In molecular epidemiologic studies, optimizing the use of available biological specimens while minimizing the cost is always a challenge. This is particularly true in pilot studies, which often have limited funding and involve small numbers of biological samples too small for assessment of recently developed biomarkers.
In this study we examined several statistical approaches for determining how many experimental subjects to use in a biomarker study and how many repeated measurements to make on each sample, given specific funding considerations and the correlated nature of the repeated measurements.
A molecular epidemiology study of DNA repair and aging in basal cell carcinoma was used to illustrate the application of the statistical methods proposed.
Our methods extend traditional designs on biomarker studies with repeated measurements to including funding constraints.