Breaking down silos is a key business initiative.
By Frank Klein
Oct. 8, 2008 | Amid a more stringent regulatory environment and soaring R&D costs, the imperative is to “fail fast and fail cheap” with new drug candidates. While there is no doubt that the pharmaceutical industry is growing in complexity, there are relatively simple steps companies can make to address logistical bottlenecks throughout the drug development cycle.
Throughout the development of a candidate drug, research teams work in isolation on tightly-defined projects, often operating independently, or even competitively. This lack of cooperation is exacerbated by the increasing complexity of tests and treatments available and a frequent lack of formal structures for sharing intelligence. This results in the development of information “silos” in which researchers are not able to access all of the intelligence previously gathered in a candidate drug’s development, and ultimately, the repetition of expensive research.
“A siloed approach to the adoption of technology has left the majority of pharmaceutical organizations with some serious business and technological challenges ahead,” wrote Markella Kordoyanni, a pharmaceutical technology analyst at Datamonitor, in the 2007 report, Trends to Watch: Pharmaceutical Technology. The objective should be to improve communication channels between laboratory scientists and those working in the pre-clinical and clinical stages of drug development, so that scientists can use clinical feedback to refine their investigations. For example, reliable comparisons of in vivo and in vitro safety study data linking subjects’ symptoms with molecular data creates a bridge between classical and molecular pathology. Biomarkers identified in the laboratory can be tracked into clinical trials—or vice-versa—providing clear proof of mechanism and proof of concept. As translational and personalized medicine become a reality, open communication between all stakeholders is critical. One area in particular in which information processing and sharing is becoming increasingly critical is proliferating digital imaging data.
Today, more than 70 per cent of all data generated in life sciences research is in the visual form, but its huge potential is only partially tapped. Pharmaceutical companies developing new drugs and therapies need to assimilate not only molecular and genomic information, but also complex, heterogeneous image data from all stages of the drug development cycle. In 2004, at the Fifth National Forum on Biomedical Imaging in Oncology, Janet Woodcock, director of FDA’s Center for Drug Evaluation and Research commented that, “There is tremendous potential for the use of imaging in drug development... from pre-clinical [applications] all the way to using surrogate markers for approval.” The drug development process is “increasingly challenging, inefficient, and costly,” and imaging technology remains “at the forefront of [the FDA’s] efforts to streamline the drug approval process.”
Manual image analysis is labor-intensive and compounded by a shortage of experts to interpret digital data. It is not unusual for major pharmaceutical companies to have over 100 different imaging instruments, and their image analysis requirements are likely to increase exponentially in the coming years. Cell proliferation and immunohistochemistry studies are a critical focus of pre-clinical pathologists. These studies can prove time-consuming, requiring scientists to manually select regions of interest and conduct cell counts. Utilizing automated image analysis technology in a 13-week oral toxicity study on 180 mice, a major pharmaceutical company reduced the time spent on the analysis of a total of 9,000 images from 16 weeks to just four. Such inefficiencies are compounded through each stage of the drug development process.
While no system can supplant highly trained specialists, adopting systems that can help bring objectivity and automation to image analysis can help streamline research and the development of more effective therapies. Currently, in most bioscience and pharmaceutical companies, image analysis tools are purchased as point solutions to solve specific problems, often imbedded in the image acquisition tools themselves. Although they may improve the speed and accuracy of analysis, many accept only particular inputs or file types, restricting their usefulness. Replacing point solutions with a single, standardized platform that handles all of an organization’s image analysis tasks can help alleviate productivity loss from inefficiencies in the analysis of digital data and expedite early candidate attrition rates.
The adoption of integrated software infrastructure for the storage and sharing of information between researchers and clinicians can be a costly and intensive process. Nonetheless, the initial cost and time outlay are justified. Implementing appropriate software platforms removes a key hurdle in the occasional breakdowns in communication that exacerbate productivity loss and waste research funds.
Frank Klein is VP of medical imaging at Definiens. He can be reached at firstname.lastname@example.org.
This article appeared in Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. Apply Today.