Purpose: Since no human data is available, the goal of this study was to select the optimal animal study and develop a dose-response relationship to predict the infectivity to humans of ingestion of T. gondii in the meat.
Methods: Relevant studies in literature were searched in the database and eligible studies should meet two criteria: i) contain ≥ 3 different doses to challenge the intermediate hosts, ii) at least 5 animals were challenged in each dose. Optimal study was selected based on the animal species, stage and genotype of T. gondii used in the experiment, and route of administration. Data from the optimal study was pooled and fitted into 4 common sigmoidal-shaped mathematic models (Exponential, Beta-poisson, Weibull and Logistic) in OriginPro 9.
Results: A mice experiment of infection with type II bradyzoites was selected to predict the dose-response curve for the following reasons: i) type II is a predominate strain in the United States, ii) bradyzoites stage is presented in the meat, and iii) oral administration in mice mimic the route of meat consumption in humans. Among 4 developed models, a Beta-poisson dose-response function (P (d) = 1-(1+d/189.484)-0.737) was selected based on its simplicity, popularity as well as the best fit (R2 = 0.993).
Significance: This study developed a dose-response model that predict the human response of consuming T. gondii contaminated meats, and could be used as an important input in a quantitative risk assessment to further estimate the public health burden.