A novel eye‐tracking digital marker outperforms plasma biomarkers in detecting cognitive impairment

Abstract INTRODUCTION Detecting and monitoring cognitive performance in older adults is critical. In this study, we evaluated the validity of an eye‐tracking tool in diagnosing cognitive impairment. METHODS We recruited 119 cognitively unimpaired (CU) individuals and 157 cognitively impaired (CI) patients who completed digital eye‐tracking tests and cognitive scales. Of them, 154, 120, 53, and 146 underwent plasma biomarker tests, amyloid‐β positron emission tomography (Aβ‐PET) scans, tau‐PET scans, and magnetic resonance imaging (MRI) scans. The diagnostic performance of eye‐tracking markers and their relationships to Alzheimer's disease biomarkers and cognition were examined. RESULTS The eye‐tracking panel exhibited better performance (area under the curve [AUC] = 0.865) in classifying CI from CU compared to plasma Aβ 42/40 (AUC = 0.699), p‐Tau 217 (AUC = 0.769), p‐Tau 217 /Aβ 42 (AUC = 0.801), glial fibrillary acidic protein (GFAP; AUC = 0.804), and neurofilament light chain (NfL) (AUC = 0.826). DISCUSSION These findings demonstrate the validity of digital eye‐tracking markers for screening patients with cognitive impairment, providing a novel digital marker for detecting cognitive decline in older adults.