Divison of Gastrointestinal Surgery, Department of Surgery, University of North Carolina at Chapel Hill, 100 Manning Drive, 4038 Burnett Womack Building, CB #7050, Chapel Hill, NC, 27599, USA.
Division of Colon & Rectal Surgery, Department of Surgery, University of Minnesota, 420 Delaware St SE, MN, 55455, Minneapolis, USA.
J Gastrointest Surg. 2023 Sep;27(9):1925-1935. doi: 10.1007/s11605-023-05755-0. Epub 2023 Jul 5.
Optimal treatment of anal squamous cell carcinoma (ASCC) is definitive chemoradiation. Patients with persistent or recurrent disease require abdominoperineal resection (APR). Current models for predicting need for APR and overall survival are limited by low accuracy or small datasets. This study sought to use machine learning (ML) to develop more accurate models for locoregional failure and overall survival for ASCC.
This study used the National Cancer Database from 2004-2018, divided into training, validation, and test sets. We included patients with stage I-III ASCC who underwent chemoradiation. Our primary outcomes were need for APR and 3-year overall survival. Random forest (RF), gradient boosting (XGB), and neural network (NN) ML-based models were developed and compared with logistic regression (LR). Accuracy was assessed using area under the receiver operating characteristic curve (AUROC).
APR was required in 5.3% (1,015/18,978) of patients. XGB performed best with AUROC of 0.813, compared with 0.691 for LR. Tumor size, lymphovascular invasion, and tumor grade showed the strongest influence on model predictions. Mortality was 23.6% (7,988/33,834). AUROC for XGB and LR were similar at 0.766 and 0.748, respectively. For this model, age, radiation dose, sex, and insurance status were the most influential variables.
We developed and internally validated machine learning-based models for predicting outcomes in ASCC and showed higher accuracy versus LR for locoregional failure, but not overall survival. After external validation, these models may assist clinicians with identifying patients with ASCC at high risk of treatment failure.
肛门鳞状细胞癌（ASCC）的最佳治疗方法是明确的放化疗。对于持续性或复发性疾病的患者，需要进行腹会阴切除术（APR）。目前预测 APR 和总生存率的模型准确性较低或数据集较小。本研究旨在使用机器学习（ML）为 ASCC 开发更准确的局部区域失败和总生存率模型。
本研究使用了 2004 年至 2018 年的国家癌症数据库，分为训练、验证和测试集。我们纳入了接受放化疗的 I-III 期 ASCC 患者。我们的主要结局是需要 APR 和 3 年总生存率。开发了随机森林（RF）、梯度提升（XGB）和神经网络（NN）基于 ML 的模型，并与逻辑回归（LR）进行了比较。使用接受者操作特征曲线下的面积（AUROC）评估准确性。
有 5.3%（1015/18978）的患者需要 APR。XGB 的 AUROC 为 0.813，优于 LR 的 0.691。肿瘤大小、淋巴血管侵犯和肿瘤分级对模型预测有最强的影响。死亡率为 23.6%（7988/33834）。XGB 和 LR 的 AUROC 分别为 0.766 和 0.748，相似。对于这个模型，年龄、辐射剂量、性别和保险状况是最具影响力的变量。
我们开发并内部验证了基于机器学习的 ASCC 结局预测模型，与 LR 相比，局部区域失败的准确性更高，但总生存率没有提高。经过外部验证后，这些模型可能有助于临床医生识别 ASCC 患者中治疗失败风险较高的患者。
