Poor prognosis of hexokinase 2 overexpression in solid tumors of digestive system: a meta-analysis

Abstract

Several previous studies have reported the prognostic value of hexokinase 2 (HK2) in digestive system tumors. However, these studies were limited by the small sample sizes and the results were inconsistent among them. Therefore, we conducted a meta-analysis based on 15 studies with 1932 patients to assess the relationship between HK2 overexpression and overall survival (OS) of digestive system malignancies. The relationship of HK2 and clinicopathological features was also evaluated. Hazard ratio (HR) or odds ratio (OR) with its 95% confidence intervals (CI) were calculated to estimate the effect size. Positive HK2 expression showed poor OS in all tumor types (HR = 1.75 [1.41-2.18], P < 0.001). When stratified by tumor type, the impact of HK2 overexpression on poor prognosis was observed in gastric cancer (HR = 1.77 [1.25-2.50], P < 0.001), hepatocellular carcinoma (HR = 1.87 [1.58-2.21], P < 0.001), and colorectal cancer (HR = 2.89 [1.62-5.15], P < 0.001), but not in pancreatic ductal adencarcinoma (HR = 1.11 [0.58-2.11], P = 0.763). Furthermore, high HK2 expression was significantly associated with some phenotypes of tumor aggressiveness, such as large tumor size (OR = 2.03 [1.10-3.74], P = 0.024), positive lymph node metastasis (OR = 2.05 [1.39-3.02], P < 0.001), advanced clinical stage (OR = 2.17 [1.21-3.89], P = 0.009) and high alpha fetoprotein level (OR = 1.47 [1.09-2.02] P = 0.013). In summary, HK2 might act as a prognostic indicator and a potential therapeutic target of these digestive system cancers.

Keywords: digestive system; hexokinase 2; meta-analysis; prognosis; solid tumors.

Figure 1. Flow diagram of the study selection process and specific reasons for exclusion in the meta-analysis

183 studies were preretrieved in accordance with the established search strategies. Of these articles, 99 were excluded because of clear lack of relevance. The remaining 84 studies were further screened out through browsing the titles and abstracts, and then 47 were removed based on the eligible criteria. After reading the full texts of 37 studies, 15 eligible studies were finally included in this meta-analysis.

Figure 2. Forest plots of the overall outcome for overall survival (OS) in solid tumors of digestive system

Hazard ratios (HRs) for each trial are represented by the squares, and the horizontal lines crossing the square stand for the 95% confidence intervals (CIs). The diamonds represent the estimated pooled effect of the overall outcome for OS in all solid tumors of digestive system. All P values are two-sided.

Figure 3. Cumulative meta-analysis of HK2 expression and OS in solid tumors of digestive system

The changes of the pooled hazard ratios (HRs) over time are represented by the squares, and the horizontal lines crossing the square stand for the coresponding 95% confidence intervals (CIs).

Figure 4. Effect of individual studies on pooled hazard ratios (HRs) for the relationship between HK2 expression and prognosis of digestive system tumors

The vertical axis at 1.75 indicates the overall HR, and the two vertical axes at 1.41 and 2.18 indicate its 95% confidence interval (CI). Every hollow round indicates the pooled HR when the left study was omitted in a meta-analysis with a random model. The two ends of every broken line represent the respective 95% CI.

Figure 5. Begg's funnel plots for potential publication bias of studies reporting overall survival included in the meta-analysis

Each included study represented by one circle. The horizontal line represented the pooled effect estimate.