| Background:Cervical cancer is a member of the important factors threatening women’s life and health seriously,and has become the leading factor of cancer death among women in 36 countries around the world.Over the past few years,the morbidity and mortality of cervical cancer have been augmenting gradually.The two main pathological types of cervical cancer are squamous cell carcinoma and adenocarcinoma.With the popularization of HPV vaccine,the morbidity of cervical squamous cell carcinoma is expected to decline,but cervical adenocarcinoma will become the main content of cervical cancer research in the future because of its relatively low HPV infection rate.Before 2018,the staging of cervical adenocarcinoma was mainly based on physical examination,but there was a high error between the clinical staging method and the final surgical staging,especially in advanced cervical adenocarcinoma.Lymph node metastasis is the primary way of cervical cancer metastasis,and pelvic lymph node is the first station of cervical cancer lymph node metastasis,which is closely related to the recurrence and prognosis of patients.By comparison to squamous cell carcinoma,cervical adenocarcinoma may have different influencing factors of lymph node metastasis,but the specific situation needs further study.FIGO included lymph node metastasis into staging in 2018,and the staging was based on imaging or pathological diagnosis.And the staging results directly affecte the choice of treatment for patients.Among them,the pathological diagnosis is realized by lymph node dissection,but this operation will afford an opportunity suffering surgical risks and complications to patients,such as increased operation time,fortified blood loss,neurovascular or ureter injury,lower limb lymphedema,lymphatic cyst and so on,and may directly affect the quality of life of patients.Imaging diagnosis is mainly r eal i zed by CT,MRI and PET-CT.Compared with pathological diagnosis,imaging diagnosis has a great a dvantage in safety.Nevertheless,owing to the restriction of nowadays scientific or technological level and the influence of other factors,the sensitivity or specificity of imaging evaluation alone are not ideal.Other clinical methods,such as the detection of serum tumor markers and HPV,have certain clinical value in the diagnosis and treatment of cervical adenocarcinoma,and may related to pelvic lymph node metastasis of cervical adenocarcinoma.Therefore,it is an urgent problem to establish a reasonable comprehensive evaluation method to improve the prediction of pelvic lymph node metastasis before clinical intervention.Purpose:Explore the risk factors of pelvic lymph node metastasis in cervical adenocarcinoma by comparing the difference of clinical information between cervical adenocarcinoma patients with positive and negative pelvic lymph node metastasis before treatment.And establish a nomogram to achieve the prediction of cervical adenocarcinoma pelvic lymph node metastases.Methods:On the basis of the inclusion and exclusion criteria,288 patients with cervical adenocarcinoma who went through extensive hysterectomy and pelvic lymph node dissection in the second Hospital of Jilin University from January 2012 to October2022 were selected.And the pathological results and medical history were recorded accurately.Taking the occurrence of pelvic lymph node metastasis as the outcome variable,the sufferers were separated into two groups: no pelvic lymph node metastasis(negative)and pelvic lymph node metastasis(positive)according to the pathological results.Using SPSS software,univariate and multivariate analysis were carried out,and anatomize the difference of clinical data amid the two groups.P<0.05 indicate statistically significant differences.R software was used to construct a nomogram for the factors with statistical significance.Area under ROC curve(AUC),calibration curve(Hosmer-Lemeshow test)and Decision Curve Analysis(DCA)were used to verify the distinguishing ability,calibration ability and clinical effectiveness of the nomogram.Results:1.Among the 288 patients with cervical adenocarcinoma,there were 237 patients with negative pelvic lymph node metastasis and 51 patients with positive pelvic lymph node metastasis.There were 227 patients(78.82%)of HPV-associated adenocarcinoma and 61 patients(21.18%)of HPV-independent adenocarcinoma.In the negative group,there were 189 patients(79.75%)of HPV-associated adenocarcinoma and 48 patients(20.25%)of HPV-independent adenocarcinoma.In the positive group,there were 38 patients(74.51%)of HPV-associated adenocarcinoma and 13 patients(25.49%)of HPV-independent adenocarcinoma.2.Univariate analysis showed that the age of the negative group was(47.61±9.48)years,and the age of the positive group was(50.61±8.97).There existed a statistical difference amid the two groups(P=0.040).The menopause rate was 33.8% in the negative group and 54.9% in the positive group.There was a statistical difference in the menopause rate between the two groups(P=0.005).In addition,there were significant differences between the negative and positive groups in preoperative staging,tumor markers CEA,CA125,NSE,HE4,radiologically suspicious pelvic lymph node metastasis rate,radiographic tumor diameter,and whether vaginal metastases were considered on imaging(P<0.001).There were no significant differences in HPV infection,gestational times and clinical symptoms between the two groups.3.In binary logstic regression analysis,preoperative staging had statistical significance for pelvic lymph node metastasis of cervical adenocarcinoma(OR=4.72,95%CI: 2.20-10.10,P<0.001).Tumor marker CEA had statistical significance for pelvic lymph node metastasis of cervical adenocarcinoma(OR=1.12,95%CI: 1.05-1.18,P<0.001).NSE had statistical significance for pelvic lymph node metastasis of cervical adenocarcinoma(OR=1.24,95%CI: 1.05-1.47,P=0.012),while imaging suspicious lymph node metastasis had statistical significance on pelvic lymph node metastasis of cervical adenocarcinoma(OR=66.03,95%CI: 6.30-691.71,P<0.001).Therefore,preoperative staging,tumor markers CEA,NSE,and radiologically suspicious pelvic lymph node metastasis are independent risk factors for pelvic lymph node metastasis in cervical adenocarcinoma.4.The pelvic lymph node metastasis profile model of cervical adenocarcinoma was successfully constructed based on preoperative staging,tumor markers CEA,NSE,and whether pelvic lymph node metastasis was suspected in imaging.Based on the ROC analysis of the nomogram,we calculated AUC=0.980,which considered that the model has a good diagnostic effect.In Hosmer-Lemeshow test,P=0.904,it showed that there was no statistical distinction amid the predicted value of the model and the actual result,and with the increase of the possibility of pelvic lymph node metastasis,the prediction probability of the model coincided with the actual occurrence probability better.The DCA analysis of the constructed nomogram shows that the threshold probability of this model has net income in the range of 0.01-0.99.Conclusions:1.According to FIGO(2018)staging,the rate of pelvic lymph node metastasis in patients with cervical adenocarcinoma diagnosed for the first time in stage IIB or before was 17.7%.2.Preoperative staging,serum tumor markers CEA,NSE and imaging suspect of the pelvic lymph node metastasis are independent risk factors for pelvic lymph node metastasis in cervical adenocarcinoma.With the increase of preoperative staging,the increase of serum tumor markers CEA and NSE concentrations,and the imaging suspect of pelvic lymph node metastasis,the actual risk of pelvic lymph node metastasis in cervical adenocarcinoma patients will increase.3.The ROC curve,calibration curve and DCA analysis were used to verify the nomogram.The results showed that the differentiation,calibration and clinical practicability of the prediction model for cervical adenocarcinoma pelvic lymph node metastasis performed well,which could effectively distinguish high-risk patients with cervical adenocarcinoma pelvic lymph node metastasis,and the prediction result of positive events for cervical adenocarcinoma pelvic lymph node metastasis was better. |