This is the conclusion of the final report of the Institute for Quality and Efficiency in Health Care (IQWiG) published on 24 November 2009.
The final evaluation of reboxetine and mirtazapine was only possible after manufacturers disclosed data that had previously been concealed. This case re-emphasizes the need for a mandatory regulation, which would require all clinical trials to be registered at the start and to have their results published after study completion.
Manufacturers only release data when public pressure mounts
In its preliminary report published at the beginning of June 2009, the only results IQWiG could present without reservation were those for bupropion XL. It was forced to add a caveat to the results for mirtazapine, because the fact could not be ignored that study data not provided by Essex Pharma might seriously bias the result. In the case of reboxetine, IQWiG abandoned its analysis of the study data that was publicly available at that time, because it was evident that the manufacturer, Pfizer, was concealing almost two thirds of all data collected in trials to date. An analysis of the available data would have produced a biased picture. Despite several requests, Pfizer had steadfastly refused to provide IQWiG with a list of all published and unpublished data.
However, after the preliminary report was published, Pfizer and Essex Pharma decided to make the unpublished data and information on trials accessible. Only then was it possible to assess all three drugs based on complete data.
The analysis of the complete data reveals that IQWiG made the right decision in abandoning its assessment of reboxetine based only on published data. The summary of results from published and non-published trials does not provide proof of benefit of reboxetine, whereas the data from the published trials appear to suggest a benefit.
Many outcomes were considered
In line with the Federal Joint Committee's (G-BA) commission, IQWiG and its external experts searched for trials that either compared each of the drugs with a sham treatment (placebo), with each other or with another antidepressant. The trials included both acute therapy and relapse prevention.
In addition to the change in depressive symptoms (remission, response) and secondary symptoms such as anxiety, pain or sleep disorders, outcomes in the benefit assessment also included mortality, suicidal tendencies, quality of life, social functioning level, and adverse effects.
Reboxetine: proof of harm, but not of benefit
IQWiG had access to a total of 17 trials on reboxetine when preparing the final report. Its analysis shows that there is no proof of benefit either for acute therapy or for relapse prevention. The patients did not respond any better to the therapy than to a placebo, nor was there an improvement in their social functioning level.
There were only two specific situations where indications of a benefit could be found. A very small trial with a total of 52 patients treated in hospital provided an indication that those receiving reboxetine responded better to the treatment than those receiving placebo. Nevertheless, it could not be shown in this trial that symptoms typical of depression completely disappeared. Another trial with outpatients also gives indications, although only for relapse prevention. In short-term acute therapy reboxetine does not alleviate symptoms of depression as well as drugs in the SSRI drug class.
Alongside the lack of evidence of benefit of reboxetine, there is also proof of harm: patients discontinued treatment more frequently due to adverse side effects, both when compared to placebo and in the comparison with fluoxetine, another antidepressant from the selective serotonin reuptake inhibitor class (SSRI).
Mirtazapine: patients respond better to therapy than to placebo
IQWiG was able to include 27 trials on mirtazapine in the assessment. As full information on these trials was now available, the Institute no longer needed to add a caveat to the results, as was the case with the preliminary report.
When compared to placebo there was proof that in acute treatment more patients experienced an improvement of depression when they were treated with mirtazapine. The prospect of a total cure was no better in the mirtazapine group than in the placebo group. However, this aspect was only investigated in 1 trial for mirtazapine. In numerous comparisons with other antidepressants, mirtazapine did not show proof of superiority. In addition, it was shown that patients treated with mirtazapine discontinued treatment more often due to side effects (adverse events) than patients treated with placebo or with some of the other antidepressants.
Bupropion: symptoms completely disappear in some patients
The search identified 7 trials on bupropion XL and the Institute was given access to the complete clinical study reports by manufacturers. There was proof of benefit for this agent compared to placebo in acute therapy and in the prevention of relapse into seasonal affective disorder. In some patients, the symptoms were reduced so much that they no longer met the criteria for a diagnosis of depression (remission). Moreover, the data provided no indications of harm.
The only antidepressant that was compared with bupropion XL in trials was venlafaxine XR. Bupropion XL showed inferiority to venlafaxine XR in acute therapy. Further information can be found in the executive summary of the final report (see below).
Obligation to publish trial results must be legally enforceable
As has been revealed in the process of preparing this report, a lack of willingness to cooperate on the part of the manufacturers leads to benefit evaluations of limited validity and causes considerable delays in producing assessments. "Deception through concealment of trial data is no trivial offense", says IQWiG's director, Peter Sawicki. "The study sponsors are depriving patients and doctors of the opportunity to make an informed decision on different therapy options. As the example of reboxetine shows, concealing trial data can lead to patients receiving a drug, for which there is not only no proof of benefit, but which can also cause harm." Moreover, it is not only the work of the Institute that is hindered, but also that of the G-BA. According to the Institute's director, "The G-BA then lacks the reliable scientific basis that it needs for its decisions on reimbursement of drugs." IQWiG's position on this issue is described in a second press release (see below).
Reboxetine was approved in Germany in December 1997. For its part, however, the German drug regulatory authority could not consider all the trials that IQWiG had analyzed, because IQWiG's report also includes trials that were completed after 1997. Pfizer also applied for approval in the USA but this was evidently not granted in 2001.
IQWiG published the preliminary results in the form of the preliminary report version 1.0 in June 2009 and interested parties were invited to submit comments. When the comments stage ended, the preliminary report was revised and sent as a final report to the contracting agency, the Federal Joint Committee, in November 2009. Documentation of the written comments and minutes of the oral debate are published in a separate document simultaneously with the final report. The report was produced in collaboration with external experts.