Biomedical Engineering Reference
In-Depth Information
designed for the treatment of angina. Unexpected and serendipitous
fi ndings in Phase I clinical trials meant that it was repositioned in 1998 to
treat erectile dysfunction and sildenafi l sold as Viagra had peak sales of
around $2B by 2008. This drug works as a PDE5 inhibitor and protects
cyclic guanosine monophosphate (cGMP), leading to smooth muscle
relaxation, causing vasodilation that increases the infl ow of blood into the
spongy, penile tissue [8]. By understanding how this drug works at the
biological level of the cells and tissue, it was then repositioned again in
2005 into pulmonary arterial hypertension (PAH). This rare lung disorder
affects patients whose arteries that carry blood from the heart to the lungs
become narrowed, making it diffi cult for blood to fl ow. This was launched
as Revatio (sildenafi l) as a lower dose, which by relaxing the smooth
muscle leads to dilation of the narrowed pulmonary arteries [9].
When new indications are considered for our AstraZeneca compounds,
having a detailed understanding about how each of our drugs works is
critical. As a large pharmaceutical company, with over 12 000 R&D
scientists that contribute to the collective knowledge about our drugs in all
of our existing and historical projects, having this in a single repository
would be incredibly valuable. Although corporate, global data repositories
do hold millions of documents that describe every fi nding about all of our
compounds, this is held in unstructured texts, project minutes, drug fi lings,
product labels and even marketing material. Unfortunately the rich,
invaluable and often tacit knowledge about how these drugs work is not
stored systematically in a structured format, but in our brains! Luckily, our
project scientists are always keen to discuss their compounds and how the
biological function can be used in new potential indications.
To capture this knowledge once and for all, a project was initiated to
describe how our compounds and drugs work at the biological level.
Historically our focus has been at the signalling pathway level, with gene
targets, traffi cking and receptor interactions. This project tried to
understand what the drug did to the cells, the tissues and the pharmacology
at the pathophysiological level. Although run as an informatics project to
store this information in new ontologies and visualise the results in new
biological process pathway maps, the information was gathered by
coordinating hundreds of interviews across the company. Similar issues
within other major companies are discussed in this topic by Harland and
colleagues (Chapter 17) and by Alquier (Chapter 16).
The technology used to build these visual network maps was
Prefuse, which easily lends itself to network objects and attaching
relationships. Rather than a static network diagram, these biological
process maps (Figure 14.8) have been designed to be built within the
Search WWH ::
Custom Search