Actigraphy as an objective measure of performance status in patients with advanced cancer.

Abstract

62 Background: Treatment decisions are often based on performance status (PS), a subjective rating of patient functioning derived from observation and patient-report. Devices that monitor physical activity, such as wrist actigraphs, can accurately measure the percent of time a person is immobile while awake (awake immobile). Thus, actigraphy may have potential to better estimate true PS. We compared actigraphy to both Eastern Cooperative Oncology Group PS (ECOG PS) and patient-reported physical activity as predictors of survival in patients with stage IV non-small cell lung cancer (NSCLC). Methods: Participants (n = 41) were ambulatory patients with stage IV NSCLC receiving care at MGH. Participants wore a watch-sized accelerometer device (ACTIWATCH 2) for three consecutive 24-hour periods (72 hours) and completed a self-report questionnaire about physical activity, scored as METS (metabolic equivalents) per week. Patients’ oncologists rated their ECOG PS (0-5) at the end of the 72-hour actigraphy period. Relationships among ECOG PS, awake immobile, and METs per week were tested with Pearson correlations. A ROC curve for 6-month survival was used to determine a meaningful cut-off for awake immobile. Unadjusted Cox regression models tested associations with survival from assessment times. Results: Participants’ ECOG PS ratings were: 0 (22%), 1 (63%), 2 (12%), and 3 (2%). ECOG PS and awake immobile were correlated (r = .42, p < .01). METS per week was correlated only with ECOG PS (r = -.35, p = .03). At time of analysis, 15 patients had died with a minimum follow up of 9 months. Among all patients, survival was predicted by ECOG PS, HR = 3.77 (95% CI 1.70-8.35), p < .01; awake immobile as both as a continuous (percentage points) and categorical ( > 23%) variable, HR = 1.04 (95% CI 1.00-1.09), p = .05 and HR = 4.12 (95% CI 1.37-12.39), p = .01, respectively; but not METS per week. Among patients with good EGOC PS (0-1), only awake immobile ( > 23%) predicted survival, HR = 5.80 (95% CI 1.39-24.12), p = .02. Moreover, within the largest ECOG group (PS1), awake immobile ( > 23%) still predicted survival, HR = 5.53 (95% CI 1.18-25.94), p = .03. Conclusions: Actigraphy, an objective measure of patient activity, may have utility in determining patient PS.