INTRODUCTION: Sleep disturbances are associated with Alzheimer's disease (AD) and Alzheimer's disease and related dementias (ADRD), but the relationship between sleep architecture, particularly rapid eye movement (REM) sleep, and AD/ADRD biomarkers remains unclear. METHODS: We enrolled 128 adults (64 with Alzheimer's disease, 41 with mild cognitive impairment [MCI], and 23 with normal cognition [NC]), mean age 70.8 ± 9.6 years, 56.9% female, from a tertiary hospital in China. Participants underwent overnight polysomnography (PSG), amyloid β (Aβ) positron emission tomography (PET), and plasma biomarker analysis: phosphorylated tau at threonine 181 (p-tau181), neurofilament light (NfL), and brain-derived neurotrophic factor (BDNF). RESULTS: After adjusting for demographics, apolipoprotein E (APOE) ε4 status, cognition, and comorbidities, the highest tertile of REM latency was associated with higher Aβ burden (β = 0.08, 95% confidence interval [CI]: 0.03 to 0.13, p = 0.002), elevated p-tau181 (β = 0.19, 95% CI: 0.02 to 0.13, p = 0.002), and reduced BDNF levels (β = -0.47, 95% CI: -0.68 to -0.13, p = 0.013), compared to the lowest tertile. DISCUSSION: Prolonged REM latency may serve as a novel marker or risk factor for AD/ADRD pathogenesis. HIGHLIGHTS: Rapid eye movement latency (REML) may be a potential marker for Alzheimer's disease and Alzheimer's disease and related dementias (AD/ADRD) pathogenesis. Prolonged REML was associated with higher amyloid beta (Aβ) burden, phosphorylated tau-181 (p-tau181), and lower brain-derived neurotrophic factor (BDNF) levels. Intervention trial is needed to determine if targeting REML can modify AD/ADRD risk. Slow-wave sleep was not associated with AD/ADRD biomarkers.