BACKGROUND: Therapy of infants with brain tumors predisposes these patients to increased risks for cognitive sequelae, especially following radiotherapy. Neuropsychological outcome gains importance for those 40-60% of patients with an atypical teratoid/rhabdoid tumor (ATRT) who survive beyond 2 years. Still, reports on cognitive late-effects in children with ATRT are scarce compared to other pediatric brain tumor groups. We analyzed neuropsychological outcome for long-term ATRT-survivors registered in EU-RHAB and infant low-grade glioma (LGG) survivors from the SIOP-LGG 2004-study and LGG-registry. PATIENTS+METHODS: Age at diagnosis of both cohorts was 0-36 months. ATRT-patients (n=13) treated with up to 54Gy radiotherapy (median age 22 months (±7.1)) were evaluated with the “ATRT-Neuropsychology” tool based on SIOPE-BTG QoS-Group recommendations at median 6.8 years (±2.8) after diagnosis. LGG-patients (n=15) treated without radiotherapy (4/15 with chemotherapy) were analyzed with the German “Neuropsychological-Basic-Diagnostic” tool 5.2 years (±0.6) post-diagnosis. RESULTS: The ATRT- vs. LGG-cohorts were comparable for median age at diagnosis, sex-ratio and tumor-localization, though they differed slightly in median age at assessment (9.5/7.2 years (±2.5/1.1)). Results of age-appropriate tests showed increased impairments for ATRT-patients in fluid intelligence (FI) (p=.006, d=1.214) and in visual-spatial processing (VSP) (p<.001, d=2.233) compared to LGG-patients. The median for neuropsychological test results of ATRT-patients spanned from considerably below the normal to the lower normal range (median=65-90), while results of LGG-patients were mostly in the lower normal range (median=83-103). Results for psychomotor speed abilities (PMS) were distinctly below the norm for both patient groups (p=.002-.007). CONCLUSION: Infant ATRT- and LGG-patients develop significant impairments in PMS abilities following multimodal treatment. Long-term survivors of ATRT suffer from additional FI and VSP deficits. Our data suggest that high malignancy requiring multimodal treatment determines the inferior cognitive outcome for the ATRT-cohort. Long-term neuropsychological monitoring (and treatment options) should be implemented as standard of care in ATRT- and LGG-trials.