Fluvoxamine and chlorimipramine in endogenous depression

Journal of Affective Elsevier Biomedical

249

Disorders, 4 (1982) 249-259 Press

Fluvoxamine

and Chlorimipramine Depression

in Endogenous

J.E.M. De Wilde’ and D.P. Doogan2 ’ Geneeskundig Institute St. Joze/, Weststrant 135, 9940 Slerdinge (Belgium) and ’ Department of Clinical Research, Duphar B V. Weesp (The Netherlands) (Received 21 December, 1981) (Accepted 29 March, 1982)

Summary Thirty patients were treated in a randomised double-blind efficacy study of fluvoxamine and chlorimipramine. The dose range for both drugs was 50-300 mg in divided daily doses. Mean daily doses were higher for fluvoxamine than chlorimipramine. Generally the baseline recordings were comparable for both drug groups. Fluvoxamine was superior to chlorimipramine in all the rating scales used without achieving statistical significance. Chlorimipramine, but not fluvoxamine, caused a significant decrease in blood pressure. There were no significant effects on ECG or laboratory variables. There was no significant relationship between plasma levels of either compound or metabolite and clinical response. Chlorimipramine exerted more unwanted effects than fluvoxamine. Autonomic effects of fluvoxamine were minimal in comparison with chlorimipramine. Chlorimipramine patients required more concurrent anxiolytic medication than fluvoxamine. Both drug groups required a significant amount of concurrent hypnotic medication.

Introduction Fluvoxamine, one of the series of 2-aminoethyloximethers Duphar BV, Weesp, The Netherlands. It was selected from

was synthesised by the group as being

Requests for reprints should be sent to: D.P. Doogan, Sandwich, Kent, Great Britain.

Research,

0165-0327/82/0000-0000/$02.75

0

1982

Department

Elsevier Biomedical

Press

of Clinical

Pfizer Limited,

250

particularly interesting in view of its potential antidepressant activity. It was also noted to be unique in its effects on neurotransmitters by being a selective 5-hydroxytryptamine (5HT, serotonin) reuptake inhibitor at brain synaptosomes. It has negligible effects by comparison, on noradrenaline (NA) and dopamine (DA) reuptake. Furthermore, from animal pharmacology data it possesses no monoamine oxidase inhibiting activity, has no amphetamine-like effects and is devoid of anticholinergic activity (Claassen et al. 1977). Fluvoxamine has no known pharmacologically active metabolites. The plasma half-life in humans is approximately 15 h and favourable safety and tolerance results in human volunteers were shown in early studies (Doogan 1980). Initial efficacy studies showed fluvoxamine to have antidepressant activity in oral doses between 50 and 300 mg/day (Itil et al. 1977; Saletu 1977; Wright and Denber 1978). In order to profile the drug and to further evaluate its efficacy double blind studies were necessary. Chlorimipramine was chosen in our study as the reference compound since it is widely accepted as being antidepressant and also the most potent of available antidepressants on 5HT reuptake (Waldemeier et al. 1976).

Materials

and Methods

Thirty hospitalised patients were evaluated in this 4-week double blind study. Fifteen patients were treated with each compound. The patient characteristics are displayed in Table 1. Inclusion criteria demanded that all patients have depression with at least 4 items of the Feighner Criteria (Feighner et al. 1977) and an initial Hamilton Rating Score (Hamilton 1967) of 17 or more on the first 17 items. No concurrent medical illness or other psychiatric disorder was allowed. Pregnant women or those of child-bearing potential were excluded. Concurrent psychotropic medication was limited to minor tranquillisers for insomnia or anxiety. Pre-treatment comparability between the two drug groups was satisfactory for both demographic and clinical characteristics. The severity of illness in all cases was either severely or extremely ill. Two fluvoxamineand one chlorimipramine-treated patients were suffering their first depressive episode. Six fluvoxamine patients and 3 chlorimipramine patients had received antidepressant therapy for the current depressive episode with poor or equivocal response. After a washout period of between 3 and 12 days patients were admitted to the drug treatment period. All patients satisfied the entry criteria. Treatment allocation was double blind, randomised and stratified by diagnosis of unipolar or bipolar depression. Bipolar illness was considered to be present in any patient with a previous history of mania. Assessments were undertaken on entry to the study, at the end of the washout period and weekly thereafter for 4 weeks. Psychiatric assessments consisted of the Hamilton Rating Scale for Depression 17 items, Clinical Global Impression and a 33-item depressive symptom checklist.

251 TABLE

1

PATIENT

CHARACTERISTICS Fluvoxamine patients (N= IS)

Chlorimipramine patients (N= 15)

4 II

IO

Age Mean (yr) SD Range

46.21 10.74 27-60

45.33 12.75 18-64

Depression Unipolar Bipolar

II (9F. 2M) 4 (2F, 2M)

IO (6F, 4M) 5 (4F. IM)

Endogenous Non endogenous

15 0

14

Sex

Male Female

Duration Mean SD

(M) (F)

of current

episode

of depression 33.0 9.2

5

(days) 34.7 14.5

Safety and tolerance were assessed by means of electrocardiogram, blood pressure, pulse, haematology, blood biochemistry and unwanted effects elicited at interview.

Results Drug and dosage All patients received medication for a minimum of 28 days. The weekly assessments at the end of the 4th treatment week were analysed and no patient was excluded from statistical analysis. The permissible dosage range for both compounds was 50-300 mg/day depending on clinical response. Patients receiving chlorimipramine had a dose titration of 50 mg, 75 mg and 100 mg on the first 3 treatment days, respectively. Where possible, all patients received drug in 3 equally divided daily doses. The mean doses of fluvoxamine were 15 1, 205, 235 and 259 mg/day for the 4 treatment weeks, respectively. The mean daily doses of chlorimipramine were 125, 186, 218 and 231 mg/day, respectively, for each of the 4 treatment weeks. Efficacy Hamilton Rating Scale for Depression (HRSD) When individual items of the HRSD were statistically

analysed

pre-treatment

252

comparability was found to be satisfactory between the drug groups. There were no statistically significant differences between the groups during any of the treatment weeks with the exception of the following: On the item “insomnia, late”, fluvoxamine patients scored less at the end of the first and third treatment weeks (P = 0.035 and 0.022 respectively, Wilcoxon). Table 2 shows the scores on the Hamilton ratings at baseline and each subsequent treatment week. For the mean total scores it is apparent that both compounds were associated with significant improvement from pre-treatment scores. The median total scores at the end of the 4th treatment week being 6 and 9 for fluvoxamine and chlorimipramine patients, respectively. The tendency was for superiority of fluvoxamine over chlorimipramine. However, the treatment differences were found not to be significant at the 5% level. The percentage improvement on the HRSD was calculated from the formula: Pre-treatment

score - end of treatment Pre-treatment score

score x 1oo

Again, an appreciable improvement was seen for both drugs at the 4th treatment week. The median percentages were 76 and 70 for fluvoxamine and chlorimipramine, respectively, without significant difference between the drugs at the 5% level. If a decrease in HRSD of 50% or less from baseline is regarded as being a criterion for 2 fluvoxamine treated and 5 chlorimipramine patients could be non-response, regarded as non-responders. There were no characteristics present in these patients

TABLE

2

RESULTS

OF HRSD

PRIOR

TO AND

DURING

TREATMENT

PERIOD

The multivariate test, tests the null hypothesis of no difference occurring between the time-response relationships for the two drugs against the alternative hypothesis that such a difference exists. Chlorimipramine

Fluvoxamine Baseline

Total HRSD Mean Median SD

29.2 29 4.3

% change from baseline Mean _ Median -

N.S. = not significant.

Treatment

Baseline

weeks

1

I

2

3

26.6 26 4.1

19.8 19 4.4

14.0 13 5.9

7.9 6 6.7

I

52.5 56

72.9 76

8.8 8

32. 37

Treatment

4

weeks 2

3

4

28.9 29 3.1

25.5 20.7 26 22 3.6 4.7 (P =0.35) N.S. Multivariate

15.5 15 5.7

11.1 9 8.5

_

11.8 28.3 15 32 (P =0.22) N.S. Multivariate

46.6 53

62. 70

I

253

which could be highlighted as being different from the remaining responding patients. Analysis of Hamilton factors (anxiety, retardation, sleep disturbance and cognitive disturbance) revealed no significant differences between either drug group. There were 4 bipolar patients in the fluvoxamine group of whom 1 did not improve satisfactorily by the 4th week. Four of the 5 bipolar patients in the chlorimipramine group did not achieve a satisfactory improvement. There was no precipitation of mania or hypomania. Clinical Global Impression (CGI) Table 3 displays the weekly scores on the severity item of the CGI. A score of 1 was regarded as normal and 7 as being extremely ill. Scores of 2-6 were of intermediate severity. There was no significant difference between the two drugs. Median severity for both drug groups was 2.0 (borderline mentally ill) by the end of 4 weeks treatment. The patients who had a less than 50% decrease in HRSD all had the highest scores in this assessment (4, 5 or 6) at week 4. The other item rated on the CGI was global change in condition at each treatment week assessment as compared with the severity of mental illness on entry to the study. This was also rated on a 7-point scale ranging from 1 = (very much improved) to 4 = (no change) to 7 = (very much worse). Rating by this item shows that both groups improved significantly. Again, the tendency was for superiority of fluvoxamine by the 4th treatment week (median score 1.0 and 2.0 for fluvoxamine and chlorimipramine, respectively). This was not statistically significant at the 5% level. TABLE

3

RESULTS

OF WEEKLY

EVALUATION

USING

Fluvoxamine Baseline

Severity

CLINICAL

GLOBAL

IMPRESSION

Chlorimipramine Treatment

weeks

Baseline

I

2

3

4

5.8 6.0

4.8 5.0

4.0 4.0

2.4 2.0

3.5 3.0

2.7 3.0

2.5 2.0

1.6 1.0

Treatment

1

weeks 3

4

6.0 5.7 4.8 6.0 6.0 5.0 Treatment week 4 (P =0.21) N.S. (Wilcoxon)

3.9 4.0

2.8 2.0

3.4 2.9 3.0 3.0 Treatment week 4 (P=O.48) N.S. (Wilcoxon)

2.3 2.0

I .9 2.0

2

score

Mean Median

6.3 6.0

Improvement

score

Mean Median

_ -

N.S. = not significant.

254

Depressive symptom checklist This scale was a checklist of 33 symptoms/items known to be associated with depression. It is not, however, a validated scale. Each item was rated on a scale of l-3. It was completed prior to treatment and at the end of the study. The distribution of pretreatment severity of all items was equal for both drugs except for items ‘blames self for depression’ (P = 0.036) ‘delusions’ (P = 0.10) and ‘overworried’ (P = 0.030). For these items fluvoxamine patients scored significantly higher. At the end of treatment there was no difference between the two drug groups except in ‘decreased motor activity’, which improved more with fluvoxamine (P = 0.036). Other items which were significantly different at the end of treatment were discounted due to poor pretreatment comparability. Curdiovasculur responses Blood pressure and pulse in both erect and supine positions were measured at baseline and weekly. Both patient groups had a decrease in heart rate (mean 8 beats for fluvoxamine and 1 beat/min for chlorimipramine). This was statistically but not clinically significant, only in the 3rd treatment week (P = 0.019 and 0.017 for supine and erect positions, respectively). Chlorimipramine patients had a significant systematic decrease in upright systolic blood pressure of 11.6 mm Hg whereas the effect was only marginal for fluvoxamine patients (2.0 mm Hg). During the 2nd, 3rd and 4th treatment weeks the difference in blood pressures was significant between the two groups (P = 0.038, 0.021 and 0.056 for the 2nd, 3rd and 4th weeks, respectively). The results of electrocardiogram recordings demonstrated that there were no changes of clinical significance except for 2 patients: 587 (chlorimipramine) and 596 (fluvoxamine). Patient 587 had a normal pretreatment recording which showed changes of mild repolarisation disturbance in the left precordial region at the 4th treatment week. Patient 596 also had a normal pretreatment ECG which by week 4 of treatment showed a flattening of T-waves in standard lead I. No causal relationship between these changes and drug therapy has been suggested. Laboratory data For haematological variables there were no detectable differences between the treatment groups except for the red blood cell count which rose slightly in both groups, (P = 0.018 multivariate test). The mean increase was 0.3 X 106/mm3 for fluvoxamine and 0.1 X 106/mm3 for chlorimipramine. In blood chemistry both groups showed a rise in bilirubin (no values were elevated above normal). There were no other clinically relevant abnormalities. Plusmu levels and clinical response Blood was taken by venepuncture for drug plasma the first intake of the drug for that day. This was done treatment weeks. The mean plasma levels of fluvoxamine were 282.3 and 4th treatment weeks, respectively (Table4). For

levels in the morning prior to at the end of the 2nd and 4th and 514.5 ng/ml chlorimipramine

for the 2nd patients the

255 TABLE 4 FLUVOXAMINE Patient number

PLASMA

LEVELS

WITH

DOSAGE

Week 2

AND CLINICAL

RESPONSE

Week 4

@g/ml)

(mg)

% HRSD change

@g/ml)

(mg)

0 HRSD change

578 580 583 585 586 589 596 597 599 602 604

313 222 376 557 42 192 403 80 295 284 341

150 225 225 150 225 225 225 225 225 225 225

42 42 41 16 37 32 27 17 42 17 12

257 149 498 918 78 711 894 175 616 506 858

150 225 225 300 300 300 300 225 300 300 300

62 88 90 19 70 76 34 87 94 75 58

Mean N=ll

282.3

211.4

29.5

514.5

265.9

68.5

Plasma

level

Daily dose

Plasma

level

Daily dose

mean plasma levels of chlorimipramine were 116.1 and 146.2 ng/ml for the 2nd and 4th treatment weeks,respectively. Mean desmethylchlorimipramine levels for the two weeks were 234.9 and 344.8 ng/ml (Table5). There appeared to be a good correlation between dosage of both drugs and the respective plasma levels. The plasma levels of chlorimipramine and desmethylchlorimipramine were in the therapeutic range (Mulgirigama et al. 1977). Statistical examination revealed no relationship between plasma levels and clinical response to either compound. The common correlation coefficient for the 2nd treatment week was r = 0.076 (P = 0.72) and for the 4th treatment week was r = 0.184 (P = 0.38). There was no apparent correlation for fluvoxamine, chlorimipramine or desmethylchlorimipramine when plotted graphically versus clinical response. The two patients who responded least to treatment (585 and 604) had the highest fluvoxamine levels of 918 and 858 ng/ml, respectively. For chlorimipramine, patients 582 and 600 had the least response by the 4th treatment week. The drug plasma levels for patient 582 were 74 and 194 ng/ml for parent drug and metabolite, respectively. For patient 600 the levels were 198 and 216 ng/ml for parent drug and metabolite, respectively. There was no correlation between plasma levels and unwanted effects. Unwanted signs and symptoms (Table 6) Fluvoxamine patients experienced dry mouth, tremor and sweating. The chlorimipramine patients also had these complaints but with greater incidence. The other unwanted complaints for chlorimipramine patients were headache, difficulty with

N= 14

Mean

573 574 576 571 579 581 582 584 590 598 600 601 603 605

Patient number

116.1

95 224 84 110 126 38 96 55 150 259 150 75 123 41

CHL

DMC

234.9

129 500 232 202 265 70 174 194 412 365 201 174 189 121

203.6

225 225 225 225 150 225 225 150 225 225 225 150 150 225

Daily dose (mg)

Plasma level

(ng/ml)

28.7

48 21 33 38 56 32 17 33 33 33 3 15 21 19

% HRSD change

(CHL) AND DESMETHYLCHLORIMIPRAMINE

Week 2

CHLORIMIPRAMINE

TABLE 5 (DMC)

146.2

124 326 166 106 74 167 74 76 148 282 158 86 173 87

CHL

(ng/ml)

Plasma

Week 4

PLASMA

level

WITH

344.8

196 519 568 306 202 198 194 260 773 543 216 223 322 247

DMC

LEVELS

241.1

225 225 300 225 225 225 225 225 300 225 225 225 225 300

(mg)

RESPONSE

63.4

89 55 61 97 78 84 28 83 81 93 6 46 46 41

% HRSD change

AND CLINICAL

Daily dose

DOSAGE

251 TABLE 6 UNWANTED

SIGNS

AND SYMPTOMS Fluvoxamine

Symptom

Number patients

Chlorimipramine

of

Severity

Number patients

Tremor

2

Mild/moderate

9

Dry mouth Sweating

3 2

Mild Mild/severe

6 6

Headache Difficult micturition Blurred vision Vertigo

0

_

1

0

_

1

of

1 1

0 0

micturition and blurred vision. which occurred in 2 fluvoxamine

Severity

Mild/severe (P (0.05) Mild (N.S.) Mild/moderate (N.S.) Moderate Mild/moderate Mild/moderate Mild

Trem.or was the most fequently occurring symptom patients and 9 chlorimipramine patients (P = 0.023).

Concurrent medication Although not recorded, per se, as unwanted symptoms, from Table 7 it is obvious that insomnia and anxiety were frequent occurrences. Hypnotic medication was required in 13 fluvoxamine and 15 chlorimipramine patients. Anxiety appeared to be more prevalent in the chlorimipramine group as tranquillisers were prescribed to 10 chlorimipramine patients and only 5 fluvoxamine patients. There were 3 chlorimipramine and 1 fluvoxamine patients with pretreatment agitation. Therefore the

TABLE 7 NUMBER

OF PATIENTS

RECEIVING

CONCURRENT

MEDICATION

Class of drug

Fluvoxamine (N=15)

Chlorimipramine (N= 15)

Hypnotic ‘) Tranquilliser b, Hypertensive ‘) Coronary vasodilator d,

13 5 1

15 10 6

1

0

a) b, ‘) d,

Includes Includes Includes Includes

flunitrazepam, nitrazepam, lorazepam. bromazepam, oxazepam. etilefrine hydrochloride. dihydroergotamine mesylate.

258

difference in frequency of prescription of tranquillisers cannot be accounted for by pretreatment complaints of anxiety or agitation. Six chlorimipramine patients and 1 fluvoxamine patient required anti-hypotensive medication during the treatment period. However, hypotension was not recorded as an unwanted effect of either drug.

Discussion

Prior to the initiation of the study, there were no commercially available specific serotonin reuptake inhibitors. Chlorimipramine, was the most potent of the then available 5HT reuptake inhibitors and was chosen as the reference compound in spite of having potent effects on reuptake of noradrenaline. These effects are mediated, not only through the parent drug but its metabolite desmethylchlorimipramine which rapidly attains higher levels in the plasma than the parent (Mellstrom and Tybring 1977). This latter finding was confirmed in the present study. There was much to point to the superiority of fluvoxamine with regard to not only efficacy but safety and tolerance. The superior efficacy was shown on all 3 depression scales. However, there were no symptoms present in the baseline period which could be regarded as being predictive of either positive or negative clinical response. With small numbers of patients per group it was unlikely that statistically significant differences between the drugs could be found. In order to calculate what size of difference necessary to be significant at the 5% level with 15 patients per treatment group the following was calculated. Estimation of the standard deviation of % change in HRSD = 25. Degree of freedom, u = 28; (Y= 0.05; Student 1 distribution = 18.7%. Therefore differences between the two treatment groups of less than 18.7% on the percentage change item of the HRSD would be not significant at the 5% level with 15 patients per treatment group. The dose of both drugs was based on clinical response and tolerance. The mean fluvoxamine dose at each treatment week was noticeably higher than for chlorimipramine. This suggests that the chlorimipramine dosage was limited by lack of tolerance (see Tables 6 and 7). Insomnia was a significant symptom in both drug groups requiring concurrent medication but anxiety, agitation and hypotension were significantly more prominent with chlorimipramine. The dose range of chlorimipramine recommended by the manufacturers 30- 150 mg/day (Martindale 1977) was exceeded in all but the first treatment week, but these patients were usually severely depressed, warranting such doses. Speculation on differences in pharmacological action and clinical activity has not previously been rewarding. However, comparing the sedative activity of the tricyclic antidepressants, it appears that secondary amines such as nortriptyline and desipramine with purer NA effects are more activating than the tertiary amines (Carlsson 1976; Rand and McCulloch 1977). The metabolite desmethylchlori-

259

mipramine, which is a potent reuptake inhibitor of both noradrenaline and serotonin (Carlsson 1976; Fuller and Wong 1977) was present in excess of the parent drug in the plasma and may have promoted more stimulating effects than chlorimipramine. Fluvoxamine, which was not a sedative compound in pharmacological tests and without amphetamine-like effects in animals, may by its serotonin effects, be less activating than compounds with noradrenaline activity. Although devoid of anticholinergic activity in animal experiments the most frequent unwanted effects of fluvoxamine were dry mouth, sweating and tremor. Whether or not these symptoms were mediated through cholinergic mechanisms is arguable but the effects were less frequent than with chlorimipramine. There was no definite correlation between plasma levels and clinical response. However, the patients with high plasma levels of fluvoxamine appeared to have a poorer response than those with lower levels. This study lends support to serotonin being important in the pathogenesis of depression (Van Praag 1980). This is especially so in view of the fact that the noradrenaline effects of fluvoxamine are negligible and unlikely to significantly contribute to its efficacy. Further controlled studies are warranted with fluvoxamine to support the evidence of efficacy suggested by this study. References Carlsson, A., The contribution of drug research to investigating the nature of endogenous depression. Pharmacopsychiatry, 9 (1976) 2-10. Claassen, V., Davies, J.E., Hertting, G. and Placheta. P., Fluvoxamine. a specific 5-hydroxytryptamine reuptake inhibitor, Brit. J. Pharmacol., 60 (1977) 505-5 16. Doogan, D.P.. Fluvoxamine as an antidepressant, Neuropharmacology, 19 (12) (1980) 12151216. Feighner, J.P., Robins, E., Guze, S.B., Woodruff, R.A.. Winokur, G. and Mmioz. R.. Diagnostic criteria for use in psychiatric research, Arch. Gen. Psychiat.. 26 (1977) 57-63. Fuller, R.W. and Wang, D.T., Inhibition of serotonin uptake, Fed. Proc.. 36 (8) (1977) 2154-2158. Hamilton, M., Development of a rating scale for primary depressive illness, Brit. J. Social. Clin. Psychol., 6 (1967) 278-296. Itil, T.M.. Bhattachyaryya, A., Polvan, N., Huque, M. and Menon. G.N., Fluvoxamine (DU 23000). a new antidepressant Quantitative pharmaco-electroencephalography and pilot clinical trials, Progr. Neuro-Psychopharmacol., 1 (1977) 309-322. Martindale, In: A. Wade (Ed.), The Extra Pharmacopoeia, The Pharmaceutical Press. London, 1977, pp. 1210. Mellstrom, B. and Tybring, G., Ion-pair liquid chromatography of steady state plasma levels of chlorimipramine and desmethyl chlorimipramine, J. Chromatogr., 143 (1977) 597-605. Mulgirigama, L.D., Pare, C.M.B., Turner, P., Wadsworth, J. and Witts, D.J.. Clinical response in depressed patients in relation to plasma levels of tricyclic antidepressants and tyramine pressor response, Postgrad. Med. J., 53 (Suppl. 4) (1977) 155-159. Rand, M.J. and McCulloch, M.W., Modes of action of antidepressants. In: G.D. Burrows (Ed.). Handbook of Studies on Depression, Excerpta Medica, Amsterdam. 1977. Saletu, B., Schjerve, M., Grunberger, J., Schanda, H. and Arnold, O.H., Fluvoxamine: A new serotonin reuptake inhibitor - First clinical and psychometric experiences, J. Neural Transm.. 41 (1977) 17-36. Van Praag, H.M., Central monoamine metabolism in depression, Part 1 (Serotonin and related compounds), Compreh. Psychiat., 21 (1980) 30-43. Waldemeier, P.C., Baumann, P., Greengrass, P.M. and Maitre, L.. Effects of chlorimipramine and other tricyclic antidepressants on biogenic amine uptake and turnover, Postgrad. Med. J., 52 (Suppl. 3) (1976) 33-39. Wright, J.H. and Denber H.C.B., Clinical trial of fluvoxamine: A new serotonergic agent, Current Therap. Res., 23 (1) (1978) 83-89.