Social Audit Ltd
P O Box 111 London NW1 8XG
Telephone/Fax 44 (0)171 586 7771
[email protected]


Mrs. Elizabeth Hopkins, Post Licensing Division
Medicines Control Agency
Market Towers, 1 Nine Elms Lane
London SW8 5NQ

15 March 2000

Dear Mrs Hopkins,

Thank you for your letter of 9 March, responding to our complaint about an advertisement for Efexor XL. It was certainly helpful to have the legislative background explained, and please would you treat our outstanding complaint (22 January, re: advertising for Cipramil/citalopram) also as a referral under regulation 3A of SI 1999 No 267.

But it wasn't helpful - especially after a wait of three months - simply to be told that the MCA is satisfied that this advertisement complied with the regulations. Your letter murmured about monitoring and "intensive surveillance" but otherwise gave absolutely no indication why the MCA should be satisfied. There are reasons for thinking it should not be. 

1. Re: Regulation 3A (1) "No person shall issue an advertisement relating to a relevant medicinal product unless that advertisement complies with the particulars listed in the summary of product characteristics"

How can this claim "comply" with the particulars in the Summary of Product Characteristics (SPC) for Efexor XL, when there is nothing whatever in that document to indicate in any way the superiority of venlafaxine over fluoxetine or any other antidepressant? Can you point me to any sentence or fragment of the SPC which would help me to understand what the MCA had in mind when it decided that this advertisement "complies"?

It seems to me that, by officially endorsing the manufacturers' claim of the superiority of their product over fluoxetine, the MCA has allowed an extension of the license terms. If the manufacturers had formally applied for a license variation in order to make these claims for superiority, they would have been required to produce evidence far more convincing and extensive than that cited in the minuscule print of the advertisement in question.

To this extent, it seems that the MCA is undermining the licensing system. It has permitted this claim of superiority pretty much by default - certainly with less scrutiny of evidence than would/should be given in the licensing process.

2. Re: Regulation 3A (2) "No person shall issue an advertisement relating to a relevant medicinal product unless that advertisement encourages the rational use of that product by presenting it objectively and without exaggerating its properties"

The MCA's view that this advertisement "is in compliance" with the regulations evidently means that the claims made may be taken at face value, as an objective and unexaggerated statement of the characteristics of the product. In turn, this suggests that the rational response to the advertisement would be to prescribe Efexor XL as a first-line treatment, in preference to fluoxetine. That is what, in effect, this decision means. And this is why I believe the MCA is again failing miserably and neglecting its duty to uphold the law:

1. I am assuming here - but please correct me if I'm wrong - that the strength of the evidence used to support a claim needs to be proportional to the significance of the message given and its intended and likely impact. In this case, an apparently compelling reason is given to use one drug in preference to another. The headline says that Efexor XL is twice as good as Prozac in eradicating the symptoms of depression, so reducing risks of relapse, impairment and suicide. The evidence therefore has to be pretty good. It is not.

The evidence for this big claim was cited to a poster presentation by Dr Richard L. Rudolph, an employee of Wyeth, the manufacturers of Efexor XL. Also cited was a review by Dr Ian Nicol Ferrier, from the Department of Psychiatry at Newcastle University. This concluded with a recommendation to treat depression to the point at which symptoms are pretty much obliterated, not simply reduced. The review was one of several papers given at a symposium sponsored by Wyeth. They all pointed in the same general direction: "to treat patients aggressively to produce full remission of their depressive syndrome". The same kind of thinking gave us a decade of Ativan (lorazepam, Wyeth) at double the appropriate dosage and I suspect there are comparable perils here.

2. I would certainly like to be reassured that the MCA had thought carefully about the implications of adopting a far more aggressive dosage strategy than is suggested in the SPC for Efexor XL. Is it really sensible to recommend such strategies, increasing risk for whole populations in order to benefit the small minority who can tolerate high doses? And is a reduction to a score of 7 on the Hamilton scale really the best way out of depression? I honestly don't know - but I can't helping wondering if someone who registers zero on the HAM-D scale wouldn't get into real trouble, if only by driving other people completely up the wall.

3. The trial which underpins all the trumpeting in this advertisement seems every bit as contrived as that Eli Lilly sponsored study we've complained about before (Rosenbaum, Fava et al, 1998). Probably the main reason it manages to 'demonstrate' that Efexor is "nearly twice as effective as fluoxetine" is that the protocol leads to pumping up the dose of venlafaxine (Efexor XL) more than the dosage of fluoxetine (Prozac). Is the MCA satisfied that this study isn't really demonstrating that relatively higher doses of Efexor XL obliterate the symptoms of depression more than relatively lower doses of Prozac?

That question arises because the study protocol called for upward adjustment of dosage - for patients who could tolerate the side effects - initially after two weeks, and then again after four weeks. Did the MCA take into account that effective dosage titration would therefore not have been possible with fluoxetine? The short half-life of venlafaxine means that dosing changes show their full effect after about four days. With fluoxetine, the very long half-life means dosing changes are not reflected in plasma for several weeks. Did the MCA decide to approve the advert for Efexor XL knowing how many patients ended up on higher or lower dosages of each drug? Was it confident that the full effects of an increased dosage of fluoxetine would be evident after only four weeks?

4. And are the risks involved in striving to treat patients at maximal dosages really worth the benefit? Clearly, the incidence and severity of many side effects tend to increase as dosage rises - including the risk of withdrawal symptoms and dependence. Did the MCA take into account, for example, that one of the centres testing patients under a similar protocol "observed a statistically significant … inverse relationship between endpoint Hamilton depression score and number of adverse events after treatment discontinuation"? (Fava et al, 1997) In other words, whatever the benefits, the risk of therapeutic dependence grows with increased dose.

5. Then, why did the MCA allow the headline claim "… is nearly twice as effective in treating depression" when this was based on a single study that was unrepresentative of others? The Ferrier review noted that "a pooled analysis of data from four studies found a significantly higher remission rate (final HAM-D score below 8) with venlafaxine XR (45%) compared with placebo (18%) or fluoxetine (37%)".

The pharmaceutical industry's own code of practice requires claims and comparisons to be based "… on an up to date evaluation of all the evidence and reflect that evidence clearly". Is it not a whopping exaggeration to suggest that the difference between 37% and 45% is comparable to "nearly twice"?

3. Re: Regulation 3A (3) "No person shall issue a misleading advertisement relating to a relevant medicinal product"

These are some of the reasons - but there are certainly others - for thinking this advertisement is misleading. The results are presented in such a way as to suggest there are very significant differences between venlafaxine and fluoxetine. In fact, the alleged superiority of Efexor XL seems to relate much more to the dosage than to the drug itself - and the benefits, if that is what they are, would be available to relatively few. The Ferrier review refers to a dose ranging study with the standard venlafaxine product (Rudolph et al., 1998) in which "the proportion of patients with a HAM-D score of 8 or less increased from 25% to 54%", dependent on use of lowest to highest recommended dose. Even if these were only ballpark figures, they clearly suggest that the apparent superiority of the drug is overwhelmingly attributable to the trial design.

I hope this is enough to persuade you that the MCA needs urgently to explain, if not completely reconsider, its position. Failing this, we shall certainly seek a referral to the Ombudsman.

Thank you for your attention; I look forward to hearing from you in the near future, at least the date by which we can expect to receive your substantive reply. 

Yours sincerely,
Charles Medawar


Contents page
List of correspondence with MCA/CSM