Just before Christmas, a new warning appeared on the US label for Paxil (paroxetine) label. For the first time, the manufacturers have admitted that withdrawal symptoms from paroxetine were frequently seen in clinical trials. So much for GlaxoSmithKline’s previous claims that withdrawal symptoms were very rare (Wheadon, 2000).

Arising out of Dr Wheadon’s claims, Social Audit made a complaint under the pharmaceutical industry’s own voluntary code of marketing practice. In September 2001, we finally got a ruling – pretty much a turndown. GSK’s remarks were considered justified, above all, on the grounds that the Company claimed that there were only 7 reports of a withdrawal syndrome among the 8,413 patients on their clinical trials database.

Social Audit[i] then appealed that decision at a hearing held by the Prescription Medicines Code of Practice Authority (PMCOPA). That was on 15^th November 2001 and, the following day, we were advised that the appeal had been upheld. In other words, GSK had failed to convince a panel of peers, professionals and patient representatives that it was justified in claiming that the risk of withdrawal and dependence was not real. It has taken over a year since the complaint was first made but, at last, publication of the findings[ii] is now due. (Now Published - 22 January 2002)

So what does the new FDA-approved label warning actually communicate? For one thing, it indicates that GlaxoSmithKline was trying to take PMCOPA for a ride. Once the case went to appeal, the Company tried to get the hearing date postponed. When the independent Chairman of PMCOPA refused,[iii] GSK decided not to appear at the hearing to defend their position at all. They would have known by then that the new warning flatly contradicted the basis of their appeal. This is what it says:

Discontinuation of treatment with Paxil: Recent clinical trials supporting the various approved indications for Paxil employed a taper phase regime, rather than abrupt discontinuation of treatment. The taper phase regimen used in GAD and PTSD clinical trials involved an incremental decrease in the daily dose by 10mg/day at weekly intervals. When a daily dose of 20mg/day was reached, spatients were continued on this dose for 1 week before treatment was stopped.

With this regimen in those studies, the following adverse events were reported at an incidence of 2% or greater for Paxil and were at least twice that reported for placebo: abnormal dreams (2% vs 0.5%), paresthesia (2% vs 0.4%) and dizziness (7% vs 1.5%). In the majority of patients, these events were mild to moderate and were self-limiting and did not require medical intervention.

During Paxil marketing, there have been spontaneous reports of similar adverse events, which may have no causal relationship to the drug, upon the discontinuation of Paxil (particularly when abrupt), including the following: dizziness, sensory disturbances (e.g. paresthesias such as electric shock sensations), agitation, anxiety, nausea and sweating. These events are generally self-limiting. Similar events have been reported for other selective serotonin reuptake inhibitors.

Patients should be monitored for these symptoms when discontinuing treatment, regardless of the indication for which Paxil is being prescribed. A gradual reduction in the dose rather than abrupt cessation is recommended wherever possible. If intolerable symptoms occur following a decrease in the dose or upon discontinuation of treatment, then resuming the previously prescribed dose may be considered. Subsequently, the physician may consider decreasing the dose but at a more gradual rate (See DOSAGE AND ADMINISTRATION).

The type size underlines the paucity of the warning that, for some people, paroxetine proves a drug of dependence and causes great distress. The evidence that paroxetine – among other SSRI antidepressants – causes such problems has been reviewed at length on the Social Audit website. Reference has been made, in particular, to the following:

1. Credible evidence of significant risk from clinical trials

2. Evidence of risk and harm published in medical journals

3. Reports to the authorities of withdrawal/dependence problems.

4. Use of definitions that undermine understanding

5. Gross understatement and omissions in drug warnings

6. Widespread misunderstanding of the problem by doctors

7. Clear and abundant evidence of problems from patients -

8. The abuse of secrecy in commerce and government

9. Historical precedents - benzodiazepines, barbiturates etc

10. Corporate Survival bias versus good common sense

In the light of this ruling, it will be interesting to see if the US Company keeps up its current Direct To Consumer Advertising campaign: “Talk to your doctor about non habit forming Paxil today”. (Readers Digest [US edition], 2001).

Interpreting the New Paxil Warning
Contributors to the discussions on this and several other dedicated websites – indeed anyone familiar with the labels on packets of corn or soap flakes - might well feel this new warning rather insults patients and consumers. But that it is not what it is intended to do. The tiny type and the obliqueness of the text is intended mainly to reassure the market – from doctors to market analysts. Paroxetine (Paxil®, Seroxat®, Aropax® etc) is GlaxoSmithKline’s best-selling product, a ‘blockbuster’ drug, accounting for about 10% of the company’s drug sales. The whole company would be at risk if paroxetine went down.

So, this is what the warning says. See also the footnotes that comment on and question what it seems to mean.

Discontinuation of treatment with Paxil: Recent[iv] clinical trials[v] supporting the various approved indications for Paxil employed a taper phase[vi] regime, rather than abrupt discontinuation of treatment. The taper phase regimen used in GAD and PTSD clinical trials involved an incremental decrease in the daily dose by 10mg/day at weekly intervals. When a daily dose of 20mg/day was reached, patients were continued on this dose for 1 week before treatment was stopped.[vii]

With this regimen in those studies[viii], the following adverse events were reported at an incidence of 2% or greater[ix] for Paxil and were at least twice that reported for placebo: abnormal dreams (2% vs 0.5%), paresthesia (2% vs 0.4%) and dizziness (7% vs 1.5%).[x] In the majority[xi] of patients, these events were mild to moderate[xii] and were self-limiting[xiii] and did not require medical intervention.[xiv]

During Paxil marketing, there have been spontaneous reports[xv] of similar adverse events, which may have no causal relationship to the drug,[xvi] upon the discontinuation of Paxil (particularly when abrupt), including the following: dizziness, sensory disturbances (e.g. paresthesias such as electric shock sensations), agitation, anxiety, nausea and sweating.[xvii] These events are generally self-limiting. Similar events have been reported for other selective serotonin reuptake inhibitors.[xviii]

Patients should be monitored[xix] for these symptoms when discontinuing treatment, regardless of the indication for which Paxil is being prescribed. A gradual reduction in the dose rather than abrupt cessation is recommended wherever possible.[xx] If intolerable symptoms occur following a decrease in the dose or upon discontinuation of treatment, then resuming the previously prescribed dose may be considered.[xxi] Subsequently, the physician may consider decreasing the dose but at a more gradual rate[xxii] (See DOSAGE AND ADMINISTRATION).

Regular visitors to this website will appreciate that neither GlaxoSmithKline (GSK), nor the big regulatory agencies (FDA, EMEA, CPMP, MCA, CSM etc), nor apparently most doctors (APA, AMA, RCPsych, RCGP etc) consider paroxetine to be a drug of dependence, even if that is what it clearly what it seems to be.

REFERENCES & NOTES

[i] Social Audit was represented at the hearing by Charles Medawar (Social Audit), Dr David Healy (a psychiatrist and expert on such drugs), and Anne Winyard (a solicitor with Leigh Day & Co, who made no charge for her firm's work). N.B. Leigh Day & Co are not involved in litigation relating to dependence on paroxetine

[ii] There will in fact be two reports of findings. One will be by the UK-based PMCOPA and the other by the international industry’s trade body, the IFPMA. Social Audit’s appeal was examined and heard by PMCOPA, the semi-autonomous regulatory arm of the Association of the British Pharmaceutical Industry, and the ABPI is a leading member of the International Federation of Pharmaceutical Manufacturers Association (which is very much run along US industry lines.)

[iii] Eight of the 15 members of the Appeal Board work in the pharmaceutical industry; the chairman and the others don’t.

[iv] These trials weren’t all that recent. Smith Kline Beecham submitted to the FDA the results of its trials on paroxetine for GAD, in May 2000. In other words, the company went on claiming there were no real withdrawal problems long after completion of these trials.

[v] These clinical trials would for most part have lasted for only a few weeks. In real life, people take antidepressants for months and years – and severity of withdrawal symptoms and dependence increase with time. The regulatory agencies should be deeply concerned that evidence of dependence is apparent even after short-term use – when the recommended minimum period of treatment of six months or more.

But the issue here is not simply that some patients suffer when they try to stop taking the drug (and that others will stay on it to avoid withdrawal symptoms). The real issue is that the evidence of the long-term effectiveness of these drugs has been based on trials in which the existence of withdrawal problems was discounted. The long-term effectiveness of SSRIs has been 'proved' by comparing what happened to people who stopped taking SSRIs with those who continued - but always counting withdrawal distress as evidence of 'relapse'. See for example the analysis of the main trial relied on by the UK Medicines Control Agency and Committee on Safety of Medicines, by Montgomery & Dunbar, (1993). The long-term effectiveness of antidepressants may turn out to be as much of a fiction for antidepressants as fot benzodiazepines; it remains very much in doubt.

[vi] If clinical trials conducted over the past few years have used a taper design, why have US physicians not been advised to do the same? Is the company’s responsibility to behave ethically or merely to comply with, in this case, the absurdly modest demands of the US Food and Drug Administration?

[vii] This particular regimen will seem incomprehensible to many people trying to get off paroxetine doses of 10mg/day and 20mg/day. Where is the evidence that any such tapering regimen would be appropriate after months or years of use?

[viii] Which studies and how were they implemented and designed? The percentage figures cited are meaningless (other than as minimum estimates of risk) unless and until the context has been made clear. In clinical practice, the incidence of withdrawal problems would be closer to 50%. See Rosenbaum et al, 1998.

[ix] If withdrawal symptoms had occurred at this frequency, why were they not reported in the label warning with paroxetine promoted for GAD and PTST? It would probably take a Congressional investigation to explain this. Indeed it would be worth taking a general look at the variety of ways in which companies tend to obscure understanding of adverse effects. Note this classic observation, for example, in the FDA’s Review and Evaluation for Paxil CR, in December 1998: “Suicide gestures and suicide attempts were coded to the preferred term “Emotional Lability”, which is not considered to be an accurate representation of these events” (Section 8.1.4.1)

[x] By omission, the implication here is that the severity of these adverse effects is comparable on active drug and placebo. Why do regulators all over the world allow companies to get away with nonsense like this?

[xi] It’s dumb just to advise what happens to a fortunate majority. What really matters to patients and doctors is what happens to the less fortunate – what the chances of misfortune are, and how to minimise the drug’s ill-effects. This is what the companies should be addressing – and it’s another disgrace that regulators permit them to ignore it.

[xii] Price and colleagues at the Medicines Control Agency/Committee on Safety of Medicines (1996) sent follow-up questionnaires to 217 doctors who had reported withdrawal reactions with paroxetine on Yellow Cards, to the MCA/CSM. They reported 21% of reactions were mild; 57.7% were moderate; and 21% were severe. Thus it would be equally true to say that most people had withdrawal reactions rated as moderate or severe – or that most people had mild to moderate reactions.

[xiii] The significance of severe withdrawal reactions has, for example, been far better explained in the following reviews:

“Most discontinuation reactions are mild and transient, but a minority are severe, of longer duration and cause considerable morbidity. Symptoms may last up to 13 weeks. With SSRIs, ataxia can cause enforced immobility, falls and absence from work while electric shock sensations can cause difficulty walking and driving. Patients may present for urgent medical help and occasionally require inpatient admission” (Haddad, 2001)

“Many discontinuation symptoms are mild and short-lasting, and manageable with reassurance and explanation. In some patients however, the symptoms are severe and make normal functioning impossible, even with slow withdrawal” (DTB, 1999)

“The CADRMP (Canadian Adverse Drug Reaction Monitoring Programme) has received 26 reports of discontinuation symptoms for the 4 SSRIs marketed in Canada … In 1 report symptoms appeared during a dose reduction (from 20mg daily of paroxetine alternating with 10mg daily) and in 3 reports (paroxetine 2, fluoxetine 1) symptoms appeared while the dose was being tapered. Three of the 26 cases resulted in hospitalisation or prolonged hospitalisation” (CADRMP, 1998)

[xiv] In the series of reports of paroxetine withdrawal followed up by the MCA/CSM (Price et al, 1996), it was also found that “112 (57.7%) had restarted paroxetine because of the withdrawal reaction … (of whom) 21 (18.7%) were known … to have been continuing with it three months or more later as a consequence.” This is clearly indicative of dependence.

[xv] Not to mention the content of several dedicated patient websites, and the fact that there have been more spontaneous reports about paroxetine withdrawal problems than for any other drug.

[xvi] On the other hand, the evidence clearly shows that what are really withdrawal symptoms are commonly misdiagnosed as a relapse.

[xvii] The key point missed here is that the combined effect of such withdrawal symptoms is typically to mimic the psychic distress that led to the drug being prescribed in the first place – which in turn means that physicians are highly liable to misdiagnose withdrawal effects, unless properly advised:

"The withdrawal of antidepressants can produce changes in mood, appetite and sleep that are apt to be incorrectly misinterpreted as indicating a depressive relapse ... The probability of depressive relapse is low in the days and weeks after the discontinuation of antidepressants, and the cumulative probability of relapse increases as a function of time when the patient is medication free ... In contrast the frequency of antidepressant withdrawal symptoms is high in the first 2 to 14 days following the last dose." (Dilsaver, 1989)

"It is unlikely that reports of withdrawal reactions represent recurrence of depressive illness rather than ADRs because of the short time interval between stopping the drug and the onset of the reaction, and the nature of the symptoms which were unlike those associated with depressive illness" (Price et al, 1996)

“Discontinuation syndromes can occur with any antidepressant; they usually start abruptly within a few days of stopping the drug and resolve quickly (usually within 24 hours) if the drug is restarted. In general, it should be possible to distinguish discontinuation syndromes from true relapse of depression, which is uncommon in the first week after stopping treatment and resolves more slowly when the drug is restarted. (DTB, 1999)

[xviii] Add: “but with paroxetine and venlafaxine apparently in a class of their own”?

[xix] Above all, patients should be warned to look out for these symptoms – and arguably need to be advised that, after they’ve taken these drugs for a bit, they may find it very hard, if not impossible, to stop.

[xx] It should be made quite clear – and this warning does anything but – that the tapering regimen used in short-term trials would be highly unlikely to be suitable in general clinical practice.

[xxi] The reference to ‘intolerable’ withdrawal symptoms as good as signals the reality of dependence. The simplest definition of drug dependence given by WHO is "a need for repeated doses of the drug to feel good or avoid feeling bad". WHO, (1998)

[xxii] The wording here seems strikingly casual considering the problems reported by many users, even after repeated attempts at withdrawal with the most prolonged tapering regimens. The warning on the Xanax (alprazolam) label comes to mind: “Some patients may prove resistant to all discontinuation regimens” – a rather mealy-mouthed way of saying they may become permanently hooked.

Charles Medawar
January 2002

	Contents page
	What's New?