Abstract BACKGROUND Children treated for cancer may experience a decline in performance for a variety of reasons, one of which is the impact of treatment. Children treated for leukemias and brain tumours are at the highest risk of developing neurocognitive sequale (Daly and Brown, 2015). METHODS In our study we used Bayesian regression to identify assessment tools that would best predict the sequale of brain tumors (BT) and acute lymphoblastic leukemia (ALL) in children and adolescents. Using data from 15 patients with BT, 15 patients with (ALL) and 9 siblings controls we compared the predictive validity of the Wechsler Intelligence Scales for Children (WISC-V), Delis-Kaplan Executive Function System (D-KEFS), Continuous Performance Test (CPT-3), the Developmental Neuropsychological Assessment (NEPSY-II) and the parent and teacher report forms of the Achenbach Assessment System (ASEBA). RESULTS We found that the best predictors of ALL were d’ scores on the CPT (BF = 20.69, pd =.05) and CPT ommission scores (BF = 17.50, pd =.05), while the best predictor a BT was the WISC-IV Coding subtest (BF = 21.17, pd =.05) in addition to CPT omission scores (BF = 12.00, pd =.08) and HRT SD scores (BF = 11.91, pd =.08). CPT d’s scores were also predictive of Methotrexate treatment (MTX) (BF = 32.61, pd = 0.03), as were WISC-V Symbol Search scores (BF = 32.61, pd = 0.03) and the NEPSY-II Auditory Attention subtest (BF = 26.06, pd = 0.04). We also noted differential sensitivity of the ASEBA teacher report form for detecting deficits identified during the neuropsychological assessment. CONCLUSIONS We feel our findings will, together with previous meta-analyses, contribute to more efficient neuropsychological assessment in patients with BT and ALL. Neuropsychological asseessment is vital for identifying potential sequale of disease and treament in pediatric oncology.