Off-label antipsychotic Rxs risky

MDs too liberal with powerful drugs, US report warns

By Owen Dyer

It's a pretty common practice, but off-label prescribing of atypical antipsychotic drugs isn't supported by enough reliable evidence, according to a report just released by the US Agency for Healthcare Research and Quality (AHRQ).

Most atypicals are approved for psychiatric conditions like schizophrenia and bipolar disorder. But their commonest off-label uses, as identified by AHRQ, run the whole gamut of mental illness: depression, obsessive-compulsive disorder, posttraumatic stress disorder (PTSD), personality disorders, Tourette's syndrome, autism and agitation in dementia. "I don't have the figures to prove it, but it's a pretty safe guess that atypicals are prescribed off-label more than almost any other class of drug," says Dr David Atkins, chief medical officer at AHRQ's Center for Outcomes and Evidence.

The report concluded that much of the research supporting such off-label prescriptions was either conducted on too few patients to constitute reliable evidence, or lacked scientific rigour. The authors also found that atypical antipsychotics can lead to adverse events including stroke, tremors, significant weight gain, sedation and gastrointestinal problems.

OFF-ROAD RULES
AHRQ's study is essentially a series of meta-analyses. A panel comprising leading experts in the most common psychiatric conditions dredged up nearly 3,000 studies, then reviewed 84 which met their criteria for size, duration and quality.

The result is a simple set of guidelines revealing what the evidence actually supports and what it doesn't. So, if you're thinking of going off the approved track with antipsychotics, here's what they recommend:

Behavioural disorders in dementia: Olanzapine, quetiapine and risperidone each get a half-nod here. A couple of studies supported the use of these drugs to treat psychosis, but a large trial in Alzheimer's patients showed the risks outweighed the benefits for treating behavioural disturbances. The report notes that a recent meta-analysis of 15 dementia trials found that patients randomized to take atypicals suffered 3.5% mortality, versus 2.3% amongst those taking placebo, a "small but statistically significant" difference. But the greatest source of danger is stroke. Five pooled trials suggested that olanzapine tripled stroke risk, a finding that in 2003 led the FDA to insist on printed warnings. Four pooled trials suggested that risperidone elevated stroke risk nearly fourfold.

Depression: Only the weakest evidence showed benefit from any atypical antipsychotic, whether used alone or in conjunction with SSRIs. A role for olanzapine was contraindicated by "moderately" strong evidence of its inefficacy.

Obsessive-compulsive disorder: There was some evidence that risperidone and quetiapine fared better than placebo, particularly in patients who don't respond to SSRI therapy, but too little data to judge the other drugs.

PTSD, personality disorders, autism: Little evidence supported the use of any atypical antipsychotic in any of these conditions. However, the combination of risperidone or olanzapine with an antidepressant and other psychotropic meds could relieve symptoms of PTSD in male veterans.

Tourette's syndrome: A very limited amount of data suggests potential benefit from risperidone.

Perhaps the largest body-blow to the second generation atypical antipsychotics was the finding that, in off-label use, the drugs don't seem to reduce the risk of tardive dyskinesia — a side effect that plagued traditional antipsychotics. Seven pooled trials of aripiprazole and risperidone in elderly dementia patients suggested an increased risk of extrapyramidal side effects — 2.5 times higher than placebo with aripiprazole and 2.8 times higher with risperidone. Eliminating that side effect was the single biggest selling-point of the whole drug class.

INFORMED RISK
Overall, the report concludes that many of the studies cited in defense of off-label use of atypicals were flawed by "highly selective enrollment criteria."

"With few exceptions," they sum up, "there is insufficient high-grade evidence to reach conclusions about the efficacy of atypical antipsychotic medications for any of the off-label indications."

Still, Dr Atkins insists the AHRQ doesn't want to discourage or condemn all off-label antipsychotic prescribing, merely to make it better informed. The practice is pervasive and here to stay, he says. "I think one reason for that is that atypicals' mechanisms of action are poorly understood. That makes it easier to hypothesize that they might help in a given condition, and harder to prove that they won't."

A CAUTIONARY TALE
Perhaps not coincidentally, Eli Lilly, maker of the best selling atypical antipsychotic olanzapine, has been in the news in recent weeks over allegations the company encouraged off-label use of the drug in its marketing campaigns, while concealing data showing a major risk of side effects including weight gain, hyperglycaemia and diabetes. The AHRQ has denied its report has anything to do with the headlines, but interestingly, the only area in which the authors found "high-quality evidence" was in the link between olanzapine and weight gain. Olanzapine users were six times as likely as placebo users to report weight gain in one trial judged sound by the AHRQ. In a head-to-head trial they were three times as likely to report weight gain as risperidone users, which itself caused weight gain in children in three pediatric trials, the report notes.

Second-generation drugs often seem to fall short of their early promise to eliminate the unwanted side effects of those that came before them. As they move from the realm of clinical trials into the real world of millions of long-term users, it soon becomes apparent that they bring their own, new problems. Such was the case with Cox-2 inhibitors, and to a lesser extent with SSRI antidepressants. Today, atypical antipsychotics are the ones getting a sceptical second look.

But Dr Atkins says it's easy to understand why physicians are willing to give these drugs a go, even outside their approved realm. "The conditions for which they're being tried are notoriously frustrating and difficult to treat, and many patients don't respond to first-choice treatments," he explains, "There's a lot of pressure to keep trying until you find something that works."