Anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis is an autoimmune syndrome in young women that is often accompanied by an ovarian teratoma (NMDAR-E teratoma). A prevailing theory implicates that the generation of autoantibodies to NMDAR on neurons in the central nervous system is triggered by neuroglial tissue in the associated teratoma. The histopathology of NMDAR-E teratomas has not been fully elucidated but limited studies have focused on alterations in neuroglial tissues and immune cell populations. We hypothesized that evidence of antibody generation in NMDAR-E teratomas could be detected by colocalized neuroglial tissue and lymphoid aggregates with germinal centers as well as by alterations in the composition and morphology of neuroglial tissues. The study compared 12 NMDAR-E teratomas (11 ovarian, 1 mediastinal) with 61 control teratomas containing neuroglial tissue from women without NMDAR-E. NMDAR-E teratomas were significantly smaller and were composed of a higher percentage of neuroglial tissue than control teratomas. Many NMDAR-E teratomas did not exhibit typical gross pathologic features of a mature cystic teratoma, but were composed of predominately solid tissue (so-called Rokitansky nodule). Colocalized neuroglial tissue and lymphoid aggregates with germinal centers were present in 11/12 NMDAR-E teratomas, predominantly within the Rokitansky nodule, but only in 4/61 control teratomas (P<0.0001). There was a relative paucity of mature neurons in NMDAR-E teratomas as well as a hypercellular astrocyte population, while there were less prominent or no differences in the presence or composition of diffuse inflammatory infiltrates, lymphoid aggregates without germinal centers, ganglion cell clusters or oligodendrocytes between NMDAR-E teratomas and control teratomas. We conclude that the presence of colocalized neuroglial tissue and lymphoid aggregates with germinal centers along with a general paucity of neurons should prompt clinical consideration for NMDAR-E even in asymptomatic women, as the symptoms may occasionally develop after an otherwise incidental oophorectomy. Tissue sampling should be directed to the Rokitansky nodule, when present, to identify neuroglial tissues; complete microscopic examination of the ovarian specimen should be considered if gross pathologic features of teratoma are not present. The significance of the altered neuroglial cell populations and potential relationship to the pathogenesis of NMDAR-E merit further study.