Patients With Locally Progressed Cervical Squamous Cell Carcinoma: The Impact Of Pretreatment Peripheral Inflammatory Index On Long-term Survival

OBJECTIVE:The purpose of this study is to examine the effect and predictive value of pre-treatment peripheral inflammatory environment on the prognosis of patients having radical concurrent radiation for stage IIb or IIIb cervical cancer.METHODS:Patients with stage IIb to stage IIIb cervical cancer who initially attended Sichuan Cancer Hospital between 2011 and 2018 and got aggressive concomitant radiation had their clinical data and blood sample results gathered.We used 140 patients as the study group.They received radical synchronous radiation therapy between 2011 and 2014 and they stayed with IIb-IIIb uterine squamous cell carcinoma.As the validation group,we used 100 IIb-IIIb uterine squamous cell carcinoma patients who underwent radical synchronous radiation therapy between 2015 and 2018.Patients’disease progression and prognosis were predicted before treatment began by utilizing the subject’s working characteristic curve to establish a cut-off value for the peripheral blood inflammation index(neutrophil lymphocyte ratio NLR,monocyte lymphocyte ratio MLR,platelet lymphocyte ratio PLR,systemic immune inflammation index SII,systemic inflammatory response index SIRI,etc.)and the nutritional prognosis index PNI.The truncation value determines whether the range is high or low.Investigate how much longer those in the high value group live compared to those in the low value group and sketch out a graph of the expected number of survivors.Both a single index and numerous indexes were used in the investigation of the elements that influence the patients’prognoses.In the second step of the process,correlation prediction models for OS and PFS were developed using linear regression(the least square approach),ridge regression,and lasso regression respectively.Goodness-of-fit of the model R~2may be predicted by drawing a prediction curve and an actual curve based on the actual value and the prediction value.Scale of the evaluative model’s effectiveness.Last but not least,we have the cross-validation diagram and Check the regression coefficient diagram of the model for any shifts in the coefficient,and carry on with the screening of the variables.RESULTS:The greatest predictor of OS in cervical cancer of all the data was the area under the PLR curve before treatment,which had an AUC=0.836 (P<0.001)(Jorden index of 0.757 and specificity of 0.957).Among the aforementioned data,the area under the pre-treatment SII curve AUC=0.913(p<0.001)was the best predictor of PFS for cervical cancer(Jordans index0.750,specificity 0.989).Age at diagnosis,tumor size,pre-treatment NLR,MLR,PLR,SII,SIRI,and PNI had an effect on the progression-free time and overall survival of patients with stage IIB-IIIB cervical squamous carcinoma,according to the analysis of individual indicators in the study and validation groups.Pretreatment PLR and SII could be used as prognostic factors affecting overall survival alone,and age at diagnosis could be used as a prognostic factor affecting progression-free survival time alone,according to the results of the combined analysis of several indicators in the study and validation groups.On the basis of linear regression and ridge regression,clinical prediction models pertaining to OS and PFS were created.Both of the models performed rather well,as shown by their goodness-of-fit R~2 values,which came in at 0.687 and0.658 respectively.Clinical prediction models based on lasso regression were constructed for overall survival(OS)and progression-free survival(PFS).The goodness-of-fit R~2 for the OS prediction model was 0.654,and the goodness-of-fit R~2 for the PFS prediction model was 0.642.A number of the variables,including age,NLR,PLR,SII,and PNI,were kept.A PFS-related clinical prediction model and variable screening based on Lasso regression were constructed,and the model performed pretty well.Age at diagnosis,tumour size,pre-treatment NLR,PLR,SII,and PNI were all kept in the model.Conclusion:In patients with stage IIB-IIIB squamous cervical cancer,the levels of pre-treatment inflammatory indicators(NLR,MLR,PLR,SII and SIRI)and nutritional prognostic index PNI had an impact on the time to progression-free disease and overall survival,with significantly shorter progression-free survival and overall survival in the high value group compared to the low value group for pre-treatment inflammatory indicators.The pre-therapy nutritional prognostic index high value group had substantially shorter progression-free survival time and overall survival than the low value group.Pretreatment PLR and SII may be utilized as independent prognostic markers for overall survival,while age at diagnosis can be used independently for progression-free survival time.OS and PFS-related clinical prediction models were developed using linear regression(least squares),ridge regression,and Lasso regression-based statistical approaches.The goodness-of-fit R~2values for the OS-related clinical prediction models were 0.722,0.687,and0.654,while the PFS-related clinical prediction models were 0.687,0.658,and0.642,with all model performance rated as excellent.In the meanwhile,Lasso regression was used to filter the OS and PFS-related variables,retaining NLR,PLR,SII,and PNI,further validating the prognostic usefulness of pre-treatment peripheral inflammation levels on the prognosis of patients with intermediate to advanced cervical cancer.