| 谭文慧,曹晓明.前列腺特异性抗原 4~10 μg/L病人前列腺癌预测模型的构建及验证[J].安徽医药,2025,29(5):977-982. |
| 前列腺特异性抗原 4~10 μg/L病人前列腺癌预测模型的构建及验证 |
| Construction and verification of a predictive model for prostate cancer in patients with PSA 4-10 μg/L |
| |
| DOI：10.3969/j.issn.1009-6469.2025.05.024 |
| 中文关键词: 前列腺肿瘤 多参数核磁 前列腺特异性抗原密度 前列腺成像报告和数据系统 前列腺活检 |
| 英文关键词: Prostatic neoplasms Multi-parameter nuclear magnetism Prostate specific antigen density Prostate imaging report-ing and data system Prostate biopsy |
| 基金项目:山西省卫生健康委科研课题( 2020087) |
|
| 摘要点击次数: 55 |
| 全文下载次数: 60 |
| 中文摘要: |
| 目的开发基于多参数核磁( mpMRI)及生化和临床参数的前列腺癌( PCa)预测模型，并评估预测模型的效能，为前列腺特异性抗原( PSA)灰区病人提供有意义的临床策略，以避免不必要的活检。方法回顾性分析 2019年 1月至 2022年 9月在山西医科大学第一医院进行过前列腺穿刺活检，并且 PSA为 4~10 μ g/L的 128例病人的临床数据，采用 7∶3随机分为训练集和测试集，进行单变量分析和多变量分析，确定 PCa的预测因子。使用 logistic回归建立预测模型，并绘制列线图。使用受试者操作特征曲线( ROC曲线)下面积，评价模型的诊断效能，通过约登指数判断模型的最佳截断值。结果年龄、前列腺特异性抗原密度( PSAD)、 PI-RADS v2.1评分为预测因子构建模型，并绘制列线图。 ROC曲线下面积表示模型的预测能力，诊断模型的 AUC值为 0.87，95%CI:(0.78，0.93)，训练集最佳截断值的灵敏度和特异度分别为 0.85和 0.79。测试集中，模型的灵敏度、特异度、阳性预测值、阴性预测值、 AUC值分别为 88.89%、75.86%、53.33%、95.65%、0.89。当风险阈值为 29%时， 76%的病人可以避免不必要的活检。结论该研究联合年龄、 PSAD和 PI-RADS v2.1评分建立了 PSA灰区 PCa的预测模型，该预测模型具有良好的诊断性能，可显著减少不必要的 PSA灰区穿刺活检。 |
| 英文摘要: |
| Objective To develop a prostate cancer (PCa) prediction model based on multi-parameter nuclear magnetic resonance(mpMRI) and biochemical and clinical parameters, and to evaluate the effectiveness of the prediction model, so as to provide meaning-ful clinical strategies for patients with prostate specific antigen (PSA) gray area to avoid unnecessary biopsy.Methods The clinical da-ta of 128 patients with PSA 4-10 μg/L in the First Hospital of Shanxi Medical University from January 2019 to September 2022 wereanalyzed retrospectively. The patients were 7: 3 randomly assgned into training set and test set. Univariate analysis and multivariateanalysis were used to determine the predictors of PCa. Logistic regression was used to establish the prediction model and draw the linechart. The area under the subject working characteristic (ROC) curve was used to evaluate the diagnostic efficiency of the model, andthe best cut-off value of the model was judged by the Jordan index.Results Age, prostate specific antigen density (PSAD) and PI-RADS v2.1 score were used as predictive factors to build a model and draw a line chart. The area under the ROC curve represented thepredictive ability of the model, the AUC value of the diagnostic model was 0.87, 95%CI:(0.78,0.93), and the sensitivity and specificityof the best truncation value of the training set were 0.85 and 0.79, respectively. In the test set, the sensitivity, specificity, positive pre-dictive value, negative predictive value and AUC value of the model were 88.89%, 75.86%, 53.33%, 95.65% and 0.89, respectively.When the risk threshold was 29%, 76% of patients could avoid unnecessary biopsies. Conclusions The study combine with age, PSAD and PI-RADSv2.1 score to establish a predictive model of PSA gray area prostate cancer.The predictive model has good diagnos-tic performance and can significantly reduce unnecessary PSA gray area biopsy. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
