This cross-sectional study aimed to examine the association between serum neurofilament light chain (sNfL) and the PCE-estimated 10-year atherosclerotic cardiovascular disease (ASCVD) score using data from the 2013-2014 United States National Health and Nutrition Examination Survey for adults aged 40-75 years. The primary outcome was the estimated 10-year ASCVD risk, dichotomized at 7.5%. We performed weighted logistic regression analyses, treating natural log-transformed sNfL [log (sNfL) ] as the primary continuous exposure. Restricted cubic spline (RCS) analysis was adopted to characterize the non-linear relationship between sNfL and the estimated 10-year ASCVD risk. Furthermore, stratified analyses were conducted by age, sex, BMI, hypertension, diabetes, and dyslipidemia. This study included 1,136 participants, among whom 472 (41.5%) were classified into the higher estimated risk category (≥ 7.5%). In multivariable-adjusted models, each 1-unit increase in log (sNfL) was associated with higher odds of elevated ASCVD risk (adjusted OR = 2.34, 95%CI: 1.67-3.26). Subgroup analyses showed consistent positive associations across most strata, though the association was not significant in the 40-60-year age group. Our findings suggest a positive, non-linear association between elevated sNfL levels and a higher PCE- estimated 10-year ASCVD risk score, indicating that sNfL may serve as a candidate biomarker associated with this estimated risk. Prospective studies with incident ASCVD outcomes are required to confirm these findings.