The objective of this study was to investigate the effect of different glass ionomer cements on secondary caries inhibition in a fully automated in vitro biofilm model. One hundred and twenty-four extracted third molars received class V cavities and were filled with one conventional (Ketac Molar/KM), and two resin-modified glass ionomer cements (Photac Fil/PF, Ketac N100/KN, 3M Espe). A bonded resin composite (Single Bond Plus/Filtek Supreme XTE) served as control. After 14days water storage at 37°C, specimens were thermocycled (10,000×5/55°C). Over a period of 10days, specimens were subjected to cariogenic challenge for 3/4/6h/day. Demineralization was caused by Streptococcus mutans (DSM 20523) alternatingly being rinsed over specimens using artificial saliva. After biological loading, teeth were cut longitudinally and demineralization depths were evaluated at the margins and at a distance of 0.5mm from the margins using fluorescence microscopy. Marginal quality was investigated under a SEM at ×200 magnification. Four-hour demineralization depths were for enamel margins (EM), enamel (E), dentin margin (DM), and dentin (D) (μm±SD): KM: EM 12±8, E 33±7, DM 56±11, D 79±6; PF: EM 19±13, E 34±13, DM 53±10, D 77±12; and KN: EM 26±5, E 38±6, DM 57±11, D 71±7. For all glass ionomer cements (GICs), demineralization depth at the margins was less compared to 0.5mm distance, with demineralization depth having been correlated to duration of cariogenic challenge (ANOVA [mod. LSD, p<0.05]). Compared to the bonded resin composite, all GICs exhibited caries inhibition at restoration margins in enamel and dentin. Fluoride-releasing GIC materials exhibit a secondary caries inhibiting effect in vitro. Glass ionomer cements have a higher secondary caries inhibiting effect than resin composites.