Department of Pediatric Hematology/Oncology and Cell and Gene Therapy, IRCCS Ospedale Pediatrico Bambino Gesù, Rome, Italy.
Department of Imaging, IRCCS Ospedale Pediatrico Bambino Gesù, Rome, Italy.
Pediatr Blood Cancer. 2021 Sep;68(9):e29110. doi: 10.1002/pbc.29110. Epub 2021 May 18.
MYCN amplification represents a powerful prognostic factor in neuroblastoma (NB) and may occasionally account for intratumoral heterogeneity. Radiomics is an emerging field of advanced image analysis that aims to extract a large number of quantitative features from standard radiological images, providing valuable clinical information.
In this retrospective study, we aimed to create a radiogenomics model by correlating computed tomography (CT) radiomics analysis with MYCN status. NB lesions were segmented on pretherapy CT scans and radiomics features subsequently extracted using a dedicated library. Dimensionality reduction/features selection approaches were then used for features procession and logistic regression models have been developed for the considered outcome.
Seventy-eight patients were included in this study, as training dataset, of which 24 presented MYCN amplification. In total, 232 radiomics features were extracted. Eight features were selected through Boruta algorithm and two features were lastly chosen through Pearson correlation analysis: mean of voxel intensity histogram (p = .0082) and zone size non-uniformity (p = .038). Five-times repeated three-fold cross-validation logistic regression models yielded an area under the curve (AUC) value of 0.879 on the training set. The model was then applied to an independent validation cohort of 21 patients, of which five presented MYCN amplification. The validation of the model yielded a 0.813 AUC value, with 0.85 accuracy on previously unseen data.
CT-based radiomics is able to predict MYCN amplification status in NB, paving the way to the in-depth analysis of imaging based biomarkers that could enhance outcomes prediction.
MYCN 扩增是神经母细胞瘤（NB）强有力的预后因素，偶尔可能导致肿瘤内异质性。放射组学是一种新兴的高级图像分析领域，旨在从标准放射图像中提取大量定量特征，提供有价值的临床信息。
在这项回顾性研究中，我们旨在通过将 CT 放射组学分析与 MYCN 状态相关联，创建一个放射基因组学模型。在治疗前 CT 扫描上对 NB 病变进行分割，随后使用专用库提取放射组学特征。然后使用降维/特征选择方法对特征进行处理，并为考虑的结果开发逻辑回归模型。
本研究共纳入 78 例患者作为训练数据集，其中 24 例存在 MYCN 扩增。共提取 232 个放射组学特征。通过 Boruta 算法选择 8 个特征，最后通过 Pearson 相关分析选择 2 个特征：体素强度直方图的平均值（p=0.0082）和区域大小非均匀性（p=0.038）。五次重复三折交叉验证逻辑回归模型在训练集上的 AUC 值为 0.879。然后将该模型应用于 21 例独立验证队列的患者，其中 5 例存在 MYCN 扩增。该模型的验证在以前未见的数据上产生了 0.813 AUC 值，准确率为 0.85。
基于 CT 的放射组学能够预测 NB 的 MYCN 扩增状态，为深入分析成像生物标志物铺平了道路，这些生物标志物可以增强对结果的预测。
