Design and statistics of pharmacokinetic drug-drug, herb-drug, and food-drug interaction studies in oncology patients

Polypharmacy is becoming increasingly prevalent in society. Patients with polypharmacy are at greater risk for drug-drug interactions, which can influence the efficacy of treatment. Especially, in oncology this is a concern since neoplasms are increasing prevalent with age, as well as polypharmacy is. Besides drug-drug interactions, also herb-drug and food-drug interactions could be present. Knowledge of these interactions is of great importance for safe and effective anti-cancer treatment, because the therapeutic window of most of these oncologic drugs are small. To study pharmacokinetic interaction effects, a cross-over pharmacokinetic study is a widely used, efficient and scientifically robust design. Yet, several aspects need to be considered when carrying out an interaction study. This includes the knowledge of the advantages and disadvantages of a cross-over design. Furthermore, determination of the end point and research question of interest, calculation of the required sample size, analysis of the generated data with a robust statistical plan and consideration of the logtransformation for some pharmacokinetic parameters are important aspects to consider. Even though some guidelines exist regarding these key issues, no clear overview exists. In this article an overview of these aspects is provided and their effect is discussed.