For the last four decades, synthesis and testing of potentially active drugs (e.g., antimetabolites) have focused on structural modification of existing metabolites as precursors of DNA and RNA synthesis. In recent years, the focus has shifted to synthesis of target-specific agents. Thus, the current emphasis of drug development is directed at inhibiting specific target(s) expressed preferentially, if not exclusively, in tumor tissues, with the ultimate goal of improving the therapeutic efficacy and selectivity of these new agents. Preclinically, proof-of-principle studies were carried out in tumors with specific expression of the intended target. With the hope of translating preclinical findings to the design of implementation of clinical trials. Thymidylate synthase (TS) continues to be a critical target for 5-fluorouracil (5-FU) and its prodrugs, UFT/LV (Orzel), capecitabine (Xeloda), and S-1, primarily because this enzyme is essential for the synthesis of 2-deoxythymidine-5-monophosphate, a precursor for DNA synthesis. While fluoropyrimidine antimetabolites have other sites of action, antifolates ZD1694 (raltitrexed, Tomudex) and AG337 (Thymitag) are more specific and potent TS inhibitors. Thus, it is hoped that pronounced and sustained inhibition of this enzyme could result in downstream regulation of molecular markers associated with sensitivity and resistance to these agents. It is also critical to recognize that the degree and duration of inhibition of the target enzyme may depend on the expression level of the target enzyme, thymidylate synthase. Correlative studies in preclinical and clinical systems demonstrated a close relationship between the enzyme level (mRNA and protein) and response to therapy of colorectal cancer patients treated with fluoropyrimidine or Tomudex. However, significant overlap was demonstrated between responders and non-responders. These data are consistent with the hypothesis that prediction of response to anticancer drugs is multifactorial, and TS is one target. Clinically, although overall response of colorectal cancer patients to a variety of TS inhibitors is similar, toxicity profiles are different. The availability of the 5-FU prodrugs offers the possibility of greater therapeutic selectivity based on the demonstration that thymidine phosphorylase, the activating enzyme for 5-FU, is expressed at a higher level in tumor tissue compared with normal tissue counterparts. It is likely that successful application of TS inhibitors will not only be based on measurement of the TS level in tumors vs. normal tissues, but on the delineation of the consequences of this inhibition on molecular markers associated with cellular proliferation, apoptosis and cell cycle regulation.