Analysis Of Risk Factors Related To Lymphatic Metastasis Of Cervical Cancer And Establishment Of Preoperative Prediction Model

Objective:Analyzing the clinical and pathological elements of cervical cancer patient with lymphatic metastasis.The lymph node metastasis of cervical cancer patients was assessed by combining the high risk features,including lymphatic enlargement in pelvic magnetic resonance imaging(MRI)and the changes of serum squamous cell carcinoma antigen(scc-Ag)before and after neoadjuvant therapy.The prediction model was established and its predictive value was analyzed to provide reference for improving the diagnostic accuracy of preoperative lymph node metastasis of cervical cancer.Materials and Methods:1.Study design: retrospective case-control study was carried out to collect clinicopathological and imaging data of cervical cancer patients,analyzing the risk factors of cervical cancer lymph node metastasis,and establishing a prediction model of lymph node metastasis.2.Subjects: Patients with primary cervical cancer,who has been treated with extensive total hysterectomy+ pelvic lymph node dissection± abdominal paraortic lymph node dissection in Lanzhou University First Hospital from March 2017 to March 2020.3.Data collection: extracting and recording the patient’s general condition,clinicopathological information and imaging data from the patient’s medical records,including name,age,FIGO stage,number of neoadjuvant chemotherapy,tumor histology type,tumor cell differentiation degree,depth of interstitial infiltration,inferior uterine segment infiltration,interstitial vascular infiltration,lymph node metastasis,Ki-67 expression,D2-40 expression,pelvic MRI stage,pelvic lymph node enlargement,serum scc-Ag within 1 week before first diagnosis and no special treatment(no neoadjuvant treatment was the same as the first diagnosis),etc.4.Statistical method: using SPSS 25.0 software for statistical analysis,according to the results of single factor analysis,then ding Logistic regression analysis and establishing prediction model.Finnally,using the working characteristics ROC curve of the subjects to obtain the best critical value of lymph node metastasis.Results:1.The results of the single-factor analysis displayed that the pelvic MRI stage,pelvic MRI lymphatic metastasis,pelvic MRI showed lymph node enlargement,tumor degree of differentiation,depth of interstitial infiltration,the change value of serum scc-Ag、Ki67、D2-40、treatment and serum scc-Ag within one week before the operation were statistically significant(P<0.05).2.The results of multivariate analysis and binary Logistic regression analysis showed that the depth of interstitial infiltration,the change value of serum scc-Ag,the pelvic MRI showed lymph node enlargement and D2-40、Ki67 were independent risk factors for lymph node metastasis of cervical cancer(P<0.05).3.The prediction model of lymph node metastasis as follows: L=the serum scc-Ag change value+7.127×the lymph node enlargement+5.386×D2-40+5.135×the optimal threshold+19.457 × Ki67;the cutoff value is 20.8261.To evaluate the prediction model,Global Omnibus test of model coefficients,(x~2=31.904,P<0.001,It’s statistically significant.The sensitivity of the model is 78.6%,79.2% specificity,The You-den index =0.578,Area under the curve area under curve is 0.838,95%confidence interval(0.765,0.912).Conclusions:1.Cervical cancer patients with pelvic MRI indicating lymph node metastasis,pelvic MRI showing lymph node enlargement,pathological examination indicating low differentiation of tumor cells,depth of interstitial invasion ≥1/2,positive D2-40,and serum scc-Ag change value ≥ 0.05 should be vigilant,given individualized neoadjuvant treatment,and ensure adequate surgical scope.2.The above lymphatic prediction model can be used in the clinic and has predictive value for preoperative lymph node metastasis.