Establishment And Comparison Of A Risk Assessment Model For Prediction Of Chinese Prostate Cancer Patients Bone Metastases At The Initial Stage Of Diagnosis

Part?Externally Validation of the Available Guidelines and the Novel Briganti's Model Addressing the Need for Staging Baseline Bone Scans in Chinese PCa PatientsObjective: Several European and American guidelines recommend that staging bone scan should be omitted in patients who are newly diagnosed with low-risk bone metastasis,which may reduce the number of unnecessary bone scans in patients with newly diagnosed PCa without reducing the BMs detection rate.However,whether these guidelines are suitable for Chinese PCa patients is questionable.The aim of our study was to externally validate the performance characteristics of the AUA/NCCN guidelines and Briganti's risk stratification tool for baseline staging bone scan in patients with newly diagnosed prostate cancer(PCa).Methods: Between 2010 and 2015,a consecutive series of 1231 patients with newly diagnosed PCa were enrolled.All patients were staged to evaluate the presence of bone metastasis(BM)with a conventional total-body Tc 99 m MDP scintigraphy performed regardless of baseline PCa characteristics.The area under the curve(AUC)estimates were used to test the accuracy of the guidelines(American Urological Association,National Comprehensive Cancer Network)and the novel Briganti's model.Results: Median age was 69(range 45–89 years),and median PSA was 20.8ng/ml(range 0.38–4348),235 patients(19.1%)had BM.The AUA/NCCN guidelines and the novel Briganti's model externally validated by Chinese newly diagnosed PCa,showed high sensitivity(91.1%-95.3%),low specificity(33.2%-47.7%),low overall accuracy(63.5%-71.7%).Conclusion: This is the first study aimed at externally validating the available guidelines and the novel Briganti's model addressing the need for staging baseline bone scans in Chinese PCa patients.All guidelines and the novel Briganti's model showed low overall accuracy and specificity.Part ? Analysis of Independent Predictors of Chinese Prostate Cancer Patients Bone Metastases at the Initial Stage of Diagnosis and Charictaristic of Extremely Low Risk GroupObjective: According to several guidelines,it is acceptable to spare a bone scan in the patients who are newly diagnosed with low-risk prostate cancer.However,who are high risk groups of bone metastasis in Chinese PCa patients is questionable.Our aim is to clarify a suitable group whereby a bone scan could be omitted at the initial staging of prostate cancer.Methods:Between 2010 and 2015,a consecutive series of 1231 patients with newly diagnosed PCa were enrolled.All patients were staged to evaluate the presence of bone metastasis(BM)with a conventional total-body Tc 99 m MDP scintigraphy performed regardless of baseline PCa characteristics.The proportion of positive bone metastases by age distribution,prostate specific antigen level at diagnosis,Gleason score,clinical T stage,bone scan and CT/MRI were evaluated.Univariate and multivariate logistic regressio n analyses were performed to identify the predictors of positive bone metastases.Results:Of all 1231 patients,235 patients(19.1%)showed bone metastases.In univariate logistic regression analyses,prostate-specific antigen level,clinical stage and Gleason score were independent predictors of bone metastasis.The multivariate analysis showed that both the prostate-specific antigen level>20 ng/ml and the Gleason score?4+3 and c T4 were independent predictors of bone metastases.Conclusion:The incidences of bone metastases in patients with a prostate-specific antigen level of ?20 ng/ml and Gleason score of ?6 were extremely low,these patients can be defined as extremely low risk group.Newly diagnosed patients with PSA>20ng/ml or Gleason?4+3 or c T4 can be defined as independent predictors of bone metastases.Part ? Establishment of A Risk Stratification Model of BoneMetastases Objective:The aim of the study was to establish a risk stratification model to determine when to perform baseline bone scan in Chinese patients with newly diagnosed PCa.The risk stratification model may help further reduce the use of staging baseline bone scan without compromising the ability of detecting bone metastasis in PCa patients.Externally validating the new model and comparing the accuracy with AUA/NCC N guidelines and a novel risk stratification tool.Methods: 1231 consecutive patients with newly diagnosed PCa were included in the study.Multivariate logistic regression analysis was performed to identify statistically significant covariates,based on which receiver operating characteristic(ROC)curves were generated to identify optimal cut-off values.These cut-off values,together with other parameters,were substituted into the regression model to calculate a discriminative PSA level.Results: The area under the ROC curve was 0.88(95% confidence interval 0.86–0.90).The sensitivity of the new risk stratification model was 90.2% and the specificity was 68.6%.However,our new model was significantly more accurate than all the guidelines and a novel risk stratification tool.Bone scan is needed for c T1–T3 PCa patients who have a biopsy Gleason score ?3+4 and a PSA >42.6ng/ml,and for c T1-T3 patients having a Gleason score of ?4+3 and PSA ?18.9 ng/ml,however,it is also advisable for c T4 PCa patients who have a biopsy Gleason score?4+3 or c T4 PCa patient having a Gleason score of ?3+4 with PSA>20.6 ng/ml.Part ? Establishment of a Nomogram for Prediction of Chinese Prostate Cancer Patients Bone Metastases at the Initial Stage of DiagnosisObjective:Establishing a nomogram to determine bone metastasis probability in Chinese patients with newly diagnosed PCa.The nomogram may help further evaluate the individualized bone metastasis probability of PCa patients.Methods: Multivariate logistic regression analysis was performed to identify predictors and coefficients,establishing a nomogram based on the multivariate logistic regression formula.Bootstrap resample validation method was used to internally validate the nomogram,which can reduce the over-fit bias.Calibration plot was mapped out to evaluated the coincidence of rediction.Results: The nomogram was internally validated by bootstrap resample validation method,the c-index was 0.88.When predicted bone metastasis probability less than 0.04 or between 0.23-0.73,the predicted value was higher tha n the actual incidence of bone metastases;when predicted bone metastasis probability was between 0.04-0.23 and 0.73-1,the predicted value was lower than the actual incidence of bone metastase.Conclusion : The nomogram can evaluate the individualized bone metastasis probability of Chinese PCa patients with accuracy of 0.88.Part ? Establishment of a Nomogram Including ALP for Prediction of Chinese Prostate Cancer Patients Bone Metastases at the Initial Stage of DiagnosisObjective : The purpose of the study is to develop a nomogram including alkaline phosphatase to decide when to perform baseline bone scan or other bone imaging examination in Chinese patients with newly diagnosed PCa.Methods: Between 2010 and 2015,770 consecutive patients with newly diagnosed PCa were treated in Shanghai Changhai Hospital and Renji Hospital(all patients included had detected serum alkaline phosphatase).A nomogram including serum alkaline phosphatase was developed using multivariate logistic regression and validated internally by bootstrapping-method(200 repetitions)to reduce the over-fit bias.The accuracy of the nomogram was evaluated by calibration.Results: In the present study,we for the first conducted a nomo gram including serum alkaline phosphatase to predict the risk of bone metastasis in Chinese patients with newly diagnosed PCa.The nomogram had an internally validated concordance index of 0.90.When predicted bone metastasis probability less than 0.08 or between 0.32-0.78,the predicted value was lower than the actual incidence of bone metastases;when predicted bone metastasis probability was between 0.08-0.32 and 0.78-1,the predicted value was higher than the actual incidence of bone metastase.Conclusion : The nomogram can evaluate the individualized bone metastasis probability of Chinese PCa patients with accuracy of 0.90.