Avian influenza: main issues

SPECIAL EDITORIAL

Avian influenza: main issues

Influenza aviária: questões centrais

Maria Rita Donalísio

Membro da Comissão de Epidemiologia da abrasco e da FCM-Unicamp. Departamento de Medicina Preventiva e Social da Faculdade de Ciências Médicas da Unicamp. Caixa Postal 6111 Campinas - SP 13083-970 [e-mail]

Although human infection by the A H5N1 avian influenza virus was first recorded in 1997 in China, its incidence has been increasing and there have been cases identified in many countries in the past few years, following the pathway of the epidemic in birds, although it has also been detected in other animal species.

Human infections by the H5N1 influenza virus were confirmed in Vietnam and Thailand in 2003 and 2004, and arrived in Cambodia, Thailand, China, Turkey, Iraq and Egypt in 2005 and 2006. From the first reports until April 6, 2006, there were 192 cases in humans with 109 deaths, almost all of them related to direct contact with infected birds (World Health Organization).

The high lethality of human cases (around 50%) causes major concerns about the real risk of a possible pandemic with severe infections, to which human susceptibility may be high and universal.

However, it is impossible to predict when the pandemic will happen, how severe it will be, or the exact identity of its virus. It can be said that, in addition to the vaccine and specific antiviral drugs, not yet available in the public health system, one of the best strategies to delay viral transmission is viral and epidemiological surveillance, that is, the early etiological detection of the first cases, blockade to transmission, and early intervention. The epidemiological surveillance system will be tested again, with the task of changing transmission pathways and the impact of a severe epidemic. Will we be ready for it?

The Revista Brasileira de Epidemiologia prepared a group of relevant questions about avian influenza, which will be published in this and in coming issues. We invited researchers and managers involved with the design of national and regional strategies to fight a possible epidemic transmission of the H5N1 influenza virus in Brazil to discuss these questions. Here follow some specific and brief comments on some core issues regarding avian influenza: 1) prospects for viral surveillance in Brazil; 2) possible mutations of the H5N1 virus and its adaptation to inter-human transmission; 3) clinical aspects of avian influenza; 4) the efficacy of antiviral therapy in human cases and 5) prospects for the production of the vaccine in Brazil.

1. Is Brazil ready to detect the arrival of the H5N1 virus and to conduct viral surveillance?

• Profª Dra. Terezinha Maria Paiva. Respiratory Virus/Virology Laboratory/Adolfo Lutz Institute São Paulo State Health Department.

Brazil is ready to carry out the viral surveillance of the influenza virus, because it has been working in this area along with the Global Surveillance Network, coordinated by the World Health Organization, since the 1950's. The country has three laboratories that are part of the Global Network: Instituto Adolfo Lutz, Instituto Evandro Chagas, and Fundação Oswaldo Cruz. These laboratories are capable of conducting classical and molecular methodologies for the rapid and specific detection of the influenza virus.

Viral surveillance in Brazil was implemented by the National Influenza Surveillance Network under the coordination of the Ministry of Health in face of the risk of a pandemic. Currently, 45 Sentinel Units, spread throughout different states of the Federation, comprise the National Influenza Surveillance Network. Public health authorities are mobilized and have already prepared the National Influenza Contingency Plan, with the participation of all sectors of the society.

2. Which are the mutations involved in the adaptation of the H5N1 avian influenza virus to humans?

• Profª Dra. Rita Catarina Medeiros Souza. Tropical Medicine Core / Federal University of the State of Pará and Respiratory Virus Laboratory / Evandro Chagas Institute /Ministry of Health.

The species barrier for the type A influenza virus is established by many viral genes, involving not only the coded genes of the glycoproteins of the viral surface, hemagglutinin (HA) and neuraminidase (NA), but also internal genes like the PB2 of the polymerase complex. The influenza virus has a segmented RNA genome, which favors the accumulation of mutations, in addition to the possibility of rearrangements of the many segments when one cell is co-infected with two or more genetically different viruses. The viruses introduced in the human population during the 1957 and 1968 pandemics, caused respectively by viruses A (H2N2) and A (H3N2), were the result of a rearrangement between segments of the genome of a virus of avian origin with viruses that had already been circulating for some time in the human population. The genes of avian origin were HA, NA and PB1, in the case of virus H2N2, and HA and PB1, for virus H3N2, suggesting that the maintenance of internal protein-coding genes of human viruses is necessary for the adaptation of the new virus to the human species. Hence, the great fear of a global expansion of avian influenza by the H5N1 virus, increasing the risk of co-infection with an influenza virus of human origin, either in humans or in another hosts, like pigs, which would favor the emergence of hybrid avian/human virus with potential risk of better adaptation and fixation in the human population. Nevertheless, very recently, it has been demonstrated that virus A (H1N1), responsible for the Spanish influenza, was a virus of exclusively avian origin, which did not undergo a rearrangement with viruses from other origins, demonstrating the adaptation potential of an avian virus to humans without needing genetic rearrangement with a human virus. In this case, genetic evolution becomes clear with the accumulation of mutations in important sites of interaction of viral proteins with the host cell. The best-documented example is that of viral hemagglutinin.

In fact, the HA of the influenza virus is capable of differently recognizing its receptors, the sialic acids that exist in the viral replication sites of many hosts. The HA of avian influenza viruses has high affinity with the terminal sialic acids of the oligosaccharide chain, when linked to a2.3 with the underlying galactose. This type of sialic acid is predominant in the epithelial cells of the intestine of wild birds, the site where viruses multiply the most in these hosts. In contrast, the HA of the influenza viruses adapted to humans fixate with greater affinity to the sialic acids bound in a2,6 with galactose, which are most abundant in the non-ciliated epithelial cells present in the human respiratory tract. These receptor preferences partly define the species barrier, preventing the free passage of the virus from one host to the other. Very recently it has been shown that there is a population of ciliated epithelial cells in the human trachea that express, at density, the receptors for the avian virus, justifying dozens of human influenza cases by virus A (H5N1) that occurred since 1997, without a mutation in the coding gene for the HA. However, when these genetic changes happen at the level of the fixation protein site to the receptor, there may occur changes in its specificity. The HA of the avian virus will then be able to increase its affinity for type 2.6 sialic acids, thus determining an improvement in viral multiplication in the human respiratory tract, especially in the trachea and pharynx, which would favor airborne transmission of the virus from one person to another.

The amino acids in positions 226 and 228 have been classically involved in this interaction specificity between the HA and its receptor. The presence of a leucine or isoleucine or valine in position 226, in addition to serine in position 228, found in the human virus, determines the greater affinity of the glycoprotein to type 2.6 sialic acids, while glutamine and glycine in positions 226 and 228, respectively, provide greater specificity to type 2.3 sialic acids. These are the last amino acids found so far in the H5 of viruses isolated both in birds an in humans, thus revealing the maintenance of the affinity of the protein for receptors 2.3. However, it has already been demonstrated that replacements in other amino acid positions that comprise of the fixation site to the receptor, for example 193, 227 and 156, may interfere in the affinity of hemagglutinin to sialic acid, which is the first step for the change in the specificity of the viral protein. As these mutations are random, it is impossible to predict how or when they will happen. However, with the rapid geographical expansion of the H5N1 virus and its continuing genetic evolution, these mutations are expected to happen, thus determining a potential risk of fixation of the virus in the human population.

3. Which clinical aspects differentiate the seasonal human influenza from infection by the H5N1 virus?

• Prof. Dr. Luiz Jacintho da Silva. Internal Medicine Department/ University of Campinas Medical School

The cases of human influenza by H5N1 are impossible to differentiate from the cases of annual human influenza by H1N1 or H3N2, except, maybe, by the higher severity and lethality of the latter, around 50%.

We should remember, however, that the clinical experience with cases of human influenza by H5N1 is small and, possibly due to the bias of reporting and diagnosis, the cases identified tend to be the most severe.

Its greater lethality is due to the higher frequency of viral pneumonia, and deaths occurred due to respiratory failure. The apparent greater severity can be explained by the trend of the virus in determining viremia.

Of the 18 patients of 1997, 11 evolved to pneumonia and six died from respiratory failure. The development of these cases, as usual with influenza, is fast, within only a few days. The incubation period is also short, from two to eight days. The time from the onset of symptoms to death has ranged from four to 30 days. Encephalitis was one of the complications found.

Like other avian influenzas, conjunctivitis was also found, although not frequent, in the first 18 cases seen in Hong Kong in 1997. Gastrointestinal symptoms, particularly diarrhea, uncommon in influenza, were frequent in these cases. The occurrence of conjunctivitis and gastrointestinal symptoms are explained by the fact that the hemagglutinin of avian influenza viruses binds to sialic acid with alpha 2-3 receptors on the surface of the cells of the host. These receptors are found in the respiratory and digestive cells of birds, and in humans they are found in the cells of the ocular and digestive mucosa, being less frequent on the mucosa of the respiratory tract, where the receptors of sialic acid are alpha 2-6, to which human influenza viruses (H1, H2, and H3) are adapted.

The clinical management of these patients should not differ from the usual management of influenza, only with greater care due to the higher lethality found so far in humans. The use of antivirals is recommended until proven otherwise.

4. How effective can antiviral drugs be for infection by the H5N1 influenza virus?

• Prof. Dr. Dirceu Bartolomeu Greco. Internal Medicine Department/Medical School/ Federal University of the State of Minas Gerais.

The discussion on the efficacy and effectiveness of the antiviral drugs available for the treatment of a possible epidemic of avian infection by H5N1 is complex and current data are scarce and contradictory.

There are, today, four drugs of two different classes available for the treatment of infection by the influenza virus.

Neuraminidase Inhibitors. There are two drugs in this class that have been approved for use after one year of age: oseltamivir in 75 mg capsules (Tamiflu® 1999) and zanamivir in blisters for inhalation with 5 mg (Relenza ®1999), which reduce the severity and duration of the symptoms of seasonal influenza, types A and B. These drugs work on one of the two main surface structures of the influenza virus, the protein neuranamidase, which is practically the same in all usual strains of influenza. For efficacy, they should be taken within no more than 48 hours after the onset of symptoms. Oseltamivir is administered orally and zanamivir is available in powder for self-administration with a plastic inhaler.

M2 inhibitors. There are also two drugs in this class, which have been available longer: amantadine (Symetrel ®1966) and rimantadine (Flumadine ®1993) for the treatment and chemoprophylaxis of type A influenza, for adults and children over one year of age.

Efficacy

Oseltamivir and zanamivir

Despite the scarce clinical data available, H5N1 may be susceptible to neuraminidase. Resistance has been rare in the regular utilization for virus A Influenza infection, although it may develop with the extensive use in a pandemic. Most H5N1 strains tested proved to be susceptible to these drugs, but their effectiveness for the treatment of the infection is still unknown. A meta-analysis published recently (Jefferson T et al, 2006) showed that a daily dose of 150 mg of Oseltamivir was efficacious in the prevention of pulmonary complications in cases of influenza (OR 0.32, 0.18–0.57).

Amantadine and rimantadine

Some strains of avian H5N1 are known to be completely resistant. However, depending on the type of virus emerging in the case of mutations for the human infection, these inhibitors may be effective.

In the meta-analysis mentioned above, the efficacy of the anti-viral drugs available was approximately 60% when they were used to prevent influenza A. No evidence was found on the efficacy of these drugs in human infection by the avian H5N1.

On the other hand, there are reports on the lack of efficacy of oseltamivir in patients infected by avian H5N1 treated at the Tropical Diseases Center of Hanoi.

Limitations to the use of neuraminidase inhibitors

The major drawbacks are related to the limited production capacity and the very high price, making these drugs unaffordable to many countries. Early diagnosis is also necessary, because these drugs act in the first 48 hours of infection. This adds to the fact that individuals with chronic conditions have a higher risk of complications when infected by influenza viruses and, moreover, several diseases, including fast progressing bacteremia may present with initial symptoms similar to influenza, which may make therapeutic decisions difficult.

It should be also taken into account that although adverse events are relatively rare, they may be severe, with a risk of increasing viral resistance to these drugs if utilized indiscriminately.

Precautions

Amantadine and rimantadine should not be used to treat seasonal influenza; likewise, due to their low efficacy, this recommendation is also extensive to neuraminidase inhibitors. Their utilization should be restricted to severe epidemic or pandemic, along with other public health measures.

Other precautions

Amantadine: may cause mydriasis, and therefore is not indicated in non-controlled glaucoma.

Zanamivir: there is risk of severe adverse events such as bronchospasm, which contraindicates its use in individuals with asthma. There have also been reports of facial and oropharynx edema.

Oseltamivir: nausea and vomiting—that are less intense when taken along with food—have been reported.

Use during pregnancy: No clinical trials have been carried out with any of these drugs during pregnancy.

In brief

The H5N1 virus is apparently susceptible to neuraminidase inhibitors in laboratory assessments. However, it is not possible to anticipate its efficacy or risk of resistance with large-scale use in the case of a pandemic. There are reports of non-efficacy in patients treated in Vietnam, although it is not possible to completely exclude the possibility of oseltamivir having been administered late in these cases. On the other hand, some strains of avian H5N1 are known to be resistant to M2 inhibitors, although it is not possible to rule out its efficacy, if and when these strains change to better adapt to man.

There are no studies on the association of these drugs.

In the case of a pandemic by an avian H5N1 virus variant and without an effective vaccine duly available, the cautious and early utilization of antiviral drugs may protect against infection and decrease complications in patients with acute disease. There are many questions as to the effectiveness, safety, possibility of emergence of resistance and on the manner of establishing monitoring mechanisms, and guaranteeing sufficient production of drugs and world access. The current WHO recommendation is to expedite efforts to develop an effective vaccine, and that countries organize to face a possible epidemic. Oseltamivir should be given early to treat respiratory infections suggestive of H5N1 in patients directly involved in bird breeding or care, in which the presence of the virus has already been proven.

5. What are the prospects for preparing a vaccine for the H5N1 virus in Brazil?

• Prof. Dr. Isaias Raw. Butantã Institute / São Paulo State Health Department.

Some months ago, the Pan-American Health Organization (PAHO) called a meeting with some countries and international manufacturers of influenza vaccine. The agenda was the possible transfer of technology from manufacturers to national institutions. The answer was obvious: the transfer, in which manufacturers had no interest, was make-believe. Whoever did not own a plant built and set up with ordered and imported equipment (such as an industrial preparative ultracentrifuge), would not be able to manufacture vaccines in less than six years, when the pandemic would have either not happened or ended!

The only Institution in a final phase of implementing a plant was Butantã. The building of approximately 12,000 m² (including technical floor), despite the usual public bidding problems, will be ready before June 2006. The equipment, purchased with funds from the Ministry of Health, has been delivered and is ready to be installed and validated until next October. The most important technicians have been trained and the remaining employees are being hired by the State.

Technology has been totally transferred from Sanofi-Aventis, as part of the purchase of bulk vaccine strains, from 2000 to 2006. The formulated and bottled vaccines purchased by the Butantã Foundation, are being delivered to the Ministry for US$ 1.60, half of the international cost, leading to savings of US$150 million, with an investment of only 10 million in the plant!

In order to define vaccines for influenza, circulating strains are monitored every year, and after serotyping by domestic reference laboratories they are sent to the Center for Disease Control and Prevention, CDC, and other international laboratories for complete RNA sequencing. This allows analyzing the annual development of the virus in the North and South Hemispheres and to order, through WHO, the seed batches and reagents for dosing, which are, in turn, supplied to manufacturers. In this fashion, vaccines are internationally defined and uniform.

The production of H5N1 vaccine is also performed with the standard strain, defined by the WHO and changed, only if necessary, during the development of a possible pandemic. This strain is reconstructed from the avian virus and it is attenuated by the replacement of two amino acids.

The production planned by Butantã uses, as is the case of most manufacturers, fertilized eggs. Every year, an egg is used to obtain a triple vaccine. Given the vaccine is currently offered free of charge for individuals over 60 years, the population at risk, and to the medical-sanitary staff exposed to the virus, the national demand has reached about 20 million doses. Brazil is the world's largest bird raiser, and using about 30 million fertilized eggs in November does not affect the supply to the domestic, neither exporter market or the price.

The initial idea that the culture of the H5N1 virus could not be done in eggs because it would kill the embryo, has been proven to be wrong with the strain produced by the WHO. Some multinational companies had bioreactors available for animal cells, but this is not a solution for Brazil. Brazil does not have experience in vaccine production using animal cell culture for human use, in order to manufacture vaccines for influenza (Butantã has a scale production, and is expanding the production of vaccines for rabies in Vero cells and erythropoietin CHO in cells with in-house technology. Other viral vaccines are not produced in the country). Culture media have to be imported (or at least their quality certified components) and at a much higher cost than eggs. If manufacturers chose an influenza vaccine produced in cells, the supply of components of culture media could not be guaranteed.

Butantã decided not to wait for the influenza plant, which was already modified to meet WHO requirements for manufacturing a vaccine for a pandemic. In record time, with small, but fundamental investments form the Health Surveillance Department, we have rebuilt an experimental laboratory to start producing H5N1 vaccine. This lab will probably begin the production of experimental batches before next June.

Brazilian innovations in the composition of the vaccine

Butantã was already concerned with the reduction in the amount of antigen for the annual influenza vaccine, in order to reduce its cost. This would allow including 1-2 year-old children every year. During the PAHO meetings, mentioned above, we presented the first results of using adjuvant, which include the traditional aluminum hydroxide and a monophosphoryl-lipid A (MPLA), produced by an original method. Both aluminum hydroxide and only 02 micrograms of MPLA allow using only one fourth dose. Aluminum hydroxide has already been tried in 60 volunteers, showing that it is possible to use one fourth of the dose of influenza A vaccine. New trials are being programmed in lab animals and volunteers, using a vaccine to be administered next April to all the elderly, and the vaccination will be repeated as soon as we have a H5N1 vaccine.

The meaning of these studies is paramount. Every year we will be able to reduce the dose by a factor of four, with a comparable reduction in immunization costs. It is fundamental to equate public and free vaccination, which will always be expanded, according to costs. The influenza vaccine costs roughly US$43 million a year, a figure that can be reduced to about 15 million. For comparison, the remaining vaccines manufactured by Butantã (DTP, hepatitis B, rabies, BCG), and which represent 83% of vaccines actually produced in the Country, cost the equivalent to US$14.3 million to the Ministry. The recombinant vaccine against hepatitis B is delivered to the Ministry for less than R$ 0.80! The imported monovalent rotavirus vaccine costs US$55 million and in 2007-8 it will be replaced by the Butantã pentavalent vaccine at a significantly lower cost.

Another implication is that Butantã's influenza vaccine plant, planned to yield 20 million annual doses, can produce 60 million vaccine doses per semester. If the H5N1 vaccine grows at the same rate, we will have 180 million doses. Tests performed in Europe have shown that this vaccine is less immunogenic, demanding up to 4 doses per individual. If the adjuvant works as expected, a 15-microgram dose will suffice, and we will have the capacity to produce 60 million doses per semester. No country has planned to vaccinate the entire population. The target of the Northern Hemisphere is to make the vaccine available for about one tenth of the population, in order to interrupt transmission, if it is the case, from one patient to those who have contact with him/her, and for healthcare staff.

Publication Dates

• Publication in this collection
  04 June 2007
• Date of issue
  Mar 2006