Liu Hailang, Tang Kun, Peng Ejun, Wang Liang, Xia Ding, Chen Zhiqiang
Department of Urology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, Hubei, People's Republic of China.
Department of Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430030, Hubei, People's Republic of China.
Cancer Manag Res. 2020 Dec 22;12:13099-13110. doi: 10.2147/CMAR.S286167. eCollection 2020.
This study aimed to develop a machine learning (ML)-assisted model capable of accurately predicting the probability of biopsy Gleason grade group upgrading before making treatment decisions.
We retrospectively collected data from prostate cancer (PCa) patients. Four ML-assisted models were developed from 16 clinical features using logistic regression (LR), logistic regression optimized by least absolute shrinkage and selection operator (Lasso) regularization (Lasso-LR), random forest (RF), and support vector machine (SVM). The area under the curve (AUC) was applied to determine the model with the highest discrimination. Calibration plots and decision curve analysis (DCA) were performed to evaluate the calibration and clinical usefulness of each model.
A total of 530 PCa patients were included in this study. The Lasso-LR model showed good discrimination with an AUC, accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of 0.776, 0.712, 0.679, 0.745, 0.730, and 0.695, respectively, followed by SVM (AUC=0.740, 95% confidence interval [CI]=0.690-0.790), LR (AUC=0.725, 95% CI=0.674-0.776) and RF (AUC=0.666, 95% CI=0.618-0.714). Validation of the model showed that the Lasso-LR model had the best discriminative power (AUC=0.735, 95% CI=0.656-0.813), followed by SVM (AUC=0.723, 95% CI=0.644-0.802), LR (AUC=0.697, 95% CI=0.615-0.778) and RF (AUC=0.607, 95% CI=0.531-0.684) in the testing dataset. Both the Lasso-LR and SVM models were well-calibrated. DCA plots demonstrated that the predictive models except RF were clinically useful.
The Lasso-LR model had good discrimination in the prediction of patients at high risk of harboring incorrect Gleason grade group assignment, and the use of this model may be greatly beneficial to urologists in treatment planning, patient selection, and the decision-making process for PCa patients.
本研究旨在开发一种机器学习辅助模型，能够在做出治疗决策前准确预测活检 Gleason 分级组升级的概率。
我们回顾性收集了前列腺癌（PCa）患者的数据。使用逻辑回归（LR）、通过最小绝对收缩和选择算子（Lasso）正则化优化的逻辑回归（Lasso-LR）、随机森林（RF）和支持向量机（SVM），从 16 个临床特征中开发了 4 种机器学习辅助模型。应用曲线下面积（AUC）来确定具有最高辨别力的模型。进行校准图和决策曲线分析（DCA）以评估每个模型的校准和临床实用性。
本研究共纳入 530 例 PCa 患者。Lasso-LR 模型显示出良好的辨别力，其 AUC、准确率、敏感性、特异性、阳性预测值（PPV）和阴性预测值（NPV）分别为 0.776、0.712、0.679、0.745、0.730 和 0.695，其次是 SVM（AUC = 0.740，95%置信区间[CI] = 0.690 - 0.790）、LR（AUC = 0.725，95%CI = 0.674 - 0.776）和 RF（AUC = 0.666，95%CI = 0.618 - 0.714）。模型验证表明，在测试数据集中，Lasso-LR 模型具有最佳辨别力（AUC = 0.735，95%CI = 0.656 - 0.813），其次是 SVM（AUC = 0.723，95%CI = 0.644 - 0.802）、LR（AUC = 0.697，95%CI = 0.615 - 0.778）和 RF（AUC = 0.607，95%CI = 0.531 - 0.684）。Lasso-LR 和 SVM 模型校准良好。DCA 图表明，除 RF 外的预测模型在临床上是有用的。
Lasso-LR 模型在预测 Gleason 分级组分配错误风险高的患者方面具有良好的辨别力，使用该模型可能对泌尿外科医生在前列腺癌患者的治疗计划、患者选择和决策过程中非常有益。
