Study Of Prediction Model For Short-term Efficacy Of Abatacept Combined With Methotrexate In The Treatment Of Rheumatoid Arthritis

Objective:Rheumatoid arthritis(RA)is an autoimmune disease,which is mainly manifested as chronic arthritis.Its symptoms mainly include joint pain,swelling,deformation and movement disorders,which ultimately leads to disability and greatly affects the quality of life of patients.The treatment principle of RA is to achieve the remission of clinical symptoms in a short time.According to the clinical indicators of patients,it is one of the main issues concerned by rheumatologists at home and abroad to predict the patients’ response to the clinical efficacy.Therefore,this study analyzes the factors related to the short-term efficacy of abatacept combined with methotrexate in the treatment of rheumatoid arthritis,builds a short-term efficacy prediction model,and carries out individualized diagnosis and treatment of RA.Methods:The RA patients who visited the outpatient and inpatient departments of the Rheumatology Department of the Zhejiang Provincial People’s Hospital from May 2020 to October 2022 were retrospectively collected,and the RA classification standard(2010ACR/EULAR standard)published by the American Association of Rheumatology(ACR)and the European Union against Rheumatology(EULAR)in 2010 was used.A total of 120 RA patients were included in this study according to the inclusion and exclusion criteria.120 patients were randomly assigned to the modeling group and the test group using R language,and 70 patients in the modeling group;There were 50 patients in the model group.All patients in the group were treated with abatacept universal methotrexate(MTX).In the course of the study,the patient’s general information was inquired and recorded in detail:age,sex,course of disease,BMI;Test indicators:C-reactive protein(CRP(0),CRP(1)and CRP(3))at the initial baseline and after 1 month and 3 months of treatment,anti-cyclic citrulline peptide antibody(anti-CCP antibody(0),anti-CCP antibody(1)and anti-CCP antibody(3)at the initial baseline and after 1 month and 3 months of treatment,rheumatoid factor(RF(0),RF(1)and RF(3)at the initial baseline and after 1 month and 3 months of treatment;Efficacy evaluation index:DAS28 scores(DAS28(0)and DAS28(3))at the initial baseline and after 3 months of treatment;Combined drug use(hydroxychloroquine,non-steroidal anti-inflammatory drugs,and iguratimod);Treatment outcome.Using R language,variables were screened according to logistic regression,independent risk factors were determined,short-term clinical efficacy prediction model was established,and the effectiveness of the model was evaluated using ROC curve.Results:A total of 120 patients were included in this study,with significantly more female patients than male patients.In general data,there was a significant statistical difference in age between the two groups of patients,while there was no significant statistical difference in gender,course of disease,and BMI;There was no significant statistical difference in the test indicators of anti CCP antibody(0),anti CCP antibody(1),RF(0),RF(1),and CRP(3);There was significant statistical significance in the combined use of eramod,but there was no significant statistical significance in the use of hydroxychloroquine and non steroidal anti-inflammatory drugs;There was no significant statistical difference in the clinical efficacy score of DAS28(3).Logistic regression algorithm is used to establish the label of the prediction model,and there is a certain correlation between different variables.Dimension reduction processing is conducted to optimize the prediction of clinical efficacy and prognostic factors.In the resulting logistic regression clinical prediction model,age,the use of Elamod,DAS28(0),CRP(1),RF(3),and anti CCP antibody(3)are the preferred predictors(P<0.05).Due to the poor timeliness of the test indicators at the end of the third month of predictive analysis treatment,RF(3)and anti CCP antibody(3)were removed.Therefore,a nomogram was ultimately generated based on independent risk factors(age,application of Elamod,DAS28(0),CRP(1)).The bootstrap method was used for internal validation of the nomogram,with 1000 self sampling times.The results showed that the C-index for internal validation was 0.837.Therefore,the area under the ROC curve(AUC)used by this model to predict the establishment of the model is 0.819,and the prediction accuracy of this model based on the external validation method is 0.978,indicating that the model has a high discrimination.The calibration diagram of the model shows a good consistency between the predicted values obtained from the nomogram and the actual observations.Conclusion:1.Age,the use of iguratimod,DAS28(0),CRP(1),RF(3)and anti-CCP antibody(3)are important factors affecting the early clinical efficacy of abatacept combined with methotrexate in the treatment of RA;2.This study screened and constructed a nomogram based on the clinical indicators of RA patients.The prediction model can predict the early clinical efficacy of RA patients after treatment with abatacept combined with methotrexate,with good differentiation.According to this model,the probability of patients reaching low disease activity can be quickly predicted.