Sunday Times: June 30, 2002

Scarred for life: Does the drug that cures acne have devastating side effects?

Richard Girling

Acne destroys teenagers' lives and affects millions. But the wonder drug that cures it may also lead to psychosis and suicide

The American explorer Elisha Kent Kane is commemorated by a bay in the Arctic, a crater on the moon and a place in medical history. In the mid-1880s he ensured lasting fame when, after surviving two winters in an iced-up ship in the Arctic, he led his crew on a desperate 900-mile trek across the ice to Greenland. Aided by Eskimos, they stayed alive by eating polar bears.

On January 5 this year, a 15-year-old student, Charles Bishop, took a Cessna light aircraft from a flight school in Florida and flew it into the 28th floor of the Bank of America Plaza in Tampa. A note in his pocket expressed sympathy for the September 11 hijackers, whose example he had intended to follow.

Other, more private, deaths have been no less tragic. In December 1997, Seumas Todd, the 20-year-old son of the actor Richard, killed himself with a shotgun. In 1998 an 18-year-old sixth-former, David Bebby, jumped from a multistorey car park in Newport, Gwent, and a 20-year-old Dublin student, Liam Grant, hanged himself from a tree.

In 2000 a young married woman, Susan Johnson, stood in the path of a high-speed train on the London-Plymouth railway line.

In New Zealand, a 19-year-old student, Hugo Wilkinson, cheerfully waved from a window as his father went out to a rugby match, then returned to his room and killed himself.

In the United States, the 18-year-old son of Congressman Bart Stupak of Michigan, a popular college football player, shot himself after a prom-night party. Later the same year, a congressional committee heard how two other normal, healthy teenagers, Daniel Baumann, 15, of Illinois, and Clay Jackson, 17, of Texas, had suddenly decided to end their own lives, and how another, 14-year-old Amanda Callais, had taken an overdose but had been saved by her mother. On his website, Congressman Stupak drew attention to a total of 54 suicides in the US between 1998 and 2000.

The thread that binds all these people together could hardly seem more innocent.

We all depend on vitamin A for the health of our skin, hair and mucous membranes. It helps night vision and our reproductive systems. We need it to develop our bones and teeth. Yellow fruit and dark green vegetables are the best sources, though we also get it from meat, fish, milk and eggs. It is the basis of all the anti-wrinkle creams with 'retinol' or 'retinoids' in their formulae. Pro-health and anti-ageing, it's exactly the kind of thing we can't get too much of.

The trouble is that anything taken in excess stops being food or medicine and becomes instead a poison. Overdose, after all, is what transforms innocent headache remedies into killers. And so it is with vitamin A. Overdose causes a condition known as hypervitaminosis A, the symptoms of which have disturbing effects on mind and body alike. This is the discovery we owe to Elisha Kane.

Why was it, he wondered, that instead of feeling elation after their life-preserving feasts of polar bear, he and his crew had become irritable and depressed? The answer, it transpired, was that the animals' livers contained so much vitamin A that it caused a psychotic reaction - classic symptoms of hypervitaminosis A.

There is a further, more precise, connection between Charles Bishop, Bart Stupak and the other young suicide victims of the late 20th and early 21st centuries. They all suffered from acne, and all had been prescribed a powerful drug called isotretinoin, a vitamin-A derivative made by the Swiss giant Hoffman-La Roche and sold in 100 countries worldwide. Its brand names are Roaccutane (UK) and Accutane (US). On both sides of the Atlantic it has made the kind of headlines normally reserved for mass murderers or airline disasters: Suicide is linked to acne pill (The Sun, May 8, 1998). My face cleared and my life went dark (The Independent, June 30, 1998). Acne drug maker faces lawsuit over pill's 'suicide link' (The Guardian, November 19, 1998). Every day I wanted to die. I have lost years (the Daily Mail, January 2, 1999). Suicide of woman depressed by acne drug (the Daily Mail, July 19, 2000). Could an acne drug have driven this boy to fly a plane into a tower? (The Independent, January 10, 2002.)

The 54 American suicides recorded by Stupak were all linked by isotretinoin and are mirroredby other reports from across the world - in the UK since 1983 there have been 15.

Since 1997 in France, and 1998 in the US and the UK, isotretinoin has come with a warning that it may cause depression and suicide. Yet depression is not the only - and perhaps not even the worst - side effect.

Women now are prescribed it only on condition that they are not pregnant, and do not become pregnant while taking it. In the US, where the hot breath of the litigation lawyer is seldom far from practitioners' necks, female patients must agree to use two methods of contraception for as long as the drug remains in their bodies. Roche itself describes the risk of birth deformity as 'extremely high', and abortion is recommended for women who become pregnant during treatment. Before the pregnancy ban, babies were born with deformed, misplaced or absent ears, abnormally wide-set eyes, enlarged heads and small chins. There were also cases of mental retardation and physical defects of the heart, brain and nervous system.

Almost all isotretinoin users can expect some side effects. Normal reactions include chapped lips, itching and peeling of the skin and dryness of the mucous membranes that may cause nosebleeds. Hair may thin. Eyes may become dry and night vision may be impaired.

'Other less common, unwanted effects,' says the leaflet given to patients, 'include headaches, feeling sick, tiredness, sweating, menstrual changes, slight loss of hearing, changes in vision (including colour vision disturbances), jaundice, liver disease, anaemia, seizures, inflammatory bowel disease, inflammation of the pancreas, systemic infections, local bacterial infections such as infection of the tissue around the base of the nail, changes in the nails, increased facial pigmentation, swellings discharging pus, swollen glands, inflammation of blood vessels (sometimes with bruising and red patches), blood in the urine, diabetes, changes in blood glucose levels... and asthma.'

Such effects are usually temporary but the risks are enough for isotretinoin to be classified as a treatment of last resort - ie, it should not be prescribed for mild outbreaks of adolescent spots but only for severe cases of potentially scarring cystic acne that have not responded to antibiotics or other treatments. In the UK it may be prescribed only by a dermatologist, not by a GP.

In reality, however, not even lip service has been paid to the idea of restricting isotretinoin to a minority of worst cases. Worldwide it is Roche's second-highest earner, with annual sales of one billion Swiss francs (434m). This hardly suggests that doctors have been holding back on the prescriptions.

Indeed, it is central to Roche's case that so much of it has been prescribed with so few lasting side effects. As early as 1992, the influential Leeds Foundation for Dermatological Research calculated that just 40% of isotretinoin prescriptions were for severe acne; the majority were moderate cases prescribed 'off-label'.

In 1995, doctors calculated that only 16% of prescriptions were for severe disease; the rest were for moderate or mild cases. Thus, it is argued, thousands more young people are exposed to the risks of liver damage, diabetes, depression, suicide and all the rest.

Inevitably it is the suicides that dominate the headlines, but some of the survivors' tales are scarcely less harrowing. The chairman of the UK branch of the Roaccutane Action Group, 25-year-old David Chow, was prescribed isotretinoin for mild acne in 1994. Not unexpectedly, he suffered the common side effects of dry eyes, dry lips and nosebleeds. Less predictably, the damage to his lips not only persisted after he finished taking the drug but went on getting worse.

'After the course,' he says, 'I continued to experience extremely dry lips. They would not only crack and bleed but the skin got to the stage where it was so thin, it turned into a white mush when touched by water.' The result has been persistent infections and years of pain, severe enough to make eating and speaking a test of willpower. Though he has studied successfully for a business degree, he is unable to work and has had to give up sport.

'Often I have had to stop speaking for days,' he says. As a direct result of the evidence he gave to the UK Medicines Control Agency (MCA) last year, the condition from which he suffers - known as persistent exfoliative cheilitis - has been added to the list of side effects against which UK patients are warned. He is now campaigning for the further addition of blepharitis (chronic inflammation of the eyelids, already covered by official warnings in Australia), plus impotence and loss of libido, which between them account for up to a half of all the e-mails the action group receives from worried patients.

More frightening still is the case of Luke Hassett, now 22, who was prescribed isotretinoin for mild acne while studying at Leeds Metropolitan University. His mother, Muriel, tells the story. 'He was a nice, popular lad,' she says, 'happy, kind-hearted and well liked.' His first year at university went well, but then early in his second year, saying nothing to his parents, he consulted his GP about acne and was referred to Leeds General Infirmary. 'He only had about five spots. His friends didn't think he had acne at all.'

His decline was precipitous. 'We became aware that he was behaving a bit oddly. He was a bit paranoid, very suspicious of the people he was sharing with and thought his computer was bugged.'

One day Luke phoned his mother at work and she knew at once that something had gone very badly wrong. 'He was paranoid,' she says simply. Without delay she drove from Manchester to Leeds to bring him home and was shocked by what she found. 'He was crashing around, bumping into things, having so much trouble with his balance that I thought he might have CJD. I couldn't persuade him to eat or drink. One of his eyes had dropped and he looked physically awful.' He had stomach pains and his skin was yellow, 'so I knew he had liver problems'. On the way back to Manchester he tried to jump out of the car on the M62.

For a while, there was hope. 'I thought that when the drug was out of his system he would get better. He slept for months. There was no pattern or rhythm to his life but he slowly got better.' So much better, in fact, that he was able to catch up with his coursework, and in the following September he returned to university. 'He was quiet and calm and we thought he was better.'

Within two weeks it had become clear that the optimism was misplaced, and Muriel Hassett drove once again to fetch her son from Leeds. This time his physical condition was worse. 'His lips were all fissured. The lining of his nose was coming away. His eyes drooped. He looked as if he'd had a stroke.' Luke accompanied her voluntarily to Leeds General Infirmary but withdrew his co-operation after a night-long wait in the hospital's casualty department.

In the morning he was sectioned under the Mental Health Act and transferred to a psychiatric unit. Bad turned to worse. Allergic reaction to drug therapy put Luke into a coma that brought him close to death. 'When he recovered from this, all the doctors rejected the idea that the problem was caused by Roaccutane. They tried all sorts of stuff on him. He used to crawl on all fours, banging his head on the floor and howling like a wolf.'

Miraculously, however, he recovered a second time and, once again, went back to university. 'He settled down nicely. We thought he was doing so well.' And this, ironically, was his undoing. He was doing so very well that he thought he could stop taking his medication. This time Muriel had to intercept him at Liverpool airport, where he was on his way to join his tutor and other students on a field trip to Amsterdam. '

I had to get his bags off the plane,' she says. He was sectioned again, and detained in the psychiatric unit at Manchester's Trafford General Hospital, where he still remains.

There is hope. 'He is great at the moment. He comes home for four hours every day on home leave, and seems to be okay.' But there is also anger. 'I cannot believe the MCA allows people to dispense this drug off-licence all the time.' With the support of the Roaccutane Action Group, Muriel Hassett is now planning legal action against Leeds General Infirmary.

Stories like this are common. In the US, more than a hundred lawsuits are taking shape, including Charles Bishop's, though none so far has come to court. In the UK, previous attempts to sue have foundered largely on economic grounds. Pauline Roberts, of the law firm Smith Llewellyn in Swansea, at one time had 80 cases on file. 'In some cases,' she says, 'the link between Roaccutane and the psychiatric injury was very clear-cut.

Nevertheless, some of the claims were complicated by the fact that the complainants were teenagers already suffering from depression. Counsel's advice was that, because the psychiatric injuries were short-lived, the claims were not economically viable - the costs would exceed the likely benefit in damages.' Even in cases of suicide there is no guarantee that any award of damages would justify the cost of an action - the value placed on a teenager's life is limited because nobody relies on them financially.

Yet even where the damage is short-term, the effects may be traumatic. To understand this, you have only to reflect on the words of the 14-year-old survivor Amanda Callais, in evidence to the US congressional committee. 'My downfall started the moment I took the first pill. I found myself feeling sad and I often cried for no reason. I hated myself more and more each day and I began to cut myself with razor blades.' To find a worse ethical outrage you'd have to look all the way back to Dr Josef Mengele. Except...

Here is another mother. Her 39-year-old son, William (not his real name), who still lives with her, developed the first signs of acne at the age of 11. By his mid-teens his face, chest and back were horribly disfigured by weeping boils. Antibiotics were prescribed in ever stronger doses but nothing worked. Huge cysts had to be surgically removed from his cheeks. 'He had the stuffing knocked out of him,' his mother says. 'He didn't look at people when he spoke to them, and he still doesn't. He was bullied throughout his school years, had no teenage social life and I'm absolutely certain it's why he hasn't married.' Once while mother and son were out walking, he summoned the courage to speak to two girls he knew from college. 'Next day they went up to him in class and told him never to speak to them again in public.'

Then the family's GP heard about a new miracle drug that had a near-100% success rate. He sent William back to his dermatologist, who agreed to prescribe Roaccutane. 'It was a saviour,' says his mother. 'It dried everything up. The acne went, but he was left with awful scarring.' Scarring so bad that, even with the acne itself long disappeared, it still affects him.

The experience has left his mother feeling sad that isotretinoin was not available soon enough to protect him from disfigurement, and angry - really, blazingly angry - that people complain to newspapers, and appear on radio and television programmes to attack a drug which, to her, is a life-saver. 'Acne ruined William's life,' she says, 'and the only thing that helped him was Roaccutane.'

Contradictions abound. The Roaccutane Action Group lists more than 40 studies that it says have established a connection between isotretinoin and psychosis. In particular it quotes the work of Tzarina Middelkoop, a researcher at University College, Dublin, who concluded that isotretinoin users were 900 times more likely to suffer depressive symptoms than patients on antibiotics.

For its part, Roche favours Susan Jick, of the Boston University School of Medicine, and other consultants in the US, who argue that isotretinoin users are no more likely to commit suicide than anyone else. 'We have done six studies, involving 33,000 patients, and have found no association,' says Daniel Kenny, international medical manager of Roche in Basel.

There are things he wants people to understand; things that, to his regret, have escaped the attention of the newspapers - the fact, for example, that there is 'no plausible biological link' between isotretinoin and the biochemistry of the brain. But then you remember that very similar things were said by the tobacco industry about lung cancer. You remember Elisha Kane. You remember, too, that Roche is not famous for its candour.

The company was reprimanded by the FDA for withholding evidence of adverse reactions to isotretinoin, failing to disclose in 1997 that the French government had ordered suicide risk to be added to the list of possible side effects, and for falsely advertising the drug as a treatment for depression.

None of this proves anything, but it is hardly calculated to pour oil on the troubled waters of suspicion. Support groups for families and patients have now sprung up in 40 countries, including Ireland, Norway, Sweden, South Africa, New Zealand and France as well as America and the UK. All insist, with mounting certainty after each new case, that there is a direct link between isotretinoin, depression and suicide.

Yet, despite all the claim and counterclaim, patients are supposed to give 'informed consent' before the drug can be prescribed. What is 'informed consent' when there is no consensus on the degree, or even existence, of danger? 'Signing a consent form,' says one senior British dermatologist, 'simply indicates that a discussion has taken place. It doesn't prove that the patient has understood anything.' There is, rather, a juggling of responsibility. The dermatologist cannot quantify the risk, so the patient (who may already be depressed, and is under the duress of a disfiguring disease) has to decide. Risk of scarring versus risk of suicide. What happens in the consulting room may be confidential, but there are shadows on the wall. Behind the doctor stands the drug company, and behind the drug company stands the lawyer. You can't take isotretinoin these days and say you haven't been warned.

Between the licensing of isotretinoin in 1982 and August 31 last year, 12m people worldwide had been treated with it - a startling total of 4.1m patient-years of exposure. This we know with reasonable certainty (the figures are Roche's own). Knowing exactly how many suffered severe or long-term damage, however, is impossible. A bad reaction becomes a statistic only if it is reported to an official body such as the FDA or Britain's Medicines Control Agency (MCA), which logs it as an Adverse Drug Report (ADR). It is accepted that the vast majority of mishaps are not recorded in this way, though there is little agreement on the amount of under-reporting. The FDA has put it as high as 99%; a more common estimate is 85-90%, while the MCA offers no figure at all.The Roaccutane Action Group says that there have been more than 500 formally reported ADRs involving suicide, attempted suicide or suicidal intent ('suicide ideation' in the jargon). In the US, the FDA recorded 54 suicides, 51 attempted suicides and 111 cases of suicide ideation between January 1998 and September 2000. In the UK, in 18 years the MCA has recorded 2,039 adverse reactions shared between 1,208 afflicted patients. Of these, 247 were psychiatric disorders, including 82 cases of depression, 14 of depression made worse, 14 of suicide ideation, 10 unsuccessful suicide attempts and 15 suicides. Nine more died of other related causes.

Trying to draw conclusions from this is like tap-dancing in treacle. Allowing for under-reporting, the 15 suicides could multiply to a hidden total of anything between 150 and 1,500. Or they might not... As the MCA points out, these are only 'suspected' associations with isotretinoin. There are many reasons why young people might become depressed or attempt suicide, and there can be no proof that isotretinoin was the cause. Roche's Daniel Kenny puts it bluntly: 'There is no direct causality,' he says. 'The patients just happened to be on the drug at the time the events occurred. All the evidence is that the suicides were caused by other factors.'

This has been Roche's line throughout all the years of argument, and it doesn't impress bereaved parents. If all the evidence pointed one way, there would be no controversy. There is evidence, and plenty of it, that isotretinoin may cause depression severe enough to lead to suicide. What is lacking is not evidence - dead children are evidence - but proof.

Roche is unarguably right, however, when it asserts that acne itself causes depression. Alison Dudley, the chief executive of the Acne Support Group, says that 75% of her 15,500 members report symptoms of depression and 15% have felt suicidal. In Boston, Susan Jick makes the link work almost to isotretinoin's advantage. She suggests that acne sufferers often blame their appearance for lack of personal success, then become depressed if their lives are not transformed after isotretinoin has done its work. By this account, the drug's only fault is that it works.

Daniel Kenny argues that young people in general are more likely to feel depressed or suicidal. Any group of adolescents, he says, will contain a higher than average number of suicides, irrespective of any treatment that they may be receiving. Young people, too, are more likely to suffer mood swings caused by alcohol or recreational drugs. The problem with this argument, in the UK at least, is that it rests on a false premise. Young people in the crucial age range of 15 to 24 are less, not more, likely than average to commit suicide. There is no statistically validated predisposition among British teenagers to kill themselves, and it is not safe to presume that the deaths of young isotretinoin patients can be marked down to problems of adolescence.

For opponents of the drug, the smoking guns are the patients whose depression lifted when they stopped taking isotretinoin, only to reappear when they started again. When this happens it is difficult to deny the possibility of a link. Difficult, but not impossible. Depression, say Roche's experts, is episodic - patients get better; then they relapse. Given the number of patients worldwide, the laws of chance dictate that a number of 'psychotic events' will indeed stop and start at the same time as isotretinoin. It is pure coincidence.

This is the kind of number-juggling that most infuriates parents, who cannot accept that their personal experiences can be denied by a computer. They have seen happy, healthy teenagers fall suddenly into fatal depression. Despite what the statisticians say, this cannot be explained in terms of 'normal' suicidal behaviour.

Professor Keith Hawton, the director of the Centre for Suicide Research at Oxford university, confirms that teenagers who kill themselves do not act spontaneously with no previous sign of depression. But still the statisticians won't let go. Even if a teenager does suddenly commit suicide, they say, it does not mean that isotretinoin is to blame, or that the victim was not already depressed. The likelier explanation is that parents and friends simply, tragically, did not recognise depression when they saw it. With teenagers, how can you tell? Mood swings are the norm and it takes a trained eye to spot the difference.

One of the saddest aspects of the story is the extent to which the debate has been poisoned by rancour. Action groups accuse Roche-funded researchers of falsifying their results - in effect, of putting profit before life. They are scandalised by money-making, and by the funding of hospital dermatology units, including the one at Leeds, by Roche.

The company's apologists in return accuse their critics of naivety and hysteria. At the FDA hearing in September 2000, a company lawyer implicitly suggested that the densely argued testimony of Liam Grant, the Irish founder of the Roaccutane Action Group, be discounted precisely because he had a personal interest and was currently involved in litigation (his son killed himself in 1997). 'I think it was important that you be aware of that,' he said, as if it changed anything.

Despite all the negative publicity, the deaths and the doubts, British dermatologists last year issued 3,000 prescriptions for isotretinoin. The man nominated by the British Association of Dermatologists (BAD) to explain the reasons for this is Richard Motley, consultant dermatologist at the University Hospital of Wales in Cardiff.

Is he aware that many prescriptions are 'off-label', for mild or moderate rather than severe acne? Does he approve? Yes, of course he is aware, and yes, he does approve. The point is not to ignore the patient's mental state but to be ruled by it. If a patient is suicidal because of a minute amount of acne, he says, then the correct prescription is the one that is going to work fastest - Roaccutane. The same applies to patients whose mild or moderate acne has not responded to other treatments. 'But,' he says, 'I wouldn't prescribe for a moderate case if the patient was unconcerned.'

Overall, BAD's message is simple. There may be a risk of psychotic damage in isotretinoin users, but it is much less than the risk of untreated acne. 'Several small, vocal pressure groups have formed around the tragic suicides of three or four patients taking the drug. Unfortunately, there are no similar representatives for the many more untreated or inadequately treated patients - including those who killed themselves because of depression caused by the appearance of their acne.'

This is a point to which dermatologists return time and again. On the balance sheet of tragedy, isotretinoin produces the least awful bottom line - it saves more lives than it costs. And, they insist, the best way to prevent severe acne is to treat moderate acne. Through its Retinoid Working Party, BAD is currently conducting yet another review of the evidence on mood change.

The working party's chairman is Mark Goodfield, a consultant dermatologist at Leeds, who acknowledges that important questions do remain to be answered. His own replies are buffered with professional tact - yes, there are some reports that suggest isotretinoin may cause depression, and, yes, others that suggest its physical side effects may sometimes be prolonged - but he is ready to accept that more and better research is needed. The problem with all the studies so far, he agrees, is that the numbers of patients surveyed were too small and, because of the under-reporting of ADRs, the figures cannot be relied upon.

It is at this point that the professional view finally converges with the protesters'. Last month, representatives of the Roaccutane Action Group met the MCA to press the case for a large-scale independent clinical trial, paid for by Roche, that would settle once and for all the question of isotretinoin's safety.

As it happens, Roche itself is discussing just such a trial with the FDA - a trial which, says Daniel Kenny, 'would be done by international collaborative effort and would involve tens of thousands of patients'. End of story? Alas, no. The trial would require volunteer acne patients worldwide to be separated into two matched groups. One would be given isotretinoin; the other would receive either a placebo or an antibiotic. Scientists would then monitor and compare their psychological health.

It would answer the question all right, but at what cost? Normally, isotretinoin is prescribed only when other treatments fail. The consequence of such a trial, therefore, would be to withhold effective treatment from half the patients involved in it. Not only would they face lifelong scarring but they would risk the depression that acne itself can cause. 'There are some ethical problems here,' Daniel Kenny admits.

There are other problems, too. Increasing the rate of depression among the 'control group' could erode any differences between the two groups and make the entire study meaningless. Kenny insists the statisticians could 'adjust for this', but would any doctor allow his patients to take the risk? 'It's certainly a tricky one,' says Mark Goodfield. 'I would be anxious at withholding the drug. There would be problems recruiting patients for such a trial.

' So: if there is a way forward, what is it? How can patients who need this 'very important, almost uniformly effective drug', as Goodfield describes it, both be persuaded to take it and be protected from side effects?

It is common sense, not science, that supplies the answers: careful prescribing, good advice and close monitoring of patients not only by their doctors but also by their friends and families. Any change in mood or behaviour should result in the immediate withdrawal of the drug. It is sensible, unexceptionable advice, but for many, for certain, it simply will not be enough.