Pharmacodynamic (PD) biomarkers play a pivotal role in anti-tumor drug development as a biochemical measurement to estimate the level of drug interaction with the target, or to assess the downstream impact of its interaction with the target. Although immunohistochemistry (IHC)-based protein biomarkers have been conventionally used as PD biomarkers, gene expression-based PD biomarkers have recently emerged as quantitative biomarkers. This review introduces examples of gene expression-based mRNA biomarkers that have been validated in preclinical or clinical studies of several anti-tumor agents including HDAC, mTOR, and B-RAF inhibitors. The measurement of PD biomarker levels in tumors has proven to be ideal; however, in clinical studies, easily accessible surrogate tissues have been used for analysis. In the present review, we also discuss the advantages and disadvantages in using surrogate tissues, such as peripheral blood mononuclear cells (PBMCs), skin tissue, and circulating tumor cells, in the assessment of PD biomarkers. PD biomarkers are generally classified into two categories: 1) target engagement biomarkers and 2) early efficacy biomarkers. This classification depends on their respective distance from target intervention. The strategies used to identify and distinguish between these two types of PD biomarkers via expression profiling are also discussed. Finally, we propose two novel approaches for PD marker identification. One approach utilizes mRNA expression profiling of tumors prior to drug treatment rather than post-treatment samples. The second method involves the application of microRNA expression profiles to determine PD effects. In conclusion, the recent advances in mRNA and microRNA profiling and the identification of gene expression-based PD biomarkers may aid investigators to drive drug development through the establishment of quantitative PD effects.