Two years ago, the pharmaceutical laboratory Hoffman-La Roche was on the point of giving up on a new drug for hypertension. The American Food and Drug Administration had noticed an anomaly in patients' electrocardiogram, and were concerned that the medicine might trigger off a cardiac arrhythmia.
The Swiss laboratory decided then to submit the problem to a small New York company, Physiome Sciences, who were developing a 'virtual' heart, in other words, a mathematical model of the electrical, chemical and mechanical activity of the organ. In remaining virtual, the researchers were able to observe the electrical activity of the heart, whilst able to check that it could not trigger off arrhythmia. Better still, they notice that other similar drugs on the market, which never triggered off arrhythmia, also registered the same anomaly.
Simulating the reaction to a drug
For the FDA to accept the simulations’ conclusions tells us much about how advanced these techniques have become. The Physiome Sciences virtual heart is the result of many years of research. Dennis Noble, Physiologist at Oxford University started work on the subject some 35 years ago. He first simulated the operation of a single cardiac muscle cell by integrating phenomenon’s such as ion exchanges through the membrane, electric polarization of the cell, reaction of certain proteins on the membranes receptors, consumption of oxygen and the cell’s production of CO2 etc. Then, by perfecting his model and benefiting in the increase in computer power, he managed to assemble several cells and simulate their interaction. For the next stage, the British researcher joined forces with two other scientists, Raymond Winslow of John Hopkins University of Baltimore and Peter Hunter of Auckland University in New Zealand. Together, they built a three dimensional model of a heart (in fact, two ventricles) inspired by the arrangement of cells in a real heart.
The function of the virtual heart, which consists of a million cells, can be visualised on a computer screen. As required the researchers can simulate a heart attack or a fibrillation. It is also possible to simulate the action of a drug, as in the case of the anti-hypertension drug from the Swiss laboratory. The physio-chemical data is then entered into the model to simulate the action of the complete heart.
It is probably here that we find the greatest potential application for the virtual heart. Pharmaceutical laboratories are, in fact, looking for techniques which will allow them to test new molecules as early as possible to reduce the number of candidates which are rejected at the clinical trials stage. And it is to exploit these possibilities that Dennis Noble and Raymond Winslow have created Physiome Sciences, with other scientists and a businessman, Jeremy Levin.
The company is already interested in other organs such as the liver, the pancreas and the kidney. In May 1999 Physiome Sciences signed an agreement with PA Consulting Group, to develop a model of the immune system. This association is not as surprising as it may at first appear. As Ken Cooper of PA Consulting in Boston explains, 'We see, in the human body, a complex hierarchy of non linear regulating cycles which, most of the time, efficiently control the function of the interconnected systems, but which, in the case of an illness functions badly. This also happens in a number of economic systems that we have already modeled and we think our methods can be applied to biological systems'.
PA Consulting, who have put 10 people on the project, will allocate five to 10 million dollars to it. The two partners hope to reach their goal within a year and a half. But they are thinking of going further still; in their minds, the model of the immune system is only one step towards a 'super model' allowing the models of different organs to be linked together in order to study their interactions. 'Whether this takes a few years or a lot longer, our ultimate goal is to develop a model, which integrates all the systems, in other words the whole human body' declares Jeremy Levin who is the head of the Board at Physiome Sciences.
Encouraging researchers to adopt a same approach
The American company is not alone in their idea of simulating organs. Entelos, in San Francisco is also working on an immune system; NaviCyte in Nevada is building a model intestine to observe drug absorption, Peter Hunter in New Zealand now specialises in the modeling of lungs to study illness such as enphecema and asthma.
All these laboratories which are working on the simulation of organs use different mathematical tools and incompatible data. This is likely to make the eventual integration of their work (into a model of a human body) difficult. For this reason, three years ago, Washington University in Seattle, launched a project called Physiome with the intention of encouraging researchers to adopt a common approach in modeling organs. This project is based on the model of the human genome code, which was itself launched almost ten years ago.
