Sage Journals: Discover world-class research

Abstract

BACKGROUND:

Nidogen-2 (NID2), a secretory basement membrane protein, has been implicated as a potential biomarker in ovarian cancer and hepatocellular carcinoma.

OBJECTIVE:

In this study, we aimed to investigate the utility of detecting serum NID2 levels for identification of esophageal squamous cell carcinoma (ESCC) patients and prediction of poor survival outcome.

METHODS:

Using an in-house NID2 enzyme-linked immunosorbent assay (ELISA), serum samples from 101 ESCC patients and 50 healthy controls were screened for their NID2 levels.

RESULTS:

The serum NID2 levels in ESCC patients (median 24.4 $\mu$ g/L) are significantly higher ( $p=$ 4.3e-09) than that of the healthy controls (median 15.85 $\mu$ g/L). The receiver operating characteristic (ROC) curve demonstrated an area under the curve of 0.756. At the threshold of 17.95 $\mu$ g/L, the sensitivity and specificity achieved are 0.76 and 0.63, respectively. Kaplan-Meier survival analysis revealed that patients with high serum NID2 levels ( $\geqslant$ 32.6 $\mu$ g/L) have significantly higher risk of death (HR $=$ 1.984, 95% CI: 1.175–3.349; log-rank p-value $=$ 0.012) compared to those with low serum NID2 levels ( $<$ 20.0 $\mu$ g/L).

CONCLUSIONS:

In conclusion, we show that detecting the elevation of serum NID2 levels has potential diagnostic and prognostic value for ESCC patients.

Keywords

Nidogen-2 enzyme-linked immunosorbent assay esophageal squamous cell carcinoma serum biomarker extracellular matrix protein

1. Introduction

Esophageal squamous cell carcinoma (ESCC) is the most common histological subtype of the esophageal cancer (EC) in Asia, accounting for more than 90% of EC in China [1]. ESCC remains a deadly cancer due to its asymptomatic nature in the early stage, resulting in late diagnosis at advanced stages [2]. As a result, there is a strong impetus for the development of non-invasive blood-based diagnostic markers to assist early detection of ESCC, to replace or add to the existing more invasive methods of ESCC diagnosis such as endoscopy and biopsy.

Several potential circulatory biomarkers have been tested and studied for use in ESCC diagnosis and prognosis. Among them, cancer-associated antigens such as the carcinoembryonic antigen (CEA), squamous cell carcinoma antigen (SCC-Ag) and cytokeratin 19 (CYFRA21-1) were reported in previous ESCC studies [3, 4, 5, 6]. However, most of these biomarkers were not satisfactory due to the low sensitivity for ESCC detection [2]. Hence, there is still an imperative need to discover alternative circulatory biomarkers to improve the diagnosis of ESCC. The potential alteration of the concentration of a secretory protein, nidogen-2 (NID2), in the serum of ESCC patients prompted this current study to evaluate the possible utility of serum NID2 level detection as an ESCC biomarker.

NID2 is a basement membrane protein that is ubiquitously expressed in the extracellular matrix (ECM) [7]. It binds the collagen IV and the laminin network in the ECM, allowing the formation of the ternary complexes [7]. Previously, our methylome data indicated that the NID2 promoter is hypermethylated in ESCC [8]. The exogenous re-expression of NID2 in ESCC cells also showed its ability to cause suppressive effects in clonogenic, migratory and invasive abilities [8]. As NID2 is a secretory protein, it can be detected in the serum or plasma, although the major source of this circulating NID2 is unknown. Kuk et al. first reported on the potential utility of serum NID2 detection as a diagnostic biomarker for ovarian cancer [9]. It was reported that similar to the pre-existing marker, cancer antigen 125 (CA125), serum NID2 level is also elevated in patients with advanced stage ovarian carcinoma [9]. In contrast, in hepatocellular carcinoma (HCC) patients, the serum NID2 level was found to be significantly lower than healthy controls or patients with benign liver lesions [10]. This discrepancy observed between patients with ovarian carcinoma and HCC suggested that there might be tissue or cancer-specificity in the regulation of the serum NID2 level. It was of interest to determine the serum NID2 level changes among the ESCC patients.

2. Materials and methods

2.1 Serum samples

A total of 101 ESCC serum samples were collected at admission from Queen Mary Hospital in years 2000–2016. Healthy control serum samples provided by the nasopharyngeal carcinoma (NPC) Area of Excellence (AoE) Tissue Bank were obtained from the Clinical Biochemistry Unit of Queen Mary Hospital. All serum samples collected were approved by the institutional review board of The University of Hong Kong (UW 14-213).

2.2 Enzyme-linked immunosorbent assay (ELISA)

An in-house sandwich ELISA (adapted from Kuk et al. [9]) was established for the estimation of NID2 concentration in the ESCC serum samples. Purified recombinant NID2 with seven two-fold dilutions ranging from 0.078125 $\mu$ g/L to 5 $\mu$ g/L was used to plot the standard curve.

A Nunc-Immuno 96 Microwell plate (Thermo Fisher Scientific, Waltham, MA, USA) was pre-coated at 4 ${}^{\circ}$ C overnight with 100 $\mu$ l of 2 $\mu$ g/ml goat polyclonal anti-human NID2 antibody (R & D Systems, Minneapolis, MN, USA) in bicarbonate/carbonate coating buffer (pH 9.6). After overnight incubation, wells were rinsed with washing buffer (phosphate-buffered saline (PBS) $+$ 0.05% Tween 20) three times, followed by addition of 200 $\mu$ l of blocking buffer (PBS with 1% bovine serum albumin) for 2 hours at room temperature. After washing, 100 $\mu$ l of standards or 10 $\times$ diluted serum samples were added to the well in duplicate and incubated overnight at 4 ${}^{\circ}$ C. 100 $\mu$ l of biotinylated anti-NID2 antibody (R & D Systems) was added at 250 ng/ml to each well after wash and incubated at room temperature for 4 hours. The wells were washed three times and 100 $\mu$ l of 1:500 diluted streptavidin-HRP (Rockland Inc, Limerick, PA, USA) was added per well and incubated for 30 minutes. The enzyme was then removed and wells were washed three times, followed by addition of 100 $\mu$ l substrate, 2,2’-azino-bis(3- ethylbenzothiazoline-6-sulfonic acid (ABTS) from Peprotech (Rockhill, NJ, USA). The plate was incubated in the dark at room temperature for 30 minutes.

An ELx808 absorbance microplate reader (BioTek Instruments, Winooski, VT, USA) was used for reading the OD at 405 nm. Successful ELISA was indicated by the quality of the standard curve (with R value close to 1). The intra-assay coefficient of variation (CV) for all samples was below 10% and inter-assay CV for selected controls was below 20%.

Figure 1.

Box plot and ROC curve based on estimated serum NID2 levels in the healthy controls and ESCC patients. (A) Box plot depicting the distribution of serum NID2 levels in the healthy controls ( $n=$ 50) and ESCC patients ( $n=$ 101). Serum NID2 levels in ESCC patients are significantly higher than that of the healthy controls ( $p=$ 4.3e-09). (B) The AUC is 0.756 (95% CI: 0.68–0.84). With the optimal cut-off of 17.95 $\mu$ g/L estimated using the Youden’s index, the sensitivity and specificity are 0.76 and 0.64, respectively.

Figure 2.

Box plot and ROC curve of serum NID2 levels of healthy controls and early stage ESCC patients. (A) Box plot comparing the serum NID2 levels between the healthy controls ( $n=$ 50) and early stage ESCC patients ( $n=$ 29). Median serum NID2 level of early stage ESCC patients was significantly higher than that of the healthy controls ( $p=$ 0.00063). (B) ROC curve show that serum NID2 level can distinguish early stage ESCC patients from the healthy controls, with an AUC of 0.725 (95% CI: 0.61–0.84). The optimal cut-off value is 18.05 $\mu$ g/L, with sensitivity and specificity of 0.71 and 0.64, respectively.

2.3 Statistical analysis

The Statistical Package for the Social Sciences (SPSS) software (version 23.0) (IBM, Armonk, NY, USA) and RStudio (version 0.99.903, Rstudio, Inc., Boston, MA, USA) were used to perform various statistical analyses for the ELISA results (Cox regression model, Kaplan-Meier survival curve, and ROC curve). Various transformations of serum NID2 levels (square-root, log, restricted cubic spline) were performed and assessed for their linearity with the survival data. The untransformed serum NID2 level ( $\mu$ g/L) was found to be the most appropriate and was used throughout the analysis. In Cox proportional hazards regression model, the NID2 level was used as a continuous variable. The proportional hazard assumption was tested using cox.zph function in survival package in RStudio and it was confirmed to fit the Cox proportional hazards regression model, $p$ -value $=$ 0.909. The variables significantly associated with overall survival in the univariate analysis were considered for the multivariable analysis, including the serum NID2 level, overall stage, N stage and R classification. The correlations between these variables were examined. There was no linear relationship between serum NID2 level, overall stage and R classification. A strong correlation between overall stage and N stage was present; therefore, the N stage was removed from the multivariable analysis. The stepwise forward selection method (likelihood ratio) was used to construct the multivariable Cox model considering three variables including serum NID2 level, overall stage and R classification. Box plots and ROC curve for the ELISA experiments were generated in RStudio. Student’s t-test was used for comparison of serum NID2 levels between two groups in the box plots. The one-way analysis of variance (ANOVA) test was used in the ELISA analysis for correlation with clinicopathological characteristics. The serum NID2 level was treated as a continuous variable in all analysis except for the Kaplan-Meier survival curve in which it was used as a categorical variable. P-values below 0.05 were considered statistically significant.

3. Results

3.1 Elevated serum NID2 levels in ESCC patients

A total of 101 ESCC sera were screened for their serum NID2 levels, together with 50 healthy control sera. Figure 1A shows the box plot depicting the distribution of the estimated serum NID2 levels from the ELISA screening of 101 ESCC patients and 50 healthy controls. Similar to the NID2 ELISA study in ovarian cancer by Kuk et al. [9], we observed statistically significant elevation of serum NID2 levels among the ESCC patients ( $p=$ 4.3e-09) (Fig. 1A). The median serum NID2 level of the 101 ESCC patients is 24.35 $\mu$ g/L, while the median serum NID2 level of the 50 healthy controls is only at 15.85 $\mu$ g/L. The receiver operating characteristic (ROC) curve of the serum NID2 levels was plotted in order to explore its potential diagnostic value (Fig. 1B). As shown, the area under the curve (AUC) of the ROC is 0.756 (95% confidence interval (CI): 0.68–0.84). The maximum Youden’s index is 0.40, corresponding to an optimal threshold of 17.95 $\mu$ g/L. At this optimal cut-off serum NID2 level, the sensitivity and specificity are 0.76 and 0.64, respectively.

To examine the value of serum NID2 levels for early detection of ESCC, we compared the NID2 level in the early stage patients (Stages I and II, $n=$ 29) and controls. The results showed that the serum NID2 levels of early stage ESCC patients were significantly elevated compared to the controls ( $p=$ 0.00063) (Fig. 2A). The AUC of the ROC for early stage ESCC detection is 0.725 (95% CI: 0.61–0.84), with the maximum Youden’s index of 0.35 (Fig. 2B). Using the corresponding threshold of 18.05 $\mu$ g/L to distinguish the early stage patients from the healthy controls, sensitivity and specificity are 0.71 and 0.64, respectively. This finding suggested that serum NID2 levels have potential diagnostic value in the early detection of ESCC.

3.2 Clinicopathological characteristics of ESCC patients and NID2 serum levels

To investigate potential correlation of this elevated serum NID2 level with the clinicopathological characteristics of the ESCC patients, the one-way analysis of variance (ANOVA) statistical test was used (Table 1). We did not observe any statistically significant correlation of this elevated serum NID2 level with any of the clinicopathological characteristics, such as gender, age, TNM stage, and differentiation status (Table 1).

Table 1
Clinicopathological characteristics of 101 ESCC patients used in the NID2 ELISA study and association test results with serum NID2 level

	Number	(%)	ANOVA
	of cases		$p$ -value
Gender
Male	74	73.3%	0.332
Female	27	26.7%
Age at diagnosis (years)
Below 60	33	32.7%	0.306
60 or above	68	67.3%
Median survival time (months)
17.3 months	101	100%
Overall stage
I	6	5.9%	0.724
II	23	22.8%
III	56	55.4%
IV	16	15.8%
Tumor (T) stage
T1	9	8.9%	0.418
T2	13	12.9%
T3	53	52.5%
T4	26	25.7%
Node (N) stage
N0	35	34.7%	0.842
N1	66	65.3%
Metastasis (M) stage
M0	85	84.2%	0.761
M1	16	15.8%
Residual disease (R) classification
R0 (complete tumor resection)	57	57.6%	0.440
R1 or R2 (microscopic/	42	42.4%
macroscopic incomplete
tumor resection)
Differentiation
Poor	26	26.3%	0.977
Moderate	64	64.6%
Well	9	9.1%

3.3 Elevation of serum NID2 level is associated with poor survival in ESCC patients

The Cox proportional hazard model was utilized to determine the effect of elevated serum NID2 level on the overall survival of the ESCC patients. Based on the univariate analysis, increasing serum NID2 level is correlated with poorer survival with a hazard ratio (HR) of 1.02 (95% CI: 1.002–1.038, $p=$ 0.026) (Table 2). Table 2 also summarizes the outcome of the univariate Cox regression analysis for the other clinicopathological parameters of the ESCC patients. Among them, the tumor stage and residual disease (R classification) were confirmed as strong independent prognostic factors and were taken into account in the multivariable analysis (Table 3). After the adjustment, serum NID2 level remained as a significant prognosis factor, independent of the tumor stage or R classification. The adjusted hazard ratio is 1.018 (95% CI: 1.000–1.036, $p=$ 0.046) (Table 3).

Table 2
Univariate analysis for Cox regression model of risk of death in ESCC patients in the NID2 ELISA study

	Univariate analysis
Characteristics	Hazard	95% CI		$p$ -value
	ratio	Lower	Upper
Gender
Female	1	–	–	–
Male	1.495	0.906	2.469	0.116
Age at diagnosis (years)
Below 60	1	–	–	–
60 or above	1.005	0.627	1.609	0.984
Overall stage
Early (Stage I/Stage II)	1	–	–	–
Late (Stage III/Stage IV)	2.986	1.727	5.164	0.00009
Tumor (T) stage
T1/T2	1	–	–	–
T3/T4	1.493	0.883	2.523	0.135
Node (N) stage
N0	1	–	–	–
N1	2.121	1.302	3.456	0.003
Metastasis (M) stage
M0	1	–	–	–
M1	1.476	0.826	2.639	0.189
Residual disease (R) classification
R0 (complete tumor resection)	1	–	–	–
R1/R2 (microscopic/	2.260	1.440	3.545	0.0004
macroscopic incomplete
tumor resection)
Differentiation
Well	1	–	–	0.894
Moderate	1.209	0.546	2.678	0.639
Poor	1.166	0.491	2.765	0.728
Serum NID2 level (continuous)
Median $=$ 24.35 $\mu$ g/L	1.02	1.002	1.038	0.026

Table 3

Multivariate analysis for Cox regression model of risk of death in ESCC patients in the NID2 ELISA study

	Multivariate analysis
Prognostic factor (s)	Hazard	95% CI		$p$ -value
	ratio	Lower	Upper
Stage	2.501	1.399	4.469	0.002
Residual disease (R)	1.766	1.000	2.836	0.019
classification
Serum NID2 level	1.018	1.000	1.036	0.046

Figure 3.

Kaplan-Meier survival curves show that high serum NID2 level is associated with poor prognosis. The patients were stratified into three groups based on the serum NID2 level: low ( $<$ 20.0 $\mu$ g/L; $n=$ 34), medium ( $\geqslant$ 20.0 $\mu$ g/L, $<$ 32.6 $\mu$ g/L; $n=$ 33), and high ( $\geqslant$ 32.6 $\mu$ g/L; $n=$ 33). The Kaplan-Meier survival curve of high serum NID2 group is well separated from the curves of low and medium serum NID2 group (Log-rank $P=$ 0.012). For those with serum NID2 level greater than or equal to 32.6 $\mu$ g/L, there is a higher risk of death (HR $=$ 1.984, 95% CI: 1.175–3.349), compared to those below 20.0 $\mu$ g/L.

We further stratified the patients into three groups [low ( $n=$ 33), medium ( $n=$ 34) and high ( $n=$ 34)] according to their serum NID2 levels (measured in $\mu$ g/L), and Kaplan-Meier survival curves were plotted (Fig. 3). Cut-off values for different groups were taken based on the lowest and highest serum NID2 levels in each group. Patients in the “low” group have serum NID2 level lower than 20.0 $\mu$ g/L, in the “medium” group have serum NID2 level equal to or higher than 20.0 $\mu$ g/L, but lower than 20.0 $\mu$ g/L; while those with serum NID2 level equal to or higher than 32.60 $\mu$ g/L were classified into the “high” group. The survival curves for patients with low and medium serum NID2 levels ( $<$ 32.60 $\mu$ g/L) are clearly separated from the curve for patients with high serum NID2 level ( $\geqslant$ 32.60 $\mu$ g/L). We found that high serum NID2 level ( $\geqslant$ 32.60 $\mu$ g/L) is significantly associated with poorer survival (HR $=$ 1.984, 95% CI: 1.175–3.349) compared to patients with low serum NID2 level ( $<$ 20.0 $\mu$ g/L), with log-rank p value of 0.012. This suggested that presence of high serum NID2 levels in ESCC patients is predictive for poor prognosis.

4. Discussion

The quest to develop clinically useful biomarkers for cancer detection has always been the core focus of translational research. In this study, we evaluated the potential utility of detecting serum NID2 alterations for the diagnosis and prognosis of ESCC. With the ELISA screening for the serum samples, it was revealed that the serum NID2 levels in 101 ESCC patients were significantly elevated, when compared to the 50 healthy controls (Fig. 1A). Based on the ROC curve (Fig. 1B), this elevation of serum NID2 level can distinguish the ESCC patients from the healthy controls, with an AUC of 0.756 (95% CI: 0.68–0.84). This value of AUC falls within the range for a “fair” biomarker [11] and is comparable to the AUC of 0.73 (95% CI: 0.66–0.80) achieved in distinguishing the ovarian cancer patients from the healthy controls, as reported previously by Kuk et al. [9]. Moreover, the use of serum NID2 level for early diagnosis of ESCC also warrants further clinical validation, as it can be used to distinguish these patients from the healthy controls.

In a meta-analysis on the diagnostic value of the conventionally used ESCC biomarkers [3], high specificity of up to 100% can be seen with the use of CEA, SCCA or CYFRA21-1 for detection [3]. However, this is at the expense of the unsatisfactorily low and highly variable sensitivities reported across various ESCC biomarker studies [3]. The choice of the optimal cut-off is heavily reliant on the trade-off made between high specificity and high sensitivity. Among them, the reportedly most promising biomarker, CYFRA21-1, achieved 100% specificity, but at the expense of reduced sensitivity at 47.9% [12]. Other than these three biomarkers, the other more recently reported biomarkers for ESCC are still under evaluation and await actual clinical application. Among them, the diagnostic potential of YKL40, a secreted glycoprotein was investigated by Zheng et al. [6]. It was reported that the serum YKL40 level produces a much better ROC curve with an AUC of 0.874, in comparison with the AUCs of CEA, CYFRA21-1 and SCCA at 0.652, 0.746 and 0.789, respectively [6]. Despite having slightly lower specificity, YKL40 offers superior sensitivity over the other three biomarkers. In our study, although our results show that NID2 could distinguish between healthy controls and ESCC patients with comparable AUC with CEA, CYFRA21-1 and SCCA, but its specificity and sensitivity in general, do not significantly outperform over these existing tumor biomarkers and the potential biomarker, YKL40.

However, to improve the diagnosis or prognosis of cancer, there is now growing recognition for a need to develop a “biomarker signature” [11]. Use of multiple tumor biomarkers in a panel would enhance both the sensitivity and specificity, allowing a more robust screening to aid clinical decisions [6, 11]. As an example, Zheng et al. also explored the different combination of biomarkers and concluded that YKL40 in combination with SCCA offers the best sensitivity and accuracy for early stage ESCC diagnosis [6]. Given the promising findings on elevated serum NID2 level reported in this study, it is warranted to further evaluate the potential benefit of adding it into panels of other ESCC tumor biomarkers.

NID2 promoter hypermethylation has been frequently investigated for its diagnostic value in various malignancies [13, 14, 15, 16], but there were no reports on its prognostic value. Both decreased and elevated serum NID2 levels have previously been reported in ovarian cancer [9] and hepatocellular carcinoma [10], respectively, suggestive of its use as diagnostic biomarker for these cancers. Both studies did not explore the association of the altered serum NID2 levels with cancer patient survival [9, 10]. To the best of our knowledge, our study is the first to investigate the potential prognostic role of NID2 in predicting ESCC patient survival. The Cox proportional hazard model reveals that serum NID2 level is a prognostic factor for ESCC, independent of the tumor stage and R classification. ESCC patients with high serum NID2 levels ( $\geqslant$ 32.60 $\mu$ g/L) were associated with significantly higher risk of death compared to those with lower serum NID2 levels. With a HR of 1.984, these patients were predicted as having substantial increase in the risk of death with every 1.0 $\mu$ g/L increase in serum NID2 level. In this study however, we were unable to assess the impact of treatments administered on the prognosis of the ESCC patients, which can be potential confounder, due to a lack of comprehensive and accurate archived treatment records and patient’s performance status. Further validation in a larger clinical sample cohort with complete follow-up and assessment on treatments and patient performance status will be warranted to determine the clinical utility of serum NID2 levels, as a prognostic biomarker for overall survival of ESCC patients.

We previously reported that the NID2 promoter is hypermethylated in ESCC and its expression was confirmed to be down-regulated at the RNA level, but no convincing evidence on the expression changes in protein level is available [8]. As NID2 was shown to suppress cell clonogenicity, migration, and invasion, it was hypothesized to be a metastasis suppressor in ESCC [8]. This might seem contradictory to the current findings of elevated serum NID2 level among the ESCC patients. However, it is possible that the ESCC cells are not the source of the elevated NID2 in the patient sera. One possibility is that the down-regulation or the absence of NID2 in the ECM surrounding the ESCC cells are sensed by the other cells, such as the stromal cells or endothelial cells, triggering a feedback mechanism to replenish the NID2 in the ECM. As a consequence, more NID2 enters the blood circulation and is ultimately represented as elevated serum NID2 level in the ESCC patients. Another possible source is from the ECM at the metastasis site, where the relatively small size NID2 could be disintegrated from the disrupted ECM network during the remodeling process, and then be carried into the bloodstream. Elevated serum levels of other cancer-specific ECM turnover proteins have also been reported in other cancers, where they often represent a unique pathological fingerprint [17, 18, 19]. As a consequence of ECM remodeling, type IV collagen-derived fragments were reportedly elevated in serum concentration in colorectal cancer patients [19], as well as in a few other cancers [20, 21]. NID2, with a role in connecting the type IV collagen and laminin in the basement membrane network, might also be subjected to a similar fate during the deregulated ECM remodeling. However, these hypotheses require further investigation to delineate the mechanism behind the elevated serum NID2 level in ESCC patients, to determine whether it is merely correlated or has a causative effect.

In this study, we demonstrated that the serum NID2 levels show significant elevation in the ESCC patients and those with relatively high serum NID2 levels ( $\geqslant$ 32.6 $\mu$ g/L) are significantly associated with poor survival outcome compared with those with very low serum NID2 levels ( $<$ 20.0 $\mu$ g/L). This study provides novel insight into the potential utility to include serum NID2 in a biomarker panel for ESCC early diagnosis and warrants further investigation to explore its translational value as a prognostic biomarker for ESCC.

Footnotes

Acknowledgments

We thank the NPC AoE Tissue Bank for providing healthy control serum samples. This work was supported by the Research Grants Council of the Hong Kong Special Administrative Region, People’s Republic of China: General Research Fund grant number 17115214 to Prof. Maria Li Lung.

References

Wang

Tang

Sun

Tang

Xie

Wang

Gao

Cox

S.B.

and Wang

J.S.

, Etiological study of esophageal squamous cell carcinoma in an endemic region: a population-based case control study in Huaian, China, BMC Cancer 6 (2006), 287.

Lee

N.P.

Cheng

C.Y.

and Luk

J.M.

, New Advances on Disease Biomarkers and Molecular Targets in Biomedicine, Humana Press, 2013.

Zhang

Zhu

Z.L.

Liu

Jin

X.Y.

Z.W.

Q.Y.

and Li

, Diagnostic Value of Multiple Tumor Markers for Patients with Esophageal Carcinoma, Plos One 10 (2015).

Zhang

H.Q.

Wang

R.B.

Yan

H.J.

Zhao

Zhu

K.L.

Jiang

S.M.

X.G.

and Yu

J.M.

, Prognostic Significance of CYFRA21-1, CEA and Hemoglobin in Patients with Esophageal Squamous Cancer Undergoing Concurrent Chemoradiotherapy, Asian Pacific Journal of Cancer Prevention 13 (2012), 199–203.

Tsuchiya

Onda

Miyashita

and Sasajima

, Serum level of cytokeratin 19 fragment (CYFRA 21-1) indicates tumour stage and prognosis of squamous cell carcinoma of the oesophagus, Med Oncol 16 (1999), 31–37.

Zheng

Xing

Liu

X.M.

Liu

Chi

P.D.

Chen

Dai

S.Q.

Zhong

Zeng

M.S.

and Liu

W.L.

, Establishment of using serum YKL-40 and SCCA in combination for the diagnosis of patients with esophageal squamous cell carcinoma, BMC Cancer 14 (2014), 490.

Kohfeldt

Sasaki

Gohring

and Timpl

, Nidogen-2: A new basement membrane protein with diverse binding properties, Journal of Molecular Biology 282 (1998), 99–109.

Chai

A.W.Y.

Cheung

A.K.L.

Dai

J.M.Y.

J.C.Y.

Chan

K.W.

Kwong

D.L.W.

W.T.

Lee

A.W.M.

Ngan

R.K.C.

Yau

C.C.

Tung

S.Y.

Lee

V.H.F.

Lam

A.K.Y.

Pillai

Law

and Lung

M.L.

, Metastasis-suppressing NID2, an epigenetically-silenced gene, in the pathogenesis of nasopharyngeal carcinoma and esophageal squamous cell carcinoma, Oncotarget 7 (2016), 78859–78871.

Kuk

Gunawardana

C.G.

Soosaipillai

Kobayashi

Zheng

Y.Y.

and Diamandis

E.P.

, Nidogen-2: A new serum biomarker for ovarian cancer, Clinical Biochemistry 43 (2010), 355–361.

10.

Cheng

Z.X.

Huang

X.H.

Wang

Chen

J.S.

Zhang

L.J.

and Chen

X.L.

, Clinical significance of decreased nidogen-2 expression in the tumor tissue and serum of patients with hepatocellular carcinoma, Journal of Surgical Oncology 105 (2012), 71–80.

11.

Borrebaeck

C.A.

, Precision diagnostics: moving towards protein biomarker signatures of clinical utility in cancer, Nat Rev Cancer 17 (2017), 199–204.

12.

Yamamoto

Oka

Hayashi

Tangoku

Gondo

and Suzuki

, CYFRA 21-1 is a useful marker for esophageal squamous cell carcinoma, Cancer 79 (1997), 1647–1655.

13.

Renard

Joniau

van Cleynenbreugel

Collette

Naome

Vlassenbroeck

Nicolas

de Leval

Straub

Van Criekinge

Hamida

Hellel

Thomas

de Leval

Bierau

and Waltregny

, Identification and validation of the methylated TWIST1 and NID2 genes through real-time methylation-specific polymerase chain reaction assays for the noninvasive detection of primary bladder cancer in urine samples, Eur Urol 58 (2010), 96–104.

14.

Guerrero-Preston

Soudry

Acero

Orera

Moreno-Lopez

Macia-Colon

Jaffe

Berdasco

Ili-Gangas

Brebi-Mieville

Engstrom

Irizarry

R.A.

Esteller

Westra

Koch

Califano

and Sidransky

, NID2 and HOXA9 Promoter Hypermethylation as Biomarkers for Prevention and Early Detection in Oral Cavity Squamous Cell Carcinoma Tissues and Saliva, Cancer Prevention Research 4 (2011), 1061–1072.

15.

Geng

J.F.

Sun

J.F.

Lin

Zhao

Y.X.

Zhang

H.Y.

Feng

Y.H.

Wang

Zhou

X.Y.

and Yu

, Methylation status of NEUROG2 and NID2 improves the diagnosis of stage I NSCLC, Oncology Letters 3 (2012), 901–906.

16.

Yegin

Gunes

and Buyukalpelli

, Hypermethylation of TWIST1 and NID2 in Tumor Tissues and Voided Urine in Urinary Bladder Cancer Patients, DNA and Cell Biology 32 (2013), 386–392.

17.

Karsdal

M.A.

Delvin

and Christiansen

, Protein fingerprints – relying on and understanding the information of serological protein measurements, Clin Biochem 44 (2011), 1278–1279.

18.

Karsdal

M.A.

Nielsen

M.J.

Sand

J.M.

Henriksen

Genovese

Bay-Jensen

A.C.

Smith

Adamkewicz

J.I.

Christiansen

and Leeming

D.J.

, Extracellular matrix remodeling: the common denominator in connective tissue diseases. Possibilities for evaluation and current understanding of the matrix as more than a passive architecture, but a key player in tissue failure, Assay Drug Dev Technol 11 (2013), 70–92.

19.

Kehlet

S.N.

Sanz-Pamplona

Brix

Leeming

D.J.

Karsdal

M.A.

and Moreno

, Excessive collagen turnover products are released during colorectal cancer progression and elevated in serum from metastatic colorectal cancer patients, Sci Rep 6 (2016), 30599.

20.

Bager

C.L.

Willumsen

Leeming

D.J.

Smith

Karsdal

M.A.

Dornan

and Bay-Jensen

A.C.

, Collagen degradation products measured in serum can separate ovarian and breast cancer patients from healthy controls: A preliminary study, Cancer Biomark 15 (2015), 783–788.

21.

Willumsen

Bager

C.L.

Leeming

D.J.

Smith

Christiansen

Karsdal

M.A.

Dornan

and Bay-Jensen

A.C.

, Serum biomarkers reflecting specific tumor tissue remodeling processes are valuable diagnostic tools for lung cancer, Cancer Med 3 (2014), 1136–1145.

Elevated levels of serum nidogen-2 in esophageal squamous cell carcinoma

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Materials and methods

2.1 Serum samples

2.2 Enzyme-linked immunosorbent assay (ELISA)

3. Results

3.1 Elevated serum NID2 levels in ESCC patients

3.2 Clinicopathological characteristics of ESCC patients and NID2 serum levels

Table 1 Clinicopathological characteristics of 101 ESCC patients used in the NID2 ELISA study and association test results with serum NID2 level

Table 2 Univariate analysis for Cox regression model of risk of death in ESCC patients in the NID2 ELISA study

Footnotes

Acknowledgments

References

Table 1
Clinicopathological characteristics of 101 ESCC patients used in the NID2 ELISA study and association test results with serum NID2 level

Table 2
Univariate analysis for Cox regression model of risk of death in ESCC patients in the NID2 ELISA study