[Expression and clinical significance of FAT1 gene in pancreatic adenocarcinoma]

Objective: To analyze the expression of FAT1 gene in pancreatic adenocarcinoma and its relationship with clinicopathological features, prognosis, and immunotherapy for pancreatic adenocarcinoma. Methods: (1) Bioinformatics analysis: based on FAT1 mRNA expression and clinical data of 179 cases of pancreatic adenocarcinoma in the TCGA database, and FAT1 mRNA expression data of 328 cases of normal pancreatic tissues in the GTEx database. We analyzed the differences in FAT1 mRNA expression in pancreatic adenocarcinoma and normal pancreatic tissues and the relationship between FAT1 mRNA expression and the degree of differentiation, clinical stage, prognosis, immune cell infiltration, and immune checkpoint-associated genes in pancreatic adenocarcinoma. FAT1-related differentially expressed genes were analyzed by applying Limma 3.40.2 software package, and GO and KEGG enrichment analysis was performed on the differentially expressed genes. Immunohistochemical (IHC) of FAT1 in pancreatic adenocarcinoma and normal pancreatic tissues was analyzed by HPA database. (2) Validation of own tissue samples: tissue samples and clinical and prognostic data of 192 patients with pancreatic ductal adenocarcinoma admitted to Zhejiang Cancer Hospital from March 8, 2010 to September 30, 2020 were collected. IHC was performed on the tissue samples to verify the protein expression of FAT1 in pancreatic adenocarcinoma and its relationship with immune-related proteins, the degree of differentiation of pancreatic adenocarcinoma, clinical staging, and prognosis. Results: (1) Bioinformatics analysis: the FAT1 mRNA expression of 179 pancreatic adenocarcinoma tissues from the TCGA database was 5.55±1.04, which was higher than that of 328 normal pancreatic tissues with FAT1 mRNA from the GTEx database (2.95±0.53, P＜0.001). FAT1-specific IHC images showed that FAT1 expression was generally high in pancreatic adenocarcinoma tissues, and FAT1 expression shifted from the cell membrane to the cytoplasm. The FAT1 mRNA expression in the highly differentiated group (31 cases), the moderately differentiated group (96 cases), and the lowly differentiated group (52 cases) were 4.99±1.46, 5.51±0.80, and 5.68±1.08, the expression of pancreatic adenocarcinoma tissues were all higher than that of normal pancreatic tissues (all P＜0.001), and the FAT1 mRNA expression of the moderately differentiated group and the poorly differentiated group were all higher than that of the highly differentiated group (all P＜0.001). The median progression-free survival time (PFS) and median overall survival time (OS) of the 90 patients in the FAT1 mRNA low-expression group were 16.5 and 24 months, respectively, which were longer than those of the 89 patients in the FAT1 mRNA high-expression group (median PFS and OS were 13 and 18 months, respectively; P-values were 0.011 and 0.005, respectively). Multifactorial Cox regression analysis showed that FAT1 mRNA expression level was an independent influencing factor for OS in pancreatic adenocarcinoma patients (HR=1.47, 95% CI: 1.09-1.99). Correlation analysis showed that FAT1 mRNA expression in pancreatic adenocarcinoma was positively correlated with B-cell infiltration, CD8+ T-cell infiltration, neutrophil infiltration, macrophage infiltration, and myeloid dendritic cell infiltration (ρ=0.27, P＜0.001; ρ=0.28, P＜0.001; ρ=0.32, P＜0.001; ρ=0.21, P=0.004; ρ=0.32, P＜0.001), and also positively correlated with mRNA expression of CD274, HAVCR2, and PDCD1LG2 (r=0.327, P＜0.001; r=0.231, P=0.002; r=0.258, P＜0.001). GO and KEGG enrichment analyses showed that FAT1 mRNA expression levels were associated with activation of the Wnt signaling pathway (P=0.029), the PI3K/Akt pathway (P<0.001), and other tumor microenvironment-related pathways. (2) Validation of own tissue samples: among 192 pancreatic adenocarcinoma tissues, FAT1 was highly expressed in 58 cases (30.21%), and the proportion of FAT1-expressing positive tumor cells was positively correlated with the combined positive score of PD-L1 and the number of CD3+ T-cells infiltration (r=0.154, P=0.032; r=0.287, P＜0.001), and the protein expression of FAT1 had no correlation with the differentiation degree of pancreatic adenocarcinoma (ρ=0.082, P=0.254). The median OS of 58 patients in the FAT1 high-expression group and 134 patients in the FAT1 low-expression group were 18.89 and 25.84 months, respectively, and the difference was not statistically significant (χ²=1.93, P=0.165). Conclusion: FAT1 gene is highly expressed in pancreatic adenocarcinoma tissues, may play an oncogenic role in pancreatic adenocarcinoma, may be an adverse influence on overall survival and progression-free survival of patients; FAT1 gene may be involved in multiple immune-related pathways and promote tumor immune escape.