Abstract: Many phenolic acids are plant metabolites widely distributed throughout the vegetal kingdom. Recent interest in phenolic acids stems from their potential as bioactive constituents and their potential protective role, through ingestion of fruits and vegetables, against oxidative damage diseases like neurodegenerative diseases and cancer. This communication aims to identify structural alerts associated with clastogenic activity that may present the phenolic acids. It was performed a QSAR study (Quantitative structure-activity relationships) that allowed to use a structure vs clastogenic activity relationship model, which encodes topological information to a substructural level, according to the TOPS-MODE approach (Topological substructural molecular design). The predictions were made using the technique of linear discriminant analysis. The used software was STATISTIC and MODESLAB. The model has presented an adequate probability of good classification for the external prediction set. It has been identified the relevant structural features to the activity under study, such as: amount and position of hydroxyl and/or methoxy groups. These structural modifications represented an indicator of the toxicity of these compounds and provide a strategy for the design of new derivatives devoid of this activity. This study can be of interest to better understand the properties of natural substances in food, and also in the development of functional foods or nutraceuticals and drug design.