Baseline Neurocognitive Functioning in Children/Teenagers and Young Adults Treated for Central Nervous System...

Central nervous system (CNS) malignancies represent approximately 30% of all paediatric and teenage and young adult (TYA) solid tumours. With improving survival rates exceed 70% at 5 years, focus has shifted towards reducing treatment-related late effects, particularly neurocognitive decline. Proton therapy (PT) has emerged as a gold standard in paediatric CNS malignancies due to its ability to spare surrounding healthy tissues. However, previous studies evaluating neurocognitive outcomes after PT have been limited by retrospective designs, inadequate baseline assessments, and inconsistent measurement tools. This study aims to describe baseline neurocognitive functioning in children, teenagers, and young adults with CNS malignancies prior to PT, providing unprecedented granular data on baseline cognitive status.

We analysed baseline neurocognitive data from 227 patients under 25 years of age (mean age: 11.8 years, 74% under the age of 16 years, 59.5% male) referred for PT for CNS/head and neck malignancies at the Christie national health service (NHS) PT centre between November 2018 and April 2022. Patients completed an age-appropriate standardised battery of neuropsychological assessments measuring multiple cognitive domains: intellectual functioning (Full Scale Intelligence Quotient [FSIQ], perceptual/fluid reasoning, working memory, visual-motor skills (visual-motor integration [VMI]), verbal semantic memory, attention, processing speed, and reading. Standard scores (mean = 100, standard deviation [SD = 15] or scaled scores (mean = 10, SD = 3) were calculated using age-based normative data. Statistical analyses examined relationships between cognitive measures and patient/previous treatments characteristics. Parametric tests (t-tests, ANOVA) and their non-parametric equivalents (Mann-Whitney U or Kruskal-Wallis) were used for group comparisons, while Spearman correlations assessed continuous-variable relationships.

Analysis revealed a pattern of specific deficits rather than global cognitive impairment. Processing speed (Coding mean = 8.63, Symbol Search mean = 8.66), motor skills (Pegboard mean = 8.03), and VMI (Beery VMI mean = 90.9) were most severely affected, with 31.25% (Coding), 25.00% (Symbol Search), 34.53% (Pegboard), and 29.01% (Beery VMI) of participants scoring >1 SD below population means. Attention and working memory showed moderate deficits (Test of Everyday Attention [TEA]/Test of Everyday Attention for Children [TEACh] mean = 9.16, Digit Span forward mean = 9.43, Digit Span backward mean = 9.42). Verbal abilities and intellectual functioning remained relatively preserved (FSIQ mean = 102.32, Vocabulary mean = 9.67), with FSIQ showing more participants scoring above average (24.56%) than below average (19.3%). Statistical analysis found minimal significant relationships between cognitive measures and patient/treatment characteristics, with only visual motor reasoning showing significant association with primary tumour site (P = 0.024) after correction for multiple comparisons.

This comprehensive assessment of baseline neurocognitive function reveals specific deficits in processing speed, motor skills, and VMI in paediatric and young adult CNS tumour patients that are evident in a significant number of patients prior to PT. By identifying these vulnerabilities before treatment initiation, clinicians can establish more accurate baselines for monitoring treatment-related cognitive changes, implement proactive interventions, and develop personalised compensatory strategies leveraging preserved cognitive strengths. This prospective approach enhances the precision of cancer-related cognitive impairment research and improves patient quality of life by addressing cognitive needs throughout the treatment pathway.