Immune response to inactivated influenza virus vaccine: antibody reactivity with epidemic influenza B viruses of two highly distinct evolutionary lineages R. Pyh~il~i*§, M . K l e e m o l a * , V. K u m p u l a i n e n t, E. V a r t i a i n e n * , S. L a p p i ~, A. P6nk/i~ a n d K. C a n t e l l *
Vaccination of adults (healthy female employees potentially capable of transmitting influenza to high-risk persons; n = 104) in autumn 1990 with a trivalent influenza virus vaccine containing B~ Yamaoata/16/88 induced a low antibody response to B/Finland/ 150/90, a recent variant of B~ Victoria/2/87-like viruses, as compared with the antibody response to B / Finland/172 / 91, a current variant in the lineage of B~ Yamagata /16 / 88-like viruses. Up to the end of the epidemic season, the antibody status declined but was still significantly better than before the vaccination. The results suggest that the vaccine strain was appropriate for the outbreak of 1990 to 1991 in Finland, but may provide unsatisfactory protection against B/Victoria/2/87-like viruses. Evidence is given that use of Madin-Darby canine kidney ( MDCK)-grown virus as an antigen in the haemaoglutination inhibition test (HI) may provide more reliable information about the protective antibodies than use of untreated or ether-treated ego-grown viruses. Significantly higher postvaccination and postepidemic antibody titres were recorded among subjects who exhibited the antibody before vaccination than among seronegative subjects. A significantly higher response rate among initially seronegative people than among seropositive people was recorded for antibody to B/Finland/150/90, but no clear evidence was obtained that the pre-existing antibody could have had a negative effect on the antibody production. Keywords:Influenza B; vaccine; immunogenicity; serological match; pre-existing antibody; MDCK-grown virus
INTRODUCTION The protection provided by influenza virus vaccines depends on the antigenic similarity between the vaccine strains and the epidemic viruses. Because of antigenic drift, the frequent appearance of new antigenic variants, WHO has recommended annual changes in the composition of inactivated influenza virus vaccines in recent years. Essential prerequisites for these recommendations are that epidemiologically important older variants have disappeared and that multiple lineages cocirculate for only short periods, thus providing an opportunity to limit the number of vaccine components. At present, trivalent influenza virus vaccines, which contain a single influenza B virus and two influenza A *National Public Health Institute, Helsinki, Finland. tHelsinki City Occupational Health Care Unit, Helsinki, Finland. tHelsinki City Environment Center, Helsinki, Finland. ~}To whom correspondence should be addressed at: National Public Health InsUtute, MannerheimlnUe 166, 00300 Helsinki, Finland. (Received 18 November 1982; revised 28 January 1992; accepted 4 February 1982) (]~84-410X/92./OeOe81-06 ~) 1 ~ Butterworth-HelnemannLM
virus components, are widely used. Some recent events in the evolution and epidemiology of influenza B viruses are a cause of fresh concern. Two highly distinct antigenic variants of influenza B viruses, Victoria/2/87-1ike (VI/87) and Yamagata/16/88like (YA/88) viruses, circulated in different parts of the world in 1988/891 , 1989/902 and 1990/913 . These variants were shown to belong to two evolutionary lineages, which share a common origin in the late 1970s4'5. In 1988, the two lineages differed by 27 amino acids out of a total of 347 in HA1, the domain of virus haemagglutinin, which is highly variable due to antigenic drift4. This distance is greater than any previously reported for cocirculating influenza B viruses or influenza A viruses of the same subtype. Since 1989, YA/88-1ike antigens have been included instead of VI/87-1ike antigens in the recommended composition of influenza vaccines t-3. Vaccination with B/Yamagata/16/88 was shown to induce antibodies against B/Victoria/2/87 when examined by haemagglutination inhibition 6 (HI) or single radial haemolysis 7 (SRH), but responses were substantially poorer than
Vaccine, Vol. 10, Issue 9, 1992 631
Influenza B virus vaccine: R. Pyhbl~ et al.
against the homologous virus. The possibility that the two influenza B virus lineages will still cocirculate and experience a further antigenic drift makes it reasonable to monitor the antibody response induced by the present HA/88-1ike vaccine to new epidemic viruses of both of the two branches. For the present study, the HI antibody responses of volunteers who received trivalent influenza vaccines (B/Yamagata/16/88 included in the composition) in autumn 1990 were analysed for new VI/87- and YA/88-1ike viruses isolated in Finland. Attention was paid to the ability of different kinds of viral antigens to detect HI antibody, to the influence of pre-existing antibody on the antibody response and to the persistence of the induced antibodies during the epidemic season following the vaccination.
Clinical infections
During the period between the collection of prevaccination and postepidemic samples, the subjects in the vaccination and control groups were asked to report any acute respiratory infections to the nursing and medical staff, and to schedule visits for the collection of acute and convalescent phase blood samples and specimens for the detection of viral antigens. For the latter purpose, nasopharyngeal secretions (NPS) were aspirated through the nostrils with a disposable mucus extractor. The serum pairs and/or NPS samples were obtained from 45 subjects as specified in Table 3. The sera were studied for HI and CF antibodies and the NPS samples for viral antigens as described below.
HI test
MATERIALS AND METHODS Vaccine schedule
The vaccine used was the Connaught inactivated trivalent split influenza vaccine (Fluzone) which contained B/Yamagata/16/88 (15/~g HA/0.5 ml) as well as two influenza A virus components. The vaccine was offered between 10 September and 28 November, 1990 to members of the staff of a Helsinki city department which provides assistance with housekeeping. The department serves elderly people and sometimes families with small children at home. Thus, the housekeepers belong to groups potentially capable of transmitting influenza to medically defined high-risk persons. The vaccine was offered to 351 subjects, 165 of whom acquired the vaccination and were cooperative enough to permit us to collect prevaccination and postvaccination blood samples. The prevaccination samples were taken at the same time as the vaccination was given. The postvaccination samples were taken 22 to 44 days after the vaccination, the great majority (93%) at 27 to 36 days. The vaccine group was followed over the epidemic season until spring 1991, when postepidemic blood samples were collected between 6 and 23 May. The final number of vaccinated subjects who were still employed in May, and who had been cooperative, was 104. With the exception of one male, they were females aged 18-60 years (median 42 years). Only three of the subjects had been vaccinated against influenza within the past 3 years. Complete antibody titrations by HI were performed with the samples of this final vaccination group. Initially, to evaluate the efficacy of the vaccination, the trial also contained a control group of 150 unvaccinated subjects. Employment, sex and age distribution (range 21-62 years, median 43 years) were as in the vaccination group. Pre-epidemic serum samples were collected at the same time as the prevaccination sera in the vaccination group. Postepidemic sera were collected similarly in spring 1991. As in the vaccination group, three of the subjects had been vaccinated against influenza within the past 3 years. These samples, too, were studied for HI antibodies. Only some of the data are worth presenting, for the low rate of influenza infections during the study period did not allow us to evaluate the efficacy of the vaccination.
632
Vaccine, Vol. 10, Issue 9, 1992
Standard microtitre techniques (0.5% hen erythrocytes, four HA units of virus) were used to detect serum HI antibodies at dilutions from 1 : 10 to 1 : 6408. Unspecific inhibitors were removed by treatment with Vibrio cholerae filtrate at a dilution of 1:5. The HI antibodies were determined for the following influenza B viruses cultivated exclusively in the allantoic cavity of embryonated eggs (E) or in Madin-Darby canine kidney (MDCK) cell cultures (M; trypsin included) and used in the HI tests as ether-treated (e) or without any treatment (no marking): (1) Yamagata/ 16/88 (E and Ee), the vaccine component; (2) Finland~172~91 (E, Ee and M), YA/88-1ike virus epidemic in Finland in 1990/91; (3) Finland/150/90 (E, Ee and M), VI/87-1ike virus epidemic in Finland in 1989/90. The treatment with ether was performed as described previously9. On the basis of the nucleotide sequence that codes for the HAl-domain of the virus haemagglutinin, Finland/150/90 was described as representing a new step in the evolution of VI/87-1ike viruses 1°. A similar analysis, not yet completed, has recorded amino acid differences between Yamagata/16/88 and Finland/172/91, e.g. at position 150 in the region analogous to antigenic site B (Ikonen, Kinnunen, Pryry and Pyhiil/i, unpublished data). Antigenic analysis of influenza viruses was carried out by HI using four HA units of virus. Antisera were produced against each of the viruses by intraperitoneal (i.p.) injection of male rats (Wistar, outbred). Booster injections were given 3 weeks later and the animals were bled 2 weeks from that. Results of the antigenic analysis are given in a dendrogram compiled following the formula11 :
where i is any antiserum, n is number of antisera, j and k are two virus strains compared and x u and xik is loglo of the titres of j and k, respectively. All possible comparisons between the virus strains were performed, and a matrix was constructed from the distance coefficients (d2). Then the two viruses with the highest resemblance were treated as a taxonomic unit and their coefficients were replaced by the arithmetic average. New matrices were formed with a reduced number of taxonomic units until all viruses were linked together.
Influenza B virus vaccine: R. Pyh~l~ et al.
ally when egg-grown virus passages are compared. Similarly, Finland/148/90 and Finland/145/90, which, like the Finland/150/90 used for HI antibody screening, represent the influenza B viruses epidemic in Finland in 1989/90, are closely related to the reference strains Victoria/2/87 and Beijing/1/87. Viruses isolated and passaged exclusively in MDCK cell cultures, Finland/ 145/90M and Finland/172/91M, are separate from the clusters of the corresponding egg-grown counterparts with distance coefficients of 0.09 and 0.14, respectively. The distance between the egg-grown and the MDCKgrown viruses is mainly due to the higher HI titres of the MDCK-grown viruses with antibodies induced not only by the homologous virus, but also by the egg-grown counterpart. On the other hand, antisera against VI/87-1ike and YA/88-1ike viruses grown in MDCK cell cultures did not cross-react with viruses of the other lineage (Table 1 ).
Other laboratory tests CF test. Complement fixing (CF) antibodies against adenoviruses, influenza A and B, parainfluenza types 1, 2 and 3, respiratory syncytial virus (RSV) and the chlamydia group were determined using a standard CF microtechnique method 12 with a fourfold or greater titre rise between the paired sera as a criterion of acute infection.
Antigen detection. Sonicated NPS samples diluted 1 : 5 in assay buffer were tested for the presence of antigens of adenoviruses (hexon antigen), influenza A (nucleoprotein), influenza B (haemagglutinin), parainfluenza viruses 1, 2 and 3 and RSV by one-incubation time-resolved fluoroimmunoassay using monoclonal antibodies x3,14. Statistical analysis
The protection rates (percentage of volunteers with antibody titre >/40) and response rates (percentage of volunteers with a fourfold or greater rise in HI antibody titre between prevaccination and postvaccination sera) were compared using the Z2 test and Fisher's exact test. Student's t test was used to compare geometric mean titres (GMT). For calculation of GMT, a titre of less than 10 was arbitrarily set at 5.
Antibody response, reactivity and persistence
Protection rates (percentage of volunteers with antibody titre ~>40), response rates (percentage of volunteers with a fourfold or greater rise in HI antibody titre between prevaccination and postvaccination sera) and GMT of prevaccination, postvaccination and postepidemic sera are given in Table 2. Most rates and titres were significantly higher for the ether-treated egg-grown viruses, and even for the untreated MDCKgrown viruses, than the titres for untreated egg-grown viruses. There is no evidence of circulation of YA/88-1ike viruses in Finland before the moment of vaccination. The outbreaks of 1989/90 and 1987/88 were due to VI/87-1ike viruses 1°. Hence, the prevaccination HI antibody status was better for the new VI/87-1ike MDCK-grown virus, Finland/150/90M, than for the corresponding YA/88like virus, Finland/172/91M. The GMTs, 15 and 10,
RESULTS Antigenic analysis Figure 1 illustrates the antigenic relationships between representatives of epidemic variants from the near past and the viruses used as antigens in the HI tests. Apart from amino acid differences (Materials and methods), Finland/172/91, the epidemic virus of 1990/91, is antigenically closely related to Yamagata/16/88, especiB/Sing/222/79
E
B/USSR/IO0/83
E
B/~rb/1/86
E
B/BeiJ/1/87
E
B/Fin/148/90 B/Vict/2/87
I |
ME Z
B/Fin/145/90
ME
B/F~n/145/90
M
B/Yam/16/88
|
E
B/Fin/172/91
E
B/F~n/172/91
M I 0.00
I
I 0.25
I
I 0.50
I
I 0.75
I
I 1.00
Figure 1 Antigenic relationship between influenza B viruses. In HI tests, antisera produced in rats against each of the 11 virus strains were used. Table 1 serves as a key for the viruses. The distance coefficient (o~; for the formula, see Materials and methods) is indicated
Vaccine, Vol. 10, Issue 9, 1992 633
Influenza B virus vaccine: R. Pyhdld et al.
respectively, differed significantly (p40 Number
Pt-V
Geometric mean titre (GMT)
Pt-E
Pt-V
Pt-E
% with significant rise ~
Virus strain"
S-
S+
S-
S+
S-
S+
S-
S+
S-
S+
S-
S+
YA/88-1ike viruses: Yam/16/88 Ee
53
51
71
100
51
94
54
168
23
100
92
92
56
92
52 ~
160
37 ~
.
98 ,
87
87
Fin/172/91
Ee
41
63
66 •
Fin/172/91
M
38
66
53
82
29
65
36
66
20
40
76
71
78
26
35
92
15
77
17
104
10
61
53
62
18
86
50
66
11
52
26
40
12
29
78
VI/87-1ike viruses: Fin/150/90 Ee Fin/150/90
M
~
97
~
31 ~/t/t/
"hAs in Table 2. = S - = Pr-V HI antibody titre of l e ~ than 10, S + = Pr-V HI antibody tltre of ;~10; *p < 0.06, **p < 0.01, * * * p < 0.001
634
Vaccine, Vol. 10, Issue 9, lg92
Influenza B virus vaccine: R. Pyhbl~ et al. Table 4 Clinical samples collected during acute infections from the 104 vaccinated and 150 unvaccinated subjects and results of laboratory tests Vaccination group
Control group
Clinical samples"
Number of patients
Aetiology~ clarified
Number of patients
Aetiology ° clarified
PS + NPS PS NPS
16 3 1
0 0 0
23 0 2
3c (13%) 0 0
Total
20/104(19%)
25/150(17%)
• P S = paired sere taken during acute and convalescent phase, NPs = nesopharyngeal secretion. bPS were studied for CF and HI antibodies and NPS for viral antigens as described in Materials and methods. =Parainfluenza 1, parainfluenza 3 and RSV (one of each)
Table 5 HI antibody titres to the epidemic influenza B viruses in the postvaccination samples of the three vaccinated subjects and in the corresponding pre-epidemic samples of the six unvaccinated controls who had a serologic.ally verified influenza B virus infection HI antibody titre to
Case number
Fin/172/90 Ee"
Fin/172/90 M"
Vaccine group 1 2 3
20 10 10
< 10 10 10
Control group 1 2 3 4 5 6
10 10 40 10 10
Vaccination of adults (healthy female employees potentially capable of transmitting influenza to high-risk persons; n = 104) in autumn 1990 with a trivalent influenza virus vaccine containing B~ Yamaoata/16/88 induced a low antibody response to B/Finland/ 150/90, a recent variant of B~ Victoria/2/87-like viruses, as compared with the antibody response to B / Finland/172 / 91, a current variant in the lineage of B~ Yamagata /16 / 88-like viruses. Up to the end of the epidemic season, the antibody status declined but was still significantly better than before the vaccination. The results suggest that the vaccine strain was appropriate for the outbreak of 1990 to 1991 in Finland, but may provide unsatisfactory protection against B/Victoria/2/87-like viruses. Evidence is given that use of Madin-Darby canine kidney ( MDCK)-grown virus as an antigen in the haemaoglutination inhibition test (HI) may provide more reliable information about the protective antibodies than use of untreated or ether-treated ego-grown viruses. Significantly higher postvaccination and postepidemic antibody titres were recorded among subjects who exhibited the antibody before vaccination than among seronegative subjects. A significantly higher response rate among initially seronegative people than among seropositive people was recorded for antibody to B/Finland/150/90, but no clear evidence was obtained that the pre-existing antibody could have had a negative effect on the antibody production. Keywords:Influenza B; vaccine; immunogenicity; serological match; pre-existing antibody; MDCK-grown virus
INTRODUCTION The protection provided by influenza virus vaccines depends on the antigenic similarity between the vaccine strains and the epidemic viruses. Because of antigenic drift, the frequent appearance of new antigenic variants, WHO has recommended annual changes in the composition of inactivated influenza virus vaccines in recent years. Essential prerequisites for these recommendations are that epidemiologically important older variants have disappeared and that multiple lineages cocirculate for only short periods, thus providing an opportunity to limit the number of vaccine components. At present, trivalent influenza virus vaccines, which contain a single influenza B virus and two influenza A *National Public Health Institute, Helsinki, Finland. tHelsinki City Occupational Health Care Unit, Helsinki, Finland. tHelsinki City Environment Center, Helsinki, Finland. ~}To whom correspondence should be addressed at: National Public Health InsUtute, MannerheimlnUe 166, 00300 Helsinki, Finland. (Received 18 November 1982; revised 28 January 1992; accepted 4 February 1982) (]~84-410X/92./OeOe81-06 ~) 1 ~ Butterworth-HelnemannLM
virus components, are widely used. Some recent events in the evolution and epidemiology of influenza B viruses are a cause of fresh concern. Two highly distinct antigenic variants of influenza B viruses, Victoria/2/87-1ike (VI/87) and Yamagata/16/88like (YA/88) viruses, circulated in different parts of the world in 1988/891 , 1989/902 and 1990/913 . These variants were shown to belong to two evolutionary lineages, which share a common origin in the late 1970s4'5. In 1988, the two lineages differed by 27 amino acids out of a total of 347 in HA1, the domain of virus haemagglutinin, which is highly variable due to antigenic drift4. This distance is greater than any previously reported for cocirculating influenza B viruses or influenza A viruses of the same subtype. Since 1989, YA/88-1ike antigens have been included instead of VI/87-1ike antigens in the recommended composition of influenza vaccines t-3. Vaccination with B/Yamagata/16/88 was shown to induce antibodies against B/Victoria/2/87 when examined by haemagglutination inhibition 6 (HI) or single radial haemolysis 7 (SRH), but responses were substantially poorer than
Vaccine, Vol. 10, Issue 9, 1992 631
Influenza B virus vaccine: R. Pyhbl~ et al.
against the homologous virus. The possibility that the two influenza B virus lineages will still cocirculate and experience a further antigenic drift makes it reasonable to monitor the antibody response induced by the present HA/88-1ike vaccine to new epidemic viruses of both of the two branches. For the present study, the HI antibody responses of volunteers who received trivalent influenza vaccines (B/Yamagata/16/88 included in the composition) in autumn 1990 were analysed for new VI/87- and YA/88-1ike viruses isolated in Finland. Attention was paid to the ability of different kinds of viral antigens to detect HI antibody, to the influence of pre-existing antibody on the antibody response and to the persistence of the induced antibodies during the epidemic season following the vaccination.
Clinical infections
During the period between the collection of prevaccination and postepidemic samples, the subjects in the vaccination and control groups were asked to report any acute respiratory infections to the nursing and medical staff, and to schedule visits for the collection of acute and convalescent phase blood samples and specimens for the detection of viral antigens. For the latter purpose, nasopharyngeal secretions (NPS) were aspirated through the nostrils with a disposable mucus extractor. The serum pairs and/or NPS samples were obtained from 45 subjects as specified in Table 3. The sera were studied for HI and CF antibodies and the NPS samples for viral antigens as described below.
HI test
MATERIALS AND METHODS Vaccine schedule
The vaccine used was the Connaught inactivated trivalent split influenza vaccine (Fluzone) which contained B/Yamagata/16/88 (15/~g HA/0.5 ml) as well as two influenza A virus components. The vaccine was offered between 10 September and 28 November, 1990 to members of the staff of a Helsinki city department which provides assistance with housekeeping. The department serves elderly people and sometimes families with small children at home. Thus, the housekeepers belong to groups potentially capable of transmitting influenza to medically defined high-risk persons. The vaccine was offered to 351 subjects, 165 of whom acquired the vaccination and were cooperative enough to permit us to collect prevaccination and postvaccination blood samples. The prevaccination samples were taken at the same time as the vaccination was given. The postvaccination samples were taken 22 to 44 days after the vaccination, the great majority (93%) at 27 to 36 days. The vaccine group was followed over the epidemic season until spring 1991, when postepidemic blood samples were collected between 6 and 23 May. The final number of vaccinated subjects who were still employed in May, and who had been cooperative, was 104. With the exception of one male, they were females aged 18-60 years (median 42 years). Only three of the subjects had been vaccinated against influenza within the past 3 years. Complete antibody titrations by HI were performed with the samples of this final vaccination group. Initially, to evaluate the efficacy of the vaccination, the trial also contained a control group of 150 unvaccinated subjects. Employment, sex and age distribution (range 21-62 years, median 43 years) were as in the vaccination group. Pre-epidemic serum samples were collected at the same time as the prevaccination sera in the vaccination group. Postepidemic sera were collected similarly in spring 1991. As in the vaccination group, three of the subjects had been vaccinated against influenza within the past 3 years. These samples, too, were studied for HI antibodies. Only some of the data are worth presenting, for the low rate of influenza infections during the study period did not allow us to evaluate the efficacy of the vaccination.
632
Vaccine, Vol. 10, Issue 9, 1992
Standard microtitre techniques (0.5% hen erythrocytes, four HA units of virus) were used to detect serum HI antibodies at dilutions from 1 : 10 to 1 : 6408. Unspecific inhibitors were removed by treatment with Vibrio cholerae filtrate at a dilution of 1:5. The HI antibodies were determined for the following influenza B viruses cultivated exclusively in the allantoic cavity of embryonated eggs (E) or in Madin-Darby canine kidney (MDCK) cell cultures (M; trypsin included) and used in the HI tests as ether-treated (e) or without any treatment (no marking): (1) Yamagata/ 16/88 (E and Ee), the vaccine component; (2) Finland~172~91 (E, Ee and M), YA/88-1ike virus epidemic in Finland in 1990/91; (3) Finland/150/90 (E, Ee and M), VI/87-1ike virus epidemic in Finland in 1989/90. The treatment with ether was performed as described previously9. On the basis of the nucleotide sequence that codes for the HAl-domain of the virus haemagglutinin, Finland/150/90 was described as representing a new step in the evolution of VI/87-1ike viruses 1°. A similar analysis, not yet completed, has recorded amino acid differences between Yamagata/16/88 and Finland/172/91, e.g. at position 150 in the region analogous to antigenic site B (Ikonen, Kinnunen, Pryry and Pyhiil/i, unpublished data). Antigenic analysis of influenza viruses was carried out by HI using four HA units of virus. Antisera were produced against each of the viruses by intraperitoneal (i.p.) injection of male rats (Wistar, outbred). Booster injections were given 3 weeks later and the animals were bled 2 weeks from that. Results of the antigenic analysis are given in a dendrogram compiled following the formula11 :
where i is any antiserum, n is number of antisera, j and k are two virus strains compared and x u and xik is loglo of the titres of j and k, respectively. All possible comparisons between the virus strains were performed, and a matrix was constructed from the distance coefficients (d2). Then the two viruses with the highest resemblance were treated as a taxonomic unit and their coefficients were replaced by the arithmetic average. New matrices were formed with a reduced number of taxonomic units until all viruses were linked together.
Influenza B virus vaccine: R. Pyh~l~ et al.
ally when egg-grown virus passages are compared. Similarly, Finland/148/90 and Finland/145/90, which, like the Finland/150/90 used for HI antibody screening, represent the influenza B viruses epidemic in Finland in 1989/90, are closely related to the reference strains Victoria/2/87 and Beijing/1/87. Viruses isolated and passaged exclusively in MDCK cell cultures, Finland/ 145/90M and Finland/172/91M, are separate from the clusters of the corresponding egg-grown counterparts with distance coefficients of 0.09 and 0.14, respectively. The distance between the egg-grown and the MDCKgrown viruses is mainly due to the higher HI titres of the MDCK-grown viruses with antibodies induced not only by the homologous virus, but also by the egg-grown counterpart. On the other hand, antisera against VI/87-1ike and YA/88-1ike viruses grown in MDCK cell cultures did not cross-react with viruses of the other lineage (Table 1 ).
Other laboratory tests CF test. Complement fixing (CF) antibodies against adenoviruses, influenza A and B, parainfluenza types 1, 2 and 3, respiratory syncytial virus (RSV) and the chlamydia group were determined using a standard CF microtechnique method 12 with a fourfold or greater titre rise between the paired sera as a criterion of acute infection.
Antigen detection. Sonicated NPS samples diluted 1 : 5 in assay buffer were tested for the presence of antigens of adenoviruses (hexon antigen), influenza A (nucleoprotein), influenza B (haemagglutinin), parainfluenza viruses 1, 2 and 3 and RSV by one-incubation time-resolved fluoroimmunoassay using monoclonal antibodies x3,14. Statistical analysis
The protection rates (percentage of volunteers with antibody titre >/40) and response rates (percentage of volunteers with a fourfold or greater rise in HI antibody titre between prevaccination and postvaccination sera) were compared using the Z2 test and Fisher's exact test. Student's t test was used to compare geometric mean titres (GMT). For calculation of GMT, a titre of less than 10 was arbitrarily set at 5.
Antibody response, reactivity and persistence
Protection rates (percentage of volunteers with antibody titre ~>40), response rates (percentage of volunteers with a fourfold or greater rise in HI antibody titre between prevaccination and postvaccination sera) and GMT of prevaccination, postvaccination and postepidemic sera are given in Table 2. Most rates and titres were significantly higher for the ether-treated egg-grown viruses, and even for the untreated MDCKgrown viruses, than the titres for untreated egg-grown viruses. There is no evidence of circulation of YA/88-1ike viruses in Finland before the moment of vaccination. The outbreaks of 1989/90 and 1987/88 were due to VI/87-1ike viruses 1°. Hence, the prevaccination HI antibody status was better for the new VI/87-1ike MDCK-grown virus, Finland/150/90M, than for the corresponding YA/88like virus, Finland/172/91M. The GMTs, 15 and 10,
RESULTS Antigenic analysis Figure 1 illustrates the antigenic relationships between representatives of epidemic variants from the near past and the viruses used as antigens in the HI tests. Apart from amino acid differences (Materials and methods), Finland/172/91, the epidemic virus of 1990/91, is antigenically closely related to Yamagata/16/88, especiB/Sing/222/79
E
B/USSR/IO0/83
E
B/~rb/1/86
E
B/BeiJ/1/87
E
B/Fin/148/90 B/Vict/2/87
I |
ME Z
B/Fin/145/90
ME
B/F~n/145/90
M
B/Yam/16/88
|
E
B/Fin/172/91
E
B/F~n/172/91
M I 0.00
I
I 0.25
I
I 0.50
I
I 0.75
I
I 1.00
Figure 1 Antigenic relationship between influenza B viruses. In HI tests, antisera produced in rats against each of the 11 virus strains were used. Table 1 serves as a key for the viruses. The distance coefficient (o~; for the formula, see Materials and methods) is indicated
Vaccine, Vol. 10, Issue 9, 1992 633
Influenza B virus vaccine: R. Pyhdld et al.
respectively, differed significantly (p40 Number
Pt-V
Geometric mean titre (GMT)
Pt-E
Pt-V
Pt-E
% with significant rise ~
Virus strain"
S-
S+
S-
S+
S-
S+
S-
S+
S-
S+
S-
S+
YA/88-1ike viruses: Yam/16/88 Ee
53
51
71
100
51
94
54
168
23
100
92
92
56
92
52 ~
160
37 ~
.
98 ,
87
87
Fin/172/91
Ee
41
63
66 •
Fin/172/91
M
38
66
53
82
29
65
36
66
20
40
76
71
78
26
35
92
15
77
17
104
10
61
53
62
18
86
50
66
11
52
26
40
12
29
78
VI/87-1ike viruses: Fin/150/90 Ee Fin/150/90
M
~
97
~
31 ~/t/t/
"hAs in Table 2. = S - = Pr-V HI antibody titre of l e ~ than 10, S + = Pr-V HI antibody tltre of ;~10; *p < 0.06, **p < 0.01, * * * p < 0.001
634
Vaccine, Vol. 10, Issue 9, lg92
Influenza B virus vaccine: R. Pyhbl~ et al. Table 4 Clinical samples collected during acute infections from the 104 vaccinated and 150 unvaccinated subjects and results of laboratory tests Vaccination group
Control group
Clinical samples"
Number of patients
Aetiology~ clarified
Number of patients
Aetiology ° clarified
PS + NPS PS NPS
16 3 1
0 0 0
23 0 2
3c (13%) 0 0
Total
20/104(19%)
25/150(17%)
• P S = paired sere taken during acute and convalescent phase, NPs = nesopharyngeal secretion. bPS were studied for CF and HI antibodies and NPS for viral antigens as described in Materials and methods. =Parainfluenza 1, parainfluenza 3 and RSV (one of each)
Table 5 HI antibody titres to the epidemic influenza B viruses in the postvaccination samples of the three vaccinated subjects and in the corresponding pre-epidemic samples of the six unvaccinated controls who had a serologic.ally verified influenza B virus infection HI antibody titre to
Case number
Fin/172/90 Ee"
Fin/172/90 M"
Vaccine group 1 2 3
20 10 10
< 10 10 10
Control group 1 2 3 4 5 6
10 10 40 10 10
