Development Of Nomogram Models Based On Peripheral Blood Score System To Predict FIGO Stage And Prognosis For Epithelial Ovarian Cancer Patients

BackgroundOvarian cancer is the most common cause of gynecologic cancer death.Epithelial ovarian cancer(EOC)accounts for about 90%of ovarian malignancies.The early stage of the EOC has a relatively better prognosis of 5 years overall survival rate 61-87%,the 5 years overall survival rate of advanced stage is 14-38%.Unfortunately,over 75%EOC patients are diagnosed at stage Ⅲ/Ⅳ.This is also the reason for bad prognosis of ovarian cancer.It is necessary to explore predictive methods of staging and prognosis to help making an appropriate treatment plan.Chronic inflammation is a significant part in cancer development.Nutritional status is also the factor associated with postoperative complication and cancer prognosis.Some inflammation and nutrition related scoring systems have been proved had clinical significance in gynecological cancer.The wide use of them limited by inconsistent cutoff values,relatively low prediction accuracy and small,weak studies.The application needs to be improved.These all provide insights into constructing a new scoring system with peripheral blood parameters and further exploring its clinical significance.Purpose1.Create a new inflammatory nutritional scoring system and explore its correlation to clinicopathological parameters;2.Develop Nomogram model based on the scoring system to predict the advanced stage before surgery and evaluate it;3.Develop Nomogram models based on the scoring system and clinicopathological parameters to predict overall survival(OS),progression-free survival(PFS)and evaluate them.Methods1.Patients selectionIt is a single-center retrospective study.We screened patients who were diagnosed with ovarian cancer and underwent primary surgery at Qilu Hospital from January 2014 to December 2018.Patients in these cases were excluded:non-epithelial,recurrent or metastatic ovarian cancer,neoadjuvant chemotherapy,infectious disease 15 days before surgery,antibiotics 1 month before surgery,malignant disease of other system,disease affecting immune or nutritional status such as hepatitis,systemic lupus erythematosus and so on.Patients in pregnancy or lactation,and underwent secondary surgery or only exploratory surgery were also excluded.Without loss of generality,those without complete clinical data or following-up data were excluded.2.Data collectionWe got patients’ clinical information from electronic medical records.(1)Basic clinical information:age,menopausal state,medical history of cancer,ASA(American Society of Anesthesiologists)grade;(2)Clinicopathological information:Federation International of Gynecology and Obstetrics(FIGO)stage,pathological pattern,histological grade,level of CA125,ascites,metastasis,residual disease after primary surgery;(3)Inflammatory and nutritional parameters obtained within 15 days before surgery:white cell count,neutrophil count,lymphocyte count,monocyte count,level of serum albumin,and plasma total cholesterol and fibrinogen;(4)Prognostic information:recurrence,metastasis,death.3.Follow-upThese patients get telephone follow-up every year after surgery until death or May 31,2021,whichever came first.The follow-up including chemotherapy,recurrence and survival.4.Indicator calculation and definitionWe calculated(1)ratio of neutrophil to lymphocyte(NLR),(2)ratio of lymphocyte to monocyte(LMR),(3)ratio of fibrinogen to lymphocyte(FLR),(4)ratio of total cholesterol to lymphocyte(TCLR),(5)Onodera’s prognostic nutritional index(ONI),(6)systemic immune-inflammation index(SII),(7)systemic inflammation score(SIS),(8)controlling nutritional status(COUNT),(9)OS,and(10)PFS.5.Statistical analysis5.1 Comparation of baseline characteristicsWe use Fisher’s exact test/Chi-square test and Kruskal-Wallis test/Mann-Whitney U test to handle categorical and continuous variables,respectively.5.2 Comparation of survival outcomesThe optimal cutoff points of ALB,NLR,LMR,FLR,and TCLR were calculated based on OS.Then the obtained cutoff values were used to dichotomize patients into two groups(high and low).The survival outcomes were delineated by Kaplan-Meier curves,Log-rank test was used to compare the differences between two groups,respectively.5.3 Construction of new scoring systemWe developed a scoring system named peripheral blood score(PBS)based on the five parameters above:For each parameter,patient with a better prognosis gets 0 point,and patient with a worse prognosis gets 1 point.Then add them up to get a total score.We created time-dependent receiver operating characteristic(ROC)curves and calculated the areas under the curve(AUC)to compare the discrimination among the scoring systems.5.4 Identifying the independent factors of FIGO Ⅲ-Ⅳ stage and construction of Nomogram modelWe used univariate and multivariate binary logistic regression analyses to select associated factors of the FIGO Ⅲ-Ⅳ stage.The independent factors were used to construct a Nomogram model and bootstrap methods of internal validation was conducted.The internal validation included discrimination and accuracy.We delineated calibration plots to evaluate accuracy and ROC curve to assess the discrimination.To evaluate the clinical practicality,we performed decision curve analysis(DCA)with the "ggDCA" package.5.5 Identifying the independent factors of OS and PFS and construction of Nomogram modelsUnivariate and multivariate Cox proportional hazards regression was used to select independent factors of OS and PFS.A Lasso regression was used to reduce data dimensionality and select variables,accompanied by 10 rounds of cross-validation.The selected variables were used to construct the Nomogram models.Then the models’performance was assessed by internal validation.The AUC of ROC curves and concordance index(C-index)curves were performed to assess discrimination.Calibration plots were performed with bootstrapping(1000 bootstrap resamples).DC A was conducted to assess the clinical benefit of models.5.6 Subgroup analysisUnivariate and multivariate Cox analyses were used to select independent factors of prognosis in FIGO Ⅰ-Ⅱ and FIGO Ⅲ-Ⅳ subgroup respectively.Note.SPSS 21.0(IBM Corporation,Armonk,NY,USA)and R(version 4.2.0)were used to analyze the data.Results1.Patients characteristics453 EOC patients were included.The FIGO stage of patients was stage Ⅰ-Ⅱ in 137(30.2%)cases,stage Ⅲ-Ⅳ in 316(69.8%)cases.279(61.6%)patients had residual disease<1 cm while 174(38.4%)had residual disease ≥1cm.162(35.8%)patients were classified into PBS group 0(PBS 0 or 1),216(47.7%)into group 1(PBS 2 or 3),and 75(16.5%)into group 2(PBS 4 or 5).2.Survival outcomes and Peripheral Blood Score(PBS)constructionThe median follow-up was 36 months and 193(42.6%)patients died during follow-up.The median OS and PFS were 63 and 40 months;The cutoff points of each peripheral blood marker variables were as follows:ALB:34.4,NLR:2.37,LMR:2.51,FLR:2.02,TCLR:3.81.Higher serum albumin,and LMR indicated favorable OS,whereas higher NLR,FLR,and TCLR indicated inferior OS(all P<0.001).The 5 parameters above were used to construct PBS.Patients were divided into three groups:PBS 0 group,PBS 1 group and PBS 2 group.The survival analysis of three PBS groups showed significantly different(P<0.001).The relationships between clinicopathological characteristics and PBS were analyzed.Among the three groups,the distribution of histological type,differentiation,FIGO stage,residual disease after surgery,CA125 levels,and ascites before surgery was very different(all P<0.05).Whereas,the distribution of age,menopause status,chemotherapy,ASA grade had no difference(all P≥0.05).3.Independent risk factors for FIGO Ⅲ-Ⅳ stage in EOC patients and construction of Nomogram modelUnivariate and multivariate logistic regression showed age,level of CA125,and PBS were independent risk factors for the FIGO Ⅲ-Ⅳ stage(OR=1.03,95CI%:1.01-1.05;OR=4.89,95CI%:2.98-8.03;OR=1.68,95CI%:1.20-2.36,respectively).A Nomogram for the FIGO Ⅲ-Ⅳ stage was created based on the independent predictors selected.The ROC curve analysis showed the model’s good efficiency in diagnosing FIGO Ⅲ-Ⅳ stage of EOC patients(AUC=0.782).The calibration plots showed that the actual and the predicted probabilities had a good consistency(P=0.986).DCA revealed that the model is beneficial to clinical practice.4.Univariate and multivariate Cox analyses for OS and PFS and construction ofNomogram modelWe selected the independent risk factors for OS and PFS by univariate and multivariate Cox analyses.We found that FIGO stage,residual disease after surgery,and PBS were independent factors affecting OS(HR=4.54,95CI%:2.54-8.09;HR=0.69,95CI%:0.50-0.96;HR=1.19,95CI%:1.07-1.33,respectively)and PFS(HR=3.47,95CI%:2.21-5.44;HR=0.76,95CI%:0.58-1;HR=1.16,95CI%:1.06-1.27,respectively).To identify the most correlated factors of OS and PFS,we performed Lasso Cox analysis and 10 rounds of cross-validation.Finally,FIGO stage,residual disease after surgery,and PBS were selected in both OS and PFS analysis.Then prognostic Nomogram models were developed based on the selected factors.The AUC of time-dependent ROC and time-dependent C-index curves were delineated to evaluate models’ discrimination.For the OS Nomogram model,the C-index was 0.68(95%CI:0.64-0.72),and for the PFS Nomogram model,it was 0.66(95%CI:0.63-0.70)after 1000 bootstrap samples to measure discrimination.The calibration curves for different years of OS and PFS rates were delineated with 1000 bootstrap samples and well overlapped with their reference lines.DCA curves showed that the Nomogram models increase net benefits.5.The subgroup analyses of FIGO stage in EOC patientsHistological pattern can affect OS,non-serous ovarian cancer had worse OS than serous ovarian cancer in FIGO Ⅰ-Ⅱ patients(HR=3.98,95%CI:1.11-14.19).For FIGO Ⅲ-Ⅳpatients,differentiation grade,residual disease,ASA grade,age and PBS were independent factors affecting OS(all P<0.05);menopause status,differentiation grade,residual disease,ASA grade and PBS were independent factors affecting PFS(all P<0.05).Conclusions1.The scoring system developed in our study,PBS can be used as an indicator to evaluate the condition and prognosis of patients with EOC.High PBS indicates advanced stage and poor prognosis;2.The prediction model predicting FIGO Ⅲ-Ⅳ stage constructed in our study based on PBS is economical and practical,it has clinical application value;3.The prediction models predicting OS and PFS constructed in our study based on PBS and clinicopathological parameters have good efficiency and practicability.Innovations and limitationsInnovations1.This study constructed a scoring system based on routine examnation,which provided a practical tool;2.The study further developed Nomogram models based on PBS to predict preoperatvie stage,OS and PFS respectively;3.This study constructed Nomogram models and validate them in the aspects of discrimination,acurracy and clinical practicability.Limitations1.It is a single-center retrospective cohort and postoperative therapies of patients have some heterogeneity,which may cause bias;2.The study only performed internal validation which proved the reproducibility of the study,but no external validation to prove its generalization;3.There were limited patients in some subgroups,which resulted in a wide range of confidence intervals(CIs);...