Characterization of brain development with neuroimaging in a female mouse model of chemotherapy treatment of acute lymphoblastic leukemia

Background: Survivors of pediatric acute lymphoblastic leukemia (ALL) exhibit abnormal neurocognitive outcomes that are possibly due to exposures to neurotoxic chemotherapy agents. This study aimed to determine the feasibility of characterizing long-term neuroanatomical changes with in vivo neuroimaging in a preclinical model of treatment for ALL. Methods: Female mice (C57BL/6) were randomly assigned to a saline control group (n=10) or a treatment group (n=10) that received intrathecal methotrexate and oral dexamethasone (IT-MTX + DEX). Mice were subsequently scanned three times on a 7T MRI at ages 3, 6, and 12 months (T1, T2, and T3, respectively), which corresponds with human age -equivalents spanning early to late adulthood. Regional brain volumes were automatically segmented, and volume change between timepoints (i.e., T1 to T2; and T2 to T3) were compared between groups (i.e., saline vs. IT-MTX + DEX). Results: Five mice in the IT-MTX + DEX group, and seven mice in the saline group completed all three scans. Between T1 and T2, volumetric change was significantly different between groups in total gray matter [estimate =2.06, 95% confidence interval (CI): 0.27-3.84], the cerebrum (estimate =1.62, 95% CI: 0.14-3.09), claustrum (estimate =0.06, 95% CI: 0.02-0.09), amygdala (estimate =0.16, 95% CI: 0.03-0.29), and striatum (estimate =0.18, 95% CI: 0.01-0.35), with the IT-MTX + DEX group exhibiting a more robust increase in volume than the saline -treated group. Between T2 and T3, group differences in structural brain development were evident for total white matter (estimate =-0.14, 95% CI: -0.27 to -0.01), and the corpus callosum (estimate =-0.09, 95% CI: -0.19 to 0.00) and amygdala (estimate =-0.05, 95% CI: -0.10 to 0.00). In contrast to the rapid brain growth observed earlier in development (i.e., T1 to T2), the IT-MTX + DEX group exhibited an attenuated increase in volume relative to the saline -treated group between T2 and T3. Conclusions: The results demonstrate feasibility of modeling pediatric ALL treatment in a preclinical model and highlight the potential of using preclinical neuroimaging models to gain insight into brain development throughout survivorship.