Construction Of Gene Mutation-associated Nomogram And Its Application In Immunotherapy For Hepatocellular Carcinoma

ObjectivesHepatocellular carcinoma(HCC)has high malignancy,and its high morbidity and mortality have attracted extensive attention from scholars at home and abroad.In the micro aspect,gene mutations play a key role in the evolution of carcinoma.From a macro perspective,the living habits,economic level and medical conditions of different countries may affect the prognosis of HCC patients.This study collected clinical cohorts from multiple countries and aimed to develop a nomogram based on gene mutations and clinical features,so as to predict the overall survival of HCC patients and provide reference for immunotherapy.MethodsThis study collected five HCC patient cohorts from four countries based on public databases,namely The Cancer Genome Atlas and International Cancer Genome Consortium.Patients in each cohort were split into two parts in a ratio of 7:3 as training and validation sets.Univariate and multivariate cox analysis were used to select significant variables for construction of nomogram.In addition,correlation assessment of the nomogram model was performed by concordance index,receiver operating characteristic curve,calibration curve and decision curve analysis.Furthermore,principal component analysis in machine learning was used to evaluate the model discrimination degree of nomogram,and support vector machine was applied to assess the key role of specific mutant genes included in the model in the occurrence and development of carcinoma.Patients were separated into high-risk and low-risk group through the risk score of nomogram.Besides,Gene Set Enrichment Analysis,immune cell infiltration assessment,Tumor Immune Dysfunction and Exclusion(TIDE)and immunophenoscore(IPS)were utilized to explore the potential mechanism of immune-related process and immunotherapy.Results1.A total of 695 HCC patients who met the criteria were selected,which contains 495 patients in the training set and 200 patients in the validation set.Nomogram was constructed based on T stage,age,nationality,mutation status of DOCK2,EYS,MACF1 and TP53.Furthermore,concordance index,receiver operating characteristic curve,calibration curve and decision curve analysis of the training and validation sets prove that the model has excellent predictive ability and the clinical benefit of the model is still good.2.Gene Set Enrichment Analysis showed that the enrichment pathways in the low-risk group were mostly normal physiological functions of hepatocytes.However,the enrichment pathways in the high-risk group were mostly malignant biological manifestations of tumors,confirming that the grouping method based on the model risk score is reasonable.At the same time,a large number of immune-related pathways were enriched differently between high and low-risk groups,indicating that the model grouping was related to tumor immunology.3.In the immune assessment of the high-risk group,the TIDE score results showed that the tumor microenvironment in the high-risk group had significantly enhanced exclusion of CD8+T cells compared with the low-risk group.Correspondingly,the CIBERSORT algorithm was employed to assess the infiltration ratio of 22 immune cells,and it was observed that the proportion of CD8+T cells in the tumor microenvironment of the high-risk group diminished significantly.At the same time,combined with the lower IPS score in the high-risk group,it is speculated that the patients in the high-risk group are not suitable for immune checkpoint inhibitors(ICI)therapy.4.In the immune assessment of the low-risk group,the TIDE score was lower than that of the high-risk group,indicating that CD8+ T cells can successfully exert anti-tumor effects in the immune microenvironment of the low-risk group.At the same time,the response prediction results of ICI treatment showed that 53.4%of patients in the low-risk group responded to ICI treatment.which was much higher than 35.2%in the high-risk group.Combined with the IPS score,it was concluded that the low-risk group was sensitive to cytotoxic T lymphocyte associated protein-4(CTLA-4)inhibitors.5.In the support vector machine model,DOCK2 has the highest marginal contribution.In the assessment of immune cell infiltration and the prediction of ICI efficacy,the increase of DOCK2 transcription level can promote the recruitment of regulatory T cells and enhance the inhibitory effect of tumor microenvironment on CD8+T cells.The group with high DOCK2 expression can benefit from ICI treatment and is also sensitive to CTLA-4 inhibitors.Conclusions1.Our research established a nomogram based on mutant genes and clinical characteristics,which is beneficial to forecast the survival prognosis of HCC patients.2.Patients were divided into high-risk and low-risk groups according to the risk score of the nomogram.T cell exclusion is a potential mechanism of HCC malignant progression in high-risk group,which recommends that ICI therapy is inapplicable to this population.However,low-risk group may be sensitive to immunotherapy,especially CTLA-4 inhibitors.