- Email: [email protected]
S.11.03 Potential rapid onset: Mechanisms of action
SI00
S.11 Can we have better antidepressants?
now pharmacological treatment strategies which produce an early onset of action in major depression. These are high doses of venlafaxine and the combination of a serotonin (5-HT) reuptake blocker with the 5-HT1A antagonist pindolol. In treatment-resistant depression, there are several agents which can be used to potentiate the therapeutic effect of the initial antidepressant drug. These include lithium, the 5-HTIA agonist buspirone and pindolol. It is noteworthy that when these strategies are effective, they generally exert their beneficial effect on depression within two weeks. Some approaches appear to have both a more rapid onset of action and a greater efficacy, such as pindolol addition and venlafaxine, whereas lithium addition does not induce a more rapid onset of action despite being effective in treatment-resistant depression. The question whether a rapid onset is related to greater efficacy of antidepressant treatment can be examined on the basis of the putative mechanisms of action of these treatments. For instance, lithium is thought to produce a rapid alleviation of depressive symptoms, in patients treated with but not responding to their antidepressant drug, by enhancing 5-HT release in certain brain regions. However, lithium is not expected to accelerate the antidepressant effect of tricyclics because the latter drugs require a few weeks of administration before sensitising postsynaptic 5-HT receptors. In contrast, pindolol, as a result of its antagonism of the cell body 5-HTIA autoreceptor, accelerates the antidepressant effect of 5-HT reuptake blockers by preventing the initial but transient decrease in firing activity of 5-HT neurons which these drugs induce. Consequently, by maintaining a normal firing rate of 5-HT neurons in the presence of 5-HT reuptake blockade, pindolol produces in some patients an accelerated response presumably by more rapidly increasing 5-HT transmission. Pindolol addition was also reported to alleviate depression in treatment-resistant patients treated with but not responding to a 5-HT reuptake blocker or a monoamine oxidase inhibitor. In such patients when a beneficial effect is obtained, one may presume that pindolol restores the firing activity of 5-HT neurons as a result of the 5-HTIA autoreceptor which may have failed to desensitise during the prior treatment. This hypothesis will only be verified when a reliable probe for the 5-HTIA autoreceptor will become available. In conclusion, a strategy with a rapid onset of action may also be endowed with a greater efficacy (i.e. pindolol), but one yielding to a superior outcome may not act more rapidly (i.e lithium). These fundamental data and clinical observations can serve as the basis to devise more rapid and effective treatments for depression.
~
Potentialrapid onset: Mechanismsof action
F. Artigas. Department of Neurochemistry, CSIC, 08034 Barcelona,
Spain The existence of pre- and postsynaptic adaptative changes of neural function after repeated antidepressant treatment has been extensively documented. Presynaptic inhibition of serotonergic (5-HT) function by SSRIs, MAOIs and tricyclic drugs has been shown to contribute to the delayed onset of antidepressant action. Antidepressant drugs of these families markedly elevate the extracellular concentration of 5-HT (5-HText) in the raphe nuclei of the midbrain, thus activating somatodendritic 5-HT1A autoreceptors and offsetting the increments of 5-HText obtained by blockade of reuptake in nerve terminals (see Artigas et al., 1996 for review). The negative feed-back is more marked in forebrain areas innervated by the dorsal raphe nucleus (DRN), such as frontal cortex or striatum, as compared to the hippocampus, which receives afferents from median rapbe (MRN) serotonergic neurones. Chronic, but not acute, treatment with clinically relevant doses of SSRIs, MAOIs and imipramine results in significant increments of 5-HText in rat frontal cortex, a finding that may be related to the desensitization of somatodendritic 5-HT1A receptors by antidepressant treatments. The combined administration of 5-HT uptake inhibitors and 5-HTIA receptor antagonists results in a potentiation of the effects of the former in rat brain. In agreement with the greater self-inhibition in DRN projections produced by SSRIs, the selective 5-HT1A antagonist WAY 100635 markedly potentiated (ca. six-fold) the elevation of 5-HText produced by 3 mg/kg paroxetine in dorsal striatum but it only doubled that produced in dorsal hippocampus, innervated preferentially by the MRN. The non-selective 5-HT]A antagonist (-)pindolol also potentiated the effects of SSRIs in striatum but to a lower extent
than WAY 100635. This effect may likely underlie the enhanced antidepressant action of SSRI + pindolol combination treatments observed in various open-label and controlled studies. Moreover, administration of desipramine further enhanced the cortical 5-HText elevation produced by the SSRI fluoxetine whereas it did not alter 5-HText in untreated animals. This observation, together with tile marked increments of rat cortical 5-HText observed after high imipramine doses, may be related to the unsurpassed therapeutic efficacy of imipramine (Bel and Artigas, 1996). These experimental and clinical data emphasize the importance of the presynaptic component of serotonergic transmission in the therapeutic action of antidepressants. Chronic antidepressant treatments elicit a plethora of changes in recepor density and intracellular messengers in various forebrain structures. Yet it is not known whether such postsynaptic adaptations also contribute to the therapeutic action of these agents. Thus, it is unclear whether clinical improvement is the immediate consequence of an enhanced activation of certain 5-HT postsynaptic receptors or whether long-term adaptive changes in limbic and cortical areas are also required. In the first instance, the presynaptic component would be entirely responsible for the delayed onset of action and the therapeutic failure. Thus, patients resistant to SSRI treatment might be viewed as those individuals with e.g., greater somatodendritic 5-HT release, hypersensitive 5-HT1A autoreceptors or hyperactive 5-HT transporter, all causes leading to a greater-than-average self-inhibition of 5-HT neurons during SSRI treatment. Those patients would have an additional difficulty to increase 5-HText in forebrain despite an effective blockade of the 5-HT transporter in nerve terminals due to any of these possible factors. Addition of 5-HTjA antagonists to the therapeutic regime would bring about a response by breaking this negative feedback and allowing 5-HT neurons to normalize their firing and terminal release. Although experimental evidence for the existence of such neurobiological differences between responder and non-responder patients is lacking, this hypothesis would be consistent with the marked improvement observed in some treatment-resistant patients after pindolol addition. Interestingly, the existence of allelic variants of of the 5-HT transporter has been described, and we have observed that patients with a high pretreatment platelet 5-HT concentration display a very poor therapeutic response to various antidepressant drugs, including clomipramine, fluvoxamine and paroxetine. This suggests that individuals with a highly efficient 5-HT transporter (at least in periphery) may be more prone to be resistant to SSRI treatment. The inability of pindolol to fully suppress the latency in the antidepressant action of SSRIs appears to suggest that postsynaptic adaptive mechanisms are also involved in the full remission of depressive symptomatology. However, several observations are in partial disagreement with this view. First, pharmacokinetic factors may contribute (i.e., slow increases in the plasmatic levels with some antidepressants). Secondly, pindolol is by no means a selective and potent 5-HTtA antagonist. In view of the data obtained in rat brain, more effective 5-HT1A antagonists would be required to better prevent self-inhibitory influences on serotonergic neurones and therefore enhance the presynaptic 5-HT function during SSRI treatments. Finally, clinical observations indicate the presence of mood changes in depressed patients whith a time course much shorter than that required for gene regulation or changes in :protein synthesis. For instance, the rapid relapse arid subsequent recovery of depressed patients treated with SSRIs when administered a tryptophan-free amino acid mixture is indicative that in certain conditions there may be a rapid switch between euthimic and depressive states (Delgado et al., 1990). Research to produce more rapid and effective antidepressant treatments should proceed along several lines. First, given the excellent therapeutic efficacy of tricyclic drugs and SNRIs, it is likely thai: the study of the 5-HT-noradrenaline interactions may provide some clues. Also, a careful study of the functional control of serotonergic neurones by the different afferent pathways to the DRN and MRN may be interesting to understand their regulatory mechanisms. Drugs activating 5-HT neurones may prove useful to either produce an antidepressant effect by themselves or to prevent the inhibitory effects produced by antidepressant treatments. Finally, a better comprehension of the postsynaptic mechanisms (receptors, circuits) involved in the therapeutic action of antidepressants would greatly help, because compounds targeted specifically at certain postsynaptic receptors would overcome the psesynaptic inhibition, responsible, at least in part, of the delayed onset of antidepressant action.
S.12
Critical issues in clinical management of obsessive compulsive disorders
References Artigas, F., Romero, L., de Montigny, C. and Blier, P. (1996) Acceleration of the effect of selected antidepressant drags in major depression by 5-HTtA antagonists. Trends Neurosci. 19, 378-383. Bel, N. and Artigas, E (1996) In vivo effects of the simultaneous blockade of serotonin and norepinephrine transporters on serotonergic function. Microdialysis studies. Journal of Pharmacology and Experimental Therapeutics, 278, 1064-1072. Delgado, P.L., Chamey, D.S., Price, L.H., Aghajanian, G.K., Landis, H. and Heninger, G.R. (1990) Serotonin function and the mechanism of antidepressant action. Reversal of antidepressant-induced remission by rapid depletion of plasma tryptophan. Arch Gen Psychiatry, 47, 411--418.
~ C a n
we identify subgroups for targeted adjunctive treatment?
Michael T. Isaac. Department of Psychological Medicine, United Medical & Dental Schools, University of London (Guy's Campus), Suite 6, Lewisham Hospital, London, UK No antidepressant currently in use exerts a significant antidepressant effect for at least two to three weeks after the patient starts taking it. Open studies suggest that, for selective serotonergic re-uptake inhibitor (SSRI) antidepressants, this latency may be reduced when the drug is taken with the 5HTIA receptor blocker pindolol. In a recent randomised, placebo controlled double blind study at Guy's and Lewisham Hospitals, eighty out-patients (mean age 36 [range 19~55], 48 female, 32 male) were recruited, each patient receiving paroxetine (20 mg o.d.) plus, randomly, either pindolol (2.5 mg t.d.s.) or placebo for six weeks. Paroxetine (open label) was offered to all, patients for a further 18 weeks. Follow-up assessment on sixty-nine patients, took place at weeks 8, 16 and 24. Taken overall, the results showed a statistically significant acceleration, observable from day 4, of the antidepressant effect of paroxetine in the group receiving pindolol. In the intention to treat population, the divergence between pindolol and placebo groups disappeared by day 28, but persisted beyond day 42 in the per protocol group. In the longer term, patients originally treated with pindolol (n = 32) showed significantly better outcome at week 24, when compared with patients taking paroxetine alone, whether they complied fully with follow-up treatment or not. Compliance with follow-up paroxetine had a significant positive effect on outcome at weeks 16 and 24 in those originally treated with paroxetine alone (n = 37). Similar findings are apparent at one year. However, it is clear that the effect is not equally shared among all patients. We found, for example, that the combination worked less well in males over the age of thirty five years, and in males and females who had either a chronic history of dysthymia/depression or who were more severely depressed. We specifically excluded substance misusers. It may be relevant that our patients were reviewed twice weekly during the first two weeks of the study, and weekly thereafter. This was intended to permit the ascertainment of an early response to medication. It is becoming clearer that the 5HT1A receptor is subject to genetic polymorphism, and such differences may be highlighted by use of a 5HT1A receptor blocker. Moreover, the minus isomer and the mixed (plus/minus) enantiomer of pindolol appear to act differently (in rats, at any rate) in their activity as partial agonists of 5HT1A receptors. We examined personality variables in 48 consecutive subjects according to a short version (TCI-125) of Cloninger et al's self-rated Temperament & Character Inventory and correlated the results with clinical responses in the trial. The results suggest that high scores in the temperament dimension of Reward Dependence and low scores in the temperament dimension of Harm Avoidance had a better outcome at 6 weeks. Patients who had received paroxetine and pindolol during the trial and who reported high Novelty Seeking and low Harm Avoidance scores had a better outcome at 6 weeks and 6 months. We suggest that temperament factors may influence outcome of antidepressant treatment. These results suggest low dopamine and low serotonin transmission. Low serotonin levels may be associated with up regulation of serotonin receptors in the mesolimbic area, which can explain the rapid acceleration effect of pindolol in these patients. Pindolol has been shown in humans to increase prolactin levels (hence influencing dopamine levels), supposedly under the influence of the serotonergic system. However, the results in
S101
the pindolol group in our study perhaps suggest that this combination can also alter brain dopamine levels via the serotonergic system. Low serotonin levels have been reported in depressed patients. In our ~tudy, a low (<20 ng.ml -I ) baseline platelet 5HT was a~sociated with a poorer response to medication at day 42, compared with a 95% fall in platelet 5HT at day 7. There was an association between high baseline platelet 5HT and moderate depression, and low baseline platelet 5HT and severe depression at day 21. A previous history of more than two treated episodes of depression, or an initially severe (MADRS >34) baseline symptom profile was associated with less acceleration and poorer overall outcome. Chronicity as such did not predict lack of acceleration, although general outcome was less good. Many of the best responders were in employment and less than forty years old. The choice of SSRI may be important. But animal studies show no differences in the physiological or neurochemical consequences of pindolol in combination with any SSRI drugs, including fluoxetine. Moreover, in comparative and systematic studies, there is little to choose between the core antidepressant effects of fluoxetine and paroxetine (or other SSRI for that matter). In addition, acceleration of antidepressant effect is notoriously difficult to evaluate. It may seem an obvious point, but one must evaluate the patient during frequently during the early stages to show an early response; it is not enough to ask the patient at, say, ten or fourteen days ff he experienced an early effect before that point. A more rapidly acting antidepressant must not, of course, show only an early response; this must be sustained. Placebo control is essential, and some protocols insist on an one-week, placebo run-in phase in an attempt to control for a placebo response. This approach has merit, but a properly controlled study should be able to allow for early placebo effects, which, in any event, are usually not sustained. Developing more predictors of the response to antidepressants is a present and future challenge to research.
S.12 Critical issues in clinical management of obsessive compulsive disorders
I
•
Predictors of response
D. Marazziti. Institute of Psychiat~ University of Pisa, Italy In the last decades, consistent advancements have been raade in the synzhesis of new psychotropic drugs whose developments has, been paralleled ;and supported by the increased knowledge in Neuroscieuces. Therefore, psychopharmacological treatments have become more specific, although their effectiveness is not always restricted to clinical diagnostic entities, t~ut to symptom clusters or psychopathological dimensions. In spite of the continuous data efflux on the mode of actions of psychotropic drugs, on their spectrum of properties and side-effects, the choice of a drug is rather empyrical and no real predictors can currently be identified. However, the careful clinical observation of the patients, coupled with the present knowledge of the neurobiology of a given disorder and with the mechanisms underlying the effects of a drug should provide us some useful information. Some clinical indexes that seem to have a negative predictive value in obsessive-compulsive disorder (OCD), but which necessitate of further validation are: comorbidity with schyzotypal personality disorder, panic disorder, tic or Tourette's syndrome, early onset and chronic course. Our group investigated the possible usefulness of a peripheral serotonergic parameter, that is, platelet serotonin (5-HT) transporter, as a predictor of response in a group of 30 drug-free OCD patients. Platelet 5-HT transporter was measured by means of the specific binding of 3H-Paroxetine (3H-Par), before (tO) and after (tl) a 2-rnonth treatment with clomipramine, fluvoxamine and fluoxetine. Symptom severity was assessed with the Y-BOCS. The results showed that at baseline the patients had a decreased number of 3H-par binding sites which increased towards normal values after the treatments which was successful in most of that cases. The number of 3H-Par binding sites were significantly and negatively correlated with symptom severity at baseline, i.e., the lower
S.11 Can we have better antidepressants?
now pharmacological treatment strategies which produce an early onset of action in major depression. These are high doses of venlafaxine and the combination of a serotonin (5-HT) reuptake blocker with the 5-HT1A antagonist pindolol. In treatment-resistant depression, there are several agents which can be used to potentiate the therapeutic effect of the initial antidepressant drug. These include lithium, the 5-HTIA agonist buspirone and pindolol. It is noteworthy that when these strategies are effective, they generally exert their beneficial effect on depression within two weeks. Some approaches appear to have both a more rapid onset of action and a greater efficacy, such as pindolol addition and venlafaxine, whereas lithium addition does not induce a more rapid onset of action despite being effective in treatment-resistant depression. The question whether a rapid onset is related to greater efficacy of antidepressant treatment can be examined on the basis of the putative mechanisms of action of these treatments. For instance, lithium is thought to produce a rapid alleviation of depressive symptoms, in patients treated with but not responding to their antidepressant drug, by enhancing 5-HT release in certain brain regions. However, lithium is not expected to accelerate the antidepressant effect of tricyclics because the latter drugs require a few weeks of administration before sensitising postsynaptic 5-HT receptors. In contrast, pindolol, as a result of its antagonism of the cell body 5-HTIA autoreceptor, accelerates the antidepressant effect of 5-HT reuptake blockers by preventing the initial but transient decrease in firing activity of 5-HT neurons which these drugs induce. Consequently, by maintaining a normal firing rate of 5-HT neurons in the presence of 5-HT reuptake blockade, pindolol produces in some patients an accelerated response presumably by more rapidly increasing 5-HT transmission. Pindolol addition was also reported to alleviate depression in treatment-resistant patients treated with but not responding to a 5-HT reuptake blocker or a monoamine oxidase inhibitor. In such patients when a beneficial effect is obtained, one may presume that pindolol restores the firing activity of 5-HT neurons as a result of the 5-HTIA autoreceptor which may have failed to desensitise during the prior treatment. This hypothesis will only be verified when a reliable probe for the 5-HTIA autoreceptor will become available. In conclusion, a strategy with a rapid onset of action may also be endowed with a greater efficacy (i.e. pindolol), but one yielding to a superior outcome may not act more rapidly (i.e lithium). These fundamental data and clinical observations can serve as the basis to devise more rapid and effective treatments for depression.
~
Potentialrapid onset: Mechanismsof action
F. Artigas. Department of Neurochemistry, CSIC, 08034 Barcelona,
Spain The existence of pre- and postsynaptic adaptative changes of neural function after repeated antidepressant treatment has been extensively documented. Presynaptic inhibition of serotonergic (5-HT) function by SSRIs, MAOIs and tricyclic drugs has been shown to contribute to the delayed onset of antidepressant action. Antidepressant drugs of these families markedly elevate the extracellular concentration of 5-HT (5-HText) in the raphe nuclei of the midbrain, thus activating somatodendritic 5-HT1A autoreceptors and offsetting the increments of 5-HText obtained by blockade of reuptake in nerve terminals (see Artigas et al., 1996 for review). The negative feed-back is more marked in forebrain areas innervated by the dorsal raphe nucleus (DRN), such as frontal cortex or striatum, as compared to the hippocampus, which receives afferents from median rapbe (MRN) serotonergic neurones. Chronic, but not acute, treatment with clinically relevant doses of SSRIs, MAOIs and imipramine results in significant increments of 5-HText in rat frontal cortex, a finding that may be related to the desensitization of somatodendritic 5-HT1A receptors by antidepressant treatments. The combined administration of 5-HT uptake inhibitors and 5-HTIA receptor antagonists results in a potentiation of the effects of the former in rat brain. In agreement with the greater self-inhibition in DRN projections produced by SSRIs, the selective 5-HT1A antagonist WAY 100635 markedly potentiated (ca. six-fold) the elevation of 5-HText produced by 3 mg/kg paroxetine in dorsal striatum but it only doubled that produced in dorsal hippocampus, innervated preferentially by the MRN. The non-selective 5-HT]A antagonist (-)pindolol also potentiated the effects of SSRIs in striatum but to a lower extent
than WAY 100635. This effect may likely underlie the enhanced antidepressant action of SSRI + pindolol combination treatments observed in various open-label and controlled studies. Moreover, administration of desipramine further enhanced the cortical 5-HText elevation produced by the SSRI fluoxetine whereas it did not alter 5-HText in untreated animals. This observation, together with tile marked increments of rat cortical 5-HText observed after high imipramine doses, may be related to the unsurpassed therapeutic efficacy of imipramine (Bel and Artigas, 1996). These experimental and clinical data emphasize the importance of the presynaptic component of serotonergic transmission in the therapeutic action of antidepressants. Chronic antidepressant treatments elicit a plethora of changes in recepor density and intracellular messengers in various forebrain structures. Yet it is not known whether such postsynaptic adaptations also contribute to the therapeutic action of these agents. Thus, it is unclear whether clinical improvement is the immediate consequence of an enhanced activation of certain 5-HT postsynaptic receptors or whether long-term adaptive changes in limbic and cortical areas are also required. In the first instance, the presynaptic component would be entirely responsible for the delayed onset of action and the therapeutic failure. Thus, patients resistant to SSRI treatment might be viewed as those individuals with e.g., greater somatodendritic 5-HT release, hypersensitive 5-HT1A autoreceptors or hyperactive 5-HT transporter, all causes leading to a greater-than-average self-inhibition of 5-HT neurons during SSRI treatment. Those patients would have an additional difficulty to increase 5-HText in forebrain despite an effective blockade of the 5-HT transporter in nerve terminals due to any of these possible factors. Addition of 5-HTjA antagonists to the therapeutic regime would bring about a response by breaking this negative feedback and allowing 5-HT neurons to normalize their firing and terminal release. Although experimental evidence for the existence of such neurobiological differences between responder and non-responder patients is lacking, this hypothesis would be consistent with the marked improvement observed in some treatment-resistant patients after pindolol addition. Interestingly, the existence of allelic variants of of the 5-HT transporter has been described, and we have observed that patients with a high pretreatment platelet 5-HT concentration display a very poor therapeutic response to various antidepressant drugs, including clomipramine, fluvoxamine and paroxetine. This suggests that individuals with a highly efficient 5-HT transporter (at least in periphery) may be more prone to be resistant to SSRI treatment. The inability of pindolol to fully suppress the latency in the antidepressant action of SSRIs appears to suggest that postsynaptic adaptive mechanisms are also involved in the full remission of depressive symptomatology. However, several observations are in partial disagreement with this view. First, pharmacokinetic factors may contribute (i.e., slow increases in the plasmatic levels with some antidepressants). Secondly, pindolol is by no means a selective and potent 5-HTtA antagonist. In view of the data obtained in rat brain, more effective 5-HT1A antagonists would be required to better prevent self-inhibitory influences on serotonergic neurones and therefore enhance the presynaptic 5-HT function during SSRI treatments. Finally, clinical observations indicate the presence of mood changes in depressed patients whith a time course much shorter than that required for gene regulation or changes in :protein synthesis. For instance, the rapid relapse arid subsequent recovery of depressed patients treated with SSRIs when administered a tryptophan-free amino acid mixture is indicative that in certain conditions there may be a rapid switch between euthimic and depressive states (Delgado et al., 1990). Research to produce more rapid and effective antidepressant treatments should proceed along several lines. First, given the excellent therapeutic efficacy of tricyclic drugs and SNRIs, it is likely thai: the study of the 5-HT-noradrenaline interactions may provide some clues. Also, a careful study of the functional control of serotonergic neurones by the different afferent pathways to the DRN and MRN may be interesting to understand their regulatory mechanisms. Drugs activating 5-HT neurones may prove useful to either produce an antidepressant effect by themselves or to prevent the inhibitory effects produced by antidepressant treatments. Finally, a better comprehension of the postsynaptic mechanisms (receptors, circuits) involved in the therapeutic action of antidepressants would greatly help, because compounds targeted specifically at certain postsynaptic receptors would overcome the psesynaptic inhibition, responsible, at least in part, of the delayed onset of antidepressant action.
S.12
Critical issues in clinical management of obsessive compulsive disorders
References Artigas, F., Romero, L., de Montigny, C. and Blier, P. (1996) Acceleration of the effect of selected antidepressant drags in major depression by 5-HTtA antagonists. Trends Neurosci. 19, 378-383. Bel, N. and Artigas, E (1996) In vivo effects of the simultaneous blockade of serotonin and norepinephrine transporters on serotonergic function. Microdialysis studies. Journal of Pharmacology and Experimental Therapeutics, 278, 1064-1072. Delgado, P.L., Chamey, D.S., Price, L.H., Aghajanian, G.K., Landis, H. and Heninger, G.R. (1990) Serotonin function and the mechanism of antidepressant action. Reversal of antidepressant-induced remission by rapid depletion of plasma tryptophan. Arch Gen Psychiatry, 47, 411--418.
~ C a n
we identify subgroups for targeted adjunctive treatment?
Michael T. Isaac. Department of Psychological Medicine, United Medical & Dental Schools, University of London (Guy's Campus), Suite 6, Lewisham Hospital, London, UK No antidepressant currently in use exerts a significant antidepressant effect for at least two to three weeks after the patient starts taking it. Open studies suggest that, for selective serotonergic re-uptake inhibitor (SSRI) antidepressants, this latency may be reduced when the drug is taken with the 5HTIA receptor blocker pindolol. In a recent randomised, placebo controlled double blind study at Guy's and Lewisham Hospitals, eighty out-patients (mean age 36 [range 19~55], 48 female, 32 male) were recruited, each patient receiving paroxetine (20 mg o.d.) plus, randomly, either pindolol (2.5 mg t.d.s.) or placebo for six weeks. Paroxetine (open label) was offered to all, patients for a further 18 weeks. Follow-up assessment on sixty-nine patients, took place at weeks 8, 16 and 24. Taken overall, the results showed a statistically significant acceleration, observable from day 4, of the antidepressant effect of paroxetine in the group receiving pindolol. In the intention to treat population, the divergence between pindolol and placebo groups disappeared by day 28, but persisted beyond day 42 in the per protocol group. In the longer term, patients originally treated with pindolol (n = 32) showed significantly better outcome at week 24, when compared with patients taking paroxetine alone, whether they complied fully with follow-up treatment or not. Compliance with follow-up paroxetine had a significant positive effect on outcome at weeks 16 and 24 in those originally treated with paroxetine alone (n = 37). Similar findings are apparent at one year. However, it is clear that the effect is not equally shared among all patients. We found, for example, that the combination worked less well in males over the age of thirty five years, and in males and females who had either a chronic history of dysthymia/depression or who were more severely depressed. We specifically excluded substance misusers. It may be relevant that our patients were reviewed twice weekly during the first two weeks of the study, and weekly thereafter. This was intended to permit the ascertainment of an early response to medication. It is becoming clearer that the 5HT1A receptor is subject to genetic polymorphism, and such differences may be highlighted by use of a 5HT1A receptor blocker. Moreover, the minus isomer and the mixed (plus/minus) enantiomer of pindolol appear to act differently (in rats, at any rate) in their activity as partial agonists of 5HT1A receptors. We examined personality variables in 48 consecutive subjects according to a short version (TCI-125) of Cloninger et al's self-rated Temperament & Character Inventory and correlated the results with clinical responses in the trial. The results suggest that high scores in the temperament dimension of Reward Dependence and low scores in the temperament dimension of Harm Avoidance had a better outcome at 6 weeks. Patients who had received paroxetine and pindolol during the trial and who reported high Novelty Seeking and low Harm Avoidance scores had a better outcome at 6 weeks and 6 months. We suggest that temperament factors may influence outcome of antidepressant treatment. These results suggest low dopamine and low serotonin transmission. Low serotonin levels may be associated with up regulation of serotonin receptors in the mesolimbic area, which can explain the rapid acceleration effect of pindolol in these patients. Pindolol has been shown in humans to increase prolactin levels (hence influencing dopamine levels), supposedly under the influence of the serotonergic system. However, the results in
S101
the pindolol group in our study perhaps suggest that this combination can also alter brain dopamine levels via the serotonergic system. Low serotonin levels have been reported in depressed patients. In our ~tudy, a low (<20 ng.ml -I ) baseline platelet 5HT was a~sociated with a poorer response to medication at day 42, compared with a 95% fall in platelet 5HT at day 7. There was an association between high baseline platelet 5HT and moderate depression, and low baseline platelet 5HT and severe depression at day 21. A previous history of more than two treated episodes of depression, or an initially severe (MADRS >34) baseline symptom profile was associated with less acceleration and poorer overall outcome. Chronicity as such did not predict lack of acceleration, although general outcome was less good. Many of the best responders were in employment and less than forty years old. The choice of SSRI may be important. But animal studies show no differences in the physiological or neurochemical consequences of pindolol in combination with any SSRI drugs, including fluoxetine. Moreover, in comparative and systematic studies, there is little to choose between the core antidepressant effects of fluoxetine and paroxetine (or other SSRI for that matter). In addition, acceleration of antidepressant effect is notoriously difficult to evaluate. It may seem an obvious point, but one must evaluate the patient during frequently during the early stages to show an early response; it is not enough to ask the patient at, say, ten or fourteen days ff he experienced an early effect before that point. A more rapidly acting antidepressant must not, of course, show only an early response; this must be sustained. Placebo control is essential, and some protocols insist on an one-week, placebo run-in phase in an attempt to control for a placebo response. This approach has merit, but a properly controlled study should be able to allow for early placebo effects, which, in any event, are usually not sustained. Developing more predictors of the response to antidepressants is a present and future challenge to research.
S.12 Critical issues in clinical management of obsessive compulsive disorders
I
•
Predictors of response
D. Marazziti. Institute of Psychiat~ University of Pisa, Italy In the last decades, consistent advancements have been raade in the synzhesis of new psychotropic drugs whose developments has, been paralleled ;and supported by the increased knowledge in Neuroscieuces. Therefore, psychopharmacological treatments have become more specific, although their effectiveness is not always restricted to clinical diagnostic entities, t~ut to symptom clusters or psychopathological dimensions. In spite of the continuous data efflux on the mode of actions of psychotropic drugs, on their spectrum of properties and side-effects, the choice of a drug is rather empyrical and no real predictors can currently be identified. However, the careful clinical observation of the patients, coupled with the present knowledge of the neurobiology of a given disorder and with the mechanisms underlying the effects of a drug should provide us some useful information. Some clinical indexes that seem to have a negative predictive value in obsessive-compulsive disorder (OCD), but which necessitate of further validation are: comorbidity with schyzotypal personality disorder, panic disorder, tic or Tourette's syndrome, early onset and chronic course. Our group investigated the possible usefulness of a peripheral serotonergic parameter, that is, platelet serotonin (5-HT) transporter, as a predictor of response in a group of 30 drug-free OCD patients. Platelet 5-HT transporter was measured by means of the specific binding of 3H-Paroxetine (3H-Par), before (tO) and after (tl) a 2-rnonth treatment with clomipramine, fluvoxamine and fluoxetine. Symptom severity was assessed with the Y-BOCS. The results showed that at baseline the patients had a decreased number of 3H-par binding sites which increased towards normal values after the treatments which was successful in most of that cases. The number of 3H-Par binding sites were significantly and negatively correlated with symptom severity at baseline, i.e., the lower