Influenza Vaccine—Live

WHY THE DISEASE IS IMPORTANT

Influenza viruses are the most common cause of lower respiratory tract infections. The number of cases of influenza that occur varies depending on the susceptibility of the population to the virus and the infectiousness of the virus during an outbreak. In addition, influenza seasons vary in length and severity. In the United States, an estimated average of 36,000 deaths were associated with influenza epidemics between 1990 and 1999, and approximately 226,000 hospitalizations between 1979 and 2001 ; approximately 90% of the deaths attributed to influenza occurred among persons age 65 years or older. In 2010, the Centers for Disease Control and Prevention (CDC) released revised estimates for influenza-associated deaths from 1976 to 2007 that ranged from approximately 3000 to 49,000 per year. Indeed, WHO estimates that seasonal influenza can result in 290,000–650,000 deaths from respiratory causes every year around the world though this estimate does not take into account influenza-related mortality from other diseases such as cardiovascular disease. Among influenza viruses, influenza type A viruses are responsible for the greatest amount of morbidity and mortality, and the nonhuman reservoir of influenza A viruses creates the potential for global pandemics as was seen in 1918, 1957, 1968, and 2009. Although influenza B does not cause pandemics, it is responsible for regional epidemics that are generally less severe than influenza A epidemics.

Influenza incidence rates vary with the nature of the epidemic virus strains and the population affected. Rates calculated from surveillance of persons with upper respiratory illness suggest that influenza is responsible for roughly 10–20% of all respiratory illnesses per epidemic year, with rates for influenza A being somewhat higher than those for influenza B. Most studies have found infection rates in preschool- and school-age children to be much higher than in adults, particularly during influenza B epidemics. ^, Consequently, families with school-age or younger children suffer disproportionately from influenza. These findings support the ideas that influenza infects persons with the lowest prior immunity and that children are a common source of influenza in the community. Furthermore, it forms the rationale for influenza vaccine strategies that target children, to both directly protect them and indirectly protect the wider population including vulnerable groups, through reduction in influenza transmission.

INFLUENZA

Influenza viruses are spread from person to person primarily via coughs and sneezes of infected persons. The incubation period for influenza is 1–4 days, with an average of2 days. Adults and children are typically infectious from the day before symptoms appear, until approximately 5 days after illness onset. Children can be infectious for a longer period, and very young children can shed virus for up to 6 days before symptom onset. Severely immunocompromised persons can shed virus for weeks.

Uncomplicated influenza illness is characterized by the abrupt onset of systemic and respiratory signs and symptoms (e.g., fever, myalgia, headache, severe malaise, nonproductive cough, sore throat, and rhinitis). Respiratory illness caused by influenza is difficult to distinguish from illness caused by other respiratory pathogens including SARS-CoV-2 on the basis of symptoms alone. Sensitivity and predictive value of clinical definitions can vary, depending on the degree of cocirculation of other respiratory pathogens and the level of influenza activity.

Influenza illness typically resolves after a few days, although cough and malaise can persist for more than 2 weeks. Influenza can exacerbate underlying medical conditions (e.g., pulmonary or cardiac disease); lead to primary influenza viral pneumonia, or to secondary bacterial pneumonia; or occur as a coinfection with other viral or bacterial pathogens. Influenza infection is also associated with encephalopathy, transverse myelitis, Reye syndrome, myositis, myocarditis, and pericarditis.

Multiple mechanisms provide resistance to influenza in humans. Acquired immunity to influenza is mediated by serum antibodies, antibodies at mucosal surfaces, and influenza-specific T cells. Antibody alone is sufficient for protection, and T cells play a role in recovery. Innate immune responses, such as production of interferon or other antiviral factors by macrophages, may also contribute to resistance to influenza infection. Immunity after natural infection provides long-lived protection against influenza virus infection and disease in the absence of antigenic shift or drift. In 2009, when the swine-origin H1N1 pandemic influenza virus (H1N1pdm09) emerged, older adults who had prior exposure to antigenically related H1N1 influenza viruses were found to have cross-reactive antibodies to the pandemic virus, and disease was not as severe in this age group as in younger adults. The H1N1pdm09 strain has now replaced previously circulating seasonal H1N1 viruses, so in this chapter, H1N1 viruses isolated after 2009 are referred to as H1N1pdm09. Immunity acquired by natural infection can provide some protection against influenza strains that have undergone antigenic drift.

VIROLOGY

Because of the complexity of the subject, it is beyond the scope of this chapter to address all aspects of influenza virology. For a comprehensive review of the biology of influenza viruses, the reader is directed to other texts.

Orthomyxoviruses have segmented, single-stranded, negative-sense RNA genomes. Influenza A and B viruses have eight RNA segments that encode at least 10 proteins ( Table 34.1 ): hemagglutinin (HA), neuraminidase (NA), three polymerase proteins (PA, PB1, and PB2), nucleoprotein (NP), nonstructural protein 1 (NS1), nuclear export protein (NEP), and matrix proteins (M1 and M2 in influenza A; M1 and BM2 in influenza B). Additional open reading frames in several gene segments encode proteins such as PB1-F2 and PA-X. The protective antibody response is directed at the two major surface glycoproteins, HA and NA. Influenza A viruses exhibit both antigenic drift and antigenic shift, ^{, ,} whereas influenza B viruses show only antigenic drift. Antigenic drift is caused by point mutations in the HA and NA genes that are driven by immune pressure and the infidelity inherent in the replication of RNA genomes by an RNA-dependent RNA polymerase that lacks proof-reading function. Although not subject to antigenic change, other genes also undergo mutational drift. The rate of drift can vary among genes and viral subtypes. Antigenic shift, however, requires complete replacement of one or both of the surface glycoprotein genes and is a consequence of the segmented genome of influenza viruses. During mixed infections, gene segments from two parent viruses can reassort to generate a virus with a novel gene constellation. ^, New antigenic changes, occurring by either mechanism, allow the virus to overcome existing immunity in previously infected and/or vaccinated hosts.

PATHOGENESIS AS IT RELATES TO PREVENTION

The pathology caused by influenza virus has been reviewed extensively. ^{, ,} Epithelial cells in the upper airway are most likely the primary site of infection and infection is usually restricted to the upper respiratory tract and trachea with bronchitis and tracheitis, although lung involvement may occur during severe influenza. Otitis media is common in children. Viremia has been reported only sporadically ^, amid numerous negative reports. Influenza B demonstrates pathology similar to that of type A and is common in children. The amount of virus present probably determines the extent and severity of symptoms. ^,

The most precise diagnosis of influenza is obtained by reverse transcriptase–polymerase chain reaction (RT-PCR) or by virus isolation in embryonated eggs or in cell cultures of nasopharyngeal or tracheal aspirates. Serologic diagnosis can be made by demonstrating fourfold rises in hemagglutination-inhibiting (HAI) antibodies.

LIVE ATTENUATED VACCINES

The segmented genome of orthomyxoviruses that permits genetic reassortment has been employed to generate vaccine strains containing the genes for the HA and NA of newly emergent wild-type (wt) viruses, while retaining other genes from attenuated strains though the incorporation of one or more internal protein genes from wild-type viruses that have been proposed and investigated in preclinical models to improve the efficacy of LAIV. ^, Attenuated master donor strains must be shown to not cause significant illness in humans and to reproducibly pass the property of attenuation on to reassortants through genes other than the HA and NA. Approaches to developing live attenuated vaccines for influenza include host-range mutants temperature-sensitive (ts) mutants, cold-adapted (ca) mutants, NS1 deletion mutants, M2 cytoplasmic tail mutants, replication-deficient viruses, genetic rearrangements (reviewed in ), and codon pair deoptimized influenza viruses, but this chapter will focus on the cold-adapted LAIV that is licensed for prevention of seasonal influenza in the United States and several other countries (see below).

HISTORY AND DEVELOPMENT OF COLD-ADAPTED (CA) INTRANASAL LIVE ATTENUATED VACCINE

The ca A/Ann Arbor/6/60 (H2N2) (A/AA ca) and ca B/Ann Arbor/1/66 (B/AA ca) viruses that are the backbones of the LAIV licensed in the United States were developed at the University of Michigan. ^, The A/AA ca virus was derived in primary chick kidney (PCK) cells from a virus originally isolated from an ill child. This virus was grown at successively lower temperatures in PCK cells followed by seven serial plaque-to-plaque purifications. This virus had the following three properties: ts, which implies that the vaccine virus replication was reduced at the higher temperature (39°C); ca, which implies that the virus grew well at a reduced temperature (25°C) compared with permissive temperature (33°C); and attenuation (att), which implies attenuated behavior in ferrets with reduced clinical signs of temperature elevation and/or weight loss and absence of detectable virus in the lungs. The B/AA ca vaccine donor strain was developed in the same manner, although it took fewer intermediate passages to achieve the final ca variant for type B virus than for type A. The wt B/AA/1/66 influenza virus was restricted for growth at 36°C at the time of isolation, and therefore the ts phenotype is defined by a reduction in replication titer at 37°C compared with the achievable titer at 25°C or 33°C. ^{, ,} Changes in the nucleotide sequence for all eight gene segments of the attenuated A/AA ca virus, compared with the nucleotide sequence of its wt counterpart, have been identified. Four major loci at base positions 1195 and 1766 of the PB1 gene, 821 of the PB2 gene, and 146 of the NP gene were identified as determinants of the ts phenotype by reverse genetics. Position 2005 of the PB1 gene was also found to contribute to the ts phenotype. ^, The ts, att, and ca phenotypes of the B/AA ca donor strain are also conferred by multiple genes: PA and NP for ts; PA, NP, and M for att; and PA, NP, and PB2 for ca. The concept underlying the use of ts viruses as live attenuated respiratory virus vaccines is that the vaccine virus will replicate in the cooler upper respiratory tract and induce immunity but will be restricted in replication and will therefore be attenuated in the warmer, core body temperature of the lower respiratory tract.

The reassortant vaccine viruses included in LAIV were previously generated by classical reassortment, achieved by coinfection with vaccine donor viruses and wt influenza viruses, but they are now derived using plasmid-based reverse genetics as discussed later ( Fig. 34.1 ). The resulting recombinant reassortant virus contains six internal gene segments of the vaccine strain combined with the two antigen-encoding gene segments, HA and NA, and it is referred to as a 6 : 2 reassortant. Numerous reassortants have been made using the A/AA ca virus and B/AA ca donor viruses, and, in all cases, attenuation, antigenicity, and genetic stability have remained unaltered and consistent. ^{, ,} Influenza A and B ca donor strains were also developed for a LAIV in Russia. The current Russian influenza A donor strain, A/Leningrad/134/47/57, contains changes in at least one of the polymerase genes as well as in the HA, NA, NP, and M genes. ^,

Reverse Genetics

The influenza virus genome comprises eight negative-sense virion (v) RNA segments that cannot function as messenger RNAs (mRNAs) to produce proteins; to initiate replication of the viral genome, viral polymerase proteins carried in with the virion must transcribe and produce mRNAs from each viral gene segment. Several systems have been developed that allow recombinant influenza viruses to be produced entirely from cloned plasmid DNAs. Hoffman and colleagues further refined these techniques and developed an eight-plasmid system. In this method, each viral complementary (c) DNA is inserted into a dual-promoter bidirectional plasmid where the viral gene is flanked by RNA polymerase I (pol I) and RNA polymerase II (pol II) promoters. Thus, eight negative-sense vRNAs and positive-sense mRNAs are generated from the eight plasmids.

This system is now used routinely to generate the seed virus for the AA ca seasonal LAIV. The HA and NA gene segments of the predicted epidemic strain are cotransfected into a cell with plasmid DNAs encoding the other six gene segments of the vaccine donor strain such as A/AA ca (see Fig. 34.1A ). The resulting recombinant virus contains six internal gene segments of the vaccine strain, combined with the two gene segments encoding the surface glycoproteins, HA and NA. Another application of reverse genetics is exemplified by the construction of live attenuated pandemic vaccine viruses by plasmid rescue. ^, Virulence determinants, such as the polybasic cleavage signal between the HA1 and HA2 subunits of the H5 HA, can be modified or removed by recombinant DNA techniques. Plasmid rescue techniques have advanced basic research on influenza virus as well as vaccine development.

CHARACTERISTICS AND FORMULATIONS OF LIVE ATTENUATED INFLUENZA VACCINE

LAIVs based on the A/AA ca and B/AA ca master donor strains were first licensed in the United States in 2003 ⁷⁷ as a trivalent vaccine. The need for a quadrivalent seasonal influenza vaccine to provide coverage against the two lineages of influenza B that circulate in the human population was a matter of debate for several years. The influenza B epidemic strain was difficult to predict, and there was often a mismatch between the vaccine strain and the circulating influenza B virus. For example, between 2001 and 2010, the influenza B component of the vaccine was of the correct lineage in five out of the 10 influenza seasons. Studies by the CDC determined that production of quadrivalent seasonal influenza vaccines would be possible, and would have added benefit in protection against influenza B disease. The clinical trials conducted with the trivalent formulation of LAIV also supported licensure of a quadrivalent LAIV (Q/LAIV), because both vaccines were manufactured using the same process, the dose of each of the component strains is the same, and both formulations are administered by the same route, in the same diluent. In addition to the clinical experience with trivalent LAIV, bridging noninferiority clinical trials were conducted in support of licensure of Q/LAIV. ^, Q/LAIV had a similar safety profile to the trivalent LAIV in adults and children and was shown to be noninferior to trivalent LAIV in terms of immunogenicity (see “Noninferiority Studies to Support Licensure of Quadrivalent LAIV” below). In early 2012, Q/LAIV was approved by the U.S. Food and Drug Administration (FDA). Q/LAIV, marketed in the United States as FluMist Quadrivalent, was available for the first time in the 2013–2014 influenza season, and contains two influenza B components, representative of both the B/Victoria and B/Yamagata lineages. Q/LAIV contains 10 ^6.5 –10 ^7.5 focus forming units (FFU) of the four component viruses and is administered in the same manner and volume as the trivalent LAIV. Q/LAIV has now replaced trivalent LAIV. Q/LAIV is manufactured and marketed by MedImmune in the United States. It is also licensed in Canada, the United Kingdom, Austria, Finland, Germany, Norway, Spain, Sweden, and Israel. In Canada, Q/LAIV is marketed as FluMist Quadrivalent; elsewhere it is marketed under the trade name Fluenz Tetra.

LAIV is sprayed into the nose using a simple syringe-like device that delivers a 0.1-mL volume of a large-particle aerosol into each nostril for a total volume of 0.2 mL ( Fig. 34.2 ). Vaccination of children requires minimal cooperation. The device is easy to use, and the vaccine is readily accepted and preferable to many over a parenteral injection by needle and syringe. Studies demonstrated self-administration of LAIV was successful, and may be an option for situations where mass immunization is necessary. ^,

The component influenza virus strains in LAIV are updated each year as recommended by the WHO and national health authorities. Monovalent bulk vaccine is produced for each reassortant 6:2 vaccine strain in specific pathogen-free hen’s eggs. The allantoic fluids are clarified by centrifugation and stabilized by the addition of stabilizing buffer: each 0.2-mL dose contains 0.188 mg/dose monosodium glutamate, 2.00 mg/dose hydrolyzed porcine gelatin, 2.42 mg/dose arginine, 13.68 mg/dose sucrose, 2.26 mg/dose dibasic potassium phosphate, and 0.96 mg/dose monobasic potassium phosphate. Each dose contains residual amounts of ovalbumin (<0.24 µg/dose) and may also contain residual amounts of gentamicin sulfate (<0.015 µg/mL) and ethylenediaminetetraacetic acid (EDTA) (<0.37 µg/dose). It is free of preservatives and is produced in specific pathogen-free hen’s eggs. Monovalent bulk preparations of each of the four strains in the formulation are combined along with the stabilizing buffer to achieve a titer of 106.5–107.5 FFU/dose and filled into the sprayer device. FluMist Quadrivalent is a refrigerator-stable (i.e., 2–8°C) formulation of Q/LAIV. The shelf life of Q/LAIV is 18 weeks.

Transmission of Live Attenuated Influenza Vaccine

LAIV strains replicate in the nasopharynx of the recipient and are shed in the respiratory secretions. Data in the published literature failed to detect transmission of vaccine viruses derived from the passaged ca master donor viruses in multiple age groups and settings. ^{, , ,} In an evaluation of direct transmissibility in a daycare setting in children 8–36 months of age, 80% of the 98 FluMist recipients shed one or more vaccine strains, with a mean shedding duration of 7.6 days. One possible transmission event was observed among the 99 placebo recipients and was confirmed by phenotype and genotype analysis. The safety profile for this child after transmission was similar to that for the other children in the study who received vaccine or placebo; the child was not ill and did not experience an SAE. An open-label, single-arm, multicenter Phase II study reported by Mallory and colleagues demonstrated that at least one component of seasonal LAIV was shed by 79% of children between 6 and 59 months of age. In this study, virus shedding was most frequently seen in children 6–23 months of age, and it occurred mostly on postvaccination days 1–11, with shedding beyond day 11 detected almost exclusively in the younger age group. Peak titers of vaccine virus were detected on postvaccination day 2, and mean peak titers were generally less than 10 ³ TCID ₅₀ (median tissue culture infective dose)/mL.

In the year after initial FDA approval, the clinical use of LAIV was restricted at some institutions because of a theoretical concern that live attenuated virus might transmit from vaccinated hospital employees to unvaccinated patient contacts. After several years of experience, this concern proved to be unfounded, and person-to-person transmission of LAIV has not been a problem. Izurieta and colleagues summarized vaccine AE reporting regarding LAIV, and no instances of recognized transmission were identified. In the one case in which transmission was suspected, and in which viral isolates were obtained, it turned out that a wt influenza virus, not vaccine virus, was involved in the transmission. The Advisory Committee on Immunization Practices (ACIP) to the CDC modified its initial recommendation to reflect the practicalities of vaccine usage in hospitals. The only concern that remains now is a theoretical one. Employees who are vaccinated should not have contact within 7 days with immunosuppressed patients who are in reverse isolation (such as bone marrow transplant recipients). All other situations are considered acceptable for using LAIV.