Physical activity is associated with exercise capacity among patients undergoing chemotherapy: a pilot randomized controlled trial
Data files
Aug 14, 2025 version files 4.53 KB
-
PLOS_data_8.13.25.csv
2.47 KB
-
README.md
2.06 KB
Abstract
Exercise intolerance is common among cancer patients undergoing active treatments.\
We conducted a pilot study in patients receiving potentially cardiotoxic therapies to investigate the\
impact of participation in a 6-month physical activity intervention (PAI) versus a healthy lifestyle\
education control intervention (HLI) on accelerometer-derived measures of PA (steps/day or minutes\
of moderate to vigorous PA [MVPA]/week) and whether these measures were associated with\
submaximal exercise capacity (6-minute walk distance [6MWD]). Participants with breast\
cancer or lymphoma (n=33) were randomized (2:1) into PAI or HLI groups. Exercise training was\
patient-centered, tailored to treatment and functional status. Objective assessments of PA (steps/day,\
MVPA/week), and submaximal exercise capacity (6MWD) were completed at baseline, 3-, and 6-\
months. Separate linear mixed-effects regression models examined changes in 6MWD, controlling\
for baseline values, a treatment group x time interaction, steps/day, or MVPA. At baseline,\
27 participants in the PAI (n=20) and HLI (n=7) groups who completed baseline and at least 1 follow\
up visit, had a mean age of 53 ±17 vs 57 ±13 years, took on average 8,330 ±2520 vs 7800 ±3830\
steps/day, did an average of 133 ±44 vs 109 ±87 minutes of MVPA/week, and had a mean 6MWD of\
496 ±94 vs 440 ±18 meters, respectively. Models adjusted for baseline values did not show any\
difference in 6MWD by intervention group or time. Further, no significant differences in steps/day or\
MVPA mins/week were found between groups over time. A post-hoc correlation analysis did find\
significant associations between both steps/day (r=0.51, p=.04) and MVPA mins/week (r=0.48,\
p≤.05) and 6MWD at 6-months. Baseline data indicate that, on average, our study\
sample was close to meeting the PA guidelines. Unexpectedly, both groups maintained their activity\
levels up to 3- and 6-months. Submaximal exercise capacity (6MWD) was also maintained (with a\
clinically meaningful but not statistically significant relative benefit in the PAI) suggesting higher PA\
levels may associate with reduced exercise intolerance after anthracycline therapy.
Dataset DOI: 10.5061/dryad.w6m905qzp
Description of the data and file structure
Participants who were between 18 to 85 years of age, diagnosed with stage I-IV Hodgkin’s or non-Hodgkin’s lymphoma, or stage I–III breast cancer, were recruited to a 6-month exercise intervention study. Data were collected at baseline, 3, and 6 months.
Files and variables
File: PLOS_data_8.13.25.csv
Description:
Variables
- pid: participant identification
- age: age in years
- race: race (White vs. non White)
- sex: Female or Male
- visit: baseline, 3 months, and 6 months
- arm: PAI, physical activity intervention. HLI, Healthy Living Intervention
- smwd: Six-minute walk distance
- daily_steps: Steps per day as measured with accelerometry,
- weekly_mvpa: Moderate to vigorous physical activity in minutes per week.
- max_rvo2: Maximal Relative Volume of Oxygen (VO2) consumption
- ef_percent: Ejection Fraction (%)
Code/software
R/RStudio software was used to analyze this data and can be downloaded for free.
The following packages were used in the analysis:
library(tidyverse)
library(table1)
library(Hmisc)
library(emmeans)
library(easystats)
library(sjPlot)
library(ggstance)
library(broom)
library(modelsummary)
library(lme4)
library(ggpubr)
library(PerformanceAnalytics)
Access information
Other publicly accessible locations of the data:
- None
Data was derived from the following sources:
- NA
Human subjects data
Direct identifiers (e.g., names, addresses, phone numbers, email addresses, social security numbers) have been removed from the uploaded data. Indirect Identifiers and potentially identifying information like demographic data (age, gender, race, ethnicity), geographic information, dates, and other variables that could be combined to re-identify individuals have been suppressed or are not part of the analytic dataset.