OBJECTIVE: To examine the interplay between subclinical atherosclerotic disease and neuroticism in explaining variance in late-life depressive symptoms. METHODS: This study was part of the Nijmegen Biomedical Study, a population-based survey; 1,517 participants aged 50-70 years were included. Depressive symptoms were measured by the Beck Depression Inventory (BDI). Principal components analysis of the BDI items yielded two factors, representing a cognitive-affective symptom cluster and a somatic-affective symptom cluster. Atherosclerotic disease was measured by the intima media thickness (IMT) of the carotid arteries and neuroticism by the revised Eysenck Personality Questionnaire. RESULTS: Multiple linear regression analyses using different measures of depressive symptoms as the dependent variable showed that neuroticism was strongly and significantly associated with the sum score of the BDI and with the two depressive symptom clusters. IMT, however, was only significantly associated with the somatic-affective symptom cluster but not with the cognitive-affective symptom cluster. Interestingly, we found a significant negative interaction between neuroticism and IMT in explaining the severity of the cognitive-affective symptom cluster but not with respect to the somatic-affective symptom cluster. CONCLUSION: The negative interaction between neuroticism and atherosclerosis indicates that neuroticism is less strongly associated with cognitive-depressive symptoms in the presence of more severe atherosclerosis. This may be explained by apathy due to cerebrovascular disease and fits with a hypothesis of vascular apathy.