Tamiflu's Effect on Children Studied

The Food and Drug Administration is examining the deaths of 12 Japanese children who took the Roche Holding AG drug Tamiflu, though it said that determining whether the medicine played a role was difficult.

In comments and its own filings to an FDA advisory committee, Roche said there was "no causal relationship between any of the reported cases of death and the use of Tamiflu," and many of the children had other conditions as well as flu.

It said that there isn't a higher rate of adverse events in children taking Tamiflu compared with all children with flu.

The FDA posted documents on Tamiflu in advance of today's meeting of an advisory committee for pediatric issues, which will look at the safety record of Tamiflu and other medicines -- a routine examination for all drugs that win marketing exclusivity tied to pediatric testing.

Tamiflu, a treatment for influenza, has become extremely high-profile in recent months as countries and individuals have sought to stockpile it as a defense against a possible flu pandemic.

Some of the deaths were sudden, with cardio-respiratory arrest listed as the cause.

There was one suicide and one case of acute pancreatitis, or an inflammation of the pancreas.

The FDA examined its own adverse-events database tracking drugs on the market, which is generally seen as carrying limited value since U.S. doctors aren't required to file reports and it is difficult to figure out the true rate or cause of a problem.

Besides the deaths, the FDA said its analysis of adverse events reported between March 22, 2004, and April 22, 2005, had turned up 32 examples of children 16 and younger who had neurological or psychiatric problems while taking Tamiflu, including delirium and hallucinations.

Of those, 31 occurred in Japan, where patients take Tamiflu much more often than in other parts of the world. In a report, the agency called the problems "concerning."

The second-biggest category of adverse events was a dozen skin and other reactions in young people, of which 11 occurred in Japan.

The FDA said that between April 2004 and March 2005, 1.8 million Tamiflu prescriptions were dispensed in the U.S., of which 456,000 were for people 16 and younger, according to data from Verispan.

Roche said yesterday that world-wide, 13 million young people have been treated with the drug since its launch in 1999, with 11.6 million of those in Japan.

David Reddy, Roche's head of pandemic planning, yesterday said there have been no deaths of pediatric patients taking Tamiflu reported in the U.S., and world-wide, there have been 59 reports of neurological or psychiatric events in young people, with a rate no higher than that of all children with flu.

"We are confident that Tamiflu can be used safely and effectively by children as well as adults," he said.

The FDA said that the adverse-event reports "provide insufficient detail to make clear assessments of causality." While the neurological and psychiatric incidents reported from Japan may reflect the country's particular concern about such problems tied to flu, the skin and sensitivity reactions "do not appear to be related" to the flu itself or a risk linked to Japanese patients in particular, the FDA said.

Write to Anna Wilde Mathews at anna.mathews@wsj.com and Jeanne Whalen at jeanne.whalen@wsj.com