Olfactory dysfunction has been reported for its association with cognitive impairment and incident dementia. However, the underlying role of neurodegeneration in this relationship is not fully understood. This prospective cohort study included participants from the Shanghai Aging Study (SAS) who were 60 years or older and diagnosed as dementia-free at baseline and completed at least one follow-up interview between 2010 and 2022. At baseline, olfactory identification function was assessed by the 12-item Sniffin' Sticks Smell test (SSST-12). Olfactory dysfunction was defined as SSST-12 < 7, 6, and 5 in three age groups of participants with (age < 70, 70 ≤ age < 80, and age ≥ 80), respectively, according to previously reported normative data of olfaction in the SAS. Baseline blood Neurofilament Light Chain (NfL) was quantified using the ultra-sensitive Single Molecular Array technology (Quanterix), and dichotomized into low and high levels according to the median concentration. Multiple cognitive function assessments were administered by the Mini-mental State Examination (MMSE) at baseline and follow-ups. We used the linear mixed-effect model to fit the MMSE trajectories in participants with and without olfactory dysfunction adjusting for baseline age, sex, education year, smoking, and APOE ε4, with person-specific MMSE intercept and slope. Stratified analysis was conducted in participants with low and high NfL. Among 1125 participants included in this study, 296 were defined as having olfactory dysfunction at baseline. Less education, higher proportion of APOE ε4 allele, and lower baseline and follow-up MMSE scores were observed in participants with olfactory dysfunction (Table 1). During the median 9.4 years follow-up, participants with olfactory dysfunction at baseline evidenced a faster MMSE decline compared with those without olfactory dysfunction (b [95% CI]: -0.08 [-0.15, -0.02], P = 0.02) (Figure 1). Such association remained statistically significant (-0.18 [-0.31, -0.06], P = 0.005) in the subgroup with high NfL, but not in the subgroup with low NfL (0.004 [-0.07, 0.07], P = 0.92) (Figure 2). Blood NfL may modify the association of olfactory dysfunction with long-term cognitive decline in community-dwelling older adults. In individuals with high potential neurodegeneration, olfactory dysfunction may be an independent indicator of future cognitive deterioration.