Predictive toxicology approaches for small molecule oncology drugs.

A daunting, unmet medical need exists for effective oncology chemotherapies, with cancer deaths in 2009 to exceed 560,000 in the United States alone. Because of the rapid demise of the majority of cancer patients with metastatic disease, oncology drug development must follow a much different paradigm than therapeutic candidates for less onerous diseases. The majority of drug candidates in development today are targeted at cancer therapy. Many of these candidate chemotherapeutic agents are active against novel targets, often presenting unique toxicological profiles. Since many of these novel targets are not unique to cancer cells, therapeutic margins may not exist. Decision making, in this event, is among the most challenging that any pharmaceutical toxicologist/pathologist or regulator will face. Nonclinical development scientists must compress timelines to present therapeutic options for cancer patients who have failed conventional therapy. In support of this goal, the U. S. Food and Drug Administration has created an oncology-specific paradigm for nonclinical testing and has introduced strategies to accelerate development and approval of successful candidates. Pharmaceutical toxicology testing strategies must not only satisfy regulation as the minimal expectation, but also attempt to reduce the current high attrition rates for oncologic candidates. A successful toxicology testing strategy represents the substance of this treatise.