Quantitative Imaging Research and Artificial Intelligence Lab, Department of Radiation Oncology Unit II, Ida B Scudder Cancer Centre, Christian Medical College, Vellore, India.
Department of Neurosurgery, Christian Medical College, Vellore, India.
Acta Neurochir (Wien). 2024 Feb 20;166(1):91. doi: 10.1007/s00701-024-05977-4.
The WHO 2021 introduced the term pituitary neuroendocrine tumours (PitNETs) for pituitary adenomas and incorporated transcription factors for subtyping, prompting the need for fresh diagnostic methods. Current biomarkers struggle to distinguish between high- and low-risk non-functioning PitNETs. We explored if radiomics can enhance preoperative decision-making.
Pre-treatment magnetic resonance (MR) images of patients who underwent surgery between 2015 and 2019 with available WHO 2021 classification were used. The tumours were manually segmented on the T1w, T1-contrast enhanced, and T2w images using 3D Slicer. One hundred Pyradiomic features were extracted from each MR sequence. Models were built to classify (1) somatotroph and gonadotroph PitNETs and (2) high- and low-risk subtypes of non-functioning PitNETs. Feature were selected independently from the MR sequences and multi-sequence (combining data from more than one MR sequence) using Boruta and Pearson correlation. Support vector machine (SVM), logistic regression (LR), random forest (RF), and multi-layer perceptron (MLP) were the classifiers used. Data imbalance was addressed using the Synthetic Minority Oversampling TEchnique (SMOTE). Performance of the models were evaluated using area under the receiver operating curve (AUC), accuracy, sensitivity, and specificity.
A total of 222 PitNET patients (train, n = 149; test, n = 73) were enrolled in this retrospective study. Multi-sequence-based LR model discriminated best between somatotroph and gonadotroph PitNETs, with a test AUC of 0.84, accuracy of 0.74, specificity of 0.81, and sensitivity of 0.70. Multi-sequence-based MLP model perfomed best for the high- and low-risk non-functioning PitNETs, achieving a test AUC of 0.76, accuracy of 0.67, specificity of 0.72, and sensitivity of 0.66.
Utilizing pre-treatment MRI and radiomics holds promise for distinguishing high-risk from low-risk non-functioning PitNETs based on the latest WHO classification. This could assist neurosurgeons in making critical decisions regarding surgery or alternative management strategies for PitNETs after further clinical validation.
世界卫生组织（WHO）在 2021 年引入了垂体神经内分泌肿瘤（PitNETs）这一术语来描述垂体腺瘤，并纳入转录因子进行亚型分类，这促使我们需要新的诊断方法。目前的生物标志物难以区分高风险和低风险的无功能 PitNETs。我们探索了放射组学是否可以增强术前决策。
使用了 2015 年至 2019 年间接受手术且可获得世界卫生组织 2021 年分类的患者的术前磁共振（MR）图像。使用 3D Slicer 在 T1w、T1-对比增强和 T2w 图像上手动分割肿瘤。从每个 MR 序列中提取 100 个 Pyradiomic 特征。建立模型来分类（1）生长激素和促性腺激素垂体瘤和（2）高风险和低风险的无功能 PitNETs 亚型。使用 Boruta 和 Pearson 相关性从 MR 序列和多序列（合并来自多个 MR 序列的数据）中独立选择特征。支持向量机（SVM）、逻辑回归（LR）、随机森林（RF）和多层感知器（MLP）是使用的分类器。使用合成少数过采样技术（SMOTE）解决数据不平衡问题。使用接收器工作特征曲线下的面积（AUC）、准确性、敏感性和特异性来评估模型的性能。
这项回顾性研究共纳入 222 例垂体瘤患者（训练集，n=149；测试集，n=73）。基于多序列的 LR 模型在区分生长激素和促性腺激素垂体瘤方面表现最佳，测试 AUC 为 0.84，准确性为 0.74，特异性为 0.81，敏感性为 0.70。基于多序列的 MLP 模型在高风险和低风险无功能 PitNETs 方面表现最佳，测试 AUC 为 0.76，准确性为 0.67，特异性为 0.72，敏感性为 0.66。
利用术前 MRI 和放射组学有可能根据最新的世界卫生组织分类区分高危和低危无功能 PitNETs。这可以帮助神经外科医生在进一步的临床验证后，针对 PitNETs 的手术或替代管理策略做出关键决策。
