In vitro toxicity testing has been used for many years, but the predictive power for either rodent, or human toxic liability has not been demonstrated, until recently.
|
The existing approaches to safety testing in the preclinical phases of drug development are not powerful enough to reliably identify potential toxic liabilities for humans. In addition, regulatory agencies and the pharmaceutical industry understand that the present “gold standard” of animal testing is not the answer, as rodent testing is about 50 percent predictive of human toxicology. The US Food and Drug Administration (FDA) has a goal to reduce, refine, and ultimately replace animal testing, and European agencies have also accelerated the search for alternative methods. Technology platforms that elucidate toxic liabilities prior to preclinical, and especially clinical phases of drug development, will save time and money and will decrease risk for both patients and the pharmaceutical industry.
For many years, simple cytotoxicity testing has been used before and in combination with animal testing. Earlier cell-based assays measured only one, or a few, cellular activities, and up to this point have not been widely embraced. Therefore, tremendous effort has been made to develop more predictive in vitro model systems that reflect either specific animal or human toxicology.
Many toxicologists want to demonstrate the potential of cell-based assays to predict known toxicity in animals, and ultimately, human toxic liabilities. Optimally, these models are inexpensive compared with animal testing and have sufficient throughput and cost-savings that can be applied earlier in the drug discovery and development process (such as late primary screening to hit-to-lead phase). This would enhance efficiency in identifying toxic compounds that have a low probability of success.
Cell-based models address these challenges. They can be designed to reflect animal or human toxic responses, and optimize recent efforts in toxicogenomics and metabolomics to select functional biomarkers that can be implemented at early stages of development. Cell-based models can be implemented at the interface with drug discovery, and provide the necessary continuity from the discovery lab to the clinic.
Currently, most major organ systems have been explored though single cell type models or co-cultures. Simple cell-based assays exist for the liver, kidney, brain, immune system, heart, and co-cultures of multiple cell types from the same organ.
Historically, cell-based assays are developed around cell lines that are well established in the research community, or from primary cells prepared from the selected organ. The advantages of cell lines are the ease with which they can be grown and maintained in culture, the relative homogeneity of cells within an assay run, and reduced cost.1 The most important disadvantage of note is the loss of some critical physiological activities that are inherent in the process of stable cell line creation.
In the case of hepatocyte cell lines, such as HepG2, the level of metabolic enzymes is greatly reduced. However, there have been valuable efforts to engineer cell lines to bring back some of the critical activities. In contrast, primary cells isolated from animals or humans more closely reflect the native functions within the living system.2 The cells need to be freshly prepared or shipped overnight from a variety of suppliers, since the full functionality of the cells has a relatively short lifetime. This is why assay profiling must directly follow preparation, because the heterogeneity of the cell population is usually higher and the cost is more substantial.
There have been two types of primary cell-based assays using intact cells and fluorescence as the primary detection method. The first is whole-plate assays, where a population average of cellular responses within each assay well is the readout. The second is high content screening (HCS), where imaging within each well of an assay plate permits single cell and sub-cellular measurements. The advantages of whole-plate reads are increased speed of measurement and smaller data sets. The disadvantages are the average response within a population of cells, where individual cellular response profiles might reflect important information, as well as the inability to analyze potentially critical spatial information within cells.
|
Advantages of HCS include the ability to analyze cellular sub-population responses, multiplex measurements using multi-color fluorescence, and analyze important spatial information, including nuclear texture and organelle redistribution, all of which create a rich data set. The disadvantages of HCS include longer readout times, and the management and analysis of data. The technologies around HCS, including throughput of assays, data management, and analysis informatics, have evolved so that the advantages of applying HCS outweighs the disadvantages. Important demonstrations of the value of the imaging approach have appeared.3,4,5
The single gene, single target, single new chemical entity strategy of the pharmaceutical industry has run its course. This over-simplification of the human system was responsible for discovering and launching important drugs, but the challenge has now become much greater.
Animals, including humans, are made up of integrated and interacting networks of genes, proteins, and metabolites that give rise to normal and abnormal functions. The cell, as the first-level system, exhibits properties that are not anticipated from detailed knowledge of the components or component functions. These “emergent” properties include life itself; since life emerges from the system, the assessment of a substance’s impact on life requires the measurement of many component functions within the same cells that capture the “systems” response. The concept of holistic biology (or systems biology) was the original reason that whole animal studies were used. Recently, whole genome profiling has led the way for exploring systems.6 Toxicogenomics is considered a logical extension of this perspective and has yielded some important insights. However, it is important to understand the functional responses of cells in addition to the genomic response to challenges.
|
Cellular systems biology (CSB) can address the complexity of the system’s response to challenges, while reducing the massive amounts of data generated into actionable indices of response, such as a safety risk index. This approach harnesses the advantages of HCS readouts, while adding more features to express the system’s response and classifier software to create a predictive tool.7(Figure 1, page 16)
Early studies have demonstrated that a CSB profile shows predictive power not evident in more simple cell-based assays.8 (Figure 2) There are now rodent and human hepatocyte models using both cell lines and primary cells coupled to classifiers that are helping to create a growing database of compounds with safety data. There must be an industry-wide initiative to create panels of the most critical cell types with their own panels of organ-specific, functional biomarkers and classifiers. Such a system has the potential to save time and money, while reducing the safety risk.
The next major step is to evolve the development of animal and human stem cell-derived systems for major organ systems.9,10,11 These cells have the potential to combine the advantages of both cell lines and primary cells. In addition, after the development of single cell type models, it is possible to create “tissue” models by building micro-3D arrays of cells, including multiple cell types. Therefore, although there are powerful CSB assays in existence today, they will continue to evolve and gain more value for the industry. The integration of panels for different cell types representing major organs will also include the development of classifiers that will analyze the full “systems” response. The vision to reduce, refine and ultimately replace animal testing is within our grasp.
References
1. O’Brien PJ, et al. High concordance of drug-induced human hepatotoxicity with in vitro cytotoxicity measured in a novel cell-based model using high content screening. Arch Toxicol. 2006;9:580-604.
2. Hewitt NJ, et al. Primary hepatocytes: current understanding of the regulation of metabolic enzymes and transporter proteins, and pharmaceutical practice for the use of hepatocytes in metabolism, enzyme induction, transporter, clearance, and hepatotoxicity studies. Drug Metab Rev. 2007;1:159-234.
3. Haskins JR, Rowse P, Rahbari R, de la Iglesia FA. Thiazolidinedione toxicity to isolated hepatocytes revealed by coherent multiprobe fluorescence microscopy and correlated with multiparameter flow cytometry of peripheral leukocytes. Arch Toxicol. 2001;7:425-38.
4. Xu JJ, et al. Cellular imaging predictions of clinical drug-induced liver injury. Toxicol Sci. 2008;1:97-105.
5. Taylor DL. Past, present, and future of high content screening and the field of cellomics. IN: Taylor DL, Haskins JR, and Guiliano KA, eds. High Content Screening: A Powerful Approach to Systems Cell Biology and Drug Discovery, Totowa NJ: Humana Press; 2007:3-18.
6. Stoughton RB, Friend SH. How molecular profiling could revolutionize drug discovery. Nat Rev Drug Discov. 2005;4:345-50.
7. Vernetti L, et al. Cellular Systems Biology Applied to Pre-Clinical Safety Testing: A Case Study of CellCiphr- Cytotoxicity Profiling In: Ekins S and Xu JJ eds. Drug Efficacy, Safety, and Biologics Discovery: Emerging Technologies and Tools. Hoboken, NJ: John Wiley and Sons; 2009:53-74.
8. Cellumen, internal data.
9. Khetani SR, Bhatia SN. Microscale culture of human liver cells for drug development. Nat Biotechnol. 2008;1:120-6
10. Yang L, et al. Human cardiovascular progenitor cells develop from a KDR+ embryonic-stem-cell-derived population.Nature. 2008; 7194:524-8.
11. Kettenhofen R, Bohlen H. Preclinical assessment of cardiac toxicity. Drug Discov Today. 2008;13(15-16):702-7.
About the Author
D. Lansing Taylor, PhD, has spent more than 25 years in academia and the industry, and is the author of more than 150 scientific papers.
This article was published in Drug Discovery & Development magazine: Vol. 12, No. 3, March, 2009, pp. 16-18.
Filed Under: Drug Discovery
