Associations between wearable device metrics and performance status in adult patients with cancer.

Abstract

427 Background: Wearable devices present a promising method for monitoring physical performance in cancer patients, potentially bridging the gap between patient experiences at home and clinical evaluations. This study assesses associations between performance status, home-based wearable device metrics, and patient-reported outcomes (PROs) among adults with cancer. (NCT03952767). Methods: This prospective cohort study enrolled 62 adults undergoing treatment for cancer, irrespective of primary site. Participants wore GENEActiv wrist accelerometer sensors (Activinsights Ltd., Kimbolton, UK) on the lumbar region and non-dominant wrist for 2 weeks, tracking gait speed and activity levels. Patients concurrently completed PRO questionnaires on physical function (PF) and fatigue (PROMIS Fatigue 7a & PROMIS Physical Function 10a). Study participants were stratified by baseline self-reported Karnofsky performance status (KPS; 100, 80-90, & ≤70). Linear regression models were used to assess associations between (1) KPS and device metrics and (2) device metrics and PROs. Covariates included age and a binary indicator for COVID period (pre/post 3/1/2020). Results: Average device wear time across the cohort was 22.2 hours/day. Significant differences in gait speed and moderate activity time were observed between the KPS≤70 group and the higher KPS groups ( Table ). Significant associations were also observed between device metrics and PROs. Each 1-hour increment moderate activity was associated with 5.5 point greater PROMIS PF score (95% CI: 3.3, 7.7; p<0.0001) and 4.6 point lower PROMIS Fatigue score (-6.7, -2.5; p<0.0001). Each 0.1 meter/sec increment in gait speed was associated with 4.6 point greater PF score (95% CI: 2.2, 7.1; p=0.0004) and 4.8 point lower Fatigue score (-6.9, -2.7; p<0.0001). Similar associations were observed between in-clinic assessments of gait speed and physical function scores. Conclusions: The associations between wearable device metrics and KPS indicate that wearable technology is both feasible and reliable for remote health monitoring in cancer patients. The associations between device metrics and PROs demonstrate that these metrics reflect patient experiences and outcomes relevant to both patients and regulatory authorities. The associations with physical function, a key PRO in the FDA’s Patient-Focused Drug Development guidance, are particularly noteworthy. Clinical trial information: NCT03952767 . Group 1(KPS 100) Group 2(KPS 80-90) Group 3(KPS ≤70) p (Group 1 vs 3) p (2 vs 3) Gait Speed (m/s) - adjusted mean, 95% CI 0.74 (0.70, 0.78) 0.72 (0.69, 0.74) 0.65 (0.60, 0.70) 0.0058 0.0210 Sedentary time (hours) 11.2 (9.8, 12.6) 10.2 (9.3, 11.0) 10.7 (9.1, 12.2) 0.5912 0.5647 Light activity (hours) 1.4 (1.0, 1.7) 1.3 (1.1, 1.5) 0.9 (0.5, 1.3) 0.0570 0.0474 Moderate activity (hours) 1.4 (1.0, 1.8) 1.2 (0.9, 1.5) 0.5 (0.0, 1.0) 0.0068 0.0159 Vigorous activity (hours) 0.1 (0.0, 0.2) 0.0 (0.0, 0.1) 0.0 (0.0, 0.1) 0.1201 0.2835