Set-up of the resting test as a valid model for the discovery of novel CNS drugs.

In the past, several drugs were developed in order to have effect on Central Nervous System (CNS), but only when they were in an advanced development phase it was showed that they could not have any effect on CNS, because they did not pass through the Blood Brain Barrier (BBB). This condition resulted in a waste of time, money and resources for researchers. With this project we decided to set-up a model which allows to test if a drug can pass through the BBB or not; its application should be feasible by using a non invasive radiological technique and it should be used even during the early development phase of a drug which is being studied to act on CNS. The model chosen to be developed was the ―resting state test‖. Resting state corresponds to the condition when the human brain is wakeful and conscious, but relaxed and in absence of any specific stimulus or task. Several studies have demonstrated that the brain, in its resting state, presents some degree of spontaneous activity measured as low-frequency fluctuations (<0.1 Hz) in the BOLD (Blood oxygen level dependent signal) fMRI signal of magnitude comparable to task-induced signals. Temporal correlations between fMRI signal fluctuations in different brain regions in its resting state have been interpreted in terms of functional connectivity (fc). These connectivity patterns were represented in a natural way as networks with brain regions. To date, at least five distinct resting states networks were found between different brain areas. Among those networks one has been identified as Default Mode Network (DMN). It can be considered unique in its response to cognitive tasks due to the fact that is characterized through high activity during rest and deactivation during tasks; it is also easily identifiable by BOLD fMRI technique and its reproducibility is reliable. Brain regions identified as parts of the resting state DMN include the posterior cingulate cortex, precuneus, medial prefrontal cortex, anterior cingulate cortex, parahippocampus and inferior parietal cortex. A series of studies published in recent years suggest that in subjects with neurological or psychiatric diseases, the DMN works differently than it does in healthy control subjects and the pathological disturbances in intrinsic activity have also been defined as related to the severity of disease. Significant effects have been also observed with antipsychotic drugs on fc in these brain regions. We hypothesized that these effects may be reflected in an increased fc and that they correlate with levels of receptor occupancy of the drug in the brain. Moreover, some fMRI studies have shown that the spontaneous fluctuations of resting state are modified during simple movements, like movement of left and/or right hands and it is possible identify a brain activity in motor areas, corresponding to the specific movements. According to these considerations it was decided to set-up the resting state model by performing a clinical trial in healthy volunteers. The primary aim of the project was to set-up the resting state test with fMRI and to test its reproducibility in healthy subjects; the secondary aim was to explore the effects of a single oral dose of alprazolam 0,75 mg on brain activity in resting state and in motor task steps. Alprazolam was chosen as drug due to the fact that, as benzodiazepine (BZD), it affects the CNS after being passed through the BBB with an anxiolitic effect. It was established to enroll a minimum of 10 healthy male and female subjects with age 18 to 50 years. Main exclusion criteria were diagnosis of mental illness or significant cardiac disease or cardiac conduction disorder and history or presence of neurological or psychiatric conditions. The radiological technique chosen was BOLD fMRI. Compared to other radiological techniques (i.e. PET) this method is safe and not invasive: it can be applied to young and old subjects and it is based on an endogen contrast. This investigation was undertaken as a double blind, single dose, crossover, randomized study. In two different occasions, each subject received an oral dose of alprazolam or placebo and underwent two fMRI scan sessions: before and 2hrs ± 30 min post dosing (approximately the t-max of alprazolam). During the first part of the fMRI exam each subject was asked to stay wakeful, motionless and with eyes closed. In the second part each subject performed the motor task which consisted of two steps: during the first one he was asked to stay wakeful, with eyes closed and to open and close the hands alternatively for 10 times; during the second one, he was asked to stay wakeful, motionless and with eyes closed for the same period of the first step. The task was repeated for 5 times. The clinical phase of the study was carried out as scheduled. The ―resting state‖ parts and the ―motor task‖ steps were performed by applying the model defined and detailed into the protocol. Alprazolam and study procedures were well tolerated. A total number of 11 subjects were enrolled into the study, because one subject did not complete both study sessions. Only data sets obtained by subjects who completed both sessions were analyzed. The functional data were analyzed by using software FSL. For resting state analysis the first step was the use of Multi Session Temporal Concatenation in MELODIC FSL component. At this level, it was also applied an F-test. The next step was the dual regression technique. The last step was to collect the different sets of spatial maps across subjects and to analyze them. The results were spatial maps defining the between-subject group-consistency and/or between-subject group-differences. For motor task step analysis, fMRI data processing was carried out using FEAT (FMRI Expert Analysis Tool) FSL application. The reproducibility of DMN across all subjects was demonstrated by the contrast of two baseline conditions: the F-Test (thresholded per p < 0,05) showed a statistically non-significant difference. The analysis performed through dual-regression method did not point out a statistically significant difference as well. A very similar condition was showed between two baseline and pre-post placebo conditions. The main result obtained by the data analysis was a diffuse and increased fc in post alprazolam maps in the posterior component of DMN and also in brain areas out of DMN. This increase of connectivity can be interpreted as a consequence of the administration of a drug with a sedative effect which inhibits tasks, increases activity at rest and makes DMN stronger. The presence of changes in brain areas out of DMN might be due to the binding of alprazolam to BZD receptors in CNS. The results obtained from motor task steps showed an evident reduction of activity in primary motor areas, bilateral primary somato-sensitive areas and bilateral cerebellar areas. These results are in agreement with characteristics of motor task paradigm as defined into the study protocol. In summary both primary and secondary aims of the study have been achieved. Resting state test was set up and its reproducibility in healthy volunteers was demonstrated. It was demonstrated that the effect of a single dose of alprazolam on brain activity in resting state consists in an increased and diffuse connectivity in DMN and in other cerebral areas and alprazolam effects in motor task step consist in a reduction of activity in motor areas activated by motor task. In conclusion the model can be considered set-up and applicable in humans during future clinical trials. The maps generated during motor task step increase knowledge of human brain topography. In the future we can schedule to carry out a similar clinical trial in patients affected by psychiatric and neurological diseases and to compare results obtained in patients against the ones obtained in healthy volunteers.