Robert Roos  News Editor

Dec 22, 2005 (CIDRAP News) – A new report says oseltamivir-resistant forms of H5N1 avian influenza virus were found in two Vietnamese girls who died of the infection, raising doubts about the antiviral drug that many countries are counting on to help protect them from a potential flu pandemic.

One of the two girls died even though she started receiving oseltamivir (Tamiflu) within the first 2 days of her illness, the recommended window for effective treatment, according to the report in today"e;s New England Journal of Medicine. The other girl was not treated until the sixth day of her illness.

The authors of the report say their findings suggest that a higher dosage, longer treatment course, or combination with other antiviral drugs may be necessary to ensure the effectiveness of oseltamivir.

Roche, the manufacturer of oseltamivir, agrees with that assessment and says that studies of the safety of a higher dosage are about to get under way.

Second study weighs in on stockpiling
In a separate article published today, a group of experts who have been monitoring resistance to oseltamivir and similar drugs says the evidence so far—including the New England Journal report—does not suggest that stockpiling of the drug is useless.

"The available data do not indicate that potential oseltamivir resistance should be a deterrent to its stockpiling for pandemic response," says the report by Frederick Hayden and other members of the Neuraminidase Inhibitor Susceptibility Network (NISN). It was published online today in Antiviral Therapy.

The report on the Vietnamese patients was prepared by a team from the Hospital for Tropical Diseases in Ho Chi Minh City, Vietnam, and Hong Kong University, with Menno D. de Jong of the Vietnamese hospital as first author.

Their study focuses on eight patients who were treated for confirmed H5N1 infection at the Hospital for Tropical Diseases between February 2004 and January 2005. Throat samples were collected from the patients for analysis at admission and later in their illness. H5N1 infection was confirmed by polymerase chain reaction.

All the patients were started on oseltamivir the day of admission to the hospital, which varied from 2 to 8 days after the onset of illness. They received the recommended regimen of 75 mg twice a day for 5 days. Four of the patients died.

The researchers did a sequence analysis of the H5N1 virus"e;s neuraminidase gene to look for resistance, signaled by the substitution of tyrosine for histidine at amino acid position 274. This mutation was found in two patients, a 13-year-old girl and an 18-year-old girl.

The 13-year-old was hospitalized the day after she fell ill with a fever and cough, which was also the day after her mother had died of H5N1 infection. Despite oseltamivir treatment, the girl"e;s condition worsened on her fourth day in the hospital, and she died of severe pneumonia on the seventh day.

The viral load in her throat was higher by the time of her death than it had been earlier, which, along other laboratory evidence, suggests that "the development of drug resistance contributed to the failure of therapy and, ultimately, the death of this patient," the report says.

The 18-year-old was hospitalized and started on oseltamivir 6 days after she had fallen ill, but she died after 2 weeks in the hospital. Nonresistant H5N1 virus was found in a sample taken 2 days after she was hospitalized, but the resistant form was found 6 days later.

Although the connection between viral resistance and the 18-year-old girl"e;s death was less clear than in the case of the 13-year-old, "The presence of replicating virus after 14 days of illness suggests an effect on the outcome," the article says.

It also says the viral load in throat specimens from the four patients who survived dropped quickly to undetectable levels during their treatment.

"Our observations suggest that at least in some patients with influenza A (H5N1) virus infection, treatment with the recommended dose of oseltamivir incompletely suppresses viral replication," the authors write. Consequently, "Strategies aimed at improving antiviral efficacy (e.g., the use of higher doses, longer durations of therapy, or combination therapy) may deserve further evaluation."

Tamiflu manufacturer, study authors comment
Roche, Swiss-based maker of Tamiflu, released a statement today agreeing that such strategies deserve consideration. The company said some data already support the safety of using a higher dosage of the drug, and clinical trials assessing the efficacy of a higher dose are scheduled. Roche is collaborating with the National Institutes of Health and the World Health Organization (WHO) on that research, the statement said.

Tran Tinh Hien of the Hospital for Infectious Diseases, one of the study"e;s authors, says Vietnamese health officials are already recommending increasing the dosage of oseltamivir for avian flu patients, according to a Reuters report published today.

"We still recommend the use of Tamiflu for bird flu cases as soon as possible and at higher doses as there is no replacement yet," he said. He added that the Vietnamese Ministry of Health has increased the treatment period to 7 days from 5 days.

The article published today by NISN, the experts who have been monitoring resistance to the neuraminidase inhibitors, say the new findings do not contraindicate stockpiling of oseltamivir, but much more research is needed. The neuraminidase inhibitors include oseltamivir and zanamivir (Relenza).

The group said no one knows how often resistance emerges in H5N1 patients being treated with oseltamivir, and the clinical consequences of such resistance are also unclear.

Many H5N1 patients treated with oseltamivir have died, but in most cases treatment was started late, after pneumonia had already developed, the report says. Some evidence suggests that the emergence of oseltamivir resistance early in treatment may lead to treatment failure, but more studies are needed.

The group also said there is no indication that H5N1 viruses now circulating in birds are resistant to neuraminidase inhibitors. Further, the likelihood that oseltamivir-resistant strains will spread in the community appears low, in contrast to the situation with two older antiviral drugs, amantadine and rimantadine, to which ordinary flu viruses are often resistant. In animal studies, the article says, the mutation that confers resistance in both H5N1 and H1N1 viruses reduces infectiousness 100-fold and reduces viral replication more than 10-fold.

The NISN statement also says that all avian and human H5N1 isolates tested so far by the Centers for Disease Control and Prevention have been susceptible to zanamivir. However, zanamivir, which is inhaled, has not been tried in human H5N1 patients.

Despite this, "Inhaled zanamivir would be a therapeutic consideration if oseltamivir resistance were likely to be present," the NISN members write. They also say the drug would be "an appropriate choice for pandemic response stockpiles."

Other experts offer opinions
To infectious disease expert Michael T. Osterholm, PhD, MPH, the report from Vietnam "reminds us again that none of us know how much drug [oseltamivir] we have in the stockpile." If a longer, higher-dose regimen is needed, a stockpile now described as 3 million treatment courses is actually smaller, said Osterholm, director of CIDRAP, publisher of this Web site.

Osterholm also said the report suggests oseltamivir resistance can have much graver consequences in H5N1 cases than in ordinary flu. In the latter, drug resistance has not been associated with treatment failure or a severe outcome, but "with H5N1 this may be a very different outcome," he said.

Osterholm called for clinical studies of the use of oseltamivir very early in the illness and at a much higher dosage than is used in ordinary seasonal flu. If that approach improves outcomes, it would have "tremendous implications for how we get Tamiflu to patients in a timely manner," he said.

A WHO official said the resistance findings are "not necessarily alarming" but do point up the need for more information, according to the Reuters report.

"What really is critical is understanding whether the way we are using the drugs contributes" to resistance, said Keiji Fukuda of the WHO"e;s global influenza program.

Some resistance is expected whenever antivirals and antibiotics are used, Fukuda said, adding that using too-small doses or too short a treatment regimen can promote resistance.

De Jong MD, Thanh TT, Khanh TH, et al. Oseltamivir resistance during ttreatment of influenza A (H5N1) infection. N Engl J Med 2005 Dec 22;353(25):2667-72 [Full text]

Hayden F, Klimov A, Toshiro M, et al. Neuraminidase Inhibitor Susceptibility Network position statement: Antiviral resistance in influenza A/H5N1 viruses. Antivir Ther 2005;10(8):873-7 [Abstract]