Department of Internal Medicine, Division of Endocrinology, Metabolism, and Diabetes, Cerrahpasa Medical School, Istanbul University-Cerrahpaşa, Kocamustafapaşa Street No:53, 34098 Fatih, Istanbul, Turkey.
Graduate School of Sciences and Engineering, Koç University, Istanbul, Turkey.
Pituitary. 2022 Jun;25(3):486-495. doi: 10.1007/s11102-022-01216-0. Epub 2022 Apr 18.
To develop machine learning (ML) models that predict postoperative remission, remission at last visit, and resistance to somatostatin receptor ligands (SRL) in patients with acromegaly and to determine the clinical features associated with the prognosis.
We studied outcomes using the area under the receiver operating characteristics (AUROC) values, which were reported as the performance metric. To determine the importance of each feature and easy interpretation, Shapley Additive explanations (SHAP) values, which help explain the outputs of ML models, are used.
One-hundred fifty-two patients with acromegaly were included in the final analysis. The mean AUROC values resulting from 100 independent replications were 0.728 for postoperative 3 months remission status classification, 0.879 for remission at last visit classification, and 0.753 for SRL resistance status classification. Extreme gradient boosting model demonstrated that preoperative growth hormone (GH) level, age at operation, and preoperative tumor size were the most important predictors for early remission; resistance to SRL and preoperative tumor size represented the most important predictors of remission at last visit, and postoperative 3-month insulin-like growth factor 1 (IGF1) and GH levels (random and nadir) together with the sparsely granulated somatotroph adenoma subtype served as the most important predictors of SRL resistance.
ML models may serve as valuable tools in the prediction of remission and SRL resistance.
开发机器学习 (ML) 模型，以预测肢端肥大症患者术后缓解、末次就诊时缓解和对生长抑素受体配体 (SRL) 的耐药性，并确定与预后相关的临床特征。
我们使用接收者操作特征曲线 (AUROC) 值的曲线下面积作为性能指标来研究结局。为了确定每个特征的重要性和易于解释，使用 Shapley 加法解释 (SHAP) 值，它有助于解释 ML 模型的输出。
最终分析纳入了 152 例肢端肥大症患者。在 100 次独立复制中得出的术后 3 个月缓解状态分类的平均 AUROC 值为 0.728，末次就诊时缓解状态分类的平均 AUROC 值为 0.879，SRL 耐药状态分类的平均 AUROC 值为 0.753。极端梯度提升模型表明，术前生长激素 (GH) 水平、手术时年龄和术前肿瘤大小是早期缓解的最重要预测因素；SRL 耐药和术前肿瘤大小是末次就诊时缓解的最重要预测因素，术后 3 个月胰岛素样生长因子 1 (IGF1) 和 GH 水平（随机和最低点）以及稀疏颗粒状生长激素细胞腺瘤亚型是 SRL 耐药的最重要预测因素。
ML 模型可以作为预测缓解和 SRL 耐药性的有价值工具。
