85.5%), positive likelihood ratio (PLR) of 1.57 (95%CI: 1.34–1.84) and negative likelihood ratio (NLR) of 0.27 (95%CI: 0.17–0.42). Meanwhile, in the present study, the serum CYR61 concentration in EGJ tumor was shown significantly lower than that in healthy control group (P<0.001), which was inconsistent with the results of high expression in the study of colon cancer, esophageal cancer and many other cancers. Therefore, we infer that the expression pattens of CYR61 differ in different types of tumors, as well as different histopathological types may lead to the difference, thereby CYR61 has certain significance for the differential diagnosis of tumors and a broad application prospect as a diagnostic biomarker of tumors.

However, there are still some limitations in the present study. It remains open to be discussed and improved. Relatively low specificity may limit the clinical application of CYR61 in the screening of asymptomatic early EGJ tumor patients, so a single detection of serum CYR61 is unable to meet the clinical demands. As reported, compared with single biomarker, combined detection of multiple serum proteins could help improve the sensitivity or specificity of gastrointestinal cancer screening [38], which provides us with a new research direction: CYR61 could be combined with other tumor markers or even other tests to diagnose EGJ tumor. Because the age and sex of normal control group were mismatching with that of EGJ tumor cases, further study could be carried out according to the corresponding age and sex. However, the P value of the variance test between the age and the concentration of serum CYR61 was 0.153 and the one between the sex and the concentration of serum CRY61 was 0.249, which showed that CYR61 has no significant relationship with the age and the sex. So, the age and the sex bias between the two groups could be reduced. In addition, due to the low clinical incidence of EGJ tumor and difficulty of diagnosing EGJ tumor as early cancer, the sample size of our study is small. Besides, incomplete clinical data and single center study are also likely bias. Our conclusion only suggested the possibility of CYR61 being a potential biomarker in the early detection of EGJ tumor. We hope further in-depth studies with large sample size, complete clinical information and well-matched age and sex controls in multiple institutions could be conducted, which could help better evaluate the diagnostic value of CYR61 as a biomarker.

## Conclusion

In summary, our study evaluated the relationship between serum CYR61 and EGJ tumor, and proved that serum CYR61 could be a potential biomarker in the early detection of EGJ tumor.

## Data Availability

The data were collected and saved in hospital’s medical history management center. Due to the legitimate protection of patients’ privacy, our information is not available on public or any private websites, but is available from the corresponding author on reasonable request.

## Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

## Author Contribution

J.J.X. and L.Y.C. initiated the project and designed the study. J.Y.Z., Y.X.Z. and Y.T.O. conceived, designed and performed the experiments, and analyzed the data. J.Y.Z., Y.X.Z. and Y.T.O. wrote the manuscript. All authors have read and approved the final manuscript.

## Funding

This work was supported by grants from the National Natural Science Foundation of China [grant number 81871921]; Natural Science Foundation of Guangdong Province-Outstanding Youth Project [grant number 2019B151502059]; and the Basic and Applied Basic Research Programs of Guangdong province [grant number 2018KZDXM033].

## Ethics Approval and Consent to Participate

The present work was approved by the Ethics Committee of First Affiliated Hospital of Shantou University Medical College, the Ethics Committee of the Cancer Hospital of Shantou University Medical College. The research has been carried out in accordance with the World Medical Association Declaration of Helsinki. Informed consents were acquired from all the participants.

## Abbreviations

• AJCC

American Joint Committee on Cancer

•
• AUC

area under the ROC curve

•
• CI

confidence interval

•
• CYR61

cysteine-rich angiogenic inducer 61

•
• EGJ tumor

esophagogastric junction tumor

•
• ELISA

•
• FNR

false negative rate

•
• FPR

false positive rate

•
• NLR

negative likelihood ratio

•
• NPV

negative predictive value

•
• OD

optical density

•
• PLR

positive likelihood ratio

•
• PPV

positive predictive value

•
• ROC curve

## References

1.
Chevallay
M.
,
Bollschweiler
E.
,
Chandramohan
S.M.
et al.
(
2018
)
Cancer of the gastroesophageal junction: a diagnosis, classification, and management review
.
1434
,
132
138
[PubMed]
2.
Bray
F.
,
Ferlay
J.
,
Soerjomataram
I.
et al.
(
2018
)
Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
.
CA Cancer J. Clin.
68
,
394
424
[PubMed]
3.
Liu
K.
,
Yang
K.
,
Zhang
W.
et al.
(
2016
)
Changes of esophagogastric junctional adenocarcinoma and gastroesophageal reflux disease among surgical patients during 1988-2012: a single-institution, high-volume experience in China
.
Ann. Surg.
263
,
88
95
[PubMed]
4.
Urabe
M.
,
Ushiku
T.
,
Shinozaki-Ushiku
A.
et al.
(
2018
)
Adenocarcinoma of the esophagogastric junction and its background mucosal pathology: a comparative analysis according to Siewert classification in a Japanese cohort
.
Cancer Med.
7
,
5145
5154
[PubMed]
5.
Okereke
I.C.
(
2017
)
Management of gastroesophageal junction tumors
.
Surg. Clin. North Am.
97
,
265
275
[PubMed]
6.
Azari
F.S.
and
Roses
R.E.
(
2019
)
Management of early stage gastric and gastroesophageal junction malignancies
.
Surg. Clin. North Am.
99
,
439
456
[PubMed]
7.
Xu
Y.W.
,
Peng
Y.H.
,
Xu
L.Y.
et al.
(
2019
)
Autoantibodies: Potential clinical applications in early detection of esophageal squamous cell carcinoma and esophagogastric junction adenocarcinoma
.
World J. Gastroenterol.
25
,
5049
5068
[PubMed]
8.
Li
J.
,
Qin
S.
,
Xu
J.
et al.
(
2016
)
Randomized, double-blind, placebo-controlled phase III trial of apatinib in patients with chemotherapy-refractory advanced or metastatic adenocarcinoma of the stomach or gastroesophageal junction
.
J. Clin. Oncol.
34
,
1448
1454
[PubMed]
9.
Anderson
L.A.
,
Tavilla
A.
,
Brenner
H.
et al.
(
2015
)
Survival for oesophageal, stomach and small intestine cancers in Europe 1999-2007: Results from EUROCARE-5
.
Eur. J. Cancer
51
,
2144
2157
[PubMed]
10.
Siegel
R.L.
,
Miller
K.D.
and
Jemal
A.
(
2017
)
Cancer Statistics, 2017
.
CA Cancer J. Clin.
67
,
7
30
[PubMed]
11.
Ychou
M.
,
Boige
V.
,
Pignon
J.P.
et al.
(
2011
)
Perioperative chemotherapy compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial
.
J. Clin. Oncol.
29
,
1715
1721
[PubMed]
12.
Lau
L.F.
and
Lam
S.C.
(
1999
)
The CCN family of angiogenic regulators: the integrin connection
.
Exp. Cell. Res.
248
,
44
57
[PubMed]
13.
Lau
L.F.
(
2011
)
CCN1/CYR61: the very model of a modern matricellular protein
.
Cell. Mol. Life Sci.
68
,
3149
3163
[PubMed]
14.
Babic
A.M.
,
Kireeva
M.L.
,
Kolesnikova
T.V.
et al.
(
1998
)
CYR61, a product of a growth factor-inducible immediate early gene, promotes angiogenesis and tumor growth
.
Proc. Natl. Acad. Sci. U. S. A.
95
,
6355
6360
[PubMed]
15.
Yang
R.
,
Chen
Y.
and
Chen
D.
(
2018
)
Biological functions and role of CCN1/Cyr61 in embryogenesis and tumorigenesis in the female reproductive system (Review)
.
Mol. Med. Rep.
17
,
3
10
[PubMed]
16.
Jeong
D.
,
Heo
S.
,
Ahn
T.S.
et al.
(
2014
)
Cyr61 expression is associated with prognosis in patients with colorectal cancer
.
BMC Cancer
14
,
164
[PubMed]
17.
N.
,
Kulkarni
P.
and
Getzenberg
R.H.
(
2012
)
Cyr61 is a potential prognostic marker for prostate cancer
.
Asian J. Androl.
14
,
405
408
[PubMed]
18.
D'Antonio
K.B.
,
Toubaji
A.
,
R.
et al.
(
2010
)
Extracellular matrix associated protein CYR61 is linked to prostate cancer development
.
J. Urol.
183
,
1604
1610
[PubMed]
19.
Gery
S.
,
Xie
D.
,
Yin
D.
et al.
(
2005
)
Ovarian carcinomas: CCN genes are aberrantly expressed and CCN1 promotes proliferation of these cells
.
Clin. Cancer Res.
11
,
7243
7254
[PubMed]
20.
Xie
D.
,
Yin
D.
,
Tong
X.
et al.
(
2004
)
Cyr61 is overexpressed in gliomas and involved in integrin-linked kinase-mediated Akt and beta-catenin-TCF/Lef signaling pathways
.
Cancer Res.
64
,
1987
1996
[PubMed]
21.
Liu
Y.
,
Zhang
F.
,
Zhang
Z.
et al.
(
2017
)
High expression levels of Cyr61 and VEGF are associated with poor prognosis in osteosarcoma
.
Pathol. Res. Pract.
213
,
895
899
[PubMed]
22.
Zhao
Z.S.
,
Li
L.
,
Wang
H.J.
et al.
(
2011
)
Expression and prognostic significance of CEACAM6, ITGB1, and CYR61 in peripheral blood of patients with gastric cancer
.
J. Surg. Oncol.
104
,
525
529
[PubMed]
23.
Tsai
M.S.
,
Bogart
D.F.
,
Li
P.
et al.
(
2002
)
Expression and regulation of Cyr61 in human breast cancer cell lines
.
Oncogene
21
,
964
973
[PubMed]
24.
Hellinger
J.W.
,
Huchel
S.
,
Goetz
L.
et al.
(
2019
)
Inhibition of CYR61-S100A4 axis limits breast cancer invasion
.
Front. Oncol.
9
,
1074
[PubMed]
25.
Maeta
N.
,
Osaki
M.
,
Shomori
K.
et al.
(
2007
)
CYR61 downregulation correlates with tumor progression by promoting MMP-7 expression in human gastric carcinoma
.
Oncology
73
,
118
126
[PubMed]
26.
Chien
W.
,
Kumagai
T.
,
Miller
C.W.
et al.
(
2004
)
Cyr61 suppresses growth of human endometrial cancer cells
.
J. Biol. Chem.
279
,
53087
53096
[PubMed]
27.
Chen
P.P.
,
Li
W.J.
,
Wang
Y.
et al.
(
2007
)
Expression of Cyr61, CTGF, and WISP-1 correlates with clinical features of lung cancer
.
PLoS One
2
,
e534
[PubMed]
28.
Wei
J.
,
Yu
G.
,
Shao
G.
et al.
(
2016
)
CYR61 (CCN1) is a metastatic biomarker of gastric cardia adenocarcinoma
.
Oncotarget
7
,
31067
31078
[PubMed]
29.
Song
Y.F.
,
Xu
Z.B.
,
Zhu
X.J.
et al.
(
2017
)
Serum Cyr61 as a potential biomarker for diagnosis of colorectal cancer
.
Clin. Transl. Oncol.
19
,
519
524
[PubMed]
30.
Malhotra
R.K.
and
Indrayan
A.
(
2010
)
A simple nomogram for sample size for estimating sensitivity and specificity of medical tests
.
Indian J. Ophthalmol.
58
,
519
522
[PubMed]
31.
Buderer
N.M.
(
1996
)
Statistical methodology: I. Incorporating the prevalence of disease into the sample size calculation for sensitivity and specificity
.
3
,
895
900
[PubMed]
32.
Rice
T.W.
,
Ishwaran
H.
,
Ferguson
M.K.
et al.
(
2017
)
Cancer of the esophagus and esophagogastric junction: an eighth edition staging primer
.
J. Thorac. Oncol.
12
,
36
42
[PubMed]
33.
Fawcett
T.
(
2006
)
An introduction to ROC analysis
.
Pattern Recognition Lett.
27
,
861
874
34.
Perkins
G.L.
,
Slater
E.D.
,
Sanders
G.K.
et al.
(
2003
)
Serum tumor markers
.
Am. Fam. Physician
68
,
1075
1082
[PubMed]
35.
Liu
Y.
,
Zhou
Y.D.
,
Xiao
Y.L.
et al.
(
2015
)
Cyr61/CCN1 overexpression induces epithelial-mesenchymal transition leading to laryngeal tumor invasion and metastasis and poor prognosis
.
Asian Pac. J. Cancer Prev.
16
,
2659
2664
[PubMed]
36.
Shen
H.
,
Cai
M.
,
Zhao
S.
et al.
(
2014
)
CYR61 overexpression associated with the development and poor prognosis of ovarian carcinoma
.
Med. Oncol.
31
,
117
[PubMed]
37.
D'Antonio
K.B.
,
Schultz
L.
,
R.
et al.
(
2010
)
Decreased expression of Cyr61 is associated with prostate cancer recurrence after surgical treatment
.
Clin. Cancer Res.
16
,
5908
5913
[PubMed]
38.
Wang
H.Y.
,
Hsieh
C.H.
,
Wen
C.N.
et al.
(
2016
)
Cancers screening in an asymptomatic population by using multiple tumour markers
.
PLoS ONE
11
,
e0158285
[PubMed]

## Author notes

*

These authors contributed to equally to this work.