ARTICLE IN PRESS. Available at journal homepage:

Transcription

1 Psychoneuroendocrinology (2007) 32, Available at journal homepage: SHORT COMMUNICATION Effects of citalopram treatment on hypothermic and hormonal responses to the 5-HT 1A receptor agonist buspirone in patients with major depression and therapeutic response Ricard Navinés a,b,, Rocío Martín-Santos b, Esther Gómez-Gil a, María J. Martínez de Osaba c, M. Luisa Imaz a,b, Cristóbal Gastó a a Institut de Neurociències. Servei de Psiquiatria. Hospital Clínic, Barcelona, Spain b Grup de Recerca Clínica en Farmacologia Humana i Neurociències, Unitat de Recerca Farmacològica (URF), Institut Municipal d Investigació Mèdica (IMIM), Barcelona, Spain c Servei d Hormonal, Hospital Clínic i Provincial de Barcelona, Barcelona, Spain Received 22 July 2006; received in revised form 11 January 2007; accepted 12 January 2007 KEYWORDS Buspirone challenge; 5-HT1A receptors; Citalopram; Major depression; Acute antidepressant treatment; Therapeutical response Summary Serotonin (5-HT) 5-HT 1A receptor seems to play an important role in the pathophysiology of major depression and in the mechanism of action of antidepressants. In vivo function of 5- HT 1A receptors can be monitored using specific pharmacological challenge tests. The present study aimed at exploring the adaptative 5-HT 1A receptor changes in depressed patients before and after 8 week treatment with citalopram. The study population consisted of 30 consecutive outpatients of both sexes aged years with major depressive disorders (DSM-IV). Basal score in the Hamilton Rating Scale for Depression (HRSD) was higher than 17. Therapeutic response was defined as a 50% decrease in the HRSD score. The hypothermic and endocrine responses (ACTH, cortisol, and prolactin) induced by the 5-HT 1A receptor agonist, buspirone (30 mg PO) were measured. After 8 weeks on citalopram, the Dmax of hypothermic response elicited by buspirone was markedly decreased (po0.001). Patients showed a decrease in responses to ACTH (Dmax p ¼ 0.005; AUC p ¼ 0.028) and cortisol (Dmax p ¼ 0.05). However, the prolactin response increased (Dmax p ¼ 0.02; AUC p ¼ 0.005). There was a significant correlation between the therapeutic effect and reductions of ACTH (r ¼ 0.883; po0.001) and cortisol (r ¼ 0.610; p ¼ 0.001) responses. Changes induced by citalopram support an alteration Corresponding author. Ricard Navinés, Grup de Recerca Clínica en Farmacologia Humana i Neurociències, Unitat de Recerca Farmacològica (URF), Institut Municipal d Investigació Mèdica (IMIM), Doctor Aiguader 80, E Barcelona, Spain. Tel.: ; fax: address: rnavines@hotmail.com (R. Navinés) /$ - see front matter & 2007 Elsevier Ltd. All rights reserved. doi: /j.psyneuen

2 412 R. Navinés et al. of 5-HT 1A receptors in major depression. A decrease in the overactivity of the HPA axis may be one factor associated with the response to citalopram. & 2007 Elsevier Ltd. All rights reserved. 1. Introduction Selective serotonin (5-HT) re-uptake inhibitors (SSRI) are widely used in the treatment of major depression. It is hypothesized that the pharmacologic effect results from adaptative changes involving serotoninergic neurotransmission (Celada et al., 2004). 5-HT 1A receptors may play an important role. These receptors act as somatodentritic autoreceptors that inhibit neuronal firing in the midbrain raphe nuclei and as postsynaptic receptors that mediate the effects of serotonin released from presynaptic terminals, mainly in cortico-limbic areas (Cowen, 2000). Presynaptic 5-HT 1A receptors are the primary target of several types of antidepressant drugs; long-term use of antidepressant drugs results in an increase in the 5-HT 1A receptors mediated hippocampal transmission (Cowen, 2000). The neuronal adaptative mechanisms of these receptors, both presynaptic and postsynaptic, may account for the delay and limited response to the antidepressant drug (Celada et al., 2004). However, elucidation of these mechanisms is hampered by many factors, such as the intrinsic complexity to study brain function, difficulties in assessing the effects of antidepressant drugs in humans, and the lack of reliable experimental models of depression. Pharmacological challenge with probe drugs measuring induced neuroendocrine alterations is a useful approach for assessing neurotransmitter receptors function. The administration of the partial 5-HT 1A agonists ipsapirone, buspirone and gepirone, and the full 5-HT 1A agonist flesinoxan, has been shown to induce hypothermia and hormonal changes in animal models and healthy subjects (Seletti et al., 1995; Cowen, 2000; Pitchot et al., 2002). Experimental data suggest that the hormonal responses to 5-HT 1A receptor agonists are mediated via postsynaptic 5-HT 1A receptors, whereas hypothermic responses induced by 5-HT 1A receptor agonists are mediates by presynaptic 5-HT 1A autoreceptors (Cowen, 2000). A mixed pre- and postsynaptic activation has also been suggested for mediation of hypothermia in rats (Lerer et al., 1999). Neuroendocrine studies with 5-HT 1A agonists in depressed patients suggest a lower functionality of pre- and postsynaptic 5-HT 1A receptors compared with healthy subjects, but results are controversial (Shapira et al., 2000; Pitchot et al., 2005; Navinés et al., 2006). These reduced 5-HT 1A responses do not normalize after antidepressant treatment with amytriptiline (Lesch et al., 1990) and electroconvulsive therapy (Shapira et al., 2000). In healthy subjects, the administration of a SSRI induces a general subsensitivity of hormone responses to the 5-HT 1A receptor agonists (Anderson et al., 1996; Sargent et al., 1997; Berlin et al., 1998; Lerer et al., 1999), including a decrease in the hypothermic response (Sargent et al., 1997; Lerer et al., 1999). To our knowledge adaptative changes in both pre- and postsynaptic 5-HT 1A receptors in depressed patients after treatment with SSRIs have not been previously investigated. The goal of the present study therefore was to investigate in depressed patients the effect of chronic treatment with citalopram on 5-HT 1A autoreceptors, using a challenge test with the 5-HT 1A receptor agonist buspirone, measuring the hypothermic response as well as prolactin (PRL), corticotrophin (ACTH) and cortisol responses and their relationship with the clinical effect. 2. Method 2.1. Participants Thirty untreated patients fulfilling DSM-IV criteria for current unipolar major depression were recruited from an outpatient Psychiatric Department of a teaching hospital in Barcelona (Spain). There were 10 males and 20 females with a mean age of , ranging from 18 to 45 years. They had no current diagnosis of substance abuse or other current or lifetime concomitant psychiatric disorder, such as anxiety disorder or bipolar disorder. None had overweight or major medical disease. Pregnant or breastfeeding women were excluded. The test challenge was performed in the first phase of the hormonal cycle. None had taken any hormonal or psychotropic medication for at least 30 days prior to buspirone challenge except for low-dose benzodiazepines. All were assessed using the Spanish version of the Structured Clinical Interview for DSM-IV Axis-I Disorders (First et al., 1995) and had a baseline score of X17 on the 21-items Hamilton Rating Scale for Depression (HRSD) (Hamilton, 1960). Patient was re-tested for the HRSD after 8 weeks of citalopram treatment. All patients underwent physical examination and laboratory test for the routine screening of hematological, biochemical and hormonal blood parameters, and standard ECG. All participants gave written the informed consent after a full explanation of the study. The study was approved by the Ethic Research Committee of the Institution Buspirone protocol Patients were admitted at 0830 h after an overnight fast. Subjects were tested seated, and indwelling catheter was inserted into the vein. After a 30-min rest period to minimize stress factors, a blood sample was taken for baseline determinations of plasma PRL, ACTH and cortisol. Then 30 mg of buspirone were administered orally. Further blood samples were taken at 60, 90 and 120 min after buspirone administration. Oral temperature was measured with a glass mercury thermometer at baseline and after the same time intervals post buspirone administration. The thermometer remained in situ for 10 min before the reading was recorded by a research nurse. The temperature of the testing room was maintained 21 1C. Every 30-min intervals subject completed 100-mm visual analogue scales (VAS) for

3 Buspirone-induced hypothermic and hormonal responses and use of citalopram 413 nausea, light-headed, cephalea and dizziness. After 8 weeks of citalopram treatment the challenge with buspirone was repeated. In women, tests were performed in the first phase of the menstrual cycle. Venous sample of 10 ml for cortisol and PRL were taken into tubes and were stored at 20 1C until analysis. Blood samples of 5 ml for ACTH determinations were collected into tubes containing EDTA, stored on ice and centrifuged within 60 min. Plasma was separated and stores at 801C for assay at a later time Citalopram treatment Following the completion of the buspirone protocol, all patients started treatment with 15 mg/day of citalopram increasing to 30 mg/day on day 7. The daily dose of 30 mg/ day was maintained for the next 8 weeks. Patients were assessed by a psychiatrist at 2-week intervals. Compliance was monitored by pill counts on every return visit Hormone assays Cortisol was determined by means or a radioimmunoassay kit (RIA) (Inmunotech, France). ACTH by a commercially available immunoradiometric assay (IRMA) (Nichols Institute, USA). PRL was determined by an RIA. Inter- and intraassay coefficients of variation over the working concentrations range of standard curves were 4% and 8% for cortisol; 5% and 11% for ACTH; and 6% and 10% for PRL Statistical analysis Group size was calculated by standard methods (Schlesslman and Stolley, 1982). A sample size of 24 subjects would have a statistical power of 80% to detect a difference between mean (SD) values of the hormonal determinations with an alpha error of 0.05, taking into account a 15% of lost to follow-up. For hypothermic and hormonal responses to buspirone the next maximal response, that is, the peak response minus baseline (D) and the area under the curve (AUC) from baseline were calculated. The differences in the maximal hypothermic and hormonal responses to buspirone before and after 8 weeks on citalopram treatment were assessed with the nonparametric Wilcoxon s t test. All tests were two-tailed with statistical significance set at po0.05. The Spearman s correlation coefficient was calculated to quantify the association of continuous variables. 3. Results Of the 30 patients included in the study, five refused buspirone challenge test after 8-week treatment with citalopram. Therefore, data presented refer to 25 patients who completed both buspirone challenge tests before and after antidepressant treatment. Table 1 shows median (25th 75th) values of basal, Dmax and AUC of hypothermic and hormonal (ACTH, cortisol and PRL) responses induced by buspirone challenge before and after antidepressant treatment. The median baseline body temperature, ACTH, cortisol and PRL values in depressed patients after 8 weeks of citalopram treatment were not significantly different than those before treatment. Following the buspirone administration we found after 8 weeks of citalopram treatment, a significant decrease in the Dmax (po0.001) but not in AUC for hypothermic response (Table 1). Figure 1 shows the temperature response to buspirone before and after 8 weeks of citalopram treatment. After buspirone challenge plasma concentrations of ACTH, cortisol, and PRL increased, both before and after citalopram treatment. Moreover, we found a decrease in Dmax (p ¼ 0.005) and in the AUC (p ¼ 0.028) for ACTH response after treatment. A decrease in Dmax (p ¼ 0.05) for cortisol response was also observed. In contrast, an increase in Dmax (p ¼ 0.02) for PRL response (p ¼ 0.05) was found (Table 1). The mean (SD) of total score of HRSD at baseline was (5.06) (range [max min] 32 17) and at 8 weeks of citalopram treatment was (4.55) (range [max min] 20 6) (paired-t test ¼ 14.79; po0.001). At the end of the study, 18 subjects (60%) showed a 50% of reduction in the HRSD score. Only decreases in ACTH and cortisol responses and improvement of depressive symptoms after 8 weeks of citalopram treatment correlated significantly, both in the 18 subjects with a 50% reduction of HRSD scores (ACTH response and HRSD scores, r ¼ 0.797, po0.001; cortisol response and HRDS scores, r ¼ 0.598; p ¼ 0.031) and in the total group of 25 subjects (ACTH response and HRSD scores, r ¼ 0.883, po0.001; cortisol response and HRDS scores, r ¼ 0.610; p ¼ 0.001). Side effects related to buspirone administration included dizziness in six subjects and slight sedation in four, which occurred about 15 min after drug ingestion. Symptoms disappeared in about min and none of the subjects had to discontinue buspirone challenge test. On the other hand citalopram was well tolerated. Only three patients presented nausea in the first days of treatment and two patients increased anxiety symptoms. All subjects continued the study without reduction or withdrawal of the antidepressant drug. 4. Discussion The attenuated hypothermic response after 8 weeks of citalopram treatment suggests desensitization of the presynaptic 5-HT 1A receptor. This finding is consistent with neuroendocrine studies in normal subjects, in which treatment with SSRIs decreased hypothermic response to the 5-HT 1A agonists (Sargent et al., 1997; Lerer et al., 1999), as well as in patients with obsessive compulsive disorders treated with fluoxetine, patients with panic disorders treated with clomipramine (Lesch et al., 1991; Broocks et al., 2003), and patients with depression treated with amytriptiline and electroconvulsive therapy (Lesch et al., 1990; Shapira et al., 2000). Nevertheless, in our study the conclusion that pre-synaptic 5-HT 1A receptor function is reduced by treatment should be tempered by the fact that the AUC for hypothermia was not significantly altered. The decrease in 5-HT 1A autoreceptors sensitivity would be expected to facilitate 5-HT neurotransmission by decreasing inhibitory feedback of these receptors at 5-HT cell bodies. These are supported by experimental date from animal studies (El Mansari et al., 2005). There are also evidences of

4 414 R. Navinés et al. Table 1 Baseline, Dmax and AUC values for hypothermic and hormone responses prior to and after 8 weeks of citalopram treatment. Data Pre-tretment n ¼ 25 Post-treatment n ¼ 25 p Median (25th 75th) Median (25th 75th) Temperature, 1C Baseline 36.4 ( ) 36.3 ( ) NS Dmax 0.65 ( ) 0.30 ( ) AUC ( ) 4323 ( ) NS ACTH, pg/ml Baseline 14.7 ( ) 13 (7.5 22) NS Dmax 2.6 ( ) 0.0 ( ) AUC ( ) ( ) Cortisol, mg/dl Baseline 11.7 ( ) 12.7 ( ) NS Dmax 1.08 ( ) 0.0 ( ) 0.05 AUC ( ) 1206 ( ) NS PRL, ng/ml Baseline 7.7 ( ) 6.9 ( ) NS Dmax 6.5 ( ) 8.8 ( ) 0.02 AUC 1426 ( ) ( ) NS Wilcoxon s test. Median differences of temperature ( C) Pre/post treatment Pre treatment Post treatment Baseline 60 Minutes 90 Minutes 120 Minutes a decreased in vivo binding of the positron emission tomography (PET) 5-HT 1A radioligands in the nucleus dorsal raphe in animal studies treated with different SSRI classes (Giovacchini et al., 2005). PET studies in depressed patients found a decrease 5-HT 1A receptor binding in the dorsal raphe nucleus at baseline which does not seem to change after SSRI treatment (Sargent et al., 2000). With regard to the hormonal response to buspirone in depressed patients after citalopram treatment, our data suggests the implications of different mechanisms involved Buspirone challenge Figure 1 Median differences of temperature (1C) at 60 (p4.05), 90 (po.001), and 120 (po.001) min since baseline, prior to and after 8 weeks of citalopram treatment. in the opposite effects on PRL and ACTH/cortisol response. The reduction in the ACTH/cortisol response supports a desensitization in post-synaptic 5-HT 1A receptors in the hypothalamus by the antidepressant treatment in agreement with neuroendocrine studies in both rats and humans (Lerer et al., 1999; Sargent et al., 1997). Although the 5-HT 1A receptors-induced cortisol response are down-regulated in unmedicated depressed patients in some studies (Shapira et al., 2000), the current results suggest that these receptors are further down-regulated after successful treatment. This is contrary to the hypothesis that SSRIs and other antidepressants cause an enhancement of 5-HT 1A function which is responsible for antidepressant effects (Celada et al., 2004). However, after antidepressant treatment the overall effect of pre and post-synaptic 5-HT 1A receptors changes should result to an increase in the 5-HT neurotransmission at post-synaptic 5-HT 1A receptors, due to the reduction in the pre-synaptic 5-HT 1A receptors inhibitory feedback at 5-HT cell bodies (Celada et al., 2004). The ACTH and cortisol response could also be explained to a reduction in the hypothalamic pituitary adrenal (HPA) axis activity. The mechanism by which changes in HPA axis function could produce a reduced cortisol response may arise from an increase in glucocorticoid receptors (GR) sensitivity and enhancing the GRmediated feedback inhibition, which leads to a decrease in the cortisol response (Holsboer and Barden, 1996). In general, the dysfunction in the HPA axis activity found in depressed patients is corrected during a clinically effective therapy with antidepressant drugs (Heuser et al., 1996; Holsboer and Barden, 1996). Favorable response to antidepressant treatment can be predicted by determining the dexamethasone suppresser status at admission (Ising et al., 2005). The correlation between the decrease in the

5 Buspirone-induced hypothermic and hormonal responses and use of citalopram 415 ACTH/cortisol response and the clinical outcome found in our study support that the reduction in the activity of HPA axis may be one factor associated to the therapeutic response. The increase in PRL response after citalopram treatment could be explained by an effect on the post-synaptic 5-HT 1A receptor or also on the dopamine-d 2 receptor. Both mechanisms have been related to buspirone PRL response (Dinan et al., 2001). The enhanced PRL response is likely to reflect facilitation in 5-HT 1A neurotransmission in depressed patients by the antidepressant treatment. Also, the enhanced PRL response may be secondary to a reduction in dopaminergic inhibition, which may be mediated by an increase in dopamine receptor sensitivity (Porter et al., 2003). The present findings should be interpreted taking into account some limitations of the study. Buspirone is not a full 5-HT 1A agonist and may induce hormone release by its action as a dopamine D2 antagonist and the ability of its metabolite 1-(2-pryrimidinyl)-piperazine (1-PP) to enhance catecholamine turnover (Shapira et al., 2000). Buspirone was selected because of its easy availability and large body of evidence on clinical safety. Although the inclusion of a placebo group in a study of depressed patients is not ethically justified, the fact that a group of healthy subjects to compare the response to citalopram with that obtained in patients with major depression was not included may be considered a limitation of the study. Finally, citalopram could increase plasma buspirone levels and the results may result from changes in plasma pharmacokinetics of buspirone (Anderson et al., 1996). Plasma buspirone levels were not measured in our study. Taken all together, the 5-HT 1A receptors changes induced by the citalopram treatment support an alteration of these receptors in major depression and in the antidepressant effect. Moreover, a decrease in the activity of the HPA axis may be one factor associated with the response to citalopram. Further neuroendocrine and neuroimaging studies are needed to correlate changes in mood and in receptor density with the effect of SSRIs on 5-HT 1A receptormediated hypothermic and neuroendocrine responses in depressed patients. Role of the funding sources Ricard Navinés was a recipient of the fellowship Beca Premio Fin de Residencia from Hospital Clinic of Barcelona. Conflict of interest The authors declare that they have no conflicts of interest. Acknowledgments Ricard Navinés was a recipient of the fellowship Beca Premio Fin de Residencia from Hospital Clinic, Barcelona. The authors thank Rafael de la Torre, Ph.D., and Magı Farré, MD, Ph.D. for fruitful comments on early version of the paper, Manel Salamero, Psychologist, Ph.D. for statistical advice and Marta Pulido, MD, for editing the manuscript and editorial assistance. References Anderson, I.M., Deakin, J.F., Miller, H.E.J., The effect of chronic fluvoxamine on hormonal and psychological responses to buspirone on normal volunteers. Psychopharmacology 28, Berlin, I., Warot, D., Legout, V., Guillemant, S., Schollnhammer, G., Puech, A.J., Blunted 5-HT1A-receptor agonist-induced corticotrophin and cortisol response after long-term ipsapirone and fluoxetine administration to healthy subjects. Clin. Pharmacol. Ther. 63, Broocks, A., Meyer, T., Opitz, M., Bartmann, U., Hillmer-Vogel, U., George, A., Pekrun, G., Wedekind, D., Ruther, E., Bandelow, B., HT1A responsivity in patients with panic disorder before and after treatment with aerobic exercise, clomipramine or placebo. Eur. Neuropsychopharmacol. 13, Celada, P., Puig, V., Amargós-Bosch, M., Adell, A., Artigas, F., The therapeutic role of 5-HT1A and 5-HT2A receptors in depression. Rev. Psychiatr. Neurosci. 29, Cowen, P.J., Psychopharmacology of 5-HT1A receptors. Nucl. Med. Biol. 27, Dinan, T.G., Scott, L.V., Thakore, J., Naesdal, J., Keeling, P.W., Impact of cortisol on buspirone stimulated prolactin release: a double-blind placebo-controlled study. Psychoneuroendocrinology 26, First, M., Spitzer, R., Gibbon, M., Williams, J., Structured Clinical Interview for DSM-IV Axis I Disorders Patients Edition (SCID-I/P, Version 2.0). New York State Psychiatric Institute, New York, NY. Giovacchini, G., Lang, L., Ma, Y., Herscovith, P., Eckelman, W.C., Carson, R.E., Differential effects of paroxetine on raphe and cortical 5-HT1A binding: a PET study in monkeys. Neuroimage 15, Hamilton, M., A rating scale for depression. J. Neurol. Neurosurg. Psychiatry 23, Heuser, I.J., Schweiger, U., Gotthardt, U., Schmider, J., Lammers, C.H., Dettling, M., Yassouridis, A., Holsboer, F., Pituitary adrenal system regulation and psychopathology during amitriptyline treatment in elderly depressed patients and normal comparison subjects. Am. J. Psychiatry 153, Holsboer, F., Barden, N., Antidepressants and hypothalamic pituitary adrenocortical regulation. Endocrine. Rev. 17, Ising, M., Kunzel, H.E., Binder, E.B., Nickel, T., Modell, S., Holsboer, F., The combined dexamethasone/crh test as a potential surrogate marker in depression. Prog. Neuropsychopharmacol. Biol. Psychiatry 29, Lerer, B., Gelfin, Y., Gorfine, M., Allolio, B., Lesch, P., Newman, E., HT1A receptor function in normal subjects on clinical doses of fluoxetine: blunted temperature and hormone responses to ipsapirone challenge. Neuropsychopharmacology 20, Lesch, K.P., Disselkamp-Tietze, J., Schimidtke, A., HT1A receptor function in depression: effect of chronic amitriptyline treatment. J. Neural. Transm. 80, Lesch, K.P., Hoh, A., Schulte, H.M., Osterheider, M., Muller, T., Long-term fluoxetine treatment decreases 5-HT1A receptor responsivity in obsessive compulsive disorder. Psychopharmacology 105, Navinés, R., Gómez, E., Martin-Santos, R., Martínez de Osaba, M.J., Imaz, M.L., Gastó, C., An increased hypothermic response in patients with major depression. Psychopharmacology 188, Pitchot, W., Wauthy, J., Hansenne, M., Pinto, E., Fuchs, S., Reggers, J., Legros, J., Ansseau, M., Hormonal and temperature

6 416 R. Navinés et al. responses to the 5-HT1A receptor agonist flesinoxan in normal volunteers. Psychopharmacology 164, Pitchot, W., Hansenne, M., Pinto, E., Reggers, J., Fuchs, S., Ansseau, M., Hydroxytryptamine 1A receptors, major depression, and suicidal behavior. Biol. Psychiatry 58, Porter, R., Gallagher, P., Watson, S., Smith, M., Young, A., Elevated prolactin responses to L-tryptophan infusion in medication-free depressed patients. Psychopharmacology 169, Sargent, P., Williamson, D.J., Pearson, G., Odontiadis, J., Cowen, P.J., Effect of paroxetine and nefazodone on 5-HT1A receptor sensitivity. Psychopharmacology 132, Sargent, P., Kjaer, K.H., Bench, C.J., Rabiner, E.A., Messa, C., Meyer, J., Gunn, R.N., Grasby, P.M., Cowen, P.J., Brain serotonin1a receptor binding measured by positron emission tomography with [11C]WAY : effects of depression and antidepressant treatment. Arch. Gen. Psychiatry 57, Schlesslman, J.J., Stolley, P.D., Case-control Studies: Design, Conduct, Analysis. Oxford University Press, New York. Seletti, B., Benkelfat, C., Blier, P., Annable, L., Gilbert, F., de Montigny, P., Serotonin1A receptor activation by flesinoxan in humans. Body temperature and neuroendocrine responses. Neuropsychopharmacology 13, Shapira, B., Newman, M.E., Gelfin, Y., Lerer, B., Blunted temperature and cortisol responses on ipsapirone in major depression: lack of enhancement by electroconvulsive therapy. Psychoneuroendocrinology 25,