(P 160) A PROSPECTIVE STUDY EVALUATING THE ASSOCIATION OF PATIENT-REPORTED OUTCOMES WITH ACTIVITY LEVEL AND SLEEP PATTERNS IN SARCOMA SURVIVORS

Objective: Sarcomas are a rare heterogeneous malignancy group affecting both sexes and all ages. Treatment is often intensive, resulting in long-term toxicities. Given the rarity of sarcoma, there is a paucity of data defining physical and emotional outcomes. The Cohort to Augment the Understanding of Sarcoma Survivorship Across the Lifespan (CAUSAL) was constructed at Vanderbilt University Medical Center (VUMC) to evaluate health-related outcomes.

Methods: Sarcoma patients treated at VUMC from 2012 – present are enrolled into CAUSAL. Patient demographics, tumor and treatment data are abstracted from the electronic medical record. Participants complete surveys including PROMIS-57 consisting of seven domains including physical function, sleep disturbance, fatigue, and satisfaction with participation in social roles, which are reported here. Participants who have completed treatment are given a FitBit to track activity level and sleep over 12 weeks. Data are synced to an online platform, Fitabase. Body Mass Index (BMI) is assessed at study entry. For each participant, daily step count, active minutes, and sleep minutes were summarized using medians. To estimate the association between PROMIS-57 scores and activity and sleep data (FitBit), linear regression models were fit controlling for age, sex, and BMI of participants.

Results: Of 736 CAUSAL participants, 421 completed the PROMIS-57. Of those, 111 participants also had FitBit data and are included in this analysis. Median step count was 6626 (quartiles: 4436, 8744). For reference, the average FitBit step count data for the US population is 8170. Median daily activity was 260 minutes (quartiles: 209.5, 317.5). Median duration of sleep was 449 minutes (quartiles: 400, 475) which was greater than the average reported by FitBit for the general US population (436 min). Linear regression analysis determined that step count was significantly associated with PROMIS physical function score (p < 0.001). A 10-point increase in physical function score is associated with an increase of 1393 (95% CI: 789, 1997) steps when adjusting for age and sex. Active daily minutes were significantly inversely correlated with PROMIS fatigue score (p = 0.012). A 10-point increase in fatigue score is associated with an average decrease of 20 (95% CI: 36, 45) minutes when adjusting for age and sex. Compared with participants who received treatment within 30 days prior to CAUSAL enrollment, participants not receiving treatment had 5.38 (95% CI: 2.96, 7.8) higher physical function score and 6.26 (95% CI: 2.81, 9.71) higher satisfaction in their social roles. There was no significant association between minutes of sleep recorded by FitBit and PROMIS sleep disturbance.

Conclusion: Using PROMIS-57, and FitBits, we demonstrated a positive association between higher step counts and self-reported higher physical function, in addition to a negative association between FitBit recorded activity level and fatigue score. We did not find a significant association between FitBit recorded sleep and self-reported sleep disturbance. Furthermore, our data suggest patients with sarcoma sleep more and walk less than the general US population. As enrollment continues in CAUSAL, we will focus on PROMIS scores over time and their association with activity level and sleep.