Scientific Objectives


The cost of drug development has increased sharply in recent years, while approval of innovative drugs has declined. This “productivity gap” represents a major challenge for the pharmaceutical industry in Europe and throughout the world. One of the main reasons for the lack of new, innovative drugs is the high attrition rate during drug development. Three factors are major contributors and represent 66% of the causes for termination of drug development projects: toxicity in animals (20%), non-acceptable unwanted effects in humans (11%) and lack of therapeutic efficacy (25%). Both lack of efficacy and unpredicted toxicity is due to the paucity of mechanism based and predictive preclinical toxicity testing approaches.


The objective of Predict-IV is to provide an improved predictivity of the non-clinical safety testing well in advance of performing animal safety testing (pre-phase I development) and clinical trials (phase I and beyond).


Recent advances in tissue and bioreactor technologies, molecular biology, toxicity modelling, and bioinformatics are leading to new approaches in the integration of toxicity testing in vitro into the early stages of drug development, with the possibility of significantly increased predictability. The project will integrate these new developments to improve and optimize cell culture models for toxicity testing and to characterize the dynamics and kinetics of cellular responses to toxic effects in vitro. The target organs most frequently affected by drug toxicity will be taken into account, namely liver and kidney. Moreover, predictive models for neurotoxicity are scarce despite the fact that the neuronal system is a frequent target of undesired side-effects of drugs. The CNS is therefore addressed by the call and included as a target system. For each target organ the most appropriate cell model will be used. The choice of these models is based on availability and biological relevance to toxicity. It is envisaged to adopt and optimize the “best choice” of established cell systems rather than to undertake development of new cellular models.


Predict-IV will combine classical in vitro toxicology with innovative technologies, profiling and modelling tools in a system biology approach (from pathology to molecular mechanisms and in vitro kinetics) from the perspective of academia, industry and SMEs. High quality standards based on sound modelling and biostatistical practices will be applied to establish the experimental design of the in vitro dynamic and kinetic experiments leading to high quality data generation for further analysis, evaluation and integration. For the introduction into pharmaceutical discovery and development, this new test strategy will ultimately need to be validated.


Participants with complementary expertise to address the widely different issues will contribute to the project. Established and optimized cell culture systems with the ability to indicate deregulation of specific cellular processes within a short time after exposure (early markers of toxicity) will be used. The combination of profiling methods such as “omics” promises further insights into cellular responses to xenobiotic insults. In this area, the project will build on the knowledge gathered from other large research efforts. In particular there will be a direct link to the on-going InnoMed/PredTox project that is studying early biomarkers of toxicity of selected pharmaceuticals in vivo. The main purpose of PredTox is the construction and delivery of an integrated database from in vivo experiments on compounds with known toxicity profiles and correlating biomarkers of early toxic effects (transcriptomics, metabolomics and proteomics) with pathology.


The approach will be evaluated using a panel of drugs with well described toxicities and kinetics in animals and partly also in humans. Some of the compounds analyzed in vivo in PredTox will also be included in the Predict-IV project. This approach will be highly advantageous as it will allow a direct comparison between the in vivo to the in vitro data. A parallel analysis of several dynamic and kinetic models with a broad spectrum of endpoints should allow for the identification of several relevant biomarkers of toxicity. Inter-individual susceptibilities will be taken into account by integrating the polymorphisms of the major drug metabolizing enzymes and correlating the observed effects in the human cell models with their genotype. Environmental influences on cellular toxicity to these compounds will also be evaluated using hypoxic stress as a relevant test model. The ultimate goal is to provide an integrated test system approach with specific safety markers to predict toxicity prior to pre-clinical animal testing. Exposure data in vitro will be provided that can be used to derive margins-of-safety.


A highly integrative approach from a multidisciplinary team with expertise in analytical chemistry, biochemistry, cellular model development, toxicogenomics, metabolomics, high-content imaging technologies, bioinformatics, kinetic modelling, toxicology and risk assessment is required. The consortium of Predict-IV represents a wide variety of specific knowledge and many partners with demonstrated excellence in their fields that will cooperate to complete specific deliverables, which will then be integrated into a coherent testing strategy.


Predict-IV addresses a number of research priorities from the sub-priority “Predicting suitability, safety and efficacy of therapies”. This project proposal is also in line with the aim of the call to mobilize and integrate research excellence necessary for the promotion of new strategies for alternative testing.


Predict-IV will aid in the scientific, technical, wider societal and policy objectives of the Health Priority in a number of ways:


- Predict-IV will build on a growing knowledge-base delivered by DGRTD FP6 projects, the progress made in pharmacological/toxicological screening and safety pharmacology and the input of an extensive list of scientific networks.


- One major contribution will be the application of cell culture systems (mainly based on human cells) to the early drug discovery and development process to accelerate drug development of pharmaceuticals and target organ specific and non-invasive endpoint determination.


- Profiling methods mainly based on metabolomics and genomics, and other endpoints indicative of the deregulation of essential cellular processes will deliver early biomarkers of human toxicity for pharmaceuticals. Improved knowledge on the toxicity mechanisms will be obtained for some model compounds.


- The development of better measurements or estimates of the real exposure of cells to drugs and/or their metabolites in the in vitro test systems will be a key element for the extrapolation of in vitro results to the in vivo situation.


- The development of consistent, 3Rs conscious strategies for predicting the kinetics of pharmaceuticals in man, should allow both early information on kinetics to be used for in vitro tests and comparison of in vitro toxic concentrations with tissue doses in vivo. These should use mostly in vitro and in silico techniques to give input data into relevant (particularly physiologically-based) kinetic models. The development of a consistent methodology to compare in vitro exposure and predicted in vivo kinetics is an important aspect for accurate toxicity prediction from in vitro toxicological test results. ECVAM will ensure that the test methods are developed according to ECVAM's criteria for inclusion of the developed strategies in future validation studies (Hartung et al., 2004).


- Predict-IV also aims to contribute to maintain Europe's position as a world-leader in the field of animal reduction in medical research and PBPK modelling.


Taken together, the Predict-IV’ approach combines biological effects (toxicodynamics) with toxicokinetics and modelling ensuring the generation of real exposure data linked to effects. This will enable a relevant extrapolation from in vitro to in vivo including predictivity for adverse effects in human. The established models can then be flexibly integrated into any drug discovery testing cascade for the selection of the best drug development candidates.