Posted by jono_in_adelaide on September 14, 2012, at 18:23:22
I was always suspicious of that Lancet article
http://www.mhra.gov.uk/Safetyinformation/DrugSafetyUpdate/CON128957
A UK and Europe-wide review of available efficacy and safety data has confirmed that reboxetine has benefit over placebo in its authorised indication. Efficacy was clearly shown in patients with severe or very severe depression. Like many antidepressants, there are limited efficacy data for the use of reboxetine in patients with mild to moderate-severity depression
Reboxetine is a highly selective and potent inhibitor of noradrenaline reuptake and has only a weak effect on serotonin. It is indicated for acute treatment of depressive illness or major depression, and for maintaining clinical improvement in patients who initially respond to treatment. Reboxetine is not widely used in the UK, reflecting NICE recommendations that some antidepressants such as reboxetine should only be used when selective serotonin reuptake inhibitors (SSRIs) do not show an effect or are not well tolerated.A meta-analysis has investigated the effect of possible publication bias on the overall assessment of the safety and efficacy of reboxetine.[1] The study was part of a German health technology assessment of three antidepressants, and concluded that reboxetine was an ineffective and potentially harmful antidepressant.[2]
As a result of the concerns raised, the totality of safety and efficacy data for reboxetine was assessed in the UK by the Commission on Human Medicines and in Europe by the Pharmacovigilance Working Party. This analysis included the published meta-analysis,1 data submitted in support of the original licence application that had shown the efficacy of reboxetine, and further data submitted by the licence holder.
The published meta-analysis1 included 13 acute-treatment trials (placebo-controlled, SSRI-controlled, or both) involving 4098 patients. Data for 3033 (74%) patients were unpublished.Seven trials looked at remission rates and noted no significant difference between those receiving reboxetine compared with placebo (odds ratio [OR] 1.17 [95% CI 0.911.51], p=0.216). Eight trials investigated response rates and after exclusion of one trial (see below), the point estimate calculated from the remaining seven showed no significant difference between reboxetine and placebo (OR 1.24 [95% CI 0.981.56], p=0.071).
The published meta-analysis1 had several limitations. A pivotal efficacy study was excluded from the efficacy analysis, but included in the safety assessment. The positive efficacy result was regarded by the researchers as a statistical outlier, which is not considered an adequate justification for omission of a valid study.
When data from this study and other earlier studies that formed part of the original licence application are included in an updated (unpublished) meta-analysis, reboxetine showed a benefit in treatment response over placebo in the treatment of depression: OR 1.47 (95% CI 1.101.97), p=0.01.
The review also explored possible reasons for the differences in the findings between the studies before and after licensing. Differences in care setting and disease severity at baseline are the most likely explanation for the difference in size of the reboxetine treatment effect. Studies in an in-patient setting and in severe depression consistently showed better efficacy results than those in out-patients. The greater effect compared with placebo seen in patients with more severe depression is also seen with other antidepressants, which is consistent with current clinical guidance that antidepressants are not recommended for first-line treatment of mild or moderate depression.
A thorough review of all available safety data has not identified any previously unrecognised safety concerns associated with reboxetine, and has confirmed the side-effect profile previously recognised.
Overall the balance of benefits and risks for reboxetine remains positive in its authorised indication.
poster:jono_in_adelaide
thread:1025659
URL: http://www.dr-bob.org/babble/20120912/msgs/1025659.html