- Email: [email protected]
Alpha-1-acid glycoprotein in major depressive disorder
Journal of Affective Disorders 59 (2000) 159–164 www.elsevier.com / locate / jad
Brief report
Alpha-1-acid glycoprotein in major depressive disorder Relationships to severity, response to treatment and imipramine plasma levels a
a,
b
c
Evaristo Nieto MD , Eduard Vieta MD *, Luisa Alvarez MD , Merce Torra MD , Francesc Colom PhD a , Cristobal Gasto´ MD a a
b
Department of Psychiatry, Hospital Clinic de Barcelona, Villarroel 170, Barcelona 08036, Spain Department of Biochemistry, Hospital Clinic de Barcelona, Villarroel 170, Barcelona 08036, Spain c Department of Toxicology, Hospital Clinic de Barcelona, Villarroel 170, Barcelona 08036, Spain Received 4 May 1999; received in revised form 20 July 1999; accepted 28 August 1999
Abstract Background: Increased plasma levels of alpha-1-acid glycoprotein (AGP) were reported in major depressive disorder. However, the relationship between AGP levels, severity of depression, treatment response and antidepressant levels are still unclear. Methods: Plasma AGP levels were measured in 36 subjects with major depressive disorder before and after a 6-week treatment with imipramine and in 30 controls. Free imipramine plasma levels of depressed patients were measured at 6 weeks. Comparative analysis between depressed patients and controls, between non-responders (N 5 12) and responders (N 5 24), and between severely depressed patients (N 5 14) and moderately depressed patients (N 5 22) were made. Results: Depressed patients had significantly higher mean values of AGP than control subjects. Imipramine non-responders and specially severely depressed patients had significantly greater increases of AGP levels during treatment than other depressed subgroups. There was no correlation between baseline AGP levels and severity of depression or free imipramine levels. Limitations: The most significant limitations of this study are the small sample size and the fact that all the subjects were out-patients. Results should not be generalized to in-patient populations. Conclusions: Depressed patients showed high baseline concentrations of AGP. AGP levels did not predict either free imipramine plasma levels or differential response after 6 weeks of treatment with imipramine. A greater increase of AGP during treatment was associated with severity of depression and treatment non-response. Clinical implications: The relationship between high plasma levels of AGP, severity of depression and lack of treatment response is clarified. The influence of imipramine levels is minimized. 2000 Elsevier Science B.V. All rights reserved. Keywords: Plasma protein; Severity; Treatment response; AGP; Depression; Imipramine
1. Introduction
*Corresponding author.
The alpha-1-acid glycoprotein (AGP) or orosomucid is an acute phase protein and its levels
0165-0327 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0165-0327( 99 )00145-7
160
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
rise during physical illness (infections, cancer, bowel and cardiac diseases etc.) and severe stress such as surgery or burning (Kremer et al., 1988). In several studies (Nemeroff et al., 1990; Healy et al., 1991; Kehoe et al., 1991; Maes et al., 1992a; Song et al., 1994; Sluzewska et al., 1996a–c; Maes et al., 1997a,b; Wong et al., 1996) increased plasma levels of positive acute-phase-proteins such as haptoglobin and AGP have been reported in major depressive disorder, and several more mental disorders (bulimia, schizophrenia, mania), including psychological stress. Although most studies found that AGP levels are elevated in major depression, others (Maes et al., 1992b; Joyce et al., 1992; Lawrence et al., 1997) did not replicate this finding. There are also discrepancies between studies with respect to the correlation between AGP levels and age, severity of illness and treatment response. A further reason for investigating AGP levels in depression is that imipramine and other tricyclic antidepressants bind to AGP (Kremer et al., 1988). Elevations in the AGP levels might effectively reduce the level of unbound drug and thereby militate against a clinical response. This may be particularly true during the initial stages of therapy when both total and free concentrations of the drug are low (Kehoe et al., 1991). Healy et al. (1991) found higher baseline AGP values in depressed nonresponders (imipramine and others) compared to depressed responders. The present study investigated AGP plasma levels and its relationship to severity of the depression, response to treatment and imipramine levels.
2. Methods
2.1. Subjects Thirty-six depressed outpatients were recruited from the Department of Psychiatry in the Hospital Clinic de Barcelona. All patients met DSM-III-R criteria for major depressive disorder and had total scores of at least 17 on the 17-item Hamilton depression rating scale. Patients were psychotropic drug-free for at least seven days and 32 (88%) were antidepressant-free for at least six months. All patients who were found to have any medical or
psychiatric comorbidity were excluded from the study. Pregnant women and bipolar patients were excluded too. The control group, age and sex-matched subjects, was recruited from the faculty of the Hospital Clinic de Barcelona. Controls were drug-free, had no evidence of mental or physical illness and were not pregnant. All subjects (patients and controls) gave informed consent to participate in the study. The Ethics Committee of the Hospital Clinic–University of Barcelona gave its approval to the study.
3. Procedures
3.1. Laboratory tests The a 1 -acid glycoprotein forms immune complexes with specific antibodies to human AGP in the human serum sample. These complexes scatter a beam of light passed through the sample. The intensity of scattered light is proportional to the concentration of AGP in the sample. The result is evaluated by comparison with a standard of known concentration. Baseline levels of AGP (AGP1) were measured by a inmunonephelometric method using reagents from Behring in the BNA (Behringwerke, Marburg,Germany). The antiserum is a liquid animal serum produced by immunization of rabbits with highly purified human AGP. Final levels of AGP (AGP2) of depressed patients were measured at a 6-week follow-up. Blood samples were taken in controls and in depressed patients at 09.00 h. Free imipramine (IMIF) plasma levels were measured by liquid chromatography in depressed patients after 6 weeks of treatment with this drug at the Laboratory of Toxicology in the Hospital Clinic. Serum samples were assayed for imipramine (total and free fraction) using a reversed-phase high-performance liquid cromatography method. The fraction of free imipramine was obtained by ultrafiltration in a centrifuge with a fixed-angle rotor at 2000 3 g, using the Centrifree Micropartition devices (Amicon Inc., Massachussets, USA). Imipramine (total and free fraction) and the internal standard (IS) trimipramine were selectively isolated from the serum by
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
solid-phase sample preparation using the bonded phase extraction columns Bio-Rad 195-7040 (BioRad, Anaheim, USA). Chromatographic separation was achieved in a Hewlett-Packard HP-1050 system (Hewlett-Packard, Waldbronn, Germany), using a reverse-phase column Bio-Rad 195-7051 (oven temperature 5 358C) and an isocratic eluent Bio-Rad 195-7070 delivered at a flow rate of 0.6 ml / min. The column effluent was monitored using ultraviolet (UV) detection at 242 nm, 0.01 AUFS and the injection volume was 100 ml. Quantitation was by peak-height ratios (imipramine / IS). Recovery was 95–98%, whereas within day and day-to-day coefficients of variation were between 4.1 and 6.4%.
3.2. Treatment of depressed patients After blood samples were taken to measure AGP1 levels, all depressed patients were treated with increasing dosages of imipramine (50 mg / day during days 1–3; 100 mg / day during days 4–7; 150 mg / day during days 8–21) until they completed 3 weeks of treatment. Then, subjects were considered to be responders if their HDRS score was ten or lower at this point. In this case imipramine dose of 150 mg / day was sustained until they completed 6 weeks of treatment. Depressed non-responders were treated with increasing doses of imipramine (200 mg / day during days 22–28 and 250 mg / day during days 29–42) until they completed 6 weeks of treatment. After 6 weeks of treatment, final decisions regarding response were made. Subjects were considered to be non responders if their HDRS score was ten or over at this point.
3.3. Severity of depressed patients According to baseline score in HDRS, 14 major depressed patients were considered to have severe depression (initial score in HDRS was higher than 30) and 22 were moderately depressed.
3.4. Statistical methods Comparative analysis of normal variables (AGP1, AGP2, DAGP and IMIF) between two groups (depressed vs. controls, depressed responders vs. nonresponders) was performed by means of an unpaired
161
two-tailed Student t-test. Correlations between AGP measures and baseline HDRS scores and free imipramine plasma levels were made by means of the Spearman’s rank correlation coefficient.
4. Results The mean age of the 36 depressed patients was 43.9 years (SD 5 14.5 and range 5 16–70). Twentynine of them (80.5%) were women and seven (19.5%) were men. There were no significant differences in age or sex between controls and depressed patients. The baseline mean HDRS score for depressed patients was 28.42 (SD 5 5.18 and range 5 18–40) and their mean free imipramine (IMIF) levels were 25.9 ng / ml (SD 5 11.1, range 5 0–58). A comparison of the baseline AGP levels between depressed patients and the control group is presented in Table 1. Table 2 shows the comparison of baseline AGP levels (AGP1), final AGP levels (AGP2), variation of AGP levels (DAGP) and free imipraTable 1 Comparison of alpha-1-acid glycoprotein levels (mg / 100 ml) between depressed patients and controls
Male Female All
Depressed patients (N 5 36)
Control subjects (N 5 30)
N
Mean
SD
N
Mean
SD
7 29 36
86.7 82.1 83.0
24.0 22.9 22.9
6 24 30
73.0 66.9 68.1
10.5 20.2 18.7
P
NS P , 0.02 P , 0.01
Table 2 Comparison of AGP1, AGP2, DAGP (mg / 100 ml) and IMIF (ng / ml) between depressed responders and depressed non-responders to imipramine after 6 weeks of treatment a Responders (N 5 24)
AGP1 AGP2 DAGP IMIF
Non-responders (N 5 12)
Mean
SD
Mean
SD
84.7 85.7 0.95 26.6
20.7 17.9 17.4 10.7
79.5 94.5 15.0 24.5
27.5 26.6 19.9 12.3
P
NS NS P , 0.05 NS
a AGP1 5 Baseline alpha-1-acid glycoprotein; AGP2 5 final alpha-1-acid glycoprotein; DAGP 5 difference between final and baseline AGP levels; IMIF 5 plasma free imipramine levels.
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
162
Table 3 Comparison of AGP1, AGP2, DAGP (mg / 100 ml) and IMIF (ng / ml) between severely depressed and moderately depressed patients a Severe depression (N 5 14)
AGP1 AGP2 DAGP IMIF
Moderate depression (N 5 22)
P
Mean
SD
Mean
SD
76.3 95.7 19.4 27.4
18.2 19.4 18.3 11.6
87.2 84.1 2 3.0 24.9
25.0 21.6 14.2 10.9
NS NS P , 0.000 NS
a
AGP1 5 Baseline alpha-1-acid glycoprotein; AGP2 5 final alpha-1-acid glycoprotein; DAGP 5 difference between final and baseline AGP levels; IMIF 5 plasma free imipramine levels.
mine levels (IMIF) between depressed responders and depressed non-responders. The only significant difference between the subgroups was that DAGP was higher in depressed non-responders. Table 3 shows the comparison of AGP1, AGP2, DAGP and IMIF between severely depressed and moderately depressed patients. The only significant difference between both subgroups was that DAGP was higher in severely depressed patients. The mean AGP levels for the four patients treated with antidepressants in the last six months were 91 mg / 100 ml. The mean AGP level for the patients who were antidepressant-free in the last six months (N 5 32) were 81.9 mg / 100 ml. Differences were not significant. We could not find any significant correlation between AGP variables (AGP1, AGP2 and DAGP) and IMIF after 6 weeks of treatment. When the 17 depressed patients treated with 150 mg / day, the 19 depressed patients treated with 250 mg / day, the 24 depressed responders, the 12 depressed non-responders, the 14 severely depressed, and the 24 moderately depressed were considered separately, correlations still remained not significant. Furthemore, there was no correlation between AGP variables (AGP1, AGP2, DAGP) and the severity of depression as assessed by HDRS scores.
5. Discussion The results obtained in our study are in agreement with previous reports of significantly higher AGP
levels in depressed patients compared to controls (Nemeroff et al., 1990; Healy et al., 1991; Kehoe et al., 1991; Sluzewska et al., 1996a,b; Maes et al., 1997a). Within the spectrum of psychiatric disorders, alterations of AGP and other acute phase proteins are not specific for major depression (Healy et al., 1991; Maes et al., 1997a,b) and may be even more pronounced in schizophrenic than in depressed patients (Maes et al., 1997a). The second finding of this study is that there was a difference between depressed patients on the basis of their response to treatment with imipramine: nonresponders had a significantly greater increase in AGP levels than responders after 6 weeks of treatment. However, our study did not replicate the previous findings of Healy et al. (1991), in the sense that depressed non-responders had significantly higher baseline (prior to treatment) levels of AGP than subsequent responders. To the best of our knowledge, the findings of Healy et al. (1991) in depressed patients have not been replicated. Some reports have analyzed the variations in AGP levels during different treatments such as electro convulsive therapy (De Vane et al., 1991) or lithium potentiation (Sluzewska et al., 1997) but none with imipramine. Levinson and Levine (1995) found mean baseline AGP levels to be significantly higher in schizophrenic non-responders treated 2 weeks with haloperidol than in responders, but in a further study, Jann et al. (1997) could not find any relationship between AGP levels and treatrment response after 2, 4 or 6 weeks of treatment. Kehoe et al. (1991) reported that increased AGP concentration reduced the free fraction of imipramine in depressed patients. We found no significant correlation between AGP levels and free imipramine levels. One possible reason for this disagreement could be that the reduction of the free fraction of imipramine does not necessarily lead to a reduction of free imipramine levels (Greenblatt et al., 1982). On the other hand, Kehoe et al. (1991) determined the free fraction of imipramine during the initial phases of therapy (from the first week to the third week of treatment), whereas we determined the free imipramine levels after 6 weeks of treatment, and steady-state levels vary considerably among individuals mainly due to variations in metabolism, but also to a smaller extent due to variations in binding to plasma proteins (Gram, 1988).
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
A major advantage of our study over previous studies was that we determined simultaneously AGP levels, free imipramine levels and clinical response whereas Healy et al. (1991) did not determine imipramine levels and Kehoe et al. (1991) did not determine clinical response. Other supplementary advantages are that our sample was bigger than the sample of Kehoe et al. and was more homogeneous with respect to treatment than the sample of Healy et al. (all our patients were treated with imipramine, whereas the patients of Healy et al. were treated with several drugs, including imipramine, lofepramine and even electro convulsive therapy). Furthermore, the study of Healy et al. has another disadvantage in that 40% of his depressed patients had been treated with antidepressant drugs in the three months prior to study and we, like Sluzewska et al. (1996c) and Baumann et al. (1982), found that the treatment, and especially resistance to treatment, in depressed patients is associated with changes in AGP levels. Another finding of this study is that there was only one difference between depressed patients on the basis of their severity. Severely depressed patients had significantly greater increments of AGP levels during 6 weeks of treatment than moderately depressed patients. The severity of depression seemed to be more deeply related to the increase in AGP during treatment than non-response. Taking only into account the 24 depressed responders, the seven more severely depressed patients had significantly greater increases during treatment (P , 0.05) than the 17 moderately depressed patients. On the other hand, our findings are in agreement with the majority of previous investigations that show a lack of correlation between baseline AGP levels and severity (according to baseline HDRS score) in depressed patients (Healy et al., 1991; Sluzewska et al., 1996b,c). Only Nemeroff et al. (1990) found a correlation between plasma AGP levels and depression severity as assessed by the Montgomery–Asberg rating scale. In summary, depressed patients showed elevated plasma concentrations of AGP. However, AGP levels did not predict either free imipramine plasma levels or a differential response after 6 weeks of treatment with imipramine. A higher increase of AGP during treatment was associated with higher baseline severity and lack of response to treatment with imipramine.
163
References Baumann, P., Tingwely, D., Schopf, J., 1982. Increased of alpha-1acid glycoprotein after treatment with amitriptyline. Br. J. Clin. Pharmacol. 14, 102–103. De Vane, C.L., Lim, C., Carson, S.W., Tingle, D., Hackett, L., Ware, M.R., 1991. Effect of electroconvulsive therapy on serum concentration of alpha-1-acid glycoprotein. Biol. Psychiatry 30, 116–120. Gram, L.F., 1988. Imipramine: a model substance in pharmacokinetic research. Acta Psychiatr. Scan. 345 (Suppl), 81– 84. Greenblatt, D.J., Sellers, E.M., Koch-Weser, J., 1982. Importance of protein binding for the interpretation of serum or plasma drug concentrations. J. Clin. Pharmacol. 22, 259–263. Healy, D., Calvin, J., Whitehouse, A.M., White, W., Wilton-Cox, H., Theodorou, A.E., Lawrence, K.M., Horton, R.W., Paykel, E.S., 1991. Alpha-1-acid glycoprotein in major depressive and eating disorders. J. Affect. Disord. 22, 13–20. Jann, M.W., Crabtree, B.L., Pitts, W.M., Lam, Y.W., Carter, J.G., 1997. Plasma alpha-one acid glycoprotein and haloperidol concentrations in schizophrenic patients. Neuropsychobioogy 36 (1), 32–36. Joyce, P.R., Hawes, C.R., Mulder, R.T., Sellman, J.D., Wilson, D.A., Boswell, D.R., 1992. Elevated levels of acute phase plasma proteins in major depression. Biol. Psychiatry 32 (11), 1035–1041. Kehoe, W.A., Kwentus, J.A., Sheffel, W.B., Harralson, A.F., 1991. Increased Alpha-1-acid glycoprotein in depression lowers free fraction of imipramine. Biol. Psychiatry 29, 489–493. Kremer, J.M.H., Wilting, J., Janssen, L.H.M., 1988. Drug binding to human alpha-1-acid glycoprotein in health and disease. Pharmacol. Rev. 40, 1–47. Lawrence, K.M., Lowther, S., Falkowski, J., Jacobson, R.R., Horton, W., 1997. Enhanced displacement of [3H]imipramine, but not [3H]paroxetine binding by plasma from depressed patients. J. Affect. Disord. 46, 127–134. Levinson, I., Levine, S., 1995. Negative correlation between alpha-1-acid-glycoprotein plasma level and response to haloperidol in the acute treatment of schizophrenia. Biol. Psychiatry 38 (3), 198–200. Maes, M., Delange, J., Ranjan, R., Meltzer, H.Y., Desnyder, R., Cooremans, W., 1997a. Acute phase proteins in schizophrenia, mania and major depression: modulation by psychotropic drugs. Psychiatry. Res. 66 (1), 1–11. Maes, M., Hendriks, D., Van Gastel, A., Demedts, P., Wauters, A., Neels, H., Janca, A., Scharpe, S., 1997b. Effects of psychological stress on serum immunoglobulin, complement and acute phase protein concentrations in normal volunteers. Psychoneuroendocrinology 22 (6), 397–409. Maes, M., Scharpe, S., Bosmans, E., Vandewoude, M., Suy, E., Uyttenbroeck, W., Cooreman, W., Vandervorst, C., Raus, J., 1992a. Disturbances in acute phase proteins during melancholia: additional evidence for the presence of an inflammatory process during that illness. Prog. Neuropsychopharmacol. Biol. Psychiatry 16 (4), 501–515. Maes, M., Scharpe, S., Van Grootel, L., Uyttenbroeck, W., Cooreman, W., Suy, E., 1992b. High alpha-1-antitripsin, hap-
164
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
toglobin, ceruloplasmin and lower retinol binding protein plasma levels during depression: further evidence for the existence of an inflammatory response during illness. J. Affect. Disord. 24 (3), 183–192. Nemeroff, C.B., Krisnan, R.R., Blazer, D.G., Knight, D.L., Benjamin, D., Meyerson, L., 1990. Elevated plasma concentrations of alpha-1-acid glycoprotein, a putative endogenous inhibitor of the tritiated imipramine binding site, in depressed patients. Arch. Gen. Psychiatry 47, 337–340. Sluzewska, A., Rybakowski, J., Bosmans, E., Sobieska, M., Berghmans, R., Maes, M., Wiktorowicz, K., 1996a. Indicators of immune activation in major depression. Psychiatry Res. 64 (3), 161–167. Sluzewska, A., Rybakowski, J., Sobieska, M., 1996b. Immune activation in endogenous depression. Psychiatr. Pol. 30 (5), 771–782.
Sluzewska, A., Rybakowsky, J.K., Sobieska, M., Wiktorowicz, K., 1996c. Concentration and microhetero-geneity glycophorms of alpha-1-acid glycoprotein in major depressive disorder. J. Affect. Disord. 39, 149–155. Sluzewska, A., Sobieska, M., Rybakowski, J.K., 1997. Change in acute-phase proteins during lithium potentiation of antidepressants in refractory depression. Neuropsychobiology 35 (3), 123–127. Song, C., Dinan, T., Leonard, B.E., 1994. Changes in inmunoglobuline, complement and acute phase protein levels in the depressed patients and normal controls. J. Affect. Disord. 30 (4), 283–288. Wong, C.T., Tsoi, W.F., Saha, N., 1996. Acute phase proteins in male Chinese schizophrenic patients in Singapore. Schizophr. Res. 22 (2), 165–171.
Brief report
Alpha-1-acid glycoprotein in major depressive disorder Relationships to severity, response to treatment and imipramine plasma levels a
a,
b
c
Evaristo Nieto MD , Eduard Vieta MD *, Luisa Alvarez MD , Merce Torra MD , Francesc Colom PhD a , Cristobal Gasto´ MD a a
b
Department of Psychiatry, Hospital Clinic de Barcelona, Villarroel 170, Barcelona 08036, Spain Department of Biochemistry, Hospital Clinic de Barcelona, Villarroel 170, Barcelona 08036, Spain c Department of Toxicology, Hospital Clinic de Barcelona, Villarroel 170, Barcelona 08036, Spain Received 4 May 1999; received in revised form 20 July 1999; accepted 28 August 1999
Abstract Background: Increased plasma levels of alpha-1-acid glycoprotein (AGP) were reported in major depressive disorder. However, the relationship between AGP levels, severity of depression, treatment response and antidepressant levels are still unclear. Methods: Plasma AGP levels were measured in 36 subjects with major depressive disorder before and after a 6-week treatment with imipramine and in 30 controls. Free imipramine plasma levels of depressed patients were measured at 6 weeks. Comparative analysis between depressed patients and controls, between non-responders (N 5 12) and responders (N 5 24), and between severely depressed patients (N 5 14) and moderately depressed patients (N 5 22) were made. Results: Depressed patients had significantly higher mean values of AGP than control subjects. Imipramine non-responders and specially severely depressed patients had significantly greater increases of AGP levels during treatment than other depressed subgroups. There was no correlation between baseline AGP levels and severity of depression or free imipramine levels. Limitations: The most significant limitations of this study are the small sample size and the fact that all the subjects were out-patients. Results should not be generalized to in-patient populations. Conclusions: Depressed patients showed high baseline concentrations of AGP. AGP levels did not predict either free imipramine plasma levels or differential response after 6 weeks of treatment with imipramine. A greater increase of AGP during treatment was associated with severity of depression and treatment non-response. Clinical implications: The relationship between high plasma levels of AGP, severity of depression and lack of treatment response is clarified. The influence of imipramine levels is minimized. 2000 Elsevier Science B.V. All rights reserved. Keywords: Plasma protein; Severity; Treatment response; AGP; Depression; Imipramine
1. Introduction
*Corresponding author.
The alpha-1-acid glycoprotein (AGP) or orosomucid is an acute phase protein and its levels
0165-0327 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0165-0327( 99 )00145-7
160
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
rise during physical illness (infections, cancer, bowel and cardiac diseases etc.) and severe stress such as surgery or burning (Kremer et al., 1988). In several studies (Nemeroff et al., 1990; Healy et al., 1991; Kehoe et al., 1991; Maes et al., 1992a; Song et al., 1994; Sluzewska et al., 1996a–c; Maes et al., 1997a,b; Wong et al., 1996) increased plasma levels of positive acute-phase-proteins such as haptoglobin and AGP have been reported in major depressive disorder, and several more mental disorders (bulimia, schizophrenia, mania), including psychological stress. Although most studies found that AGP levels are elevated in major depression, others (Maes et al., 1992b; Joyce et al., 1992; Lawrence et al., 1997) did not replicate this finding. There are also discrepancies between studies with respect to the correlation between AGP levels and age, severity of illness and treatment response. A further reason for investigating AGP levels in depression is that imipramine and other tricyclic antidepressants bind to AGP (Kremer et al., 1988). Elevations in the AGP levels might effectively reduce the level of unbound drug and thereby militate against a clinical response. This may be particularly true during the initial stages of therapy when both total and free concentrations of the drug are low (Kehoe et al., 1991). Healy et al. (1991) found higher baseline AGP values in depressed nonresponders (imipramine and others) compared to depressed responders. The present study investigated AGP plasma levels and its relationship to severity of the depression, response to treatment and imipramine levels.
2. Methods
2.1. Subjects Thirty-six depressed outpatients were recruited from the Department of Psychiatry in the Hospital Clinic de Barcelona. All patients met DSM-III-R criteria for major depressive disorder and had total scores of at least 17 on the 17-item Hamilton depression rating scale. Patients were psychotropic drug-free for at least seven days and 32 (88%) were antidepressant-free for at least six months. All patients who were found to have any medical or
psychiatric comorbidity were excluded from the study. Pregnant women and bipolar patients were excluded too. The control group, age and sex-matched subjects, was recruited from the faculty of the Hospital Clinic de Barcelona. Controls were drug-free, had no evidence of mental or physical illness and were not pregnant. All subjects (patients and controls) gave informed consent to participate in the study. The Ethics Committee of the Hospital Clinic–University of Barcelona gave its approval to the study.
3. Procedures
3.1. Laboratory tests The a 1 -acid glycoprotein forms immune complexes with specific antibodies to human AGP in the human serum sample. These complexes scatter a beam of light passed through the sample. The intensity of scattered light is proportional to the concentration of AGP in the sample. The result is evaluated by comparison with a standard of known concentration. Baseline levels of AGP (AGP1) were measured by a inmunonephelometric method using reagents from Behring in the BNA (Behringwerke, Marburg,Germany). The antiserum is a liquid animal serum produced by immunization of rabbits with highly purified human AGP. Final levels of AGP (AGP2) of depressed patients were measured at a 6-week follow-up. Blood samples were taken in controls and in depressed patients at 09.00 h. Free imipramine (IMIF) plasma levels were measured by liquid chromatography in depressed patients after 6 weeks of treatment with this drug at the Laboratory of Toxicology in the Hospital Clinic. Serum samples were assayed for imipramine (total and free fraction) using a reversed-phase high-performance liquid cromatography method. The fraction of free imipramine was obtained by ultrafiltration in a centrifuge with a fixed-angle rotor at 2000 3 g, using the Centrifree Micropartition devices (Amicon Inc., Massachussets, USA). Imipramine (total and free fraction) and the internal standard (IS) trimipramine were selectively isolated from the serum by
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
solid-phase sample preparation using the bonded phase extraction columns Bio-Rad 195-7040 (BioRad, Anaheim, USA). Chromatographic separation was achieved in a Hewlett-Packard HP-1050 system (Hewlett-Packard, Waldbronn, Germany), using a reverse-phase column Bio-Rad 195-7051 (oven temperature 5 358C) and an isocratic eluent Bio-Rad 195-7070 delivered at a flow rate of 0.6 ml / min. The column effluent was monitored using ultraviolet (UV) detection at 242 nm, 0.01 AUFS and the injection volume was 100 ml. Quantitation was by peak-height ratios (imipramine / IS). Recovery was 95–98%, whereas within day and day-to-day coefficients of variation were between 4.1 and 6.4%.
3.2. Treatment of depressed patients After blood samples were taken to measure AGP1 levels, all depressed patients were treated with increasing dosages of imipramine (50 mg / day during days 1–3; 100 mg / day during days 4–7; 150 mg / day during days 8–21) until they completed 3 weeks of treatment. Then, subjects were considered to be responders if their HDRS score was ten or lower at this point. In this case imipramine dose of 150 mg / day was sustained until they completed 6 weeks of treatment. Depressed non-responders were treated with increasing doses of imipramine (200 mg / day during days 22–28 and 250 mg / day during days 29–42) until they completed 6 weeks of treatment. After 6 weeks of treatment, final decisions regarding response were made. Subjects were considered to be non responders if their HDRS score was ten or over at this point.
3.3. Severity of depressed patients According to baseline score in HDRS, 14 major depressed patients were considered to have severe depression (initial score in HDRS was higher than 30) and 22 were moderately depressed.
3.4. Statistical methods Comparative analysis of normal variables (AGP1, AGP2, DAGP and IMIF) between two groups (depressed vs. controls, depressed responders vs. nonresponders) was performed by means of an unpaired
161
two-tailed Student t-test. Correlations between AGP measures and baseline HDRS scores and free imipramine plasma levels were made by means of the Spearman’s rank correlation coefficient.
4. Results The mean age of the 36 depressed patients was 43.9 years (SD 5 14.5 and range 5 16–70). Twentynine of them (80.5%) were women and seven (19.5%) were men. There were no significant differences in age or sex between controls and depressed patients. The baseline mean HDRS score for depressed patients was 28.42 (SD 5 5.18 and range 5 18–40) and their mean free imipramine (IMIF) levels were 25.9 ng / ml (SD 5 11.1, range 5 0–58). A comparison of the baseline AGP levels between depressed patients and the control group is presented in Table 1. Table 2 shows the comparison of baseline AGP levels (AGP1), final AGP levels (AGP2), variation of AGP levels (DAGP) and free imipraTable 1 Comparison of alpha-1-acid glycoprotein levels (mg / 100 ml) between depressed patients and controls
Male Female All
Depressed patients (N 5 36)
Control subjects (N 5 30)
N
Mean
SD
N
Mean
SD
7 29 36
86.7 82.1 83.0
24.0 22.9 22.9
6 24 30
73.0 66.9 68.1
10.5 20.2 18.7
P
NS P , 0.02 P , 0.01
Table 2 Comparison of AGP1, AGP2, DAGP (mg / 100 ml) and IMIF (ng / ml) between depressed responders and depressed non-responders to imipramine after 6 weeks of treatment a Responders (N 5 24)
AGP1 AGP2 DAGP IMIF
Non-responders (N 5 12)
Mean
SD
Mean
SD
84.7 85.7 0.95 26.6
20.7 17.9 17.4 10.7
79.5 94.5 15.0 24.5
27.5 26.6 19.9 12.3
P
NS NS P , 0.05 NS
a AGP1 5 Baseline alpha-1-acid glycoprotein; AGP2 5 final alpha-1-acid glycoprotein; DAGP 5 difference between final and baseline AGP levels; IMIF 5 plasma free imipramine levels.
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
162
Table 3 Comparison of AGP1, AGP2, DAGP (mg / 100 ml) and IMIF (ng / ml) between severely depressed and moderately depressed patients a Severe depression (N 5 14)
AGP1 AGP2 DAGP IMIF
Moderate depression (N 5 22)
P
Mean
SD
Mean
SD
76.3 95.7 19.4 27.4
18.2 19.4 18.3 11.6
87.2 84.1 2 3.0 24.9
25.0 21.6 14.2 10.9
NS NS P , 0.000 NS
a
AGP1 5 Baseline alpha-1-acid glycoprotein; AGP2 5 final alpha-1-acid glycoprotein; DAGP 5 difference between final and baseline AGP levels; IMIF 5 plasma free imipramine levels.
mine levels (IMIF) between depressed responders and depressed non-responders. The only significant difference between the subgroups was that DAGP was higher in depressed non-responders. Table 3 shows the comparison of AGP1, AGP2, DAGP and IMIF between severely depressed and moderately depressed patients. The only significant difference between both subgroups was that DAGP was higher in severely depressed patients. The mean AGP levels for the four patients treated with antidepressants in the last six months were 91 mg / 100 ml. The mean AGP level for the patients who were antidepressant-free in the last six months (N 5 32) were 81.9 mg / 100 ml. Differences were not significant. We could not find any significant correlation between AGP variables (AGP1, AGP2 and DAGP) and IMIF after 6 weeks of treatment. When the 17 depressed patients treated with 150 mg / day, the 19 depressed patients treated with 250 mg / day, the 24 depressed responders, the 12 depressed non-responders, the 14 severely depressed, and the 24 moderately depressed were considered separately, correlations still remained not significant. Furthemore, there was no correlation between AGP variables (AGP1, AGP2, DAGP) and the severity of depression as assessed by HDRS scores.
5. Discussion The results obtained in our study are in agreement with previous reports of significantly higher AGP
levels in depressed patients compared to controls (Nemeroff et al., 1990; Healy et al., 1991; Kehoe et al., 1991; Sluzewska et al., 1996a,b; Maes et al., 1997a). Within the spectrum of psychiatric disorders, alterations of AGP and other acute phase proteins are not specific for major depression (Healy et al., 1991; Maes et al., 1997a,b) and may be even more pronounced in schizophrenic than in depressed patients (Maes et al., 1997a). The second finding of this study is that there was a difference between depressed patients on the basis of their response to treatment with imipramine: nonresponders had a significantly greater increase in AGP levels than responders after 6 weeks of treatment. However, our study did not replicate the previous findings of Healy et al. (1991), in the sense that depressed non-responders had significantly higher baseline (prior to treatment) levels of AGP than subsequent responders. To the best of our knowledge, the findings of Healy et al. (1991) in depressed patients have not been replicated. Some reports have analyzed the variations in AGP levels during different treatments such as electro convulsive therapy (De Vane et al., 1991) or lithium potentiation (Sluzewska et al., 1997) but none with imipramine. Levinson and Levine (1995) found mean baseline AGP levels to be significantly higher in schizophrenic non-responders treated 2 weeks with haloperidol than in responders, but in a further study, Jann et al. (1997) could not find any relationship between AGP levels and treatrment response after 2, 4 or 6 weeks of treatment. Kehoe et al. (1991) reported that increased AGP concentration reduced the free fraction of imipramine in depressed patients. We found no significant correlation between AGP levels and free imipramine levels. One possible reason for this disagreement could be that the reduction of the free fraction of imipramine does not necessarily lead to a reduction of free imipramine levels (Greenblatt et al., 1982). On the other hand, Kehoe et al. (1991) determined the free fraction of imipramine during the initial phases of therapy (from the first week to the third week of treatment), whereas we determined the free imipramine levels after 6 weeks of treatment, and steady-state levels vary considerably among individuals mainly due to variations in metabolism, but also to a smaller extent due to variations in binding to plasma proteins (Gram, 1988).
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
A major advantage of our study over previous studies was that we determined simultaneously AGP levels, free imipramine levels and clinical response whereas Healy et al. (1991) did not determine imipramine levels and Kehoe et al. (1991) did not determine clinical response. Other supplementary advantages are that our sample was bigger than the sample of Kehoe et al. and was more homogeneous with respect to treatment than the sample of Healy et al. (all our patients were treated with imipramine, whereas the patients of Healy et al. were treated with several drugs, including imipramine, lofepramine and even electro convulsive therapy). Furthermore, the study of Healy et al. has another disadvantage in that 40% of his depressed patients had been treated with antidepressant drugs in the three months prior to study and we, like Sluzewska et al. (1996c) and Baumann et al. (1982), found that the treatment, and especially resistance to treatment, in depressed patients is associated with changes in AGP levels. Another finding of this study is that there was only one difference between depressed patients on the basis of their severity. Severely depressed patients had significantly greater increments of AGP levels during 6 weeks of treatment than moderately depressed patients. The severity of depression seemed to be more deeply related to the increase in AGP during treatment than non-response. Taking only into account the 24 depressed responders, the seven more severely depressed patients had significantly greater increases during treatment (P , 0.05) than the 17 moderately depressed patients. On the other hand, our findings are in agreement with the majority of previous investigations that show a lack of correlation between baseline AGP levels and severity (according to baseline HDRS score) in depressed patients (Healy et al., 1991; Sluzewska et al., 1996b,c). Only Nemeroff et al. (1990) found a correlation between plasma AGP levels and depression severity as assessed by the Montgomery–Asberg rating scale. In summary, depressed patients showed elevated plasma concentrations of AGP. However, AGP levels did not predict either free imipramine plasma levels or a differential response after 6 weeks of treatment with imipramine. A higher increase of AGP during treatment was associated with higher baseline severity and lack of response to treatment with imipramine.
163
References Baumann, P., Tingwely, D., Schopf, J., 1982. Increased of alpha-1acid glycoprotein after treatment with amitriptyline. Br. J. Clin. Pharmacol. 14, 102–103. De Vane, C.L., Lim, C., Carson, S.W., Tingle, D., Hackett, L., Ware, M.R., 1991. Effect of electroconvulsive therapy on serum concentration of alpha-1-acid glycoprotein. Biol. Psychiatry 30, 116–120. Gram, L.F., 1988. Imipramine: a model substance in pharmacokinetic research. Acta Psychiatr. Scan. 345 (Suppl), 81– 84. Greenblatt, D.J., Sellers, E.M., Koch-Weser, J., 1982. Importance of protein binding for the interpretation of serum or plasma drug concentrations. J. Clin. Pharmacol. 22, 259–263. Healy, D., Calvin, J., Whitehouse, A.M., White, W., Wilton-Cox, H., Theodorou, A.E., Lawrence, K.M., Horton, R.W., Paykel, E.S., 1991. Alpha-1-acid glycoprotein in major depressive and eating disorders. J. Affect. Disord. 22, 13–20. Jann, M.W., Crabtree, B.L., Pitts, W.M., Lam, Y.W., Carter, J.G., 1997. Plasma alpha-one acid glycoprotein and haloperidol concentrations in schizophrenic patients. Neuropsychobioogy 36 (1), 32–36. Joyce, P.R., Hawes, C.R., Mulder, R.T., Sellman, J.D., Wilson, D.A., Boswell, D.R., 1992. Elevated levels of acute phase plasma proteins in major depression. Biol. Psychiatry 32 (11), 1035–1041. Kehoe, W.A., Kwentus, J.A., Sheffel, W.B., Harralson, A.F., 1991. Increased Alpha-1-acid glycoprotein in depression lowers free fraction of imipramine. Biol. Psychiatry 29, 489–493. Kremer, J.M.H., Wilting, J., Janssen, L.H.M., 1988. Drug binding to human alpha-1-acid glycoprotein in health and disease. Pharmacol. Rev. 40, 1–47. Lawrence, K.M., Lowther, S., Falkowski, J., Jacobson, R.R., Horton, W., 1997. Enhanced displacement of [3H]imipramine, but not [3H]paroxetine binding by plasma from depressed patients. J. Affect. Disord. 46, 127–134. Levinson, I., Levine, S., 1995. Negative correlation between alpha-1-acid-glycoprotein plasma level and response to haloperidol in the acute treatment of schizophrenia. Biol. Psychiatry 38 (3), 198–200. Maes, M., Delange, J., Ranjan, R., Meltzer, H.Y., Desnyder, R., Cooremans, W., 1997a. Acute phase proteins in schizophrenia, mania and major depression: modulation by psychotropic drugs. Psychiatry. Res. 66 (1), 1–11. Maes, M., Hendriks, D., Van Gastel, A., Demedts, P., Wauters, A., Neels, H., Janca, A., Scharpe, S., 1997b. Effects of psychological stress on serum immunoglobulin, complement and acute phase protein concentrations in normal volunteers. Psychoneuroendocrinology 22 (6), 397–409. Maes, M., Scharpe, S., Bosmans, E., Vandewoude, M., Suy, E., Uyttenbroeck, W., Cooreman, W., Vandervorst, C., Raus, J., 1992a. Disturbances in acute phase proteins during melancholia: additional evidence for the presence of an inflammatory process during that illness. Prog. Neuropsychopharmacol. Biol. Psychiatry 16 (4), 501–515. Maes, M., Scharpe, S., Van Grootel, L., Uyttenbroeck, W., Cooreman, W., Suy, E., 1992b. High alpha-1-antitripsin, hap-
164
E. Nieto et al. / Journal of Affective Disorders 59 (2000) 159 – 164
toglobin, ceruloplasmin and lower retinol binding protein plasma levels during depression: further evidence for the existence of an inflammatory response during illness. J. Affect. Disord. 24 (3), 183–192. Nemeroff, C.B., Krisnan, R.R., Blazer, D.G., Knight, D.L., Benjamin, D., Meyerson, L., 1990. Elevated plasma concentrations of alpha-1-acid glycoprotein, a putative endogenous inhibitor of the tritiated imipramine binding site, in depressed patients. Arch. Gen. Psychiatry 47, 337–340. Sluzewska, A., Rybakowski, J., Bosmans, E., Sobieska, M., Berghmans, R., Maes, M., Wiktorowicz, K., 1996a. Indicators of immune activation in major depression. Psychiatry Res. 64 (3), 161–167. Sluzewska, A., Rybakowski, J., Sobieska, M., 1996b. Immune activation in endogenous depression. Psychiatr. Pol. 30 (5), 771–782.
Sluzewska, A., Rybakowsky, J.K., Sobieska, M., Wiktorowicz, K., 1996c. Concentration and microhetero-geneity glycophorms of alpha-1-acid glycoprotein in major depressive disorder. J. Affect. Disord. 39, 149–155. Sluzewska, A., Sobieska, M., Rybakowski, J.K., 1997. Change in acute-phase proteins during lithium potentiation of antidepressants in refractory depression. Neuropsychobiology 35 (3), 123–127. Song, C., Dinan, T., Leonard, B.E., 1994. Changes in inmunoglobuline, complement and acute phase protein levels in the depressed patients and normal controls. J. Affect. Disord. 30 (4), 283–288. Wong, C.T., Tsoi, W.F., Saha, N., 1996. Acute phase proteins in male Chinese schizophrenic patients in Singapore. Schizophr. Res. 22 (2), 165–171.