Prognostic significance of CXCR7 in cancer patients: a meta-analysis

Abstract

Background: CXC chemokine receptor 7 (CXCR7) is frequently overexpressed in a variety of tumors. Nevertheless, whether CXCR7 can be used as a tumor prognosis marker has not been systematically assessed. The current meta-analysis was performed to obtain an accurate evaluation of the relationship between CXCR7 level and the prognosis of cancer patients.

Methods: Embase, Web of Science, and PubMed were systematically searched according to a defined search strategy up to June 11, 2018. Then, the required data were extracted from all qualified studies which were screened out based on the defined inclusion and exclusion criteria. Finally, the hazard ratios (HR) with 95% confidence intervals (CI) were used to evaluate the prognostic significance of CXCR7 in tumor patients.

Results: A total of 28 original research studies comprising 33 cohorts and 5685 patients were included in this meta-analysis. The results showed that CXCR7 overexpression was significantly related to worse overall survival (OS) (HR 1.72; 95% CI 1.49-1.99), disease-free survival (DFS) (HR 5.58; 95% CI 3.16-9.85), progression-free survival (PFS) (HR 2.83; 95% CI 1.66-4.85) and recurrence-free survival (RFS) (HR 1.58; 95% CI 1.34-1.88) in cancer patients. Furthermore, for certain types of cancer, significant associations between higher CXCR7 expression and worse OS of glioma (HR 1.77; 95% CI 1.43-2.19), breast cancer (HR 1.45; 95% CI 1.28-1.63), esophageal cancer (HR 2.72; 95% CI 1.11-6.66) and pancreatic cancer (HR 1.46; 95% CI 1.12-1.90) were found. However, for lung cancer and hepatocellular cancer, there was no significant relationship between CXCR7 expression level and OS, (HR 2.40; 95% CI 0.34-17.07) and (HR 1.37; 95% CI 0.84-2.24) respectively.

Conclusions: Increased CXCR7 level could predict poor prognosis of tumor patients and might be regarded as a novel prognostic biomarker for tumor patients.

Keywords: CXCR7; Cancer; Meta-analysis; Prognosis.

Fig. 1

The flow diagram indicated the process of study selection

Fig. 2

Meta-analysis of the pooled HRs of OS for cancer patients

Fig. 3

Results of subgroup analysis of pooled HRs of OS for cancer patients. a Subgroup analysis stratified by type of cancer. b Subgroup analysis stratified by sample size. c Subgroup analysis stratified by follow-up time. d Subgroup analysis stratified by the region. e Subgroup analysis stratified by source of HR. f Subgroup analysis stratified by NOS score

Fig. 4

Meta-analysis of the pooled HRs of OS for glioma (a), breast cancer (b), esophageal cancer (c), pancreatic cancer (d), lung cancer (e) and hepatocellular cancer (f)

Fig. 5

Meta-analysis of the pooled HRs of PFS (a), RFS (b) and DFS (c) for cancer patients

Fig. 6

Sensitivity analysis plot of pooled HRs of OS (a), RFS (b), DFS (c) and PFS (d) for cancer patients with abnormally expressed CXCR7

Fig. 7

Egger’s test (a) and Begg’s test (b) for publication bias. Trim and fill analysis of the eligible studies for the present meta-analysis (c)