RHOA protein expression correlates with clinical features in gastric cancer: a systematic review and meta-analysis

Abstract

Background: Gastric cancer (GC) is one of the most fatal cancers worldwide and is generally only detected after it has progressed to an advanced stage. Since there is a lack of comprehensive data on RHOA protein expression of patients with GC, this study utilized a systematic review and meta-analysis to address the limitation. The objective of this meta-analysis was to link GC clinical features with RHOA protein high- vs. low-expressing patients with GC.

Methods: The PubMed and Web of Science were used for a systematic literature review of GC related to RHOA. The included studies were obtained from two literature databases from past to Aug 31, 2021, by searching keywords. This meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. The odds ratios (ORs) and 95% confidential intervals (CIs) for clinical features were estimated according to the high and low protein expression levels of RhoA. The mean effect sizes of ORs were obtained using the random-effects and fixed-effects models of meta-analysis. Heterogeneity of the studies was assesed by using statistics: τ², I²; and Q values. The symmetry of funnel plots were inspected for publication bias.

Results: Finally, 10 studies including 1,389 patients with GC (735 RHOA-positive and 654 RHOA-negative) were eligible for our meta-analysis to estimate associations between the protein expression and clinical features (e.g., Union for International Cancer Control [UICC] stage progression, differentiation, Lauren histological classification, and vascular invasion). In our meta-analysis, RHOA positive expression was determined to have a statistically significant association with UICC stage progression (P = 0.02) and poorly differentiated status (P = 0.02). The association between RHOA positivity and Lauren subtypes was not statistically significant (P = 0.07).

Conclusions: This meta-analysis suggested that RhoA protein expression in patients with GC was associated with clinical features: UICC stage progression and poorly differentiated status. Our findings are inconclusive but indicate that high RHOA protein expressing patients with GC could predict advanced UICC stages. A large prospective cohort study is required for validation in future.

Keywords: Gastric cancer; Immunohistochemistry; Meta-analysis; RHOA.

Fig. 1

Flow of systematic publication selection processes

Fig. 2

Meta-analysis on clinical parameters. The first column indicates study names; the second column indicates experimental group; the third column indicates control group; the fourth column indicates forest plot; the fifth column indicates odds ratios (ORs) of RHOA protein high- vs. low-expressing patients in the experimental group vs. the control group, and 95% confidence intervals (CI); and the sixth column indicates weight. Events indicate RHOA protein high expressing patients with GC (equivalently, RHOA positive patients with GC). Given a clinical feature, one overall pooled effect size of OR for RHOA high- and low-expressing patients was obtained. Also, heterogeneity was measured by between-study variance τ², Higgins’ I² and Cochran’s Q-tests. A Union for International Cancer Control (UICC) stages III–IV (experimental group) vs. I–II (control group). In each group, events (i.e., RHOA protein high-expressing patients with GC) were obtained from each study. The overall effect estimate indicates that the OR of RHOA protein high expression over low expression between the two groups was greater than one. Thus, RHOA protein high expressing patients in the experimental group (i.e., stages III–IV) are more prevalent than in the control group (I–II). In other words, RHOA protein positivity is likely to be advanced UICC stages (i.e., UICC stage progression). B Poorly vs. “well plus moderately differentiated” types. C Lauren subtypes diffuse vs. Lauren intestinal. D Vascular invasion statuses of yes vs. no

Fig. 3

Sensitivity analyses of the meta-analysis results. A UICC stages III–IV vs. I–II. B Poorly vs. “well plus moderately differentiated” types

Fig. 4

Funnel plots for inspecting publication biases. A UICC stages III–IV versus I–II. B Poorly vs. “well plus moderately differentiated” types