Based On The Clinical Data Of 80 Patients With Advanced Gastric Cancer, A Study On The Prediction Model Of Gastric Cancer Prognosis Was Constructed

Objectives:(1)Based on the collection of related data,picked out Ⅳ period a risk factor for gastric cancer patients survival time and protection factors,preliminary build with nomogram for presentation Ⅳ stage gastric cancer prognosis prediction model.(2)Preliminary definition of "benefit(advantage)population for advanced gastric cancer combined with traditional Chinese and western medicine"(hereinafter referred to as "beneficial population")and "non-benefit(disadvantaged)population for advanced gastric cancer with integrated traditional Chinese and western medicine"(hereinafter referred to as "non-beneficial population")Based on this,the patients were divided into three categories:"beneficiary group1","non-beneficiary group" and "middleman group",and a discriminative model of the stage Ⅳpatient population category was initially constructed.Methods:Based on December 1,2016 to December 31,2018 in the Dongzhimen Hospital of Beijing University of Chinese Medicine,Beijing Huanxing Cancer Hospital,and Beijing Sanhuan Cancer Hospital patients with advanced gastric cancer,the sifting into standard of 80 patients with advanced gastric cancer,its general information,clinical disease information collection,TCM clinical data,etc.,in February and a follow-up observation.Deaths caused by cancer as the main event,OS Ⅳ period as the main outcome indicators,using Cox regression analysis filter factors affecting Ⅳ period the prognosis of gastric cancer patients survival time,and based on prognostic factors,categories,using multivariable Cox regression analysis and linear discriminant analysis,construct the nomogram survival prediction and the discriminant model of categories.Results:During the research period from December 1st,2016 to December 31st,2018,93 cases of advanced gastric cancer were included in the study group,13 cases were excluded according to the exclusion and withdrawal criteria,and 80 cases were finally included in the statistical analysis.As of December 31st,2018,62 of the 80 patients had died and 18 had shed.80 patients were diagnosed Ⅳ period after the survival time in the range of 1.7～28.9 months,an average Ⅳ period survival time 11.8 months,the median phase Ⅳ survival time 11.6 months.(1)Univariate Cox regression analysis showed that HR with BMI greater than 18kg/m2 was 0.41(95%CI:0.22-0.74,P=0.003),HR with KPS score of>70 was 0.47(95%CI:0.28-0.81,P=0.007),HR with medio-high differentiation was 0.54(95%CI:0.31-0.94,P=0.03),and HR with second-line and follow-up treatment was 0.25(95%CI:0.13-0.45,P<0.001),HR for 3 months or more after receiving TCM treatment was 0.33(95%CI:0.19-0.56,P<0.001),all of which were prognostic protective factors.HR with diabetes was 2.08(95%CI:1.01-4.29,P=0.048),HR with signet-ring cell carcinoma was 2.38(95%CI:1.15-4.91,P=0.019),HR with hemoglobin≤94g/L was 2.81(95%CI 1.61-4.92,P<0.001),and HR with serum albumin≤38.2g/L was 2.16(95%CI:1.3-3.6,P=0.003),all of which were prognostic risk factors.(2)Multivariable Cox regression analysis showed that:the KPS score>70 HR of 0.46(95%CI:0.24 0.9,P=0.024),with high-differentiation in HR of0.48(95%CI:0.23 0.99,P=0.047),the second line and subsequent treatment of the HR of 0.30(95%CI:0.14 0.65,P=0.002),the intervention of traditional Chinese medicine more than 3 month in the HR of 0.27(95%CI:0.14 0.53,P<0.001),are all independent protective factors related to prognosis.(3)Nomogram:According to the four independent prognostic factors,a model is constructed to predict the survival probability of patients with advanced gastric cancer and a nomogram is drawn.The results shows that the consistency index of the model is 0.79,indicating that the model has achieved high accuracy.The calibration curve shows that the model has good fitting performance.(4)Discrimination Model:The linear discriminant model was completely correct in predicting the categories of 57 patients with a total prediction accuracy of 71.25%.81.08%of the beneficiaries were accurately predicted to be the beneficiaries,83.33%of the non-beneficiaries were accurately predicted to be the non-beneficiaries,and 59.46%of the intermediate population were accurately predicted to be the intermediate population.The linear discriminant model has a high recognition rate for the beneficiary and the non-beneficiary.CONCLUSION:Better physical condition and differentiation,Chinese medicine treatment for 3 months or more,and second-line and follow-up treatment for qualified patients are independent protective factors for prognosis of advanced gastric cancer.