PET studies of the serotonin transporter in the human brain

Abstract

The serotonin (5-HT) system attracts considerable attention in research on the pathophysiology and drug treatment of neuropsychiatric disorders. The brain imaging technique Positron Emission Tomography (PET) allows for examination of neurotransmission systems in the human brain in vivo. Though suitable radioligands are available for the serotonin 5-HT1A and 5HT2A receptor subtypes, imaging studies of the 5HT transporter (5-HTT), a key target for drug treatment of mood and anxiety disorders, has been limited due to lack of optimal radioligands.

In autoradiographic studies on rat and post mortem human brain slices the radioligand 3H-labelled MADAM has been shown to bind specifically in regions known to have high 5-HTT density. The first aim of the present thesis was therefore to evaluate 11C-labelled MADAM for in vivo quantification of 5-HTT in the human brain using PET. The second aim was to apply this method in control subjects and to address questions relevant for the pathophysiology and drug treatment of major depressive disorder (MDD).

In the first study, a total of four cynomolgus monkeys were examined. The selectivity and reversibility of [11C]MADAM binding was examined. Pretreatment with citalopram resulted in decreased [11C]MADAM uptake in 5-HTT rich regions. Pretreatment with GBR 12909 and maprotiline did not affect [11C]MADAM uptake, thus confirming that [11C]MADAM binds selectively to the 5-HTT in the primate brain.

In the second study [11C]MADAM binding could be described by standard compartment models using an arterial input function. The cerebellum was evaluated as reference region in simplified quantitative approaches showing that the rank order of regional binding potential (BP) values was in good agreement with the rank order reported in binding studies on human brain tissue post mortem. The use of simplified quantitative approaches in clinical studies was thus supported.

In the third study the reproducibility of [11C]MADAM binding was studied in a test-retest design tailored to be relevant for future applied studies. Analysis of test-retest data indicate that the reliability of [11C]MADAM binding measurements is good to excellent for averaged regions. The study supports clinical studies also on small regions such as the raphe nuclei, although pooling of data may be required for bilateral regions to improve accuracy.

The fourth study examined expression levels of 5-HTT and 5-HT1A receptors in control subjects, i.e. two proteins suggested in mode of action of anti-depressive drugs. A trend towards correlation between [11C]MADAM and [11C]WAY 100635 binding was found in the raphe but not in hippocampus or neocortex. Also, the ratio in the raphe nuclei showed a wide range. This finding suggests that expression of these two proteins may be co-regulated in the raphe nuclei. The wide range for binding ratio represents a possible explanation for the clinical variability in response to antidepressant drug treatment.

In the fifth study the use of [11C]MADAM to determine 5-HTT occupancy was examined by comparing 10 mg escitalopram, and 20 mg of the racemate citalopram in a two-way cross over, double blind design. Citalopram gave higher occupancy than escitalopram - although the subjects were given equimolar amounts of S-citalopram, the hypothesised active compound in both citalopram and escitalopram. The hypothesis that the lower response rate of citalopram may in part be dependent on R-citalopram interacting with 5HTT could thus find indirect support.

In summary, [11C]MADAM was shown to be a suitable radioligand for quantitative studies of 5-HTT in the living human brain, and can readily be applied for studies on the pathophysiology and pharmacology of neuropsychiatric disorders.