Our objective is to develop methods for utilizing highly repeated remotely monitored data, clinical assessments, and DNA methylation-based immunophenotyping. We used data from a pilot clinical trial examining the impact of physical activity tracking and counseling with dietary supplementation in older adults with obesity. Elevated body mass index (BMI) contributes to excess cancer burden and many specific cancers are directly associated with obesity. Obesity promotes inflammatory cytokines, such as TNF-alpha, hormonal dysregulation, such as estrogen and adipokines, in addition to impairments in insulin and cell growth regulators, all of which impact cell proliferation and cancer. Although physical activity is a key component in obesity interventions, specific relationships between physical activity, inflammatory, and immune markers are less clear. We examine integration methods for testing the association of physical activity data, biomarkers, and immune cell proportions inferred from DNA methylation in a trial investigating the impact of a multicomponent obesity intervention. Older obese (>= 65 years, BMI >= 30 kg/m2) who received primary care services at Dartmouth-Hitchcock were recruited (n = 15). All participants received regular in-person nutrition counseling, group exercises, and health coaching while being monitored through a wrist-worn physical activity monitor, and received additional protein supplementation. Baseline and 12-week measurements were taken for 12 clinical assessments, such as objective measures (i.e., hand grip strength and 5 times sit-to-stand), subjective (i.e., Late-Life Function and Disability Instrument), 32 blood-based biomarkers (i.e., leptin and TNF-alpha), and DNA methylation of whole blood. Physical activity and steps were measured continuously throughout the study period using a Fitbit Alta HR. To compare analytical methods, we used mixed-effects models, difference in difference, and elastic net regression models to examine the association of physical activity, immunomethylation phenotypes, and clinical outcomes. We followed 15 older adults (100% white, females), whose mean age was 73.4 years (SD = 6.0) years and mean BMI = 35.5 kg/m3 (SD = 4.68). All clinical assessments and blood-based biomarkers were collected on all 15 participants for both baseline and follow-up timepoints. An average of 92.2 days of data (steps, distance, total calories, activity calories, or minutes in sedentary, light activity, or fairly active, and very active) were recorded (SD = 9.5). We collected DNA methylation data on 10 participants at baseline and 14 participants in follow-up. Through developing newly linked data sources longitudinally, we will be able to examine new subpopulations of phenotypes that could elucidate the relationship between BMI and cancer impacting cancer screening, diagnosis, and treatment.

Note: This abstract was not presented at the conference.

Citation Format: Curtis L. Petersen, Brock C. Christensen, John A. Batsis. Connecting remote patient monitoring, DNA-based immunophenotyping, and clinical outcomes data [abstract]. In: Proceedings of the AACR Special Conference on Modernizing Population Sciences in the Digital Age; 2019 Feb 19-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(9 Suppl):Abstract nr A07.