A key part of drug discovery and development is the characterization and optimization of the safety and efficacy of drug candidates to identify those that have an appropriately balanced safety-efficacy profile for a given indication. The therapeutic index (TI)--which is typically considered as the ratio of the highest exposure to the drug that results in no toxicity to the exposure that produces the desired efficacy--is an important parameter in efforts to achieve this balance. Various types of safety and efficacy data are generated in vitro and in vivo (in animals and in humans), and these data can be used to predict the clinical TI of a drug candidate at an early stage. However, approaches to systematically and quantitatively compare these types of data and to apply this knowledge more effectively are needed. This article critically discusses the various aspects of TI determination and interpretation in drug development for both small molecule drugs and biotherapeutics.

MLN2704 is an antibody-chemotherapeutic conjugate designed to target prostate-specific membrane antigen (PSMA). PSMA is a transmembrane receptor whose expression is largely restricted to prostatic epithelium and prostate cancer cells with its expression level increasing during the progression of malignancy. MLN2704 consists of a de-immunized, monoclonal antibody that is specific for PSMA conjugated to drug maytansinoid 1 (DM1), a microtubule-depolymerizing compound. After antibody binding to PSMA and the subsequent cellular internalization of this complex, DM1 is released leading to cell death. MLN2704 has an approximate half-life of 39 hours in scid mice bearing CWR22 tumor tissue, and the antibody effectively penetrates xenograft tumor tissue. Optimization of dosage and schedule of MLN2704 administration defined interdependency between these conditions that maximized efficacy with no apparent toxicity. Tumor growth delays of ϳ100 days could be achieved on the optimized schedule of one dose of 60 mg/kg MLN2704 every 14 days for five doses (q14d؋5). The unconjugated antibody (MLN591) demonstrated essentially no antitumor activity and DM1 alone or a non-PSMA targeted antibody-DM1 conjugate was only weakly active. Furthermore, we show that MLN2704 is active in a novel model of osteoblastic prostate cancer metastasis.

Drug transporters, including efflux transporters (the ATP binding cassette (ABC) proteins) and uptake transporters (the solute carrier proteins (SLC)), have an important impact on drug disposition, efficacy, drug-drug interactions and toxicity. Identification of the interactions of chemical scaffolds with transporters at the early stages of drug development can assist in the optimization and selection of new drug candidates. In this review, we discuss current in vitro and in vivo models used to investigate the interactions between drugs and transporters such as P-gp, MRP, BCRP, BSEP, OAT, OATP, OCT, NTCP, PEPT1/2 and NT. In vitro models including cell-based, cell-free, and yeast systems as well as in vivo models such as genetic knockout, gene deficient and chemical knockout animals are discussed and compared. The applications, throughput, advantages and limitations of each model are also addressed in this review.

No abstract