Diagnostic accuracy of plasma amyloid-β42 and Montreal Cognitive Assessment score as composite index for Alzheimer's disease in resource-limited settings: An exploratory study

Background Accurate diagnosis of Alzheimer's disease (AD) in Indonesia remains challenging due to limited access to cerebrospinal fluid (CSF) analysis and PET imaging. Plasma amyloid-β42 (Aβ 42 ) may offer a less-invasive and affordable alternative, particularly when combined with cognitive assessment. However, its diagnostic performance in Indonesian populations is not well established. Objective To exploratorily assess the diagnostic accuracy of plasma Aβ 42 , alone and as a composite index combined with Montreal Cognitive Assessment-Indonesian version (MoCA-INA) scores, as potential alternatives to CSF biomarkers for AD diagnosis in a resource-limited setting. Methods A cross-sectional study was conducted involving 25 AD patients and 20 cognitively normal controls. Plasma and CSF biomarker levels were measured using ultrasensitive immunoassay. Cognitive function was evaluated using MoCA-INA. Receiver operating characteristic curve analysis was used to assess diagnostic accuracy. Results Plasma Aβ 42 with the level of 9.6 pg/ml showed moderate diagnostic accuracy (AUC = 0.78, sensitivity = 69.2%, specificity = 70.6%), while its combination with MoCA-INA Score improved accuracy (adjusted AUC = 0.93, sensitivity = 80.0%, specificity = 84.0%). No significant correlation was found between plasma and CSF biomarkers. Plasma p-tau181 was excluded from the final composite index because it showed an unexpected direction of change. Conclusions Plasma Aβ 42 combined with MoCA-INA as a composite index, demonstrated potential diagnostic performance for AD in resource-limited settings. However, these exploratory findings should be interpreted cautiously and require further validation in larger and independent cohorts before clinical application.