A meta-analysis of 12 second-generation antidepressants suggests that the treatments that shows the best balance in terms of efficacy, tolerability and cost are Pfizer’s Zoloft and Lundbeck/Forest Laboratories’ Cipralex/Lexapro.
The meta-analysis, which has been published in The Lancet, was carried out by Andrea Cipriani of the University of Verona, Italy and colleagues. The researchers reviewed 117 studies from 1991 to 2007 involving 25,928 people with major depression.
They found that Zoloft (sertraline), Cipralex (escitalopram), Organon’s Remeron (mirtazapine) and Wyeth’s Effexor (venlafaxine) were “significantly more efficacious” than Eli Lilly/Boehringer Ingelheim’s Cymbalta (duloxetine), Lilly’s Prozac (fluoxetine), Solvay’s Luvox (fluvoxamine), GlaxoSmithKline’s Paxil/Seroxat (paroxetine) and Pfizer’s Edronax (reboxetine).
The latter was “significantly less efficacious than all the other antidepressants tested”, the researchers wrote. Escitalopram and sertraline showed the best profile of acceptability, leading to significantly fewer discontinuations than did duloxetine, fluvoxamine, paroxetine, reboxetine, and venlafaxine.
The reviewers noted that “clinically important differences exist between commonly prescribed antidepressants for both efficacy and acceptability in favour of Cipralex and Zoloft. They added that “sertraline might be the best choice” when starting treatment because it has “the most favourable balance between benefits, acceptability and acquisition cost”.
Dr Cipriani said that “there is a rational hierarchy when prescribing antidepressants” but stressed that all of the drugs were effective and that the differences between them were small. However, Sagar Parikh, a psychiatrist at the University of Toronto, wrote a commentary in The Lancet saying that “such findings have enormous implications”.
He added that “now a clinician can identify the four best treatments, identify individual side-effect profiles, explore costs and patients’ preferences and collaborate in identifying the best treatment.”
