INTRODUCTION
Neuroblastoma (NB) is the most common extracranial solid tumor of childhood and arises in the developing sympathetic nervous system, resulting in tumors of the adrenal glands and/or sympathetic ganglia. NB exhibits different clinical presentations and outcomes, in function of tumor biology, ranging from extremely aggressive behavior to spontaneous regression during infancy [1]. Treatment decisions in NB are based on prognostic factors such as age at diagnosis, the tumor stage (according to the International Neuroblastoma Staging System (INSS) [2] or the International Neuroblastoma Risk Grouping Staging System (INRGSS) [3], the histology according to the International Neuroblastoma Pathology Classification (INPC) [4, 5], and molecular and cytogenetic characteristics [6]. The molecular profiling of NB is based on biopsies or surgical specimens following resection. However, a significant challenge for the accuracy of molecular profiling may arise because of tumor heterogeneity and partial sampling of lesions. MYCN amplification occurs in around 20% of NB and is par excellence associated with poor survival with treatment implications [7].
MYCN amplification occasionally shows intratumoral heterogeneity (hetMNA) and may depend on the analyzed amount of tumor material and on the applied techniques. More rarely, heterogeneity may occur spatially within the tumor and metastases or temporally, during disease evolution. Due to the unclear biological and clinical impact of hetMNA, no specific therapeutic strategies exist for those patients [8, 9]. Standard radiological imaging can identify macroscopic evidence of hetMNA, providing useful information on disease status. Computed tomography (CT) and magnetic resonance imaging (MRI) are recommended for diagnosis and staging of NB, although there currently is no consensus about the optimal imaging modality for assessing local disease. Both techniques are routinely used, depending on availability and other factors. CT uses ionizing radiation and contrast medium, but it is characterized by very fast acquisitions thereby eliminating the need for sedation. MRI requires long acquisition times and is considered superior in patients with intraspinal extension [10].
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 [11]. To date radiomics is mainly used in oncology to predict diagnosis, survival, progression of disease and gene mutation (i.e. radiogenomics) [12-15]. Since radiomics has the potential for predicting molecular characteristics, it can potentially be used for determination of mutational or amplification status of specific genes, as suggested by promising preliminary experiences also in the specific field. The purpose of this study was to evaluate if radiomics features extracted from standard staging CT scans could help in predicting MYCN amplification status in patients affected by NB.