Study On The Application Of Prognosis Model Based On Multiple Serological Indicators For Advanced Colorectal Cancer

Objective: Radiotherapy,chemotherapy,targeted therapy,and immunotherapy significantly improved the survival time of patients with metastatic colorectal cancer,but the long-term survival is still unsatisfactory.Therefore,a simple and effective clinical evaluation index is urgently needed to predict the prognosis of m CRC.Although many prognostic factors of m CRC have been reported,there is no unified prognostic evaluation system except TNM stage.In this study,we will explore the prognostic factors of m CRC,and build the prognosis model of m CRC to provide the basis for individualized precision medicine of m CRC.Methods: A total of 94 patients with stage IV colorectal cancer were enrolled in the Department of Gastroenterology and Oncology of Affiliated Zhongshan Hospital Dalian University from October 1,2008,to October 1,2016.Clinical and laboratory indicators of each patient have been collected.94 m CRC patients were randomly divided into a training cohort(70 patients)and a validation cohort(24 patients).SPSS20.0 was used for statistical data analysis,Cox regression analysis was used to analyze the relationship between statistical variables and prognosis,all factors with P<0.05 were incorporated into the model,each factor was assigned by nomogram,the receiver operating characteristic curve(ROC)was used to compare the model with TNM stage and independent prognostic factors.Results: 1.Kaplan-Meier survival analysis showed that Circulating tumor cells(CTCs),Fibrinogen(FIB),Albumin(ALB),Eastern cooperative oncology group(ECOG)score,Carbohydrate antigen 19-9(CA19-9),Neutrophil to lymphocyte ratio(NLR),Platelet to lymphocyte ratio(PLR),D-dimer(D-D),Carcinoembryonic antigen(CEA),treatment and TNM stages were correlated overall survival in m CRC patients(P<0.05).Hemoglobin(Hb),Sex,Age were not relevant to overall survival of patients(P>0.05).2.Cox regression analysis in the training cohort showed that: FIB,CA19-9,CTCs,ECOG,and ALB were independent prognostic factors.Thus,the above five factors were included into nomogram model,and then calculating concordance index(C-index)of training cohort and validation cohort.The results showed that,C-index of the training cohort was 0.835,and C-index of validation cohort was 0.714,which showed that the prediction ability of the model was better.3.Prognostic scores of all patients were divided into four groups by quartiles: low-risk group(total scores<85),moderate-risk group(85?total scores<196),high-risk group(196?total scores<285),extremely high-risk group(total scores?285).Using Kaplan-Meier survival analysis,the median survival time of four groups were 31,24,16 and 7 months respectively,the differences were statistically significant(P<0.001).4.Calculation of area under the curve(AUC)by ROC curve,AUC of nomogram model,TNM stage,CTCs,FIB,ECOG,ALB and CA19-9 were 0.803,0.667,0.658,0.694,0.776,0.383 and 0.686,respectively.These results indicated that the prediction ability of this model was better.Conclusion: The model in this study includes 5 factors that have great a influence on the prognosis of m CRC,which can predict the survival time of patients,improve the accuracy of TNM stage for prognosis prediction.At the same time,the model is convenient and simple,can be widely used in clinical practice.