Daily Step Counts Predict Outcomes for Locally Advanced Non-Small Cell Lung Cancer Patients Treated with Concurrent Chemoradiotherapy

Abstract

Wearable devices can provide objective physical activity data for cancer patients. In this analysis, we explore the prognostic value of baseline step count data for patients treated with definitive chemoradiotherapy for locally advanced non-small cell lung cancer (NSCLC). Locally advanced NSCLC patients who wore a commercial fitness tracker as part of a clinical trial during a course of definitive, concurrent chemoradiotherapy were included in this analysis. For each study subject, a baseline step count average was defined as the average daily step total over the time period between study enrollment and the completion of the first week of radiotherapy. Based on published data describing healthy subjects’ step counts as a function of age, study subjects were categorized as inactive (below the 25th percentile), moderately active (25th to 75th percentile), or highly active (above the 75th percentile). Fisher’s exact test was used to examine activity level as a predictor of hospitalization during radiotherapy and of completing the planned radiotherapy course without delay exceeding one week. Median progression-free survival (PFS) and overall survival (OS) durations were estimated using the Kaplan-Meier method. Physical inactivity was tested as a predictor of PFS and OS using Cox proportional hazards models. Fifty subjects met eligibility criteria. The mean age was 67, and the median baseline daily step count average was 5,354 (range: 540 to 18,854). Nine subjects (18%) were categorized as highly active, 23 subjects (46%) were categorized as moderately active, and 18 subjects (36%) were inactive. Eleven subjects (22%) had ECOG performance status (PS) 0, 33 (66%) had PS 1, and 6 (12%) had PS 2. Inactive subjects were more likely to be hospitalized during the radiotherapy course than other subjects (50% v. 9%, p=0.004) and less likely to complete radiotherapy without delay exceeding one week (67% v. 97%, p=0.006). Median PFS duration was 5.3 months for inactive subjects and 18.3 months for other subjects (HR for inactivity = 4.52, 95% CI 2.02 to 10.12, p<0.001). Median OS duration was 15.0 months for inactive subjects and not reached for other subjects (HR for inactivity = 3.88, 95% CI 1.45 to 10.43, p=0.007). PS was associated with activity level, but PS was not a significant predictor of PFS or OS. Baseline activity level measured using wearable devices may help identify NSCLC patients who are fit for concurrent chemoradiotherapy and can predict clinical outcomes in this setting. Objective activity data could supplant PS to inform clinical decisions in Oncology.