The 2nd Annual Think Tank was held in the auspicious surroundings of the Alzheimer’s Saal in Munich from 25th-26th November 2013. Please keep scrolling to read the full report:
Once more Munich did its best to tempt us from our purpose, this time with snow and the “Christkindlmarkt” (Christmas market), complete with “Gluhwein”, in Marienplatz. However, we were not to be deterred and 27 enthusiastic discussants met for the 2nd CINP ThinkTank, devoted to “Biomarkers in Psychiatry: Fact or Fantasy?” Once again, thanks to the efforts of Hans-Jürgen Möller, the generosity of Peter Falkai and the organisational skills of CINP head office, the meeting took place in Alzheimer’s Saal, in the Psychiatric Department at Ludwig-Maximilians University.
This second ThinkTank was opened by Tom Insel, who set the tone for the meeting. Having run over the definition of biomarkers, he provided some examples of their use in other fields of medicine, and posed the question “why has it taken us so long to develop clinical tests”? Drawing on a paper he co-wrote with Shitij Kapur, he stimulated extensive discussion about the potential significance of biomarkers, for the better and also the worse. Tom cautioned that biomarkers have a darker side, for example; is there any use in predicting the future onset of a disorder when no reliable therapeutic intervention is available?
Despite having been co-opted for Toms’ presentation, Shitij Kapur switched smoothly into a brief outline of NEWMEDS and how a private/public consortium, funded by the European Union, of academics and industrials can achieve advances in knowledge, which neither group could do on their own. Shitij provided an overview of some of the projects that have been undertaken (CNV effect in the general “healthy” population) and some of the spinoffs that are now freely available (touchscreen tests for mice).
Next Brian Dean shared some of his experiences with biomarker discovery, both from his time in the endocrine lab where they successfully developed and marketed a diagnostic test for thyroid-stimulating hormone receptor antibodies and more recently in the field of psychiatric research. One of the points Brian made was that many of the issues we’re facing have been successfully overcome in other disciplines and, despite the specific challenges of CNS disorders, maybe we could learn from their approaches rather than re-inventing the wheel.
In the final discussion of the first day, Sabine Bahn presented what she described as a case report for what we’d been discussing – the difficulties of discovering biomarkers and developing them into a diagnostic test. She described the identification, validation and development of a battery of circulating biomarkers with potential diagnostic specificity for schizophrenia. Although the test was marketed, a Company takeover resulted in it being withdrawn because it was perceived as being too costly to be an attractive diagnostic aid, highlighting the importance of market forces. Disease biomarkers as a rule need to be cost effective and widely available.
Daniel Javitt started the ball rolling on the second day, with a rounded presentation containing a mix of pre-clinical, clinical, neurochemical and neurophysiological data related to the use of neuroimaging as biomarkers, focussing specifically on glutamatergic transmission. This included magnetic resonance spectroscopy (MRS) to quantify markers of glutamate metabolism in regions like the hippocampus, measures of blood volume and neurophysiological markers of sensory processing like mismatch negativity. Collectively, the tests he described could suggest routes towards early prediction of risk for psychosis and possibly for novel treatment strategies.
The baton was ably taken up by Alessandro Bertolino, who addressed the relationships between genetic variation, neural circuitry and behaviour. In doing so he presented a series of studies, looking at different nodes of the signalling networks, showing associations between genetic sequence, gene expression levels, performance on memory tasks and sometimes, but not always, diagnosis.
As the day drew to a close we heard from Chris Turck who discussed the local (Munich) approach to identifying markers that predict response to antidepressants. In the process, they developed a number of different mouse models, including breeding programs based on specific behavioural responses, and the identification of responders/non-responders to the effect of antidepressants in the forced swim test.
The final, closely linked, discussion of the ThinkTank was led by Hans-Jürgen Möller who shared work on MRI based prediction of disorders. This talk focussed on the potential use of brain volume changes for identifying at risk individuals who will go on to develop psychosis. He also described different brain maturation patterns that distinguish major depressive disorder and schizophrenia. In additions to these findings, characteristic patterns of cognitive disturbances help identify at risk individuals who develop psychosis.
The meeting was summarised by Elizabeth Scarr and Mark J. Millan, who briefly touched on the key points of each presentation. Several of the major themes that prevailed during the two days of discussion were:
· The potential for different uses of biomarkers within psychiatry – diagnosis, prognosis (risk), response to treatment, avoidance of adverse event.
· The necessity of standardising measurements and tests across groups/centres.
· The need for partnerships, possibly academic/industry/government, to successfully identify and validate markers so that reliable clinical tests can be developed.
· That there are a number of potential markers under investigation but they are not yet at the stage where they can be rolled out on a large scale.
Thanks to all who attended the 1st Think Tank meeting in Munich from 13-15 September 2012. The meeting followed a more holistic approach to discussion from previous CINP meetings to allow more input from all attendees throughout the meeting. This proved to be a great success. To read the full report, please visit our 2012 Think Tank page.