Approximately 2,400 children and adolescents are diagnosed with acute lymphoblastic leukemia (ALL) each year in the United States. The probability of 5-year overall survival is now at 90%, so there is a compelling need to minimize neurotoxicity and improve the quality of life for childhood ALL survivors. Even with the elimination of radiation therapy as a component of most ALL therapies, survivors remain subject to increased cognitive impairments secondary to disease and treatment. While neurocognitive performance for ALL survivors as a whole appears normal, a disproportionate number of survivors have impaired performance in attention (44%) and working memory (66%). In our previously funded study, we determined the prevalence of leukoencephalopathy and carefully characterized the structural changes apparent on MRI during treatment and later neurocognitive deficits in attention and memory. We demonstrated that methotrexate exposure results in 1) white matter damage preferentially in the frontal lobes and 2) specific neurocognitive deficits associated with frontal lobe functioning. Imaging alone was unable to accurately predict later neurocognitive deficits. We now hypothesize that genetic polymorphisms in the coding of key folate pathway enzymes mediate the vulnerability of patients treated for ALL to methotrexate neurotoxicity. We will prospectively test the hypothesis that genetic polymorphisms in the folate pathway will result in a neuroimaging phenotype (differing degrees of myelin disruption) early in therapy which can be incorporated into a model to identify those patients at greatest risk of developing specific neurocognitive deficits in frontal lobe functioning at completion of therapy (Aim 1). Variable methotrexate toxicity related to genetic differences in methotrexate metabolism causes altered rates of cortical thinning in frontal cortex over the course of therapy which will result in deficits in neurocognitive measures of frontal lobe functioning (Aim 2).Disrupted myelin and abnormal cortical thickness diminish the efficiency of neural processing, especially in prefrontal cortex, which leads ultimately to altered patterns of brain activity and therapy-induced cognitive deficits (Aim 3). We build upon the experience of the previous study to propose a shift in the research paradigm through development of a neurocognitive late effects risk model, which will greatly advance current research, and potentially clinical practice, by establishing the relationship between genetic polymorphisms in the folate pathway and frontal cortex structure and function.