Potential of facial biomarkers for Alzheimer’s disease and Obstructive sleep apnea in Down syndrome and general population

Background Down syndrome (DS) is caused by trisomy 21, leading to increased risks of Alzheimer’s disease (AD) and Obstructive sleep apnea (OSA). Traditional diagnostic methods for AD and OSA, like cerebrospinal fluid analysis and polysomnography, are invasive and challenging for people with DS. Facial morphology has emerged as a non-invasive biomarker for these conditions. Therefore, we conducted a comprehensive 3D analysis of facial shape variation in DS and euploid control (EU) populations, assessing the effect of critical factors such as age, sex, and facial size, to investigate its association with AD and OSA. Methods Facial shape differences among groups were analyzed from the coordinates of 21 landmarks automatically registered on 3D facial models generated from head magnetic resonance images in a cross-sectional sample of 131 individuals with DS and 216 EU controls from 18 to 90 years old, including subjects diagnosed with AD and OSA. We used Procrustes ANOVA and MANOVA tests to quantify the amount of facial shape variation attributable to sex, age, and facial size, and quantified facial shape differences among diagnostic groups using geometric morphometrics. Results Besides facial shape differences between DS and EU individuals, our results detected significant interactions between diagnosis and sex, and between diagnosis and age, indicating sex-dependent differences and an altered pattern of facial shape change over adulthood in DS, with females presenting more severe alterations. Multivariate regression analyses showed that facial shape significantly correlated with the concentration ratio between amyloid beta peptide 1- 42 and amyloid beta peptide 1-40 (Aβ1-42/Aβ1-40) in cerebrospinal fluid, a common biomarker for AD diagnosis. In the DS population, facial shape differences between individuals with and without AD diagnosis did not achieve statistical significance after adjusting for age and facial size, but significant shape differences were detected in the EU population. Regarding OSA, facial shape significantly correlated with the apnea-hypopnea index, and individuals with DS and severe OSA presented a significantly different facial morphology in comparison to individuals with DS and no signs of OSA, suggesting that facial morphology could be associated with sleep respiratory disturbances. Conclusions Overall, our study underscores the potential of facial morphology as a diagnostic biomarker for the early detection and clinical management of AD and OSA.