LAUSR

BACKGROUND Some adult survivors of childhood cancers develop frailty at higher rates than expected based on their chronological age. This study examined the incidence of frailty among survivors at ten or more years after diagnosis, frailty prevalence 5 years later, and risk factors for becoming frail. METHODS Frailty was measured at study entry and five years later. Logistic regression tested the associations of several factors with having frailty at five years for all participants and separately by sex and by study entry frailty status. Cox models evaluated the hazard of death associated with entry frailty considering covariates. RESULTS Cancer survivors (range = 0-22 years at diagnosis, median = 7 years) were age 18-45 years (median = 30 years) at study entry. Frailty prevalence increased from 6.2% (95% confidence interval [CI] = 5.0%-7.5%) to 13.6% (95% CI = 11.9%-15.4%) at 5 years. Risk factors for frailty at follow-up among all survivors included chest radiation ≥20 Gy (odds ratio [OR] = 1.98, 95% CI = 1.29-3.05), cardiac (OR = 1.58, 95% CI = 1.02-2.46) and neurological (OR = 2.58, 95% CI = 1.69-3.92) conditions, lack of strength training (OR = 1.74, 95% CI = 1.14-2.66), sedentary lifestyle (OR = 1.75, 95% CI = 1.18-2.59), and frailty at study entry (OR = 11.12, 95% CI = 6.64-18.61). The strongest risk factor for death during follow-up was prior frailty (OR = 3.52, 95% CI = 1.95-6.32). CONCLUSIONS Prevalent frailty more than doubled at 5 years after study entry among adult childhood cancer survivors. Frailty at entry was the strongest risk factor for death. Because treatment exposures cannot be changed, mitigation of other risk factors for frailty, including lack of strength training and sedentary lifestyle may decrease risk of adverse health events and improve longevity in survivors.