Department of Hepatobiliary Surgery, Songshan General Hospital, Chongqing, China.
Department of Stomatology, Children's Hospital of Chongqing Medical University, Chongqing, China.
Front Public Health. 2022 Oct 18;10:1028905. doi: 10.3389/fpubh.2022.1028905. eCollection 2022.
According to statistics, patients with high-risk prostate cancer (PC) account for about 15% of prostate cancer diagnoses, and high-risk patients usually have a poor prognosis due to metastasis and recurrence and have a high mortality rate. Therefore, the accurate prediction of prognostic-related risk factors in middle-aged high-risk PC patients between 50 and 65 can help reduce patient mortality. We aimed to construct new nomograms for predicting cancer-specific survival (CSS) and Overall survival (OS) in middle-aged high-risk PC patients.
Data for patients aged between 50 and 65 years old and diagnosed with high-risk PC were obtained from the Surveillance, Epidemiology, and End Results (SEER) database. Univariate and multivariate Cox regression models were used to identify independent risk factors for CSS and OS in patients. Nomograms predicting CSS and OS were developed based on multivariate Cox regression models. The concordance index (C-index), the area under the receiver operating characteristic curve (AUC), and the calibration curve are used to detect the accuracy and discrimination of the model. Decision curve analysis (DCA) is used to detect the potential clinical value of this model.
Between 2010 and 2018, 1,651 patients diagnosed with high-risk PC and aged 50-65 years were included. In this study, the training group ( = 1,146) and the validation group ( = 505) were randomly assigned in a ratio of 7:3. The results showed that M stage, Gleason (GS) and surgical mode were independent risk factors for CSS; marital status, T stage, M stage, surgical mode, and GS were independent risk factors for OS. The C-index for predicting CSS in the training and validation groups are 0.84 and 0.811, respectively; the C-index for predicting OS in the training and validation groups are 0.824 and 0.784, respectively. The AUC and the calibration curves also showed good accuracy and discrimination.
We constructed new nomograms to predict CSS and OS in middle-aged high-risk PC patients. The prediction tools showed good accuracy and reliability, which can help clinicians and patients to make better clinical decisions.
据统计，高危前列腺癌（PC）患者约占前列腺癌诊断的 15%，由于转移和复发，高危患者通常预后不良，死亡率较高。因此，准确预测 50-65 岁中年高危 PC 患者的预后相关风险因素有助于降低患者死亡率。我们旨在构建新的列线图来预测中年高危 PC 患者的癌症特异性生存（CSS）和总体生存（OS）。
从监测、流行病学和最终结果（SEER）数据库中获取年龄在 50-65 岁之间且诊断为高危 PC 的患者数据。使用单因素和多因素 Cox 回归模型来确定患者 CSS 和 OS 的独立危险因素。基于多因素 Cox 回归模型开发预测 CSS 和 OS 的列线图。一致性指数（C 指数）、接收者操作特征曲线下面积（AUC）和校准曲线用于检测模型的准确性和区分度。决策曲线分析（DCA）用于检测该模型的潜在临床价值。
2010 年至 2018 年间，共纳入 1651 例诊断为高危 PC 且年龄在 50-65 岁的患者。在本研究中，训练组（n=1146）和验证组（n=505）以 7:3 的比例随机分配。结果表明，M 期、Gleason（GS）分级和手术方式是 CSS 的独立危险因素；婚姻状况、T 期、M 期、手术方式和 GS 分级是 OS 的独立危险因素。训练组和验证组预测 CSS 的 C 指数分别为 0.84 和 0.811，预测 OS 的 C 指数分别为 0.824 和 0.784。AUC 和校准曲线也显示出良好的准确性和区分度。
我们构建了新的列线图来预测中年高危 PC 患者的 CSS 和 OS。预测工具具有良好的准确性和可靠性，可帮助临床医生和患者做出更好的临床决策。
