Histogenesis of Atypical Teratoid Rhabdoid Tumors: Anatomical and Embryological Perspectives

Objectives: Atypical teratoid/rhabdoid tumors (ATRTs) are rare, malignant central nervous system (CNS) neoplasms that predominantly affect infants and young children. While ATRT arises throughout the CNS, its extracranial counterpart, malignant rhabdoid tumor, occurs in other organs. A single-institutional cohort is reviewed to map anatomic distribution of pediatric ATRTs and to integrate a literature review to contextualize ATRT histogenesis from anatomical and embryological perspectives. Methods: A retrospective review was conducted on a cohort of 50 pediatric patients with ATRT treated over 20 years. Demographic, surgical, and neuroimaging data were correlated to define tumor location, extent, and compartmental involvement. A focused literature review synthesized molecular subclassifications and proposed cells of origin/cytogenesis. Results: Of the 50 ATRTs, 18 (36%) were infratentorial, 15 (30%) supratentorial, 11 (22%) in the pineal region, and 6 (12%) in the spinal compartment. Among infratentorial tumors, 10 were centered in the fourth ventricle, with or without extension into the cerebellopontine angle (CPA) cistern; the remainder arose in the CPA. Among ATRTs of the cerebral hemispheres, 3 showed bi-hemispheric involvement crossing the falx cerebri. ATRTs of the pineal region predominantly originated from the superior medullary velum. These topographic data were corelated with embryological and molecular information available in the literature. Conclusions: ATRTs arise across diverse neuroanatomical compartments-including intraparenchymal, intraventricular, extra-axial, and extradural sites-underscoring biological heterogeneity. Inactivation of SMARCB1 is the defining molecular event and principal oncogenic driver, although the upstream mechanisms precipitating these alterations remain incompletely resolved. Molecular subgroups-ATRT-TYR, ATRT-SHH, and ATRT-MYC-display distinct age distributions and anatomic predilections, implicating developmental context in tumor initiation. The characteristic cellular admixture of rhabdoid cells with mesenchymal and/or epithelial differentiation, together with intra- and extra-axial and occasional extradural presentations, supports a model in which at least a subset of ATRTs may originate from neural crest-derived lineages, despite little or no neural crest contribution to brain parenchyma development. Neural plate border progenitors with bipotent features represent a plausible intraparenchymal cell of origin. Definitive resolution of these origins and the mechanisms of SMARCB1 disruption will require integrated approaches. Further investigations are warranted to clarify these mechanisms.