Trial Outcomes & Findings for Web-Based Physical Activity Intervention in Improving Long Term Health in Children and Adolescents With Cancer (NCT NCT03223753)

Recruitment period: 08/14/2017-1/10/2025; first enrollment 11/29/2017, last enrollment 4/26/2024. 118 COG sites opened the protocol, with 66 enrolling 1 or more participant in clinic.

Enrollment and randomization occurred concurrently.

Web-Based Physical Activity Intervention in Improving Long Term Health in Children and Adolescents With Cancer

PCI is calculated in beats per meter from the average heart rate during the walking test minus the resting heart rate divided by meters walked per minute. Differences in PCI will be compared between groups at 24 weeks, the end of the intervention. Intent to treat analysis of variance (either transforming the data or employing a non-parametric equivalent if the data are not normally distributed) controlling for the stratification factors (sex, age, and treatment related risk group) will be used for comparison. General linear mixed models will be utilized to evaluate the effects of group assignment on changes in PCI over time to account for repeated measures on individual children, and for potential random effects such as original treating institution. Changes from baseline will be estimated as differences between model-based least squares means at specified time points (e.g., 24 weeks minus baseline) derived from the mixed model.

Markers of Cardiometabolic health include blood pressure, body mass index, waist to height ration, fasting insulin, glucose, and lipid levels. Differences in these markers between randomized groups and the effects of group assignment on changes over time will be evaluated.

Markers of an Inflammatory state include High sensitivity C-reactive protein, interleukin-6, and tumor necrosis factor alpha. Differences in these markers between groups and the effects of group assignment on changes over time will be evaluated.

Will be assessed using Pediatric Quality of Life (PedsQL) 4.0 Generic Core Scale. The 23-item PedsQL Generic Core Scales were designed to measure the core dimensions of health as delineated by the World Health Organization, as well as role (school) functioning. Scores range from 0-100, higher is better. There are four scales (Physical Functioning, Emotional Functioning, Social Functioning, and School Functioning). Emotional, Social and School Functioning can be combined to create a psychosocial summary (summed and divided by total number of items). Differences between groups and the effects of group assignment on changes over time will be evaluated.

Will be assessed using Pediatric Quality of Life Multidimensional Fatigue Scale. The PedsQL Multidimensional Fatigue Scale was designed as a generic symptom-specific instrument to measure fatigue in patients with acute and chronic health conditions as well as healthy school and community populations. It has three dimensions, general fatigue, sleep/rest fatigue and cognitive fatigue. Scores range from 0-100 and items are summed and divided by the total to get the mean. Differences between groups and the effects of group assignment on changes over time will be evaluated.

Will be assessed using parent report. Differences between groups and the effects of group assignment on changes over time will be evaluated.

Markers of cardiometabolic health include blood pressure, body mass index, waist to height ration, fasting insulin, glucose, and lipid levels. It will be determined if the effect of an interactive web-based physical activity intervention on markers of cardiometabolic health is mediated by changes in fitness among children and adolescents following treatment for acute lymphoblastic leukemia. A causal inference approach will evaluate whether group assignment results in an improved outcome in the presence of a lower PCI. This requires decomposing the averaged total effect into indirect and direct effects of group assignment of on the cardiometabolic outcomes. A principal stratification method will be used that classifies participants into strata based on their performance on the PCI (mediator variable) for each group of the randomized intervention. Mediation analyses would then be based on intent-to-treat effects of group assignment on cardiometabolic health outcome within each strata.