Digital biomarkers signal potential for trials, clinical care in pediatric lung disease
Click Here to Manage Email Alerts
Digital biomarkers derived from portable spirometers and smartwatches show promise as candidate endpoints for use in clinical trials or clinical care of pediatric asthma and cystic fibrosis, researchers reported.
“Noninvasive measurements with digital and portable devices for home use may provide ... new biomarkers. Physical activity has been shown to be related to asthma severity, and it is plausible that heart rate and parameters related to sleep also correlate well with an increase in disease activity,” Matthijs D. Kruizinga, MD, from the Juliana Children’s Hospital at Haga Teaching Hospital, The Hague, and the Centre for Human Drug Research and Leiden University Medical Centre, the Netherlands, and colleagues wrote in European Respiratory Journal. “Such parameters can be easily and objectively obtained by consumer devices like a smartwatch. Several physical activity- and heart rate-derived digital biomarkers, such as daily step count, step count taken during the most active hour per day and daytime or nocturnal heart rate, have been proposed and evaluated in healthy children, and these candidate digital biomarkers exhibited reasonable intra-subject variability.”
The researchers aimed to clinically validate physical activity, heart rate, sleep and FEV1 as digital biomarkers, measured by portable spirometry and smartwatches, in children with asthma and cystic fibrosis.
The prospective cohort study included 30 children with controlled asthma (mean age, 10.5 years; 67% boys), 30 children with uncontrolled asthma (mean age, 10.5 years; 67% boys)and 30 children with cystic fibrosis (mean age, 9.7 years; 47% boys). The researchers used data from 128 healthy children (mean age, 11.1 years; 46% boys) for comparison.
All participants wore a smartwatch, performed daily at-home spirometry and completed a daily symptom questionnaire for 28 days. Overall median compliance was 88%.
Compared with healthy controls, daily physical activity was lower for patients with controlled asthma (difference = 731 steps; P = .049), uncontrolled asthma (difference = 1,264 steps; P < .001) and cystic fibrosis (difference = 847 steps; P = .019).
In addition, children with uncontrolled asthma had a significantly higher nocturnal heart rate compared with healthy controls and patients with controlled asthma or cystic fibrosis. All patient groups had higher daytime heart rate compared with healthy controls.
Among children with uncontrolled asthma or cystic fibrosis, days with higher symptom scores were associated with lower physical activity, with a 15% decrease in step count per point increase in symptom score for uncontrolled asthma (P = .045) and a 1-point increase associated with a 3% decrease in step count for cystic fibrosis (P = .002).
Patients with uncontrolled asthma had a 0.25 lower FEV1 z score for each point increase in symptom score (P = .05) and patients with cystic fibrosis had a 0.07 lower FEV1 z score for each point increase (P = .005). Patients with controlled asthma had an average daytime heart rate that was 1.6 bpm higher (P = .02) and a 1.2 bpm higher nocturnal heart rate (P = .07) per point increase. Among patients with uncontrolled asthma, daytime heart rate was 1.6 bpm higher (P = .05) and nocturnal heart rate was 2.8 bpm higher (P = .001) per point increase. Patients with cystic fibrosis had a 0.4 bpm higher nocturnal heart rate (P = .049) per point increase, but there was no observed effect on daytime heart rate.
“The sample size for all analyses is limited and future studies should include larger cohorts to increase the generalizability and robustness of the current findings,” the researchers wrote.
The identified candidate biomarkers also appeared to describe pulmonary exacerbations, according to the researchers.
“Digital biomarkers derived from remote monitoring with a smartwatch and portable spirometer show promise in pediatric lung disease. Physical activity, heart rate and pulmonary function monitoring is tolerable, can differentiate patients from controls and is correlated to symptom scores,” Kruizinga and colleagues wrote.