3.1 Characteristics of relevant studies
Totally, data were collected on 29,972 participants (13,793 cancer subjects and 16,179 controls) from 30 case-control studies on FGFR4 G388R polymorphism. The most common types of cancer were prostate cancer (6 studies, n = 4,610), breast cancer (6 studies, n = 3,008), hepatocellular carcinoma (HCC) (3 studies, n = 2,481), oral squamous cell carcinoma (OSCC) (2 studies, n = 2,396), colorectal cancer (CRC) (3 studies, n = 2,349), lung cancer (2 studies, n = 899) (Table 1). In subgroup analysis by ethnicity, 15 studies were based on European population, 13 studies on Asian descendant, 1 study on Latins, and 1 study on Africans. For FGFR4 V10I variant, 7 studies with 9,369 subjects (4,377 cases and 4,992 controls) were identified in total. 4 studies focused on Asian population (n = 5,055), 2 on Europeans (n = 4,087), and 1 on Africans (n = 227). In stratification analysis by cancer type, 2 studies concentrated on prostate cancer. There was only one study each on cervical cancer, OSCC, breast cancer, HCC, and skin cancer. Moreover, minor allele frequencies (MAF) from the main worldwide population were also investigated. For FGFR4 G388R polymorphism, global population, 0.300; African descendants, 0.110; Americans, 0.310; East Asians, 0.463; Europeans, 0.294; South Asians, 0.390 (Fig. 1A). For FGFR4 V10I variant, global population, 0.229; African descendants, 0.018; Americans, 0.320; East Asians, 0.1; Europeans, 0.180; South Asians, 0.210 (Fig. 1B).
3.2 Main results
The overall result showed that FGFR4 G388R variant was associated with elevated susceptibility to cancer under homozygous comparison (OR = 1.21, 95%CI = 1.03–1.43, Pheterogeneity < 0.001, P = 0.020) and recessive genetic model (OR = 1.21, 95%CI = 1.04–1.41, P value for heterogeneity < 0.001, P = 0.012, Table 2). In stratification analysis by cancer type, similar positive findings were indicated in prostate cancer (allelic contrast, OR = 1.17, 95%CI = 1.07–1.29, Pheterogeneity = 0.183, P = 0.001; heterozygous, OR = 1.16, 95%CI = 1.02–1.32, Pheterogeneity = 0.714, P = 0.025; dominant model, OR = 1.20, 95%CI = 1.06–1. 35, Pheterogeneity = 0.892, P = 0.004, Fig. 2A) and breast cancer (allelic contrast, OR = 1.26, 95%CI = 1.13–1.41, Pheterogeneity = 0.622, P < 0.001; heterozygous comparison, OR = 1.25, 95%CI = 1.07–1.47, Pheterogeneity = 0.186, P = 0.005; homozygous comparison, OR = 1.73, 95%CI = 1.35–2.20, Pheterogeneity = 0.960, P < 0.001; recessive model, OR = 1.32, 95%CI = 1.14–1.54, Pheterogeneity = 0.197, P < 0.001; dominant model, OR = 1.46, 95%CI = 1.17–1.83, Pheterogeneity = 0.986, P = 0.001). In subgroup analysis by ethnicity, we observed that FGFR4 G388R variant was also correlated with increased risk to cancer in Asian descendants (RR vs. GG, OR = 1.29, 95%CI = 1.01–1.66, P value for heterogeneity < 0.001, P = 0.043; RR vs. RG + GG, OR = 1.28, 95%CI = 1.02–1.60, Pheterogeneity < 0.001, P = 0.034, Fig. 3A). For FGFR4 V10I polymorphism, no significant correlation was found when all studies pooled together (I vs. V, OR = 0.94, 95%CI = 0.85–1.04, P value for heterogeneity = 0.049, P = 0.227; IV vs. VV, OR = 0.97, 95%CI = 0.89–1.07, Pheterogeneity = 0.169, P = 0.601; II vs. VV, OR = 0.92, 95%CI = 0.72–1.17, P value for heterogeneity = 0.020, P = 0.488; II + IV vs. VV, OR = 0.95, 95%CI = 0.87–1.04, P value for heterogeneity = 0.147, P = 0.300, Fig. 2B; II vs. IV + VV, OR = 0.90, 95%CI = 0.74–1.11, P value for heterogeneity = 0.020, P = 0.328). Similar findings were indicated in subgroup analysis by cancer type. In stratification analysis by ethnicity, no positive association was identified in Asian population (II vs. IV + VV, OR = 0.85, 95%CI = 0.66–1.08, Pheterogeneity = 0.021, P = 0.184, Fig. 3B) and European descendants (II vs. IV + VV, OR = 1.08, 95%CI = 0.83–1.42, Pheterogeneity = 0.825, P = 0.563).
3.3 In silico and IHS analysis of FGFR4 expression
We used in silico tools to investigate whether the G388R and V10I mutation could affect the protein function of FGFR4. Polyphen2 bioinformatics analysis showed that FGFR4 G388R is predicted to be possibly damaging with a score of 0.700 to the protein function (Fig. 4A). Nevertheless, the V10I variation is predicted to be benign with a score less than 0.001 to the FGFR4 protein function (Fig. 4B). We also utilized online database to assess the expression of FGFR4 in prostate cancer participants. As described in Fig. 5A, the FGFR4 expression is elevated in prostate cancer, as compared to control. The Cancer Genome Atlas (TCGA) samples were also adopted to investigate the effect of FGFR4 expression on overall survival (OS) time. No significant difference on the OS time can be observed between the high FGFR4 expression group and low expression group (P > 0.05, Fig. 5B). In order to demonstrate the expression of FGFR4 in prostate cancer tissues, we applied IHS to evaluate its expression among prostate cancer patients in our centers. The feature distribution from prostate cancer volunteers has been provided in our previous article (Zhang et al., 2020). As described in Fig. 6, the expression of FGFR4 is up-regulated in more advanced cases, as compared to ones in early stage (T3 + T4 versus T1, P < 0.05). Moreover, the gene-gene correlation of FGFR4 was also assessed. At least 24 genes can participate in the interaction with FGFR4 (Fig. 7A). Among them, CORIN (corin, serine peptidase, Fig. 7B) is predicted to positively correlated with FGFR4.