Consistent Physical Activity and Future Neurocognitive Problems in Adult Survivors of Childhood Cancers: A Report From the Childhood Cancer Survivor Study

PURPOSE To investigate longitudinal associations between physical activity (PA) and neurocognitive problems in adult survivors of childhood cancer. METHODS A total of 12,123 5-year survivors diagnosed between 1970 and 1999 (median [range] age at diagnosis, 7 [0-21] years, time since diagnosis at baseline, 16 [6-30] years) and 720 siblings self-reported PA and neurocognitive problems. PA was collected at baseline, and PA and neurocognitive data were obtained 7 (1-12) years and 12 (9-14) years later. PA consistency was defined as any combination of >= 75 minutes of vigorous or 150 minutes of moderate activity per week on all surveys. Multiple linear regressions, conducted separately for CNS and non-CNS survivors, identified associations between PA consistency and neurocognitive outcomes (expected mean, 50; standard deviation [SD], 10). Mediating effects of body mass index (BMI) and chronic health conditions (CHCs) were evaluated. RESULTS Survivors were less likely than siblings to report consistent PA (28.1% v 33.6%) and more likely to report problems in Task Efficiency (T-scores mean SD: siblings, 50.0 +/- 0.4; CNS, 61.4 +/- 0.4; non-CNS, 53.3 +/- 0.3), Emotion Regulation (siblings, 51.4 +/- 0.4; CNS, 54.5 +/- 0.3; non-CNS 53.4 +/- 0.2), and Memory (siblings, 50.8 +/- 0.4; CNS, 58.9 +/- 0.4; non-CNS, 53.5 +/- 0.2; all P < .001). Survivors of CNS cancers (52.8 +/- 0.3) also reported poorer Organization than siblings (49.9 +/- 0.4; P < .001). After adjusting for age at diagnosis, age at questionnaire, emotional distress, and cancer treatment exposures, consistent PA was associated with fewer neurocognitive problems compared with consistent inactivity for both CNS and non-CNS groups (T-score differences ranging from -7.9 to -2.2) and larger neurocognitive improvements over time (-6.0 to -2.5), all P <= .01. BMI and severe CHCs partially mediated the PA-neurocognitive associations, but the mediation effects were small (change in beta <= 0.4). CONCLUSION Adult survivors of childhood cancer who report more consistent PA have fewer neurocognitive problems and larger improvements in these concerns many years after treatment.