3.1 Literature Search
A total of 379 articles were initially retrieved from a database search, of them, 133 were from PubMed, 136 from Embase, and 110 from Web of Science. And then 198 duplicated articles were removed, the left 181 articles were evaluated by viewing the titles and abstracts and 159 were excluded according to the inclusion and exclusion criteria. Finally, twelve articles[8, 10, 18–27] were enrolled in this meta-analysis. The detailed process of selection was illustrated in Fig. 1.
3.2 Study Characteristics
The main characteristics of included studies are listed in Table 1. Twelve eligible articles containing 1461 patients who were diagnosed with 8 different types of cancers including ESCC[18], prostate cancer[10], HCC[20, 27], clear cell renal cell carcinoma (RCC)[21, 25], lung adenocarcinoma (LUAD)[22], CRC[26], CCA[8], and bladder cancer[19, 23, 24]. The studies were published from 2018 to 2021. Across the 12 studies, TRIP13 expression was detected in tumor tissue, detected methods included quantitative reverse transcription PCR (qRT-PCR) and immunohistochemistry (IHC) staining. All studies reported data on OS, however, only ten studies reported data on the correlation between TRIP13 expression and clinicopathological features. HR and its 95%CI were extracted directly in eight studies and estimated from survival curves indirectly in the other four studies. The NOS scores of all included studies were equal to or more than 6.
3.3 The association between TRIP13 expression and overall survival
As shown in Fig. 2, 12 studies including 1461 patients reported HRs for OS. The results indicated that high expression of TRIP13 was significantly associated with poor OS. There was no significant heterogeneity between studies (I2<50%, P>0.05 ), and the fixed-effects model was used to combine the HR (1.91, 95%CI 1.67-2.16, P<0.001).
In addition, subgroup analysis for cancer type, the detected method for TRIP13 expression, extract data method, race, and the sample size was performed and results were listed in Table 2. The subgroup analysis revealed that TRIP13 overexpression could predict unfavorable OS in digestive system (HR=1.90, 95%CI 1.47-2.33, P < 0.001) and urinary system (HR=2.05, 95%CI 1.69-2.42, P < 0.001). And we also found that TRIP13 significantly related to unfavorable OS in the studies with sample size < 100 (HR = 1.88, 95%CI 1.50-2.25, P < 0.001), as well as those with sample size ≥100 (HR = 1.94, 95%CI 1.61-2.27, P < 0.001). For the detection method of TRIP13 expression, the overall HR for the qRT-PCR group was 1.73 (95%CI 1.24-2.22, P < 0.001), for the IHC staining group was 1.98 (95%CI 1.69-2.26, P < 0.001). Whatever the extract method of HR, TRIP13 overexpression had a significant influence on OS (direct extraction group: HR=1.80, 95%CI 1.49-2.10; P < 0.001; indirect extraction group: HR=1.71, 95%CI 1.71-2.56, P < 0.001).
A sensitivity analysis was conducted to assess the reliability and stability of results, as shown in Fig.3, the pooled HR for OS was not influenced, which meant increased credibility. Additionally, Egger’s test and Begg’s test were applied to evaluate publication bias, the funnel plot was symmetry (Fig.4) and the P value of Egger’s test and Begg’s test was 0.707 and 0.891 respectively, which indicated no publication bias in this meta-analysis.
3.4 The association between TRIP13 and clinicopathological features
The association between high TRIP13 expression and clinicopathological features was summarized in Table 3. Meta-analysis results showed that high TRIP13 expression was correlated to T3+T4 tumor (pooled OR=2.86, 95%CI 1.38-5.94, P=0.005), lymph node metastasis (pooled OR=3.72, 95%CI 2.65-5.20, P<0.001), advanced TNM stages (pooled OR=2.57, 95% CI 1.87-3.53, P<0.001), big tumor diameter (pooled OR=1.92, 95%CI 1.04-3.53, P=0.037) and distant metastasis (pooled OR=2.64, 95%CI 1.42-4.91, P=0.002). However, high TRIP13 levels were not correlated to gender (pooled OR=1.00, 95%CI 0.76-1.31, P=0.985), age (pooled OR=1.25, 95%CI 0.99-1.58, P=0.058), and histological grade (pooled OR=1.18, 95%CI 0.91-1.54, P=0.214).
To explain the heterogeneity found in tumor size and tumor diameter, we performed sensitivity analysis and found that heterogeneity was caused by Li W’s study[22], however, the stability of results wasn’t significantly altered.
Publication bias was found in TNM stages (Egger’s test: P=0.001, Begg’s test: P=0.015), other clinicopathological parameters did not show evidence of publication bias. Therefore, the “trim-and-fill” method with a fixed-effect model was performed to assess the impact of potential publication bias in TNM stages. The adjusted result continued showing a significant association between high TRIP13 expression and advanced TNM stages (corrected OR = 2.27, 95%CI 1.70-3.03, P < 0.001).