CXCR3 expression as a prognostic factor in gastric cancer: a meta-analysis.

BACKGROUND: CXC chemokine receptor 3 (CXCR3) plays a critical role in tumorigenesis, and CXCR3 expression is associated with prognosis in many cancers. Recently, CXCR3 expression is recognized as a potential prognostic factor for patients with gastric cancer. In this study, we analyzed the prognostic significance of CXCR3 expression in gastric cancer.
METHODS: We conducted a meta-analysis after selecting eligible studies through a literature search. We calculated pooled results to assess the associations between CXCR3 expression and overall survival (OS) and clinicopathological factors for gastric cancer.
RESULTS: The pooled hazard ratio (HR) with 95% confidence interval (CI) between high expression of CXCR3 and OS was 0.46 (95% CI 0.30-0.71, P<0.001), suggesting that high expression of CXCR3 was associated with a favorable OS. High expression of CXCR3 was significantly correlated with younger age [odds ratio (OR) 0.67, 95% CI: 0.49-0.91, P=0.011], lower tumor grade (OR 0.46, 95% CI: 0.29-0.73, P=0.001), absence of lymph node metastasis (OR 0.47, 95% CI: 0.31-0.71, P<0.001), and lower Tumor-Node-Metastasis stage (OR 0.51, 95% CI: 0.35-0.74, P<0.001).
CONCLUSIONS: High expression of CXCR3 was associated with better survival and may be a potential prognostic factor for patients with gastric cancer patients. 2021 Translational Cancer Research. All rights reserved.

Introduction

Gastric cancer is the fourth most prevalent cancer, and the second most common cause of cancer-related mortality worldwide (1,2). The prognosis of gastric cancer is rather unfavorable, with a five-year survival rate below 30% (2). Genetically and biologically, gastric cancer is a heterogeneous disorder, and several prognostic factors have been suggested, but their use is limited (3).

CXC chemokine receptor 3 (CXCR3) is a G-protein-coupled receptor shown to influence immune responses, generation of the vascular system, and the repair of wounds (4). Recently, CXCR3 has been shown to play a crucial role in tumorigenesis, and the expression of CXCR3 is associated with the prognosis of many cancers, such as colorectal, breast and ovarian cancer, renal cell carcinoma, and glioblastoma (5-11).

Although CXCR3 expression as a prognostic factor has been studied in gastric cancer, it has been demonstrated in smaller samples and has not been systematically assessed (12-16). Therefore, we performed a meta-analysis to establish the prognostic significance of CXCR3 expression in gastric cancer.

We present the following article in accordance with the PRISMA reporting checklist (available at http://dx.doi.org/10.21037/tcr-20-2862).

Methods

Literature search

We searched the literature using the following keywords: (CXCR3) and (cancer or tumor or carcinoma) and (prognostic or prognosis or survival). The search was conducted on June 1, 2020 using PubMed, Embase, and the Cochrane library, and the manual search was performed at the same time.

Selection criteria

It was included in the analysis only if the following criteria were met in the literature: (I) CXCR3 expression was identified using immunohistochemistry in the human cancer tissue of the stomach, (II) the hazard ratio (HR) and 95% confidence interval (CI) were established between CXCR3 expression and patient survival. Excluded were duplicated literature, reviews, conference abstracts, case reports, and non-English articles.

Data collection and quality assessment

The entire literature contained in the analysis was reviewed to gather the primary data for each literature. The primary data collected included the following: first author, year of publication, country, sample size, sex of patients, study and follow-up period, and CXCR3 expression cut-off. Two authors compiled the data, and differences were revised through consensus.

Quality evaluation was conducted on the literature included in the analysis, and the Newcastle-Ottawa Scale was used for quality evaluation. Quality evaluation was also performed independently by two authors, and disagreement resolution was revised through consensus.

Statistical analysis

Mata-analysis was conducted using StataSE12 (Stata, College Station, TX, USA). Pooled HR with 95% CI: was calculated to evaluate the correlations between CXCR3 expression and patient survival, and the I2 value was used to determine the heterogeneity between the included literature. The funnel plot was plotted to verify publication bias visually. Egger’s test was performed to validate the funnel plot’s statistical significance. Sensitivity analysis was performed to determine the reliability of the pooled results. Pooled odds ratio (OR) with 95% CI: was calculated to assess the associations between CXCR3 expression and clinicopathological factors. Statistical significance was defined as a P<0.05.

Results

Search results and primary data of studies

illustrates the process of selecting the included literature. All studies were published in China, and the study period was from 2008 to 2013. For a total of 716, the number of samples varied from 96 to 192. The included studies’ quality evaluation score was between six and eight, indicating good quality ().

Figure 1

Flow diagram of the study selection process.

Table 1

Basic data of the included studies

Study	Country	Sample size	Gender (male/female)	Study period	Median follow-up (months)	Clinical outcome	CXCR3 detection	Cut-off value of CXCR3 expression	NOS
Chen et al. [2019]	China	156	114/42	2008–2013	21.5	OS	IHC	Staining scores with intensity and extent (≥2)	8
Chen et al. [2018]	China	169	124/45	2008–2013	22	OS	IHC	Moderate staining and more than 25% of cells staining positive (≥2)	8
Zhou et al. [2016]	China	103	72/31	2006–2010	NA	OS	IHC	Positive staining	6
Hu et al. [2015]	China	96	69/27	2008–2013	NA	OS	IHC	Moderate staining and more than 25% of cells staining positive (≥2)	7
Li et al. [2015]	China	192	138/54	2008–2013	NA	OS	IHC	Moderate staining and more than 25% of cells staining positive (≥2)	7

CXCR3, CXC chemokine receptor 3; IHC, immunohistochemistry; NA, not available; NOS, Newcastle-Ottawa Scale; OS, overall survival.

Association between high expression of CXCR3 and overall survival (OS)

The included studies’ heterogeneity was so considerable that the pooled HR was calculated using a random-effects model (I2=62.7%, P=0.030). The results revealed that the elevated expression of CXCR3 was associated with a favorable OS in gastric cancer (HR 0.46, 95% CI: 0.30–0.71, P<0.001) ().

Figure 2

Forest plot of the association between CXCR3 expression and overall survival. CXCR3, CXC chemokine receptor 3.

Association between high expression of CXCR3 and clinicopathological factors

High CXCR3 expression was strongly associated with younger age (OR 0.67, 95% CI: 0.49–0.91, P=0.011), lower tumor grade (OR 0.46, 95% CI: 0.29–0.73, P=0.001), absence of lymph node metastasis (OR 0.47, 95% CI: 0.31–0.71, P<0.001), and lower Tumor-Node-Metastasis (TNM) stage (OR 0.51, 95% CI: 0.35–0.74, P<0.001). However, no substantial association was identified with sex, tumor size, Lauren classification, and tumor stage () ().

Table 2

The association between CXCR3 expression and clinicopathological factors in gastric cancer

Factor	Number of studies	Number of patients	Pooled OR (95% CI)	P value	Heterogeneity
					I2 (%)	P value	Model
Age (old vs. young)	5	716	0.67 (0.49–0.91)	0.011	0.0	0.983	Fixed
Gender (male vs. female)	5	716	0.90 (0.63–1.27)	0.541	0.0	0.418	Fixed
Tumor size (≥5 vs. <5 cm)	3	428	0.69 (0.45–1.04)	0.077	0.0	0.997	Fixed
Tumor grade (PD vs. WD MD)	3	428	0.46 (0.29–0.73)	0.001	0.0	0.599	Fixed
Lauren classification (diffuse vs. intestinal)	4	613	0.98 (0.71–1.36)	0.914	0.0	0.949	Fixed
Tumor stage (III IV vs. I II)	5	716	0.57 (0.19–1.71)	0.315	88.7	<0.001	Random
Lymph node metastasis (present vs. absent)	3	517	0.47 (0.31–0.71)	<0.001	0.0	0.427	Fixed
TNM stage (high vs. low)	4	613	0.51 (0.35–0.74)	<0.001	0.0	0.826	Fixed

CI, confidence interval; CXCR3, CXC chemokine receptor 3; MD, moderately-differentiated; OR, odds ratio; PD, poorly-differentiated; TNM, tumor-node-metastasis; WD, well-differentiated.

Figure 3

Forest plot of the association between CXCR3 expression and clinicopathological factors. (A) Age, (B) sex, (C) tumor size, (D) tumor grade, (E) Lauren classification, (F) tumor stage, (G) lymph node metastasis, and (H) tumor-node-metastasis stage. CXCR3, CXC chemokine receptor 3.

Publication bias

According to the funnel plot, it appeared that there were small study effects, but Egger’s test did not prove it (P=0.232) (). The filled funnel plot also showed that the data did not change (HR 0.46, 95% CI: 0.30–0.71, P<0.001) ().

Figure 4

Funnel plot (A) and filled funnel plot (B) of the association between CXCR3 expression and overall survival. CXCR3, CXC chemokine receptor 3.

Sensitivity analysis

Sensitivity analysis indicated that the study published by Zhou et al. (15) affected the overall results (HR 0.38, 95% CI: 0.29–0.50) (). However, the overall results were not substantially different (HR 0.44, 95% CI: 0.34–0.57), even after each study was excluded ().

Figure 5

Sensitivity analysis of the association between CXCR3 expression and overall survival. CXCR3, CXC chemokine receptor 3.

Discussion

This study showed a strong association between high expression of CXCR3 and a better prognosis than low expression in gastric cancer. We also demonstrated a correlation between CXCR3 expression and clinicopathological factors, which revealed a significant association between high CXCR3 expression and younger age, lower tumor grade, absent lymph node metastasis, and lower TNM stage.

CXCR3 is an interferon-inducible chemokine receptor expressed in various cells (17). CXCR3 comprises three isoforms, CXCR3-A, CXCR3-B, and CXCR3-Alt, in humans with distinct roles in cell biology and tumorigenesis (17). Of these, CXCR3-A and CXCR3-B are two well-studied isoforms (18). Cancer cells can control the expression of CXCR3 isoforms in a way that benefits overall and can decrease their proliferation and survival by overexpressing CXCR3-B and suppressing CXCR-A (18). Indeed, Hu et al. (16) demonstrated that CXCR3-B mRNA was substantially higher than CXCR3-A mRNA in gastric cancer tissue. CXCR3-B mRNA was significantly lower in patients with metastasis than in patients without metastasis. Therefore, our findings that high expression of CXCR3 indicates a favorable prognosis may have been derived from increased CXCR3-B in gastric cancer tissue. However, the mechanism underlying CXCR3 isoform expression remains to be elucidated, so further research is needed.

This analysis has some limitations. First, the number of studies included was small, making it challenging to analyze in various ways, including subgroup analysis. Second, all the included studies were published in China, so we need to consider whether this could apply to other regions.

However, in this study, we first presented the significance of CXCR3 expression as a prognostic factor in gastric cancer.