The introduction of human in vitro data to validate animal-based test methods will enable a more informed choice as to their use and will likely enhance the quality of the resulting decision making.
In some cases, there is no useful animal model for establishing proof-of-concept for efficacy. Cystic fibrosis results from the presence of a mutant form of the gene encoding the epithelial ion transporter CFTR. Although cystic fibrosis causes problems with epithelial function in organs throughout the body, it is the effects on the lung that are usually the cause of death. Attempts to develop a useful animal model have failed. Even CFTR knockout mice do not display the broncho-pulmonary symptoms that are so characteristic of cystic fibrosis.
We do not intend to suggest that in vitro data alone can provide all the necessary information for clinical go/no-go decision making. In light of these examples, however, we believe that efforts should be directed toward human tissue studies in vitro to better understand the relevance of in vivo test systems.
It must also be accepted that in vitro testing has its limitations, at least in part because isolated tissues can never fully represent the complex integrated biological systems operating in vivo. Indeed, there are examples of diseases for which efficacy and side effect testing can really only be undertaken in vivo, such as psychiatric disorders. In such cases, as with the choice of species for ADME/safety testing, scientists have a responsibility to establish, as far as possible, the relevance of the model(s) chosen. This can in part be achieved by comparing profiles of expression and function of key targets and biochemical pathways in relevant human tissues with those in candidate animal models.
To make human in vitro testing available to the biopharmaceutical industry, Asterand (www.asterand.com) has established XpressBANK™ and ProCURE™, through which it provides human tissues for research programs, and PhaseZERO®, through which it provides human tissue-based research services. These not only aid the identification and validation of human native targets and biomarkers but also contribute to an understanding of the action, disposition, and safety of potential new medicines.
For example, the antidiabetes drug Rezulin (troglitazone) was withdrawn from the commercial market in 2000 as a result of hepatotoxicity in some patients. In a subsequent published PhaseZERO study, troglitazone and two other compounds of the same class, rosiglitazone and pioglitazone, were tested for hepatotoxicity on human hepatocytes.
Troglitazone demonstrated a narrow therapeutic window when comparing the concentrations required to induce frank hepatotoxicity in vitro with the maximum blood concentrations in patients. This narrow window was not observed with rosiglitazone or pioglitazone. Early PhaseZERO profiling of troglitazone could have identified a high risk of liver toxicity and provided a means of identifying a safer compound of comparable efficacy.