About CANMAT           Help & Resources           Clinical Research          CME         
 
cme
interviews
Positron Emission Tomography: Use in Diagnosing Mood Disorders
Dr. Shitij Kapur


Dr Shitij Kapur
Deputy Head for Research, Schizophrenia Division
Research Scientist, Positron Emission Technology (PET) Centre
Centre for Addiction and Mental Health, Clarke Division

 
q How new is the use of PET imaging in the evaluation of mood disorders and perhaps psychiatric illnesses in general?
a I'd say about 10 years old. The technology itself was developed in a few labs around the world who were interested in the technology and then it went around to other labs who were interested in illnesses. So in the last 10 years, the first studies in each of the psychiatric illnesses have been forthcoming.
   
q What's the principle of using PET imaging in the exploration of mood disorders? What is PET showing?
a Insofar that mood disorders have anything to do with the brain, PET technology is very relevant. It is very relevant because PET can tell you different aspects of brain function. You can use PET to find out the distribution of blood flow in different regions of the brain. You can see how different illnesses and medications affect this blood flow. You can use PET to see glucose metabolism in different regions in the brain, and you can again see how different illnesses and drugs affect glucose metabolism. And finally, you can look at specific chemicals - transporters and receptors - so, for example, if you are interested in the serotonin 2 receptor - you can actually see and measure the number of these receptors using PET. As well, you can see if medications work on those receptors and how their action on receptors correlates with their clinical effects.
   
q By identifying these physiological abnormalities in a patient with say, for example, bipolar disorder, does that give you a new understanding of how that illness develops and why people become ill in the way they do with a certain disorder?

a This research is very new but I would say the research as a field has made many important contributions. The first is a broad philosophical contribution. Today, people might not think it was much of a contribution because it has become so obvious. But not so long ago, people had a very difficult time conceding that illnesses like mood disorders were actually brain illnesses. So through very persuasive pictures, PET imaging has convinced everyone, even the strictest of the skeptics, that indeed, mood disorders, broadly defined, are brain illnesses. This is not to say that psychological therapies may not work - they may well work and even better than medications - but they are brain illnesses. The other valuable insight that we are beginning to receive is that even after you treat the depressive episode, for example, there is not a complete restitution of brain chemistry or brain function to the normal state. So while the patient may come into complete remission, there is perhaps a lingering effect in the brains of these patients which suggests that this illness, while it may reverse itself on the surface, some underlying neurobiological vulnerability may continue. And that might be what's responsible for the relapse of the illness.
   
q Do you think it may be possible to develop better drugs that are capable of reversing the biochemical abnormalities in the brain to therefore promote longer remissions or prevent relapses?

a That would certainly be the hope and at this point in time, I think there is every reason to keep that hope alive. One must admit that we do not know precisely what it is that triggers or brings on depression. But as a concept, there is no doubt that if we understood the biology of the onset of depression better, we would be able to make better drugs to deal with it.
   
q What specifically are you using your PET studies to explore here at the Clarke Institute?

a The work at the Clarke in the area of depression is being done by a group led by Dr Sid Kennedy and also Dr Jeff Meyer, and what we are looking at here is the serotonin system, and in particular the serotonin receptors. The reason why this is of interest is because various lines of evidence show that there are serotonin abnormalities in depression. And if you look at all the antidepressant treatments, they in one way or another work in the serotonin system. What has been unclear is where along the serotonin system, beginning from cell to the synapse to the postsynaptic and then to the secondary and tertiary messengers, where along this chain is there a problem. And one very valuable target is at the level of the synapse and it's called the serotonin receptor, and in particular the serotonin 2 receptor. We have been focusing our efforts to look at what is the status of the serotonin 2 receptor in patients with depression, and how do the medications affect these receptors in patients as well as in normal controls - and that has been the focus of our work in this area.
   
q And what have your findings been to date?
a At this point in time, our findings in patients are actually in the process of completion so I don't think I can comment on them. But to give you one example: one of the theories used to be that all of the antidepressants - even though they look very different in their chemical structures - have one common property and that common property is that all of them down-regulate the serotonin receptor. There is lots of animal evidence to show that regardless of the actual structure of the antidepressant, all of them do this one common thing. Now that is very important because it tells you that this is a kind of bottleneck and it is something that is very important in neurobiology. So in normal subjects we have been pursuing that finding. And our results show that actually, the SSRIs (the Prozac class of antidepressants), at least in normal controls over a limited period of time, do not show this down-regulation, which is a bit of a problem. I wish these results were positive but nonetheless, they are important, because they tell us that theory, derived largely from animal findings, is not easily confirmed in humans. So we need to look elsewhere, and before of course we have the final word on this, we have to repeat this same study in patients because it is likely that while the drug may not do anything dramatic to the serotonin receptors in normal subjects, it might do something to the serotonin receptors of depressed patients. And that study is currently underway and hopefully we will have the results from that study within the next few months.
   
q So are you saying that the 5-HT2 receptor seems to be the key receptor in depression?

a It was one of the hypotheses. One of the problems we have in many psychiatric illnesses, but in depression in particular, is that the drugs on the surface look very different. On the surface, I mean some of them block norepinephrine reuptake, some of them block serotonin reuptake, and these are distinct mechanisms in the brain. Nonetheless, they seem to work for all patients and a patient who responds to a serotonin drug may also respond to a norepinephrine drug, not 100%, but there is a large overlap. So one has to think that these drugs are doing something at a deeper level - maybe there is some common funneling pathway - and that is very important to know because if we knew what that was, we could design drugs that could go after exactly that pathway rather than doing it in an indirect fashion. And therefore the serotonin 5-HT2 is one such attractive area to look at.
   
q I guess none of this work would have been previously possible before PET. Or have we ever been able to image the brain in any useful way before PET?
a In a biochemical sense, absolutely not. A lot of hypotheses about what these drugs do to the brain are actually hypotheses about what these drugs do to the normal rat brain. So not only are there species problems - lower species versus humans - there is also a sort of trait problem, in that those are in any sense "normal" rats, they are not depressed rats, because the concept of depression in a rat would be very hard and the current models of depression look nothing like the models in humans. So one of the first enterprises is to look in humans and see how much of our hypotheses generated from animals actually hold out in humans. And from the first wave of human studies will then come the idea of the next wave of animal studies.
   
q Are you using PET in various anxiety disorders?
a Yes. We are using PET in anxiety disorders and in the area of schizophrenia. In particular, one of the areas of push in anxiety disorders is work being done and led by Dr (Jacques) Bradwejn, who used to be at the Clarke but who is currently at the University of Ottawa. And that has to do with the study of CCK, which is a peptide that induces anxiety attacks in patients with anxiety disorders and panic disorders, who have even more of an exaggerated response than normal subjects, so they have a sensitivity to this. Dr Bradwejn has been doing some work to understand which particular brain regions in these patients respond abnormally, to understand how one could construct a model of the neurobiology of anxiety disorders.
   
q Taking your own work in depression, how would the physiological abnormalities or changes that you see in patients with depression perhaps differ from those say, for example, with bipolar disorder? Or are there differences?

a There is not that much work done on bipolar disorder at this point in time to permit a comparison, but there is a lot of work done in schizophrenia. Even though we tend to think of depression and schizophrenia as very different illnesses, at the brain level they may be perhaps more similar than different and this may have important implications because our bible of diagnosis puts them in different chapters, and never the twain shall cross. But if anyone sees patients, they are always struck by the tremendous degree of overlap. In fact, the beauty of diagnostic systems is not that they apply to most patients, it's that in clinical practice you find that they don't, and there is tremendous overlap and you find that you are force-fitting patients into one or the other [diagnosis]. And that is because our diagnostic systems are categorical - you either have this diagnosis or not - you can't have degrees of the problem, whereas the brain in all likelihood is likely to be dimensional. To give you an example, one set of symptoms that patients with depression often would show would be lethargy, lack of motivation, lack of drive, lack of desire and social interaction, and patients with schizophrenia also show these symptoms. When patients with schizophrenia show them, we tend to call them "negative" symptoms, we don't tend to call that depression in schizophrenia. Early brain imaging results show that whether you get this complex of symptoms in depression or whether you get them in schizophrenia, they actually reflect a low functioning of the left prefrontal cortex of your brain. This is not to say that depression and schizophrenia are all the same but the many overlaps in symptoms between the two illnesses may derive from the fact that, at the level of the brain, the same brain regions are involved at least in some part of both illnesses.
   
q Do you use PET then to look at the effect drugs have on specific regions identified as being abnormal?
a A lot of work is being done in that direction. It's still early, there aren't too many good studies which have been able to demonstrate that but yes, that is precisely where the field is headed to see what is the effect of these drugs acutely, and that is because they don't really work acutely, and also what is the effect at 4 to 6 weeks, because what we might want to distinguish are the acute effects from the chronic effects because the chronic effects are the therapeutic ones.
   
q As you were suggesting earlier, if you can see the action of an antidepressant is less than complete, or not as targeted to that specific abnormality, then it would be feasible to fashion a molecule which perhaps has more potency in that particular area?

a Yes. That is the grand design. Of course one has to move towards it in small, small steps, which is being done.
   
q Do you think that this information that mental illnesses are clearly biochemical illnesses - you can see it on functional study that there is something physically wrong in the brain - helps patients better accept their diagnosis of a mood disorder or perhaps a more difficult psychiatric illness?

a Absolutely. As you may notice, I have never said up to now that we have a diagnostic test. So in some sense while it has greatly helped science, these imaging techniques have not reached the patient by the bedside as yet - I think they might - but they are not there yet. However, I've been impressed by how many patients and family groups are so relieved and so thankful that we are understanding and able to allocate some of the causal blame of the illness to the brain. It's a function of Western society in the late 20th century, that if things are in some sense identified with some physical origin, that is seen as a more positive thing as things deemed to be entirely psychological - which is perhaps why patients have told me how valuable they find this broad concept to help understand what they are going through, to help explain to others what they are going through. I think it has led to the removal of some of the stigma, though there is still more than there ever should be. And finally, it does help fight for parity with respect to various insurance aspects of this illness and various other coverage aspects of this illness.
   
q What do you find most exciting about working with PET imaging in this particular field of mood disorders?

a The mind-brain correlation. Because what you find at the level of the brain (like PET alone) after some time is just a series of pictures, not very different from postmortem slices. What makes it really revolutionary in my mind is that you can talk to the live person and find out what they are feeling and find out whether they are going to get better or not, which you can't really do with postmortem data. So it is that aspect of PET - the ability to correlate living human chemistry or living human biology to experiences - that I find is perhaps what distinguishes PET from other technologies and is also what I find most grabbing and most interesting.
   
q It sounds as if PET is going to make a significant contribution in being able to identify which areas of the brain are responsible for which sets of emotions.

a Yes and the way to go about it is to study normal mood first. There are studies underway and our centre is involved in some of them under Dr Michael Gemar and Dr Zindel Segal giving mood challenges. Although the difficulty that field seems to have run into is that there isn't one good way to induce a sad mood, and it seems that every different way one induces a sad mood leaves a slightly different imprint. So in time what I think will happen is that we will look at the overlapping areas after a lot of studies have been done, and we will perhaps be able to find what are the core areas responsible for mood induction. But that is the basic work that is being done and hopefully will lead to a better understanding of depression.
   
   
   


Over one million Canadians suffer from some form of depressive illness.