THE JOURNAL OF INFECTIOUS DISEASE. VOL. 136, SUPPLEMENT. DECEMBER 1977 © 1977 by the University of Chicago. All rights reserved.
Neuraminidase Content of Influenza Vaccines and Neuraminidase Antibody Responses after Vaccination of Immunologically Primed and Unprimed Populations From the World Health Organization Collaborating Center for Influenza, Respiratory Virology Branch, Virology Division, Bureau of Laboratories, Center for Disease Control, Atlanta, Georgia
Alan P. Kendal, Gary R. Noble, and Walter R. Dowdle
on influenza virus replication was probably a consequence of viral aggregation and binding of virus to infected cells, rather than of inhibition of enzymatic activity [5]. This hypothesis subsequently received experimental support [6, 7]. Results of studies with animals [4, 8, 9] are consistent in indicating that preexisting N AAb can ameliorate the clinical course of influenza. This effect has also been observed for humans when the effect of natural infection was studied [10, 11] or when volunteers were artificially challenged with either wild-type or attenuated influenza viruses [12-14]. In several of these studies, only low levels of virus shedding were observed for those volunteers with reduced illness [12-14], a finding suggesting that although infection occurred, dissemination of virus was reduced by NAAb; this finding is consistent with the results of studies in eggs, tissue culture, and animals [1-9]. A retrospective analysis of naturally occurring Hong Kong influenza in Tecumseh, Mich., showed that the attack rate of influenza (determined by seroconversion to antibody to H3 he-
Studies with cells or animals as hosts have demonstrated that antibodies to neuraminidase (NAAb) probably do not cause true neutralization of influenza virus but limit viral replication by reducing the spread of virus from infected cells [1-4]. After the observation that NAAb caused flocculation of influenza viruses, the suggestion was made that the effect of NAAb
We thank Dr. Robert P. Cunningham and The Southern Bell Telephone Company employees, Atlanta, Georgia, for participation in a vaccine trial; Drs. R. Brackett and F. Brandon (Park, Davis and Company, Indianapolis, Indiana) for making available to us sera and protocols from their vaccine trials and for providing samples of investigational vaccines; Ms. July Galphin for technical assistance; and Dr. David Alling (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland) for statistical analysis. Use of trade names is for identification only and does not constitute endorsement by the U.S. Public Health Service or the U.S. Department of Health, Education, and Welfare. Please address requests for reprints to Dr. Alan P. Kendal, Respiratory Virology Branch, 7-112, Center fur Disease Control, Atlanta, Georgia 30333.
S4l5
Downloaded from http://jid.oxfordjournals.org/ at Cambridge University on July 30, 2015
Vaccines prepared with influenza Ajswinejl976j37-like virus contained neuraminidase activity comparable to that of H3N2 vaccines, whereas little neuraminidase activity could be detected in influenza A/New Jersey /76 vaccines. In single-dose vaccine studies, Ajswinejl976j37.like split-virus vaccine induced antibody to neuraminidase (NAAb) in about 20% of children younger than 18 years, and this NAAb response was better than or equal to the antibody response to the vaccine's hemagglutinin (HAAb). In immunologically primed adults, the NAAb response to a single dose of HswlNl or H3N2 vaccine was 38%-54%, but this value was about 1.7-fold lower than that for HAAb response. Children six to 17 years old given two doses of Ajswinejl976j37-like vaccine had an overall 70% NAAb response, whereas children of similar age had an overall 30%-48% NAAb response to two doses of AjVictoria/3/75 (H3N2) virus. The findings support a hypothesis that the NAAb response to influenza vaccines can be suppressed if subjects receiving the vaccine are immunologically primed to its hemagglutinin component. NAAb responses in the absence of HAAb responses occur in only about 5% of vaccinated persons who are immunologically primed to both the hemagglutinin and neuraminidase of the vaccine.
Kendal, Noble, and Dowdle
8416
Materials and Methods
Vaccines. Details of year of manufacture, manufacturer, type (whole-virus or split-virus), and strain composition of vaccines are given in table 1. Vaccines no. 1, 2, 3, 5, 6, and 7 were taken from lots being used in the immunization program. Vaccines no. 4 and 8 were those used in the trials conducted during the summer of 1976 under the auspices of the U.S. Public Health Service and the Department of Defense. Vaccine no. 9 was prepared from the same virus bulk concentrate used to make vaccine no. 8. The effective dilutions of vaccines no. 8 and 9 from the original concentrate were similar (F. Brandon, personal communication). Vaccines no. 10, II, and 13 were used in vaccine trials conducted at Southern Bell Telephone and Telegraph Co., Atlanta, Ga., during the fall of 1975. Vaccine no. 17 was used in a trial conducted in 1968-1969 [17]. Other vaccines were from stocks kept for research purposes in this laboratory. All vaccines were stored at 4 C continuously. 1975-1976 vaccine trials. All trials with HswIN I or A/Victoria/3/75 vaccines conducted under the auspices of the U.S. Public Health Service and Department of Defense followed the relevant protocols [18]. Trials conducted with HswlNI vaccine by Parke, Davis and Company (PD) were done at a private pediatric practice, St. Petersburg, Fla., and at Michigan Reformatory, Ionia, Mich. In these trials, vaccines were all administered im in O.5-ml volumes with either a syringe or jet injector. Blood samples were taken three or four weeks after the first vaccination and two weeks after the second vaccination. In trials at the Southern Bell Telephone and Telegraph Co., vaccines were given im by syringe in 0.5-ml volumes. Serum was collected three weeks later. Sera were collected before vaccination in all trials except those in which the subjects were younger than six years old and those in which children six to 17 years old received a single dose of HswlNI vaccine. Serologic methods. All HAl tests were done in the Respiratory Virology Branch, Virology Division, Bureau of Laboratories, Center for Disease Control (Atlanta, Ga.). HAl results for the
Downloaded from http://jid.oxfordjournals.org/ at Cambridge University on July 30, 2015
magglutinin) was reduced from about 39% in subjects lacking detectable NAAb to about 21% in those possessing the highest levels of NAAb [15]. Couch et a1. have similarly shown in a prospective study of immunized and challenged volunteers that lack of infection correlated with presence of high titers of NAAb [13]. The above studies nearly all related t9 natural or artificially contrived circumstances where any possibility for effect of antibodies to hemagglutinin (HAAb) on viral infection was either minimized or totally excluded. Such circumstances occur rarely in nature for humans; the advent of H3N2 Hong Kong influenza in 1968, after an era of infection with H2N2 Asian influenza, was the only such instance clearly documented. Thus, the normal protective effect of NAAb may be far less than the results of the above-described studies imply, since in most years HAAb induced by immunization and/or prior infection would be expected to provide the major source of protection. The organization of large-scale vaccine trials as part of the National Influenza Immunization Prog~am provided an opportunity to determine the NAAb response relative to the HAAb respons~ in various population groups, including persons born after viruses antigenically related to HswlNI viruses had ceased to infect humans. This was therefore the first instance in which NAAb and HAAb responses could be compared in subjects who were either primed or unprimed to both the hemagglutinin and neuraminidase antigens in a vaccine. Results obtained, therefore, enable us to determine the relative effectiveness of influenza vaccine hemagglutinin and neuraminidase antigens in stimulating an antibody response at a time of complete antigenic shift in influenza virus and to compare results with those obtained either when an antigenic shift occurred only in the hemagglutinin or when no antigenic shift occurred in either the hemagglutinin or the neuraminidase. To ensure that vaccines used in the present trials contained neuraminidase antigen activities comparable to each other and to such activities in vaccines used in prior studies, we also measured these activities.
Neuraminidase Antibody Responses
5417
Table 1. Characteristics of the influenza virus vaccines tested.
Laboratory no.
Year*
Hemagglutinin (eCA units/ 0.5 ml)+
Neuraminidase activity (X 10 3 )11 Protein (lJg/0.5 ml)§
Units/ Units/0.5100 ml dose CCA units
Units/ p.g of protein
Manufacturert
Virus strains
1976-1977
PD-S
A/New jersey/76 (HswlNl)
200
22.5
Neuraminidase Content of Influenza Vaccines and Neuraminidase Antibody Responses after Vaccination of Immunologically Primed and Unprimed Populations From the World Health Organization Collaborating Center for Influenza, Respiratory Virology Branch, Virology Division, Bureau of Laboratories, Center for Disease Control, Atlanta, Georgia
Alan P. Kendal, Gary R. Noble, and Walter R. Dowdle
on influenza virus replication was probably a consequence of viral aggregation and binding of virus to infected cells, rather than of inhibition of enzymatic activity [5]. This hypothesis subsequently received experimental support [6, 7]. Results of studies with animals [4, 8, 9] are consistent in indicating that preexisting N AAb can ameliorate the clinical course of influenza. This effect has also been observed for humans when the effect of natural infection was studied [10, 11] or when volunteers were artificially challenged with either wild-type or attenuated influenza viruses [12-14]. In several of these studies, only low levels of virus shedding were observed for those volunteers with reduced illness [12-14], a finding suggesting that although infection occurred, dissemination of virus was reduced by NAAb; this finding is consistent with the results of studies in eggs, tissue culture, and animals [1-9]. A retrospective analysis of naturally occurring Hong Kong influenza in Tecumseh, Mich., showed that the attack rate of influenza (determined by seroconversion to antibody to H3 he-
Studies with cells or animals as hosts have demonstrated that antibodies to neuraminidase (NAAb) probably do not cause true neutralization of influenza virus but limit viral replication by reducing the spread of virus from infected cells [1-4]. After the observation that NAAb caused flocculation of influenza viruses, the suggestion was made that the effect of NAAb
We thank Dr. Robert P. Cunningham and The Southern Bell Telephone Company employees, Atlanta, Georgia, for participation in a vaccine trial; Drs. R. Brackett and F. Brandon (Park, Davis and Company, Indianapolis, Indiana) for making available to us sera and protocols from their vaccine trials and for providing samples of investigational vaccines; Ms. July Galphin for technical assistance; and Dr. David Alling (National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland) for statistical analysis. Use of trade names is for identification only and does not constitute endorsement by the U.S. Public Health Service or the U.S. Department of Health, Education, and Welfare. Please address requests for reprints to Dr. Alan P. Kendal, Respiratory Virology Branch, 7-112, Center fur Disease Control, Atlanta, Georgia 30333.
S4l5
Downloaded from http://jid.oxfordjournals.org/ at Cambridge University on July 30, 2015
Vaccines prepared with influenza Ajswinejl976j37-like virus contained neuraminidase activity comparable to that of H3N2 vaccines, whereas little neuraminidase activity could be detected in influenza A/New Jersey /76 vaccines. In single-dose vaccine studies, Ajswinejl976j37.like split-virus vaccine induced antibody to neuraminidase (NAAb) in about 20% of children younger than 18 years, and this NAAb response was better than or equal to the antibody response to the vaccine's hemagglutinin (HAAb). In immunologically primed adults, the NAAb response to a single dose of HswlNl or H3N2 vaccine was 38%-54%, but this value was about 1.7-fold lower than that for HAAb response. Children six to 17 years old given two doses of Ajswinejl976j37-like vaccine had an overall 70% NAAb response, whereas children of similar age had an overall 30%-48% NAAb response to two doses of AjVictoria/3/75 (H3N2) virus. The findings support a hypothesis that the NAAb response to influenza vaccines can be suppressed if subjects receiving the vaccine are immunologically primed to its hemagglutinin component. NAAb responses in the absence of HAAb responses occur in only about 5% of vaccinated persons who are immunologically primed to both the hemagglutinin and neuraminidase of the vaccine.
Kendal, Noble, and Dowdle
8416
Materials and Methods
Vaccines. Details of year of manufacture, manufacturer, type (whole-virus or split-virus), and strain composition of vaccines are given in table 1. Vaccines no. 1, 2, 3, 5, 6, and 7 were taken from lots being used in the immunization program. Vaccines no. 4 and 8 were those used in the trials conducted during the summer of 1976 under the auspices of the U.S. Public Health Service and the Department of Defense. Vaccine no. 9 was prepared from the same virus bulk concentrate used to make vaccine no. 8. The effective dilutions of vaccines no. 8 and 9 from the original concentrate were similar (F. Brandon, personal communication). Vaccines no. 10, II, and 13 were used in vaccine trials conducted at Southern Bell Telephone and Telegraph Co., Atlanta, Ga., during the fall of 1975. Vaccine no. 17 was used in a trial conducted in 1968-1969 [17]. Other vaccines were from stocks kept for research purposes in this laboratory. All vaccines were stored at 4 C continuously. 1975-1976 vaccine trials. All trials with HswIN I or A/Victoria/3/75 vaccines conducted under the auspices of the U.S. Public Health Service and Department of Defense followed the relevant protocols [18]. Trials conducted with HswlNI vaccine by Parke, Davis and Company (PD) were done at a private pediatric practice, St. Petersburg, Fla., and at Michigan Reformatory, Ionia, Mich. In these trials, vaccines were all administered im in O.5-ml volumes with either a syringe or jet injector. Blood samples were taken three or four weeks after the first vaccination and two weeks after the second vaccination. In trials at the Southern Bell Telephone and Telegraph Co., vaccines were given im by syringe in 0.5-ml volumes. Serum was collected three weeks later. Sera were collected before vaccination in all trials except those in which the subjects were younger than six years old and those in which children six to 17 years old received a single dose of HswlNI vaccine. Serologic methods. All HAl tests were done in the Respiratory Virology Branch, Virology Division, Bureau of Laboratories, Center for Disease Control (Atlanta, Ga.). HAl results for the
Downloaded from http://jid.oxfordjournals.org/ at Cambridge University on July 30, 2015
magglutinin) was reduced from about 39% in subjects lacking detectable NAAb to about 21% in those possessing the highest levels of NAAb [15]. Couch et a1. have similarly shown in a prospective study of immunized and challenged volunteers that lack of infection correlated with presence of high titers of NAAb [13]. The above studies nearly all related t9 natural or artificially contrived circumstances where any possibility for effect of antibodies to hemagglutinin (HAAb) on viral infection was either minimized or totally excluded. Such circumstances occur rarely in nature for humans; the advent of H3N2 Hong Kong influenza in 1968, after an era of infection with H2N2 Asian influenza, was the only such instance clearly documented. Thus, the normal protective effect of NAAb may be far less than the results of the above-described studies imply, since in most years HAAb induced by immunization and/or prior infection would be expected to provide the major source of protection. The organization of large-scale vaccine trials as part of the National Influenza Immunization Prog~am provided an opportunity to determine the NAAb response relative to the HAAb respons~ in various population groups, including persons born after viruses antigenically related to HswlNI viruses had ceased to infect humans. This was therefore the first instance in which NAAb and HAAb responses could be compared in subjects who were either primed or unprimed to both the hemagglutinin and neuraminidase antigens in a vaccine. Results obtained, therefore, enable us to determine the relative effectiveness of influenza vaccine hemagglutinin and neuraminidase antigens in stimulating an antibody response at a time of complete antigenic shift in influenza virus and to compare results with those obtained either when an antigenic shift occurred only in the hemagglutinin or when no antigenic shift occurred in either the hemagglutinin or the neuraminidase. To ensure that vaccines used in the present trials contained neuraminidase antigen activities comparable to each other and to such activities in vaccines used in prior studies, we also measured these activities.
Neuraminidase Antibody Responses
5417
Table 1. Characteristics of the influenza virus vaccines tested.
Laboratory no.
Year*
Hemagglutinin (eCA units/ 0.5 ml)+
Neuraminidase activity (X 10 3 )11 Protein (lJg/0.5 ml)§
Units/ Units/0.5100 ml dose CCA units
Units/ p.g of protein
Manufacturert
Virus strains
1976-1977
PD-S
A/New jersey/76 (HswlNl)
200
22.5
