Page 163 - Visions of the Future Chemistry and Life Science

The making of the virtual heart 147

8.5.2 Added value for drug and device development
Drug development is currently largely based on trial and error. This is an
exceedingly time-consuming process, and some of the associated errors
have proved quite costly for patients involved. Even if distressing conse-
quences of clinical testing could be avoided, the economical costs of bring-
ing a new drug to market are prohibitive: close to US$0.5 billion.
Also, the fact that only an estimated 10 per cent of pre-clinically tested
lead-compounds are likely to ever reach the market must discourage com-
panies from investing into new drug development, in particular for pathol-
ogies that are not deemed to constitute a profitable market. Thus, from the
point of view of a commercial drug developer, ideal targets are chronic and
non-lethal complaints that affect people in the developed world at the
prime of their financial viability. In other words, it is ‘more economical’ to
come up with a treatment for obesity, baldness or impotence, rather than
to tackle a rare but lethal disease that affects small patient groups or people
in underdeveloped regions of the world.
Analytical computer models clearly have the potential to improve this
situation, as they may help:
• to speed-up drug development by in silico screening for early identifi-
cation of promising lead compounds;
• to simplify the assessment of complex pre-clinical data and predict
(patho-)physiological (side-)effects of drugs;
• to cut the associated financial and ethical costs;
• to reduce the risk of clinical testing.

The above may not be enough, though, as it will be crucial to change
the whole approach to drug development. What is needed is a method to
identify the desired drug effect and (sub-)cellular target for pharmacological
intervention before directed compound synthesis and testing commence.
Virtual organs will form the basis for this novel approach.
Similar concepts apply to the world of medical devices. In future, suc-
cessful products will increasingly be tuned to flow with the stream of
human physiological function, even to mimic it in fine detail. Modelling
and computation are set to make major contributions, since:
• devices become sufficiently ‘intelligent’, with their on-board comput-
ing power, to use analytical descriptions of (patho-)physiological organ
function;