Ovarian central nervous system-type tumors: integrated neuropathological and methylation-based classification reveals...

Central nervous system (CNS)-type tumors may occur in the ovary, often associated with a mature teratomatous component. Because of their rarity, little is known about the tumor types historically designated within the primitive neuroectodermal tumors (PNET) terminology and whether they share histopathological and molecular features akin to those of their CNS counterparts. Herein, we retrospectively investigated data from 13 ovarian tumors, initially diagnosed as either "PNETs" or CNS-type neoplasms. For each tumor we performed comprehensive histopathologic, genetic and epigenetic analyses and retrieved clinical data when available. Integrated diagnoses were established after a central review of histopathological and molecular data, the following entities were identified: four embryonal tumors with multilayered rosettes (non C19MC-altered), three medulloblastomas, SHH-activated, three ependymomas not elsewhere classified, one Ewing sarcoma, one sarcoma, DICER1-mutant, and one peripheral neuroblastoma. Interestingly, none of the ETMRs harbored a C19MC amplification or DICER1 mutation. The three medulloblastomas, SHH-activated were histopathologically and molecularly similar to their CNS counterparts. Ependymomas did not show any classifying molecular alteration and presented a distinct epigenetic profile when compared with CNS ependymomas. These results indicate that ovarian "PNETs" comprise a heterogeneous spectrum of CNS and extra-CNS embryonal or non-embryonal tumor types, and that brain tumor methylation classifiers may be used to classify these tumors. Moreover, these components are characterized by distinct molecular alterations from primary CNS tumors, without C19MC alterations for ETMRs, with an overrepresented SHH-subgroup for medulloblastomas, and with an epigenetic profile distinct from CNS counterparts in ovarian ependymomas. These data need to be confirmed before they can be incorporated into future patient personalized treatment.