The US Food and Drug Administration (FDA) has denied instituting any dramatic policy change in the way drugs in clinical trials are assessed for potential risk of suicidiality. “We are not requiring this information broadly as part of drug applications,” the agency stressed.

Last week a report in The New York Times claimed the FDA had quietly introduced “new rules” requiring drug manufacturers to study closely whether patients became suicidal during clinical studies. The newspaper called this “one of the most profound changes of the past 16 years to regulations governing drug development”.

The FDA says, though, it has been asking for this information “for some time” on a case-by-case basis, when other drugs in the same class “have been linked to suicidiality or when signals in clinical trial data suggest this might be a risk”.

As The New York Times noted, the issue goes back to the FDA’s still controversial meta-analysis showing that the average risk of suicidal behaviour or ‘suicidal ideation’ in more than 4,400 children and adolescents taking antidepressants was 4% compared with 2% on placebo. The finding led to the introduction of a ‘black box’ warning on antidepressants about increased risks of suicidal thoughts and behaviour in children and adolescents in October 2004.

One thing that convinced initially sceptical agency officials that the suicide risk with antidepressants was serious was a re-analysis of clinical trial data requested from Dr Kelly Posner and colleagues from Columba University and the New York State Psychiatric Institute’s Division of Child & Adolescent Psychiatry, the newspaper pointed out. Dr Posner’s team “created a detailed questionnaire called the Columbia Suicide Severity Rating Scale, now adopted by the drug agency as an often mandatory test to be used in clinical trials”, it added.

In fact, the Columbia Suicide Severity Rating Scale (C-SSRS) is a prospective version of the Columbia Classification Algorithm of Suicide Assessment (C-CASA), the standardised rating system used in the FDA’s meta-analysis of paediatric antidepressant trials. The agency says it contracted with Columbia University to develop this tool in 2001.

According to a paper published by Dr Posner and colleagues in The American Journal of Psychiatry in July 2007 (Am J Psychiatry 164: 1035-1043), the algorithm was also used to classify suicidal adverse events in a subsequent FDA analysis of antidepressant safety in adults and has “been mandated to be applied to all anticonvulsant trials and other centrally acting agents and non-psychotropic drugs”.

Rimonabant risks
Dr Posner’s system was more recently employed in the agency’s review of Sanofi-Aventis’ anti-obesity drug, rimonabant (Acomplia), The New York Times said. In June 2007 an FDA advisory committee voted 14-10 against approving rimonabant after concluding that the drug’s benefits did not outweigh the risk of psychiatric adverse effects, including suicide and seizures.

The newspaper also cited the potential psychiatric problems that emerged this month from a Phase II trial of Merck & Co’s cannabinoid-base obesity therapy taranabant, as well as growing fears that “drugs used to treat epilepsy, seizures and mood disorders may have similar effects”.

However, the FDA downplayed the suggestion of any formal rule change or that the agency, as The New York Times put it, “for years was inattentive to the psychiatric effects of new medicines”. On a case-by-case basis, the FDA may either ask sponsors prospectively to collect and assess information on psychiatric side-effects or retrospectively ask for further analysis of adverse-event data, noted Crystal Rice from the agency’s Center for Drug Evaluation and Research.

“These determinations are based on factors such as the class of the drug, what may be known from other members of the class, findings from animal studies, or signals of imbalances in psychiatric adverse events reported in controlled clinical trials,” Rice added. “This type of data collection and analysis is not new, rather FDA has been looking at this for some time for certain types of drugs.”

According to Rice, the agency has asked companies to analyse their data for suicidiality using the Columbia University tool only “a handful of times” since 2001. In two cases the tool has “yielded information that led us to list suicidiality among the possible side-effects in the labelling of a drug”, she said.

Rice acknowledged the FDA was “taking a proactive approach to better data collection for possible psychiatric side-effects for certain types of drugs” while emphasising there was no “requirement that this be done in all clinical trials”. The agency may ask a drug sponsor on a case-by-case basis “to collect and categorise data on psychiatric side-effects during clinical trials if FDA thinks there might be a cause for concern”, she commented.