| Part I Survival Benefit of Palliative Surgery in Metastatic Colorectal Cancer Patients and the Construction of Prognostic-predicting ModelObjective: We aimed to explore the survival benefits of palliative resection of primary tumor for metastatic colorectal cancer,and construct the prognosticpredicting model in surgical as well as non-surgical groups.Methods: The patients who were diagnosed as metastatic colorectal cancer between 2010 and 2015 from the Surveillance Epidemiology and End Results database were reviewed retrospectively.Propensity score matching(PSM)was used to control the clinical and tumor characteristics between the surgical and non-surgical groups.Unmatched clinical features were stratified to compare the survival rate between the two groups.Overall survival between the two groups were evaluated by Kaplan–Meier estimator and log-rank tests.Univariate Cox regression method was used to select the significant variables predicting the prognosis.The OS prediction model was established by multi-factor Cox regression and visualized by nomogram.The concordance index(C-index)and calibration were used to assess the predictive performance of the nomogram.Results: 21464 patients met the study requirements were obtained through inclusion criteria.Following 1:1 matching by PSM analysis,7621 pairs of patients were generated,no significant difference in baseline characteristics including age,sex,race and tumor location were observed between the two groups.KaplanMeier survival analysis revealed that the overall survival of the surgery group was significantly higher than the non-surgery group whether before or after matching(P < 0.05).Subgroup analysis were conducted by tumor differentiation,chemotherapy retained,T and N stage,patients who received palliative surgery yield better overall survival than those with no surgery in the stratified analyses(P<0.05).By univariate Cox regression,age,marital status,tumor location,histologic grade,with or without chemotherapy,number of lymph nodes examined,lymph node ratio,T and N stage,CEA level were all significant predicting variables of overall survival in the surgery group which were included in the predictive models.In the non-surgery group,age,marital status,tumor location,histologic grade,CEA level were significant predicting variables selected by the univariate Cox regression.The nomogram of the two models were displayed.The model in surgery group displayed a acceptable discriminatory capacity with a C-index of 0.723 and the 95% confidence interval was 0.686–0.764,while C-index in non-surgery group was 0.620(95%CI 0.586-0.676).The result of C-index and calibration curves in surgery group showed good linear regression.Conclusions: Palliative resection of the primary tumor have survival benefits in metastatic colorectal cancer patients.The prognostic predicting model presented a reliable predictive ability for metastatic colorectal cancer patients who received palliative surgery.Part II Mechanism of Tumor Mutational Burden in Predicting the Efficacy of Immunotherapy for Colorectal Cancer and Construction a Prognostic Risk Score Model Based on Immune-related GenesObjective: Analyze the immune characteristics of tumor mutation burden(TMB)in colorectal cancer(CRC),and explore its predictive effect on CRC immunotherapy.Calculate the differentially expressed immune genes in different TMB groups,and construct a prognostic risk score model based on immunerelated genes.Methods: Gene expression profile,mutation and clinical data of colorectal cancer patients were obtained from The Cancer Genome Atlas(TCGA)database.Divide the samples into high and low TMB groups according to the cutoff value of TMB expression level of 20 mutations/Mb to identify differentially expressed genes(DEGs).Functional enrichments analysis were performed to identify the biological functions of the DEGs.And the biological functions of genes in high TMB group were analyzed by(Gene set enrichment analysis,GSEA)method.Then,immune cell infiltration signatures were calculated by the CIBERSORT algorithm.Finally,the prognostic immune-related genes in CRC patients were determined by univariate and multivariate Cox proportional risk regression models,then the prognostic risk score model based on the expression of immunerelated gene was constructed.Overall survival between the high and low risk score groups were evaluated by Kaplan–Meier estimator and log-rank tests.Results: The gene set enrichment analysis resulted that DEGs in the high and low TMB groups were enriched in immune-related pathways,including cytokine-cytokine receptor interaction,chemokine signaling pathway,cell adhesion molecules,natural killer cell mediated cytotoxicity,antigen processing and presentation as well as Th17,T1,T2 cell differentiation.Gene set enrichment analysis in the high-TMB level group showed that DEGS were enriched in immune-related pathways,such as antigen processing and presentation,Toll-like receptor signaling and natural killer cell-mediated cytotoxicity.The result of CIBERSORT algorithm showed that a higher infiltration level of CD8+ T cells,CD4+ T cells,activated NK cells,M1 Macrophages and T follicular helper cells were observed in the high-TMB level group.Furthermore,univariate and multivariate Cox proportional hazards analysis established an immune-related prognostic signature consisting of AMH,DHX58,EPOR and TNFRSF19.The area under the receiver operating characteristic(ROC)curve was 0.707 in the model sample.Then the patients were classified by prognostic risk score based on the 4-gene signature,Kaplan–Meier curves showed that the high-risk score group seems to have significantly worse outcome compared to the low-risk score group(P<0.05).Univariate and multivariate Cox analysis showed that the risk score based on immune genes was an independent prognostic factor for CRC(HR=3.53,95% CI=2.136-5.836,P<0.001).Conclusions: Our data demonstrate that the high TMB levels are associated with the immune signature in CRC.A prognostic risk score model based on immune-related genes can be used as an independent predictor for CRC. |
PDF Full Text Request