Escitalopram versus nortriptyline: How to let the clinical GENDEP data tell us what they contained
Version of Record online: 10 MAR 2013
© 2013 John Wiley & Sons A/S. Published by Blackwell Publishing Ltd
Acta Psychiatrica Scandinavica
Volume 127, Issue 4, pages 328–329, April 2013
How to Cite
P., B., P., A., E. R., L., C., C. and R. W., L. (2013), Escitalopram versus nortriptyline: How to let the clinical GENDEP data tell us what they contained. Acta Psychiatrica Scandinavica, 127: 328–329. doi: 10.1111/acps.12067
- Issue online: 10 MAR 2013
- Version of Record online: 10 MAR 2013
- European Commission Framework 6 . Grant Number: LSMB-CT-2003-503428
GENDEP is an acronym for Genome-based Therapeutic Drugs for Depression. The GENDEP trial is a partially randomized multi-centre clinical study comparing the antidepressive effect of escitalopram and nortriptyline.
In their first report  the authors made a factor analysis across all three primary outcome rating scales. This produced three factors with items from the different scales. When using these factors, the GENDEP data were not able to discriminate with clinical significance between the antidepressive effects of the two therapeutic drugs for depression .
We have made an attempt to let the data from the Hamilton Depression Scale (HAM-D), which was actually the first listed GENDEP primary outcome scale, tell us what they really show when using clinimetrically based analyses . We have consequently used both the full HAM-D17 and its six-item subscale (HAM-D6) which measures the pure antidepressive effect . These six items are those which experienced psychiatrists use to give global assessments of depression severity . The HAM-D6 includes the core symptoms of depressed mood, guilt, work and interests, psychomotor retardation, psychic anxiety and general somatic. In several drug trials, the HAM-D6, in contrast to HAM-D17, has been able to detect the pure antidepressive effect of antidepressants .
In our GENDEP analysis we focused on the 765 patients with major depression who had a complete rating scale data set at baseline. After 8 weeks of therapy 569 patients were completers (74%) and after 12 weeks, 496 patients (65%), this is quite acceptable .
In total, 427 of the 765 patients received escitalopram (mean dose 18 mg daily) and 338 received nortriptyline (mean dose 105 mg daily).
At baseline 38 patients had doubtful depression (HAM-D17 < 13); 92 patients had a mild depression (HAM-D17 = 13–17); 405 patients had moderate depression (HAM-D17 = 18–24) and 230 patients had severe depression (HAM-D17 = 25–52). The distributions of patients receiving escitalopram or nortriptyline within these four categories were equal. Rasch analysis (3) of the HAM-D6 across the GENDEP rating occations showed acceptable transferability.
Based on the HAM-D6 as a continuous outcome measure, effect size statistics were used for the evaluation of response to treatment .
Furthermore, any drug–drug difference reported through computing number needed to treat (NNT) was used for obtainment of remission after 12 weeks of therapy, applying a HAM-D6 cut-off score of <5 .
The intention to treat approach was applied when evaluating response (effect size) and remission (NNT). The last observation carried forward method (LOCF) was used, but for effect size the mixed model was also used.
The results showed an effect size of 0.31 (LOCF) and 0.28 (mixed model) in favour of escitalopram. This level is of clinical significance when comparing two different classes of drugs for depression .
We found a remission rate of 70% for mild depression in the escitalopram-treated patients and 46% in the nortriptyline-treated patients, resulting in an NNT of 5. For moderate depression the remission rates were 49% and 34%, respectively, resulting in an NNT of 7. For severe depression the remission rates were 42% and 22%, respectively, resulting in an NNT of 5. These NNT ranges between 5 and 7 are of clinical significance and equal an effect size of approximately 0.3 . In other words, the superiority of escitalopram over nortriptyline was independent of baseline depression severity .
By making the GENDEP data set tell us what they actually are about rather than making factor analysis tell them how they ought to have behaved, we were able to demonstrate the superiority of escitalopram over nortriptyline in pure antidepressive effect, no matter whether the data were analysed as continuous measures (effect size) or categorical measures (remission).
The GENDEP study was funded by the European Commission Framework 6 grant. EC Contract Ref.: LSMB-CT-2003-503428. H Lundbeck A/S provided both nortriptyline and escitalopram free of charge for the GENDEP study. The sponsor had no role in the design and conduct of the study, in data collection, in analysis, interpretation or writing the report.
The authors would like to thank the GENDEP steering group for having access to the data set.The results reported here do not necessarily, however, represent the line of thoughts of the GENDEP steering group.
Declaration of interest
Per Bech, Peter Allerup, Erik Roj Larsen and Claudio Csillag have no conflicts of interest to declare. For the last 3 years Rasmus W. Licht has served on the advisory boards of Bristol-Myers Squibb, Astra-Zeneca and MSD; has received honoraria for lectures for Eli Lilly, Janssen-Cilag, GlaxoSmithKline, Bristol-Myers Squibb, Pfizer, Astra-Zeneca and Lundbeck, and honoraria for other services for Astra-Zeneca and Lundbeck.
- 3Clinical psychometrics. :Wiley Blackwell, 2012..