Objectives: The aim of this study was to compare the caries preventive effect of highly fluoridated dentifrices and gels on sound dentin as well as on artificial dentin caries-like lesions. Methods: Bovine dentin specimens (n = 240), with 2 different surfaces each (1 sound surface [sound treatment (ST)] and one caries lesion [demineralized treatment (DT)]), were prepared and randomly allocated to one highly (6 x 120 min demineralization/day [H]) and one lowly cariogenic (6 x 60 min demineralization/day [L]) pH-cycling model. Treatments during pH-cycling (28 days) were: brushing 2x/day with: 0 ppm F [H-0/L-0], 1,450 ppm F [H-1,H-450/L-1,L-450], 2,800 ppm F [H-2,H-800/L-2,L-800], 5,000 ppm F [H-5,H-000/L-5,L-000], 5,000 ppm F plus TCP [H-5,H-000+TCP/L-5,L-000+TCP], and 12,500 ppm F [H-12,H-500/L-12,L-500] containing dentifrices/gels. Dentifrice/gel slurries were prepared with deionized water (1:2 wt/wt). Differences in integrated mineral loss ( increment increment Z) and increment lesion depth were calculated between values before and after pH-cycling using transversal microradiography. Results: The correlation between Delta Delta Z(DT) and F- was strong for the highly (r(H) = 0.691; p < 0.001) and moderate (r(L) = 0.500; p < 0.001) for the lowly cariogenic model, indicating a fluoride dose-response for both. Significant differences for Delta Delta Z(DT) and Delta Delta Z(ST) could be found between H-0, H-1,H-450, H-5,H-000, and H-12,H-500 as well as L-0, L-5,L-000, and L-125,L-000 (p <= 0.046; analysis of covariance [ANCOVA]). Except for 0 ppm F-, no significant difference in Delta Delta Z(ST) and Delta Delta Z(DT) could be found between the highly and lowly cariogenic model (p >= 0.056; ANCOVA). Conclusion: For both pH-cycling conditions a dose-response for fluoride could be revealed. For elderly people with exposed root surfaces, the use of gels containing 12,500 ppm F instead of regularly (1,450 ppm F) or highly (5,000 ppm F) fluoridated dentifrices should be further investigated, as it offered higher caries-preventive effects in vitro.