Cars & Clinical trials: driving simulators Vs. on-the-road-driving-tests

For many of us, driving a car represents freedom and independence and economical power. What has this to do with CHDR, you might ask? Well, many pharmacological agents that affect the central nervous system are of concern when it comes to driving safety. Those medications (and also drugs of abuse) can induce fatigue, impair vision and reduce vigilance leading to periods of inattention. This poses a risk for the operation of dangerous equipment like a car. Although those risks are often already known at a drug-group level, for clinical pharmacologists like us, it is important to estimate the dose response relationship for an individual investigative drug early in its development. Additionally there may be certain groups, like the elderly or the young drivers, which are much more at risk when impaired by medicines, alcohol or drugs.

The effect of investigative drugs on driving performances is often assessed by using a standardized on-the-road-driving-test, which was developed in the 1980s and has been applied extensively in over 50 studies since. In short, subjects are instructed to drive a car over a 100 km highway. The car is equipped with two camera’s that constantly monitor the position of the car within the traffic lane. The outcome is mainly measured by the alterations in the standard deviation of lateral position (SDLP), which has proven to be a suitable parameter for vehicle control and traffic safety.

However, the on-the-road-driving-test is inflexible, labor intensive and therefore costly, it may take weeks to complete a full study while testing a single dose only, making the test less suitable for the assessment of dose response effects and certainly not suitable for a quick adaption in the study design. Nevertheless, the on-the-road-driving-test is often required for registration purposes, but a negative study outcome may cause a significant restriction in its indication as specified in the SMPC. Therefore, an on-the-road-driving-test may not be the single most suitable tool for drug-induced driving risks.

Luckily, the SDLP can also be measured in an in- house driving simulator in a much more practical and reproducible manner. Other parameters that are of importance in driving a car safely -, i.e. alertness, memory, coordination amongst others – can be added effortlessly. Also, driving simulation makes it possible to study the temporal relationship between medication and driving performance and work out the pharmacokinetic dynamic relationships. Of course all tests are surrogate endpoints for the occurrence of accidents but this ‘real’ endpoint can never be assessed. The current generation of driving simulators have, however, been validated positively against real driving behavior.

At CHDR we are currently studying the sensitivity of a driving simulator test battery to the effects of ethanol and a benzodiazepine, trying to compare these results with those of the NeuroCart™, a validated and established test battery developed by CHDR that quantifies a large range of drug-sensitive CNS-functions that are also relevant for every-day performance. Results from this study will allow us to determine whether a driving simulator is suitable to assess impairment in performance due to investigative drugs in a laboratory setting. This will set the standard for future studies in the important area of drug effects on real life performance.

Rob Zuiker

2 thoughts on “Cars & Clinical trials: driving simulators Vs. on-the-road-driving-tests

  1. This is basically a biomarker-surrogate emdpoint-clinical endpoint problem. A perfromance test like the neurocart is a biomarker for brain performance, a on road driving test or a simulator is perhaps a surrogate for the real clinical endpoimt, which is of course the number of accidents. Contrary to other clinical endpoints this can never be directly validated so we have to rely on intermediate validation, for instance as you are doing by testing alcohol as a gold standard, assuming that if the test picks up an alcohol effect it is also a predictor for road accidents.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s