Objective: The aim of this study was to evaluate the remineralization of residual demineralized dentin underneath restorations by fluoride-releasing liners in vitro. Method and Materials: Two glass-ionomer cement liners, Lining Cement (GC) and RK-141A (GC), and two light-cured resin liners with and without 5 wt% NaF were used in this study. The resin liners were made by mixing TEGDMA and bis-GMA. Class 5 cavities were prepared on extracted human premolars. Demineralized dentin was promoted by using a bacterial caries induction system at the cavity floor. Cavities were restored with a resin composite following application of each liner. Specimens were then stored for 4 weeks at 37°C and 100% humidity. After storage, the specimens were cut through the restorations, and axial sections of about 80-µm thickness were prepared. Contact microradiographs were taken, and the radiopacity of the demineralized dentin layer underneath the liner was analyzed. Results: The ratios of radiopacity of demineralized dentin under the fluoride-releasing liners were significantly higher than those under the resin liner without fluoride release. Although the amounts of fluoride release from the glass-ionomer cement liners were lower than that from the resin liner with fluoride release, there were no significant differences in the ratios of radiopacity in demineralized dentin layers among the 3 liners with fluoride release. Conclusion: Although the fluoride released from liners can enhance remineralization of demineralized dentin, the amount of fluoride release does not affect the ratio of radiopacity in the remineralized dentin. Keywords: caries, demineralized dentin, fluoride, glass-ionomer cement, liner, remineralization, resin