A30-01 Longitudinal Sensor-based Monitoring of Oxygenation Metrics: An Evaluation of Oxygen Desaturation Index in a Phase 3 Open-label Extension Study

Abstract Rationale Due to the night-to-night variability of obstructive sleep apnea (OSA), multiple nights of measurements may be necessary to reliably evaluate response to therapy. Here we employed a Sleep Data Management System (SDMS) as an optional component in a phase 3 open-label extension (OLE) study to evaluate the number of nights required to reliably measure oxygen desaturation index (ODI4) using a proximal finger-mounted pulse-oximeter of ring formfactor. Methods Participants completing either parent phase 3 clinical trial (SynAIRgy or LunAIRo) of AD109 (combination of the novel antimuscarinic, aroxybutynin 2.5 mg and selective norepinephrine reuptake inhibitor, atomoxetine 75 mg) in OSA were eligible to enroll in the OLE (NCT06566820) evaluating long-term safety. Nightly, participants could record and upload data (oxygen saturation, pulse-rate, and motion) to the SDMS, which calculated oxygenation metrics including ODI4 which were analyzed as exploratory outcomes. A running average approach was used to determine the number of nights required for an individual’s mean ODI4 to stabilize. Mean ODI was recalculated as each subsequent night’s ODI4 was added. Stability was defined when running average change is ≤ 1 event/hour and the remaining 80% of nights are stable, without assuming sequential recordings. In supportive analysis, population-level single night reliability was assessed using Intraclass-Correlation-Coefficient (ICC(A,1)) analysis, and the Spearman-Brown formula was used to predict nights (ICC(A,k)) required for excellent reliability (ICC0.9). Results Of 602 participants enrolled in the OLE; 474 (79%) opted to use SDMS, and 431 (91%) uploaded at least one oximetry recording. The cohort was 51.2% male (n = 243), mean age (SD) 56.6±10.6 years. A total of 28813 recordings (≥ 4hrs) were collected, with an average of 68±70 nights per participant and mean recording duration of 7.1 hrs.The ODI4 stability analysis cohort consisted of participants with ≥ 5 recordings on treatment in the Full Dose period (N = 300, Fig-1A), enabling stability analysis of ODI4 values across individual nights (Fig-1B) and weeks (Fig-1C).On average, using running average analysis, participants reached stability after a median of 4 nights and at the 90th percentile after 9 nights. In variability analysis, 4 nights were required to achieve ICC0.9. Conclusions A high proportion of eligible participants (79%) opted to utilize SDMS. Similarly, a high proportion of participants met or exceeded protocol recommendations for SDMS use. We estimate that stability can be achieved in 4 nights on average with similar results showing excellent ICC from variability analysis, and near full stability was achieved in 9 nights across the cohort. This abstract is funded by: Apnimed, Inc.