New drugs: do the calculations first, before testing on patients

Eight out of ten experimental drugs do not make it past the test phase on patients. Piet Hein van der Graaf, Professor of Systems Pharmacology, would like to prevent this waste of time and money by using new measurement and calculation methods. Inaugural lecture on 27 May.

A new medicine costs more than 1.5 billion euros to develop

Piet Hein van der Graaf, Professor of Systems Pharmacology and Director of the Leiden Academic Center for Drug Research

Piet Hein van der Graaf, Professor of Systems Pharmacology and Director of the Leiden Academic Center for Drug Research

In recent years, the productivity of research in the pharmaceutical industry has been stagnating. Fewer and fewer new drugs make it to the market and the costs keep climbing. ‘From start to finish the development track of a new drug costs on average more than 1.5 billion euros,’ says Van der Graaf. ‘This figure is so incredibly high because out of every ten projects, eight are doomed to fail. And that generally happens when the drug is tested on real patients, after ten years or more of research. It is not until this point that it becomes clear that the drug is not effective, or not effective enough.’ He wants to develop new measurement and calculation methods which will make it possible to see the effects of a drug at an earlier stage in the testing process.

First calculate the effects

Van der Graaf recently moved to Leiden University from drug manufacturer Pfizer. In addition to his position as professor, Van der Graaf is also Director of the Leiden Academic Center for Drug Research (LACDR). ‘There are very few academic research centres in the world with such a broad expertise as ours: we make molecules, we develop new forms for administering drugs and we test active substances. In addition, through our collaboration with organisations such as the Centre for Human Drug Research, we can also test prototype drugs on humans.’


As a Professor of Systems Pharmacology, Van der Graaf combines biophysical, chemical, pharmaceutical and mathematical approaches. This kind of multidisciplinary approach is necessary for developing advanced biomarkers: indicators that show whether substances reach their destination and do the work they are supposed to do. ‘In concrete terms, this means that we are far more likely to first calculate the effects in computer models in order to increase the chances of success in studies on patients.’

Crucial conditions

Van der Graaf recently studied approximately one hundred drug projects to check whether in the earlier phases of testing three crucial conditions had been measured: whether the dosage was appropriate, whether the active substance sufficiently adhered to the area where it is supposed to be and whether the active substance was able to do its work by sufficiently activating or inhibiting the relevant receptors (the proteins that a drug can bind to). These are the factors that determine the effect of a drug. However, Van der Graaf found that in a majority of the research projects these effects had not been measured before the drugs were tested on patients.

Shock wave

How is this possible? ‘Many pharmaceutical researchers are used to developing drugs first rather than first making the relevant calculations using mathematical models. And then there is the fact that there are not always good biomarkers available. For example, it isn't easy to determine whether a drug works in the brain. This calls for new measurement methods and concepts to be developed.’ The results of his study caused a ‘shock wave’ in the pharmaceutical industry, says Van der Graaf. Because they fail to meet these three conditions, many projects fail and the researchers are left retrospectively trying to guess why: is the drug not working because of the components chosen? Is the dosage wrong? Or has the wrong protein been chosen for the manipulation?

More collaboration

Van der Graaf is not only advocating more collaboration between disciplines, he also argues that universities and companies have to collaborate more closely and share their respective knowledge, in particular in the early stages of drug research. ‘Openly available knowledge speeds up the process because more parties can simultaneously work on developing drugs. The goal is not to keep our knowledge to ourselves, but to cure patients. Luckily new ecosystems for drug research are becoming available, such as the Bioscience Park in Leiden, where companies and the University can develop new ways of working together.’

Developing prototype drugs

The LACDR has a lot of expertise at its disposal. Could the Centre not bring the drugs to market themselves? ‘No. That is not the role of an academic centre and it is something best left to commercial companies. What we can do is develop concepts and prototypes. These are drugs that are not designed for a specific purpose, but whose effect on the body we know precisely, for instance how they inhibit an enzyme.’

And this can have important consequences. As an example, Van der Graaf mentions the sildenafil molecule which was first developed by Pfizer to treat heart and vascular diseases such as high blood pressure. Now it is being used for erection disorders and lung problems. These effects were discovered more or less by accident during the clinical studies. ‘We need to build this kind of adaptability in from the start.’

(23 May 2013)

Inaugural lecture Prof. P.H. Van der Graaf - Pharmaceutical Sciences, Systems Pharmacology
Date: Monday 27 May 2013 
Faculty: Science
Location: Academy Building, Rapenburg 73, 2311 GJ Leiden

See also

Studying in Leiden

Bio-Pharmaceutiscal Sciences

Life Science & Technology


Bio-Pharmaceutical Sciences


Last Modified: 20-06-2013