Development and Validation of Nomograms to Predict Cancer-Specific Survival and Overall Survival in Elderly Patients With Prostate Cancer: A Population-Based Study

• DOI: 10.3389/fonc.2022.918780
• Publication Date: 2022-06-23T04:27:29Z
• Authors: Zhaoxia Zhang; Chenghao Zhanghuang; Jinkui Wang; Xiaomao Tian; Xin Wu; Maoxian Li; Tao Mi; Jiayan Liu; Liming Jin; Mujie Li; Dawei He
• Keywords: Nomogram; Univariate; Univariate analysis

Prostate cancer (PC) is the most common non-cutaneous malignancy in men worldwide. Accurate predicting survival of elderly PC patients can help reduce mortality patients. We aimed to construct nomograms predict cancer-specific (CSS) and overall (OS) patients.Information on aged 65 years older was downloaded from Surveillance, Epidemiology, End Results (SEER) database. Univariate multivariate Cox regression models were used determine independent risk factors for Nomograms developed CSS OS based a model. The accuracy discrimination prediction model tested by consistency index (C-index), area under subject operating characteristic curve (AUC), calibration curve. Decision analysis (DCA) test clinical value compared with TNM staging system D'Amico stratification system.135183 2010-2018 included. All randomly assigned training set (N=94764) validation (N=40419). revealed that age, race, marriage, histological grade, stage, surgery, chemotherapy, radiotherapy, biopsy Gleason score (GS), prostate-specific antigen (PSA) PC. C-index 0.883(95%CI:0.877-0.889) 0.887(95%CI:0.877-0.897), respectively. 0.77(95%CI:0.766-0.774)and 0.767(95%CI:0.759-0.775), It showed proposed has excellent discriminative ability. AUC curves also good discriminability. DCA have potential value.We new been internally validated reliability doctors make better decisions.