Endocrine Unit and Neuroendocrinology Research Center, Medical School and Hospital Universitário Clementino Fraga Filho-Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brazil.
Neuroendocrine Unit-Instituto Estadual do Cérebro Paulo Niemeyer, Secretaria Estadual de Saúde , Rio de Janeiro, RJ, Brazil.
J Clin Endocrinol Metab. 2021 Jun 16;106(7):2047-2056. doi: 10.1210/clinem/dgab125.
Artificial intelligence (AI), in particular machine learning (ML), may be used to deeply analyze biomarkers of response to first-generation somatostatin receptor ligands (fg-SRLs) in the treatment of acromegaly.
To develop a prediction model of therapeutic response of acromegaly to fg-SRL.
Patients with acromegaly not cured by primary surgical treatment and who had adjuvant therapy with fg-SRL for at least 6 months after surgery were included. Patients were considered controlled if they presented growth hormone (GH) <1.0 ng/mL and normal age-adjusted insulin-like growth factor (IGF)-I levels. Six AI models were evaluated: logistic regression, k-nearest neighbor classifier, support vector machine, gradient-boosted classifier, random forest, and multilayer perceptron. The features included in the analysis were age at diagnosis, sex, GH, and IGF-I levels at diagnosis and at pretreatment, somatostatin receptor subtype 2 and 5 (SST2 and SST5) protein expression and cytokeratin granulation pattern (GP).
A total of 153 patients were analyzed. Controlled patients were older (P = .002), had lower GH at diagnosis (P = .01), had lower pretreatment GH and IGF-I (P < .001), and more frequently harbored tumors that were densely granulated (P = .014) or highly expressed SST2 (P < .001). The model that performed best was the support vector machine with the features SST2, SST5, GP, sex, age, and pretreatment GH and IGF-I levels. It had an accuracy of 86.3%, positive predictive value of 83.3% and negative predictive value of 87.5%.
We developed a ML-based prediction model with high accuracy that has the potential to improve medical management of acromegaly, optimize biochemical control, decrease long-term morbidities and mortality, and reduce health services costs.
人工智能（AI），尤其是机器学习（ML），可用于深入分析第一代生长抑素受体配体（fg-SRL）治疗肢端肥大症时的反应生物标志物。
建立肢端肥大症患者对 fg-SRL 治疗反应的预测模型。
纳入经初次手术治疗未能治愈且术后接受 fg-SRL 辅助治疗至少 6 个月的肢端肥大症患者。如果患者的生长激素（GH）<1.0ng/ml 且年龄校正的胰岛素样生长因子（IGF）-I 水平正常，则认为患者得到控制。评估了 6 种 AI 模型：逻辑回归、k 最近邻分类器、支持向量机、梯度提升分类器、随机森林和多层感知器。分析中包含的特征包括诊断时的年龄、性别、诊断时和治疗前的 GH 和 IGF-I 水平、生长抑素受体亚型 2 和 5（SST2 和 SST5）蛋白表达以及角蛋白颗粒模式（GP）。
共分析了 153 例患者。得到控制的患者年龄更大（P=0.002）、诊断时 GH 水平更低（P=0.01）、治疗前 GH 和 IGF-I 水平更低（P<0.001），且更常存在颗粒密集（P=0.014）或高度表达 SST2（P<0.001）的肿瘤。表现最佳的模型是支持向量机，其特征包括 SST2、SST5、GP、性别、年龄以及治疗前 GH 和 IGF-I 水平。该模型的准确率为 86.3%，阳性预测值为 83.3%，阴性预测值为 87.5%。
我们开发了一种基于 ML 的预测模型，其具有较高的准确率，有可能改善肢端肥大症的医疗管理，优化生化控制，降低长期发病率和死亡率，并降低医疗服务成本。
