A heartbeat inside a computer

Print Friendly Version of this pagePrint Get a PDF version of this webpagePDF




Two scenarios. The first one could be entitled: "A marvelous new drug is launched on the market." This is a well-known situation. We know it from our own experience, from the stories told by our relatives or from media coverage, sometimes overly sensational. The second scenario, with the heading "Scientists are working on a therapeutically promising substance" also affects our lives, although it is not as widely publicized and much less spectacular. What is the connection between these scenarios, pharmacists and… mathematicians?

Let us have a closer look at scenario 1. "Success! A new drug has been launched! Over ten years of research, with costs sometimes exceeding a billion dollars but it was worth it. There is hope for effective, safe therapy for a huge number of patients and the manufacturer is starting to make profits." However, after a few years of the new drug's presence on the market, deaths of patients are noted, caused by an abnormal heart rhythm (arrhythmia). The drug, which helps a lot of people and has been considered safe so far, has to be withdrawn from the market immediately. Is it really necessary? Maybe it would be enough to indicate and modify the therapy of sensitive individuals who are more vulnerable to the risk of adverse effects of the drug?

Scenario 2 develops in completely different conditions. "The (apparent!) peace and quiet of a research laboratory. The new substance that we are testing is extremely promising. No drug known before treats (enter an organ and its function here) like this one. This may be a breakthrough discovery! We only need the results of one more standard laboratory test that will evaluate the potential cardiotoxicity (adverse effects for the heart) and we will start clinical trials." Unfortunately, the test result is positive; the analyzed substance shows adverse effects so we have to start all over again. Is it really necessary? Maybe the effect observed on a laboratory scale will not occur in population?

Real difficulties

Two different scenarios, yet the questions remain the same. How to evaluate whether an adverse effect (such as arrhythmia) will occur after the administration of a given drug (this is the easier task, although it is not simple at all) and first of all (here comes the real challenge): who will be more likely to suffer from such adverse effects: Mr. Smith, who is taking the given drug (e.g., analgesic) for his arthritis or Joanne who has asthma and takes the drug for her migraines?

Early evaluation of the potential threat of arrhythmia is very difficult. First, the available test methods using cell lines or laboratory animals have some serious limitations making it impossible to evaluate the influence of the analyzed substance on the complex processes regulating the heart activity in humans in a reliable way. Second, instances of abnormal heart rhythm and death caused by the use of a drug are very rare and such an effect may not manifest itself in the relatively small, selected group of patients participating in clinical trials before the drug is marketed. Additionally, the effect of the drug and the connected risk of occurrence of arrhythmia depend not only on the administered dose, but also on many other factors, including sex, age, the presence of mutations of certain genes, the level of hormones, illnesses from which the patient suffers, concurrent use of other drugs, heart rhythm and even the time of day. During research carried out in compliance with standard protocols it is impossible to evaluate all possible cases. However, what is impossible in the real world can be achieved in virtual reality.

Virtual twin

The system developed at the Unit of Pharmacoepidemiology and Pharmacoeconomics of the Faculty of Pharmacy (Jagiellonian University Medical College) enables us to create a fragment of the virtual world. The computer system for the evaluation of risk of cardiotoxic effects of drugs allows us to simulate clinical trials, i.e., to conduct experiments on virtual patients. These tests can be carried out already at a very early stage of the drug's development, even before the substance is synthesized by chemists, when the new molecule still exists only in the computer's memory. The application of dedicated computer algorithms allows us to predict whether and how the molecule designed will affect the heart's activity in various individuals, utilizing information derived exclusively from its chemical structure. Conducting such tests accelerates the process of drug design, shortening the period of waiting for new therapies. It enables us to minimize the risk of arrhythmia in healthy volunteers and in ill patients participating in clinical trials, and to limit the financial risk connected with the failure of the works on the new therapeutic substance. What is important is that this refers also (or even first of all) to drugs in which we do not expect adverse effects for the heart as they are intended to act on completely different organs (e.g., analgesics, anti-atherosclerotics, antibiotics, etc.) yet they can still adversely influence the functioning of this organ.

The group of virtual, simulated patients may also include people who would not qualify for actual clinical trials. These are people for whom, in real conditions, the risk connected with using the drug is assessed based on data collected only after the drug has been marketed. Obtaining such information is enabled by a virtual population generator, which "creates" patients in the computer's memory by composing them from elements, making sure to take into account the fact that every person is different, i.e., to reflect the various characteristics of patients and related possible differences in response to the drug. The thickness of the heart wall, size of heart cells, heart rhythm, concentration of the most important ions in plasma, but also genetic constitution – these are some of the elements that make us different and influence the effects of drugs.

As a result, it is also possible to use the created system in a different way, which will probably enable us to avoid or predict the negative consequences listed in the first scenario above. Preparing a "virtual twin" in computer memory, based on the profile of a real patient, will enable the physician to check how administering the drug will affect heart function and evaluate whether the patient may safely use the drug or not, even before including the drug into therapy. The system will evaluate how taking several drugs (so-called polytherapy) or diet will affect the therapeutic result.

"Do not be surprised if, in a few years, during a visit to the clinic, you will be asked to switch on the ‘virtual twin' you brought on a memory stick in the form of a key chain so that the doctor can check whether the drug is safe for you before prescribing it," foresees Sebastian Polak, PhD, coordinator of the work on the development of the system simulating the influence of drugs on heart function.

See also: www.tox-portal.net


Research team: Sebastian Polak, PhD; Barbara Wiśniowska, PhD; Aleksander Mendyk, PhD; Miłosz Polak, MSc, Eng.; Anna Glinka, MSc – all from the Jagiellonian University and Kamil Fijorek, MSc – from Kraków University of Economics