Role of models in assessing new agents for caries prevention--non-fluoride systems.

While fluoride is an effective anti-caries agent, the search for more effective alternative therapies continues. A wide range of non-fluoride anti-caries agents has been postulated, and this paper reviews some of the pre-clinical models that have been utilized in their evaluation and some of the pitfalls that must be avoided. Using data on the potential anti-caries efficacy of phosphopeptides obtained from casein, the caution that must be applied in extrapolating laboratory data to predict clinical performance is discussed. Evaluation strategies that focus on only one potential mode of action (e.g., inhibition of demineralization) may overestimate the true clinical efficacy which may arise from a combination of two or more effects (e.g., inhibition of demineralization and stimulation of remineralization). Although laboratory and in situ data predict anti-caries efficacy for sodium trimetaphosphate in combination with fluoride, this was not found in three-year clinical trials. A possible reason for this, the lack of suitable calibration methods, is discussed. Finally, some comments on the appropriateness of laboratory evaluation strategies are made.