Frailty and neurocognitive decline in young adult survivors of childhood cancer: A longitudinal analysis from the St. Jude lifetime cohort.

Abstract

10555 Background: Among young adult childhood cancer survivors, 8% meet the criteria for frailty, an aging phenotype associated with poor health. Frailty is associated with neurocognitive decline in the elderly general population, but this association has not been examined in young adult survivors of childhood cancer. Methods: Childhood cancer survivors (N = 845, mean [SD] age 30 [7] years, 22 [7] years post diagnosis, 52% male) were clinically evaluated for prefrailty/frailty (defined as ≥2/≥3 of muscle wasting, muscle weakness, low energy expenditure, slow walking speed, exhaustion) and completed neuropsychological assessments at baseline and five years later. Linear regression models estimated mean differences in neurocognitive decline in prefrail/frail survivors vs. non-frail survivors adjusting for age, sex, race, CNS therapy (cranial radiation, intrathecal chemotherapy, neurosurgery), and baseline neurocognitive performance. P-values were adjusted for multiple comparisons using false discovery rate (FDR). Results: 18% and 6% of survivors were prefrail and frail at baseline. Baseline frailty was associated with declines in visual-motor processing speed, short-term memory, and sustained attention (Table). Prefrailty and frailty were associated with declines in focused attention and executive function (Table). No significant associations were observed between prefrailty or frailty and decline in global cognition, academics, motor processing speed, long-term memory, verbal learning, or verbal fluency despite significant baseline cross-sectional associations. Conclusions: Young adult prefrail and frail survivors had greater declines in attention and executive function compared to non-frail survivors, domains commonly associated with aging. These findings suggest that interventions designed to mitigate components of frailty may also mitigate or prevent neurocognitive decline. [Table: see text]