Quantitative MRI outcomes in child and adolescent leukemia survivors: Evidence for global alterations in gray and white matter

• Published in: NeuroImage clinical 2020-01, Vol.28, p.102428, Article 102428
• Main Authors: van der Plas, Ellen, Spencer Noakes, T. Leigh, Butcher, Darci T., Weksberg, Rosanna, Galin-Corini, Laura, Wanstall, Elizabeth A., Te, Patrick, Hopf, Laura, Guger, Sharon, Spiegler, Brenda J., Hitzler, Johann, Schachar, Russell J., Ito, Shinya, Nieman, Brian J.
• Format: Article
• Language: English
• Subjects: Acute lymphoblastic leukemia; Chemotherapy; Gray matter; Late effects; Magnetic resonance imaging; Regular; White matter
• ISSN: 2213-1582; 2213-1582
• DOI: 10.1016/j.nicl.2020.102428

•Studies of cognition in ALL survivors consistently show evidence of mild neurocognitive difficulties.•Structural MRI showed widespread differences between survivors and controls.•Gray and white matter volumes were respectively 5 and 6% smaller in survivors.•Survivors also exhibited altered white matter diffusion overall. Cure rates for pediatric acute lymphoblastic leukemia (ALL) have reached an all-time high (>90%); however, neurocognitive difficulties continue to affect quality of life in at least a subset of survivors. There are relatively few quantitative neuroimaging studies in child and adolescent ALL survivors treated with chemotherapy only. Use of different outcome measures or limited sample sizes restrict our ability to make inferences about patterns of brain development following chemotherapy treatment. In this study, we used magnetic resonance imaging (MRI) to evaluate brain outcomes in ALL survivors, comparing against a group of typically developing, cancer free peers. Participants included 71 ALL survivors, on average 8 years after diagnosis and 8–18 years of age, and 83 typically developing controls. Anatomical MRI was performed to evaluate brain structure; diffusion and magnetization transfer MRI were used to examine brain tissue microstructure. Successful MRI scans were acquired in 67 survivors (94%) and 82 controls (99%). Structurally, ALL survivors exhibited widespread reductions in brain volume, with 6% less white matter and 5% less gray matter than controls (p = 0.003 and 0.0006 respectively). Much of the brain appeared affected – 71 of 90 evaluated structures showed smaller volume – with the most notable exception being the occipital lobe, where no significant differences were observed. Average full-scale IQ in the survivor and control groups were 95 (CI 92–99) and 110 (CI 107–113), respectively. Using data from the NIH Pediatric MRI Data Repository, we evaluated the extent to which elevated IQ in the control group might affect the structural differences observed. We estimated that two thirds of the observed brain differences were attributable to ALL and its treatment. In addition to the structural changes, survivors showed, on average, globally lower white matter fractional anisotropy (-3%) and higher radial diffusivity (+5%) (p < 10−6), but no differences in magnetization transfer ratio. Neuroanatomical alterations in late childhood and adolescent ALL survivors treated with chemotherapy-only protocols are widespread, with white matt