White matter anisotropy in post-treatment childhood cancer survivors: Preliminary evidence of association with neurocognitive function

Abstract

Purpose: We aim to determine if the loss of white matter fractional anisotropy (FA), measured by diffusion tensor magnetic resonance imaging (DTI), in post-treatment childhood medulloblastoma (MED) and acute lymphoblastic leukemia (ALL) survivors correlate with intelligence quotient (IQ) scores. Materials and Methods: MED and ALL survivors (n = 30; 20 male, 10 female; age range, 6.0 to 22.1 years; mean, 13.1 years) were recruited for DTI and IQ tests. In this cross-sectional study, age-matched normal control (n = 55; 32 male, 23 female; age range, 6.0 to 23 years; mean, 12.1 years) DTI was obtained to compute percentage difference in white matter FA (AFA%) for each patient compared with the age-matched control group. Multivariate regression analysis was performed to determine the relationships between AFA%, age at treatment, irradiation dose, time interval from treatment, and full-scale IQ (FSIQ), verbal IQ (VIQ), and performance IQ (PIQ). Receiver operating characteristics curves were used to determine the best AFA% cutoffs for predicting FSIQ, VIQ, and PIQ of less than 85. Results: AFA% had a significant effect on FSIQ (adjusted r 2 = 0.439; P < .001), VIQ (adjusted r 2 = 0.237; P = .028), and PIQ (adjusted r 2 = 0.491; P < .001) after adjusting for the effects of age at treatment, irradiation dose, and time interval from treatment. The best AFA% value to predict less than 85 scores in FSIQ, VIQ, and PIQ was -3.3% with specificities of 100% and sensitivities ranging from 77.8% to 87.5%. Conclusion: Our preliminary findings suggest that white matter FA may be a clinically useful biomarker for the assessment of treatment-related neurotoxicity in post-treatment childhood cancer survivors. © 2006 by American Society of Clinical Oncology.