Prospective and external validation of an objective performance status (OPS) in patients with metastatic solid malignancies using wearable accelerometry.

12012 Background: Performance status (PS) assessment is critical for clinical trial eligibility assessment and treatment selection. However, PS is based on the subjective impression of the clinician and is often inaccurate. Wearable accelerometers may allow clinicians to more objectively assess PS. In this analysis of 2 separate prospective studies, we define and externally validate an “Objective Performance Status” (OPS) by measuring the association between daily physical activity and overall survival among patients with metastatic cancer. Methods: We first measured daily physical activity (in step counts) using a wearable accelerometer over a prescreening period (median 14 d, range 3-28 d) in a Spanish observational prospective study (PIC123-18FJD) embedded during the screening period for a phase 1 clinical trial in patients with locally advanced unresectable or metastatic solid and hematological malignancies. A multivariable Cox proportional hazards model was used to derive an OPS by measuring associations between mean daily steps with overall survival (OS), adjusting for patient demographics. We subsequently externally validated this OPS cutoff in a separate prospective cohort of patients with metastatic non-small cell lung and gastrointestinal cancers who wore a wearable accelerometer continuously for 6 months enrolled in a randomized trial of proactive symptom monitoring at a large academic center in the United States (NCT04616768) Results: Full data was available for 123 patients (70 in OPS derivation cohort; 53 external validation cohort). The cut-off selected to define the OPS was determined by univariate survival analysis to be mean daily step count = 1200 meterswith poor OPS (≤1200 m/day, of whom 46% had clinician-recorded ECOG ≥1) had greater mortality than patients with good OPS (> 1200 m/day, 85% with ECOG 0) (3. 2 vs. 11. 2 mos median OS, adjusted hazard ratio 5. 76, 95% CI 2. 98-11. 1, p≤0. 001). In the external validation cohort, poor OPS strongly predicted mortality compared to good OPS (unadjusted HR 5. 22, 95% CI 1. 24-21. 88, p = 0. 02) and was a better predictor of mortality than clinician- (HR = 1. 00, 95% CI 0. 24-4. 17, p = 1. 0) or patient-reported ECOG (HR = 1. 16, 95% CI 0. 14-9. 45, p = 0. 89). Conclusions: The OPS is an independent, externally validated prognostic factor for survival and could serve as an objective surrogate for traditional methods of PS assessment in clinical trials and choice of therapy. Clinical trial information: NCT04616768 (external validation trial). Concordance between OPS and clinician-reported ECOG. ECOG = 0 ECOG≥1 Kappa (95%CI) Spanish trial Distance (meters) ≥1200 36 (85. 7%) 15 (53. 6%) 0. 34 (0. 12-0. 56) < 1200 6 (14. 3%) 13 (46. 4%) External validation trial Distance (meters) ≥1200 17 (85. 0%) 22 (33. 3%) 0. 15 (-0. 04-0. 35) < 1200 3 (15. 0%) 11 (66. 7%)