Longitudinal associations between peripheral neuropathy and neurocognitive decline in long-term survivors of pediatric cancer treated without central nervous system (CNS) directed therapy.

Abstract

10054 Background: Survivors of pediatric cancer treated without CNS-directed therapy are at risk of neurocognitive impairment and peripheral neuropathies. Using the St. Jude Lifetime Cohort, we investigated longitudinal associations with neuropathy to inform contributors to long-term cancer-related neurocognitive decline. Methods: Survivors of pediatric cancer (N=899 mean age 32 [SD 12], 24[10] years from diagnosis) treated without CNS-directed therapy completed neurocognitive testing and medical assessments at two timepoints (T1, T2 4.4[2.3] years apart). Peripheral sensory and motor neuropathy subscales combined with functional tasks were severity graded using a modified CTCAE (no neuropathy = none or grade 1 at T1 and T2; new onset = new grade 2 or 3 at T2; persistent = grade 2 or 3 at T1 & T2). Linear regression models estimated change in neurocognitive Z-score associated with neuropathy groups, adjusted for diagnosis age, time between visits, sex, race/ethnicity, high-dose intravenous methotrexate & baseline neurocognitive Z-score. Results: New onset sensory (n=33, 3.6%) and motor (n=35, 3.9%) neuropathy was associated with a 0.33 standard deviation (SD) decline in visual-motor processing speed (Table). New onset motor neuropathy was associated with a 0.50 SD decline in sustained attention as well as significant declines in verbal recall, working memory, and cognitive flexibility (p’s<0.05). Persistent sensory neuropathy (n=25, 2.8%) was associated with large declines in focused attention and cognitive flexibility (p’s<0.05). Persistent motor neuropathy (n=18, 2.0%) was associated with a 0.50 SD decline in sustained attention, verbal recall, and working memory (p’s<0.05). Conclusions: Peripheral neuropathies are associated with future neurocognitive decline in adult survivors of pediatric cancer. Research is needed to understand shared mechanisms (e.g. pain, inflammation) to inform targeted interventions to improve neuropathic symptoms and neurocognitive function. [Table: see text]