Validation in Applied Epidemiological Studies

1c Validation in Applied Epidemiological Studies. The final stage in biomarker validation is to apply it in the context of an etiological or clinical study. Common study designs used for these validations are case-control or nested case-control studies. These studies measure the prevalence of the exposure or effect in the study population, refine measurements of inter- and intra-variability, determine sensitivity and specificity at the population-based level, and identify more potential confounders and effect modifiers. 

Example. Validation of a potential biomarker to measure HCA adducts in blood. 

A laboratory-based assay needs to be developed in the lab to measure HCA adducts. 

The assays specificity and sensitivity is determined by using human samples spiked with the HCA-adduct and experimental models (i.e., rodents), fed known doses of the HCA of interest. Spiked human samples are used because HCAs are very potent mutagens and because some specific HCAs have been classified as reasonably anticipated to be human carcinogens by the National Toxicology Program. Therefore human studies involving voluntary exposure to purified HCAs are prohibited. 

The biomarker is validated in a cross-sectional population-based study to determine if there are correlations between self-reported HCA exposure (for example, food frequency questionnaire using photographs similar to the case-control study reported by Butler et al.) and HCA measurements. 

---