Robert Roos  News Editor

Jun 8, 2007 (CIDRAP News) – Antibodies gleaned from four Vietnamese patients who survived H5N1 avian influenza were used successfully to prevent and treat H5N1 infection in mice, suggesting that the same approach might be useful in humans, according to a recent report by an international team of researchers.

The authors, reporting in the journal Public Library of Science (PLoS) Medicine, said the human antibodies, when used to treat sick mice, were effective not only against the H5N1 strain that infected the Vietnamese patients, but also against a later strain that came from Indonesia.

Anthony S. Fauci, MD, director of the National Institute of Allergy and Infectious Diseases (NIAID), said the study suggests that antibody treatment could be a valuable weapon against pandemic influenza.

"If the success of this initial study is confirmed through further laboratory and clinical trials, human monoclonal antibodies could prove to be valuable therapeutic and prophylactic public health interventions for pandemic influenza," Fauci commented in a news release. The NIAID was a sponsor of the study.

Antibody treatment, also known as passive immunotherapy, has been used to prevent chickenpox, hepatitis A and B, rabies, and respiratory syncytial virus infections. A crude form of it was used during the "Spanish flu" pandemic of 1918-19, when blood products from recovering patients were sometimes given to sick patients. A recent analysis of studies conducted at the time indicated that the treatment reduced mortality by about half (see link below).

The new study was conducted by a team from Vietnam, Switzerland, and the United States, with Kanta Subbarao of the NIAID in Bethesda, Md., as the senior author.

The antibodies used in the study were gleaned from four Vietnamese who were infected with a "clade 1" strain of H5N1 between January 2004 and February 2005 and donated blood after their recovery. Swiss researchers on the team extracted antibody-producing white blood cells, called memory B cells, from the samples and used Epstein-Barr virus to induce them to continuously produce large amounts of antibody, according to the NIAID release and the PLoS Medicine report.

Team members at the NIAID lab then screened 11,000 antibody-containing samples and found a few that, in the lab, neutralized H5N1 virus strains from 2004 and later. The researchers purified the B cells from these and created four monoclonal antibodies (mABs) that secreted H5N1-specific neutralizing antibodies, the NIAID said.

To find out if the antibodies could prevent H5N1 infection, the researchers treated groups of five mice with either of two H5N1 antibodies at one of three doses. Two other groups of mice were treated with human antibodies against diphtheria or anthrax to serve as controls. A day later, all the mice were given lethal intranasal doses of H5N1 virus.

All the control mice died within a week, while all the mice that received an H5N1 antibody called FLA5.10 survived, regardless of dose. The effectiveness of the second antibody, called FLA3.14, depended on the dose; 80% of mice that received the highest dose survived, but with the two lower doses, fewer survived.

The control mice were found to have high levels of H5N1 virus in the lungs and evidence of the virus in the brain and spleen as well, according to the report. The treated mice, in contrast, had 10- to 100-fold lower levels of virus in the lungs and no detectable virus in brain tissue.

To test the antibody method as a treatment for already-sick mice, the researchers exposed mice to a lethal dose of the 2004 strain of H5N1 from Vietnam. A total of 60 mice were given one of the four antibody preparations 24, 48, or 72 hours after infection, and control mice were treated with human diphtheria antibodies. All the control mice died within 10 days, while 58 of the 60 treated mice survived.

Encouraged by these findings, the team went on to try the antibodies as treatment for mice infected with an H5N1 strain collected from Indonesia in 2005 (a clade 2 strain). The mice were treated 24 hours after infection. Mice treated with a control antibody and with one of the H5N1 antibodies died, but those treated with the other three H5N1 antibody preparations survived.

"These data provide proof of concept that mAb therapy for at least 72 h[ours] postinfection in the mouse model can markedly improve survival from highly virulent H5N1 infection," the authors write. "Importantly, these data also imply it is possible to obtain significant cross-protection against a Clade II H5N1 virus using a mAb elicited by a Clade I virus."

They add, "Potentially, a cocktail of these cross-reactive antibodies mAbs could represent an adjunctive treatment option against [human] H5N1 infection."

They caution that it is not certain that the antibodies used in this study would be effective against an emerging H5N1 pandemic virus. But they say they are encouraged by the antibodies"e; broad neutralizing activity and the "moderate doses" required to confer protection in mice.

The NIAID said the researchers now plan to scale up the protection of H5N1 antibodies, test them in other animal species, and, if those studies are successful, test them in humans.

Other experts hail the findings as promising, but there are reservations about how much of a practical impact passive immunotherapy could have in a flu pandemic.

Gregory Poland, MD, a flu vaccine expert at the Mayo Clinic in Rochester, Minn., called the report "a very promising development." If further studies support the approach, it could be used along with antiviral drugs to treat human patients with severe H5N1 disease, and also to protect people who might be exposed to H5N1 and have not been vaccinated, he said.

"If [H5N1] cases were going to be admitted to the University of Minnesota Hospital and healthcare workers had to go in and take care of them, this might be a useful interim measure," said Poland, who is a professor of medicine in infectious diseases at the Mayo College of Medicine and directs the Mayo Vaccine Research Group and Program in Translational Immunovirology.

"One can think of it a bit like how we used to use immunoglobulin to protect against hepatitis A before we had vaccine," Poland added. "But something like this could not and would not supplant vaccines as the cornerstone of prevention of disease."

He listed other caveats as well. One big if is whether the findings in mice will be duplicated in humans. Though the researchers demonstrated "cross-clade protection," "that may or may not be true in a human model," he said. Another question is whether the antibody preparations could be produced in large amounts.

Assuming the treatment turns out to be effective, the timing of its use will be important, Poland said. "If you give it too far in advance of exposure, there won"e;t be any circulating in the bloodstream. If you give it too long after exposure, it won"e;t help. . . . And we don"e;t know what the effect would be of giving this and a vaccine in close proximity."

Michael T. Osterholm, PhD, MPH, an infectious disease expert and pandemic preparedness advocate, agreed that the findings are promising from the scientific standpoint but expressed doubts about the practical potential to exploit them in the face of a pandemic.

"In theory it makes obvious sense—it"e;s an extension of the immunization process," said Osterholm, director of the University of Minnesota Center for Infectious Disease Research and Policy, which publishes CIDRAP News.

The main question on the scientific side is whether the antibody treatment would have broad enough activity to be effective against an emerging pandemic strain of H5N1, which could differ from the strain used in producing the antibodies, he said.

But on a practical level, Osterholm said, "Once a pandemic hits, there won"e;t be time or materials to obtain and stockpile large volumes of this. What in theory is ideal, is in practice maybe a nightmare. I don"e;t see how the plasmapharesis community is going to be able to quickly gear up to actually make lots of this antibody and then move it into the clinical setting in a timely way to have much impact. . . . The idea of trying to create millions of immunotherapy treatments is a stretch."

Concerning Poland"e;s idea of using passive immunotherapy to protect healthcare workers who could be exposed to H5N1 patients, Osterholm commented, "Now there"e;s a more targeted application. But that"e;s going to be a hell of an ethical dilemma. Are you going to get a consensus in the community to actually allow giving it to healthcare workers and not others?"

Another flu expert, Dr. William Schaffner of Vanderbilt University, called the study by Subbarao"e;s team "a very lovely, elegant proof of principle," according to a recent Associated Press (AP) report.

Schaffner suggested that the possible applications of this principle include not just H5N1 influenza, but also ordinary seasonal flu, which kills thousands of people each year.

"This has the dual potential of being useful potentially in a pandemic, but perhaps more so on an annual basis," he told the AP. "That"e;s where I think the real excitement is."

Simmons CP, Bernasconi NL, Suguitan AL, et al. Prophylactic and therapeutic efficacy of humamn monoclonal antibodies against H5N1 influenza. PLoS Med 2007;4(5):0928-36 [Full text]