The Expression Of GSTO1、LAMA5 And ITGA6 In Epithelial Ovarian Cancer Tissues And Its Relationship With Clinicopathological Features,prognosis And Immune Cell Infiltration

Objective: To investigate the expression levels of Glutathione S-transferase Ω1(GSTO1),Laminin α5(LAMA5),Integrin α6(ITGA6)in Epithelial Ovarian Cancer(EOC)and their relationship with clinicopathological features,prognosis and immune cell infiltration of patients.Methods:1.Quantitative analysis of N-glycosylation in Ovarian Cancer(OC)highly metastatic cells and parental cells: the research group sent three pairs of OC highly metastatic cells and parental cells(presented by Professor Wang Zehua’s team from Union Hospital,Tongji Medical College,Huazhong University of Science and Technology)to BGI for protein N-glycosylation detection and analysis(detection No.F19FTSNWWLJ7959＿J200109064).The results showed that there were significant differences in Nglycosylation modification among GSTO1,LAMA5 and ITGA6 in tumor pathways.Based on this,our group conducted in-depth studies on GSTO1,LAMA5 and ITGA6.2.Bioinformatics methods were used to explore the expression of GSTO1,LAMA5 and ITGA6 in OC and their relationship with clinicopathological features,prognosis and immune cell infiltration:(1)c Bioprotal database was used to investigate the gene mutations of GSTO1,LAMA5 and ITGA6 genes in OC.(2)GEPIA2 and GEO databases were used to analyze the expression levels of GSTO1,LAMA5 and ITGA6 between OC tissues and control group.(3)TCGA database was used to analyze the correlation between the expression levels of GSTO1,LAMA5 and ITGA6 and clinicopathological features of patients.(4)Kaplan-Meier plotter database was used to analyze the relationship between GSTO1,LAMA5 and ITGA6 gene expression and prognosis of patients.(5)STRING database was used to analyze the interaction among GSTO1,LAMA5 and ITGA6.(6)GSEA functional enrichment analysis was used to analyze the possible mechanism of GSTO1,LAMA5 and ITGA6 genes in OC.(7)GEPIA2 database was used to analyze the correlation between GSTO1,LAMA5,ITGA6 expression and PD-1,PD-L1,CTLA4 expression.TIMER database and Cibersort algorithm were used to analyze the relationship between the expression of the three genes and immune cell infiltration.Kaplan-Meier plotter database was used to analyze the relationship between the expression of the three genes and the prognosis of patients with different immune cell infiltration status.3.110 cases of EOC patients in the First Affiliated Hospital of Medical College of Shihezi University were selected as the experimental group From January 2010 to May 2022,and the pathological paraffin blocks were obtained to make tissue microarray.Thirty cases of normal fallopian tube fimbrial epithelial tissues were selected as the control group.4.Immunohistochemistry(IHC)staining was used to detect the expression of GSTO1,LAMA5 and ITGA6 in 110 EOC tissues and 30 normal controls.According to IHC staining results,the patients were divided into high expression group and low expression group,≥4 points were defined as high expression,< 4 points were low expression.5.We collected the clinicopathologic data of 110 EOC patients including age,FIGO stage,tumor histological Type(Type),tumor Grade(Grade),Lymph node metastasis(LNM),Lymphovascular space invasion(LVSI),CA125,HE4.The relationship between the expression of GSTO1,LAMA5,ITGA6 and the clinicopathological features of EOC was analyzed.6.A total of 110 EOC patients were followed up by telephone or outpatient/inpatient medical record system to obtain the recurrence and survival information of patients,and to calculate the Progression free survival(PFS)and Overall survival(OS)of patients.The relationship between the expression of GSTO1,LAMA5 and ITGA6 and the prognosis of patients was analyzed.Univariate and multivariate COX risk assessment models were used to analyze the related factors affecting the prognosis of patients.7.The clinicopathological data and IHC staining scores of GSTO1,LAMA5 and ITGA6 genes were used as research indicators to predict the recurrence or survival of patients in 1,3 and 5 years by drawing the Nomogram.Four algorithms of machine learning models were used to establish OC malignant risk prediction models.Results:1.Quantitative analysis of protein N-glycosylation: there was N-glycosylation modification in the tumor pathway between OC high-transformant cells and parental cells,among which GSTO1,LAMA5 and ITGA6 showed significant differences in N-glycosylation modification.2.Bioinformatics database results:(1)The c Bioprotal database showed that the mutation frequency of LAMA5 in OC was 15%,GSTO1 and ITGA6 were 1.9% and 6%.(2)GEPIA2 database showed the expression of LAMA5 in OC tissues was higher than that in the control group,and the difference was statistically significant(P < 0.05).There was no significant difference in GSTO1 and ITGA6 between the two groups(P > 0.05).GEO dataset analysis showed that the expression of LAMA5 and ITGA6 in OC was higher than that in normal control group,and the difference was statistically significant(all P < 0.001).(3)TCGA database showed that the expression levels of GSTO1 and ITGA6 in OC were correlated with FIGO stage,and the expression level of LAMA5 was correlated with tumor differentiation(all P < 0.05).(4)The results of Kaplan-Meier plotter database showed that high expressions of GSTO1 and LAMA5 were associated with poor prognosis of patients,and high expressions of GSTO1 and LAMA5 shortened the OS of patients(all P < 0.05).The high expression of ITGA6 significantly prolonged PFS and OS(all P< 0.001).(5)STRING database results showed that LAMA5 interacted with ITGA6.Analysis of gene expression data in the GEO database showed that LAMA5 and ITGA6 were co-expressed in OC tissues(P < 0.001).(6)Functional enrichment analysis showed that the high expressions of GSTO1,LAMA5 and ITGA6 genes in OC were not only related to the classical tumor signaling pathway,but also related to the immune function pathway.(7)GEPIA2 database showed that GSTO1,LAMA5 and ITGA6 genes were correlated with different immune checkpoints(PD1,PDL-1 and CTLA4)(all P < 0.05).Both TIMER database and CIBERSORT algorithm suggested that the expression levels of the three genes in OC were related to the infiltration of multiple immune cells(all P < 0.05).Searching the Kaplan-Meier plotter database found that the expression of the three genes in different immune cell subsets in OC patients was different,suggesting that different immune cell infiltration may be one of the reasons for different prognosis.3.Data verification results of 110 EOC patients included in the study:(1)The results of IHC showed that the expression of GSTO1 was high in 89 cases(80.9%)and low in 21cases(19.1%).There were 83 cases(75.5%)with high LAMA5 expression and 27 cases(24.5%)with low LAMA5 expression.There were 97 cases(88.2%)with high ITGA6 expression and 13 cases(11.8%)with low ITGA6 expression.Compared with the control group,the expression of GSTO1,LAMA5 and ITGA6 was increased in EOC(all P =0.000).(2)The expression of GSTO1 was correlated with the pathological types of EOC patients(χ2=4.133,P=0.042).The expression of LAMA5 was correlated with pathological type and tumor grade(χ2=14.858,15.707,both P =0.000).ITGA6 expression was associated with lymph node metastasis(χ2=6.654,P=0.036).(3)High expression of LAMA5 shortened PFS(P=0.046),while high expression of ITGA6 prolonged PFS(P=0.024),high expression of GSTO1 shortened PFS and OS(P=0.773,0.498),but the differences were not statistically significant.(4)The Nomogram prediction model showed that the clinicopathological indicators included in the prediction and the expression of GSTO1,LAMA5 and ITGA6 genes had good predictive efficacy for the prognosis of EOC patients.The human-computer interactive machine learning model showed that the clinicopathological indicators and the three genes included in the study had good clinical practice value for the malignant risk assessment of EOC patients.(5)Univariate and multivariate Cox risk assessment model showed that age,FIGO stage and low expression of ITGA6 were independent risk factors for PFS in EOC patients.Age and FIGO stage are independent risk factors for OS of EOC patients.Conclusions:1.Compared with the control group,GSTO1,LAMA5 and ITGA6 were highly expressed in EOC and correlated with some clinicopathological features.2.The high expression of LAMA5 in EOC is related to the poor prognosis of patients,and the high expression of ITGA6 in EOC has a better survival outcome of patients,and ITGA6 is an independent factor affecting the prognosis of patients.3.The high expression of GSTO1,LAMA5 and ITGA6 in EOC is correlated with the infiltration of a variety of different immune cells.The different prognosis outcomes of patients with high expression of LAMA5 and ITGA6 in EOC may be caused by the changes of immune microenvironment caused by the infiltration of different immune cells.