Logo: to the web site of Uppsala University

Background: Long-term follow up of neurocognitive late effects in paediatric brain tumour survivors (PBTS) is essential, particularly after radiotherapy (RT). Processing speed impairments are among the most prevalent late effects, yet their relation to underlying oculomotor control and visual attention remains unclear. Eye movement metrics reflect the integrity of neural systems supporting attention, visual processing, and executive control, and may serve as sensitive potential biomarkers of neurocognitive impairment, complementing neuropsychological assessment. To explore this, we employed a pro–antisaccade task to examine fundamental oculomotor and executive control processes. The primary aim was to assess whether PBTS showed impaired pro-antisaccade performance compared with age-matched controls. Within the PBTS group, we also examined associations between saccadic latency, processing speed, and RT doses to brain structures important for visual and neurocognitive networks.

Methods: This long-term case-control study included 21 PBTS treated with proton and/or photon RT 8-20 years earlier and 50 age-matched controls. PBTS were aged 15-34 years (mean 23; 12 males and 9 females), and controls were aged 18-34 years (mean 26; 25 males, 24 females, and one non-binary participant). Eye movements were measured with an eye-tracker, and participants were instructed to look either towards an upcoming target (pro-saccades) or away from it (anti-saccades). Between-group differences were analysed using independent t-tests (Cohen´s d effect sizes) and modified t-tests for single samples to identify individual impairments. Within the PBTS group, we further explored associations between saccadic latency, processing speed, and mean RT doses to established and potential new organs at risk.

Results: PBTS demonstrated significantly longer pro- and antisaccade latencies than controls, with effect sizes ranging from small to large and the greatest impairments in those who received whole-brain RT. Modified t-tests revealed that nearly half of the PBTS performed in the impaired range on the anti-saccade task relative to the control reference group. Within the PBTS group, longer saccadic latency correlated with slower processing speed and with higher mean RT doses to the optic nerves and pons.

Conclusion: PBTS demonstrated long-term impairments in pro- and antisaccade performance many years after RT, with longer saccadic latencies linked to higher RT doses and slower processing speed. These patterns suggest lasting disruptions in visual attention, oculomotor, and executive networks. Eye-tracking offers a sensitive measure for detecting subtle neurocognitive late effects that can complement neuropsychological assessment and may help inform targeted rehabilitation strategies.