This from Matthew Holford (head of the Seroxat Secrets legal department!), writing to Sarah Walk at the MHRA.
I’ve got to say that Matt makes a very, very good point here. The reply will be most illuminating, I feel.
We have, then, satisfied ourselves that the EWG was correct in asserting that patients on Seroxat are not dependent on the drug, to the extent that the DSM IV and ICD 10 systems are used to assess the characteristics of dependence. We have also satisfied ourselves that withdrawal is unquestionably a characteristic of dependence, and that withdrawal is also unquestionably a characteristic of Seroxat use.
The question that occurs to me immediately concerns the volume of patients, who may anticipate experiencing withdrawal, when they use the drug. I understand that the current PIL states 30% (it used to be 25%, and before that 0.7%, and less than four years ago, the figure was 0.1%), although I have seen estimates as high as 80%. Is the MHRA aware of any trials, which have been undertaken specifically to assess the extent of this issue?
Now, given that these figures have been revised so radically over such a short space of time, the question that occurs to me is “why?” What was it that caused GSK and the MHRA to come to the conclusion that the information provided on the PIL needed re-wording? I understand that spontaneous patient reports are not viewed as sufficiently compelling, irrespective of the importance allegedly attached to the Yellow Card scheme. And I understand that, in the absence of a perceived (ie, as perceived by it) problem, GSK initially declined to conduct trials to investigate the prevalence of withdrawal. So why the sudden change in perception?
Clearly, we should be grateful that the manufacturer and the regulator are so concerned with patient health that their continuous monitoring of SSRIs has permitted them to identify an issue, albeit that it is now acknowledged to be 300 times worse than originally perceived (or 30,000%, whichever you prefer). Would the MHRA have licensed the drug, had it been aware of the proximity and magnitude of the risk, at the time?