[Ip-health] Financial Times: Seroxat lessons

[Spring Gombe]

Seroxat lessons
Published: March 9 2008 18:47 | Last updated: March 9 2008 18:47

Efforts by British medicine regulators to prosecute GlaxoSmithKline
for allegedly delaying and playing down warnings about suicidal
feelings in children taking Seroxat, its antidepressant, have
collapsed after more than four years. Everyone involved should draw
the lessons far more swiftly.

The regulatory framework has improved, partly in response to the GSK
case. Drug companies are now required to register and report to
regulators more extensively on side-effects identified during clinical
trials conducted within the European Union. They also voluntarily and
publicly disclose details of trials.

Yet some of the loopholes that applied in 2003 are still present. GSK
avoided prosecution because of an absurd distinction: the rules were
focused on reporting side-effects of drugs when tested in line with
their authorised use, rather than in studies of new or alternative
=93off label=94 uses =96 where doctors (sometimes justifiably) prescribe at
their own discretion, despite lack of formal regulatory approval.

Most importantly, British and EU-wide rules must therefore be
tightened to ensure that drug companies are obliged to report all
potential new risks they identify with their drugs, regardless of why
or where the trials were conducted. They should notify regulators in
confidence swiftly, even ahead of more detailed scrutiny of the data.
Companies have an ethical and legal responsibility to comply with the
spirit of the rules.

However, the Seroxat case is complicated because there were no stat
istically significant signs of suicidal feelings in any of the nine
individual clinical trials that GSK conducted in children over several
years up until 2002. The risk was only identified in a subsequent
=93meta-analysis=94 that pooled all the different sets of results.

Transparency of data is critical so that independent experts can study
results. GSK=92s own public register, for instance, includes the
original inconclusive paediatric studies but not the meta-analysis
that identified the problems. That should change. But the dilemma is
that some companies conduct fresh studies =96 to test new authorised
uses for their drugs or attempt to prove better efficacy than rival
treatments =96 while others do not. More data may provide evidence to
boost sales, but may also unearth risks that would not otherwise have
come to light.

That suggests the need for a debate on broader public funding of
additional trials and analyses =96 or obligations on companies to run
them =96 at least for drugs that are widely used, or offer powerful
potential benefits or risks.

Copyright The Financial Times Limited 2008
--------------------------
posted by
Spring Gombe
spring.gombe@keionline.org