YouTube Facebook LinkedIn Google+ Twitter Xinginstagram rss  

Systems Pathology for the 21st Century

By Irena Melkinova

April 15, 2005 | Ensuring the safety of drugs brought to market is one of the most crucial and expensive undertakings of any biopharmaceutical company. Veterinary pathologists have always played an important role in this process, conducting pre-clinical safety studies in various animal species to assess potential toxicity before introducing a drug in humans. During the past decade, the advent of combinatorial chemistry, advances in high-throughput technologies, and the genomics revolution produced a glut of drug candidates and drug targets for evaluation by pathologists, moving pathology assessment earlier into the discovery process and, consequently, putting a lot of pressure on already limited resources.

Despite major technological advances in the industry, the way pathologists operate has remained largely unchanged: Pathologists still spend hundreds of hours looking down a microscope at thousands of tissue slices to make a diagnosis. With increasing R&D costs, and pressures to improve productivity, reduce attrition rates, and produce drugs with increased efficacy and safety, the time is ripe for serious advances in the way pathology assessment is carried out. A more comprehensive approach to safety evaluation, akin to systems biology in drug discovery, is necessary. We call this  approach systems pathology. In a recent white paper sponsored by SAIC, Life Science Insights explored the challenges and opportunities this presents for the biopharma industry and technology vendors.

The systems pathology approach integrates multiple biological data streams — from “-omics,†to anatomic pathology across various species, to clinical and medical information — enabling the comprehensive understanding of biological processes for drug development. A combination of two major components constitutes the systems pathology concept: a high-throughput pathology platform with sophisticated informatics support for computer-assisted diagnostics; and informatics applications for data integration, visualization, analysis, and mining.

The platform for high-throughput pathology includes equipment for automated pathology slide reading and software for data processing. Slides are scanned and digital images stored in a database. Analysis software processes visual information and distinguishes between normal and abnormal specimens. Abnormal samples are flagged. Sophisticated algorithms that perform quantitative histomorphometric data analysis allow pathologists to make more objective computer-assisted diagnoses than currently possible. Such a platform will improve pathologists’ efficiency and arm them with quantitative tools to identify subtle changes on a consistent basis.

Extraction Required

The data-integration component of a systems pathology platform and accompanying informatics tools should help uncover patterns of behavior within the data in relation to treatment, detect patterns that relate to and explain disparate endpoints (for example, pathology and toxicogenomics data), and allow the discovery of biological networks that control observed events. Extracted information will be instrumental for identifying and validating new drug targets, surrogate markers of efficacy and toxicity, and the mechanisms of toxicity.

Several technology companies are developing tools and services that could be incorporated into a systems pathology platform. However, no company alone has a full solution at this point. High-throughput pathology systems that include digital image capture and data-analysis informatics are in development by Icoria, Aperio Technologies, LifeSpan Biosciences, Systems Pathology Corp., Trestle, and Histometrix, among others. Gene Logic and Iconix Pharmaceuticals offer databases of integrated toxicogenomics information. Of these, Icoria and Systems Pathology Corp. are also working on integration of -omics data with biochemical and phenotypic profiling and, at this point, are the closest to a systems pathology concept. What differentiates Systems Pathology Corp. from the rest of the field is the close involvement of experienced pathologists in the development of bioinformatics algorithms for histopathology data processing and analysis. This element is crucial for winning end-users’ acceptance.

Systems pathology represents an enabling technology that has the potential to revolutionize the way pathology assessment is performed.

Irena Melnikova is senior research analyst at Life Science Insights. E-mail:

Click here to login and leave a comment.  


Add Comment

Text Only 2000 character limit

Page 1 of 1

For reprints and/or copyright permission, please contact Angela Parsons, 781.972.5467.