Acta Pzdiatr Scand 65: 171-176, 1976
MEASLES VACCINATION VIII. The Occurrence of Antibodies against Virus Envelope Components after Imnzunization vtlith Inactil-luted Vaccine. Effects of Rel-vaccination btith Live Measles Vaccine ERLING NORRBY and RUTGER LAGERCRANTZ From the Department of Virology, Karolinska lnsritutet und rhe Paediatric Clinic, Karolinska Sjukhuset, Stockholm, Snaeden
ABSTRACT. Norrby, E. and Lagercrantz, R. (Department of Virology, Karolinska Institutet and the Paediatric Clinic, Kardinska Sjukhuset, Stockholm, Sweden). Measles vaccination VIII. The occurrence of antibodies against virus envelope components after immunization with inactivated vaccine. Effects of revaccination with live measles vaccine. Acta Paediatr Scand 65: 171, 1976.-Children immunized with 4 doses of formalin-inactivated vaccine and/or purified hemagglutinin prepared from Tween 80-ether (TE) treated material were subjected to a follow-up 8-9 years after the last dose of vaccine. 11 out of 27 children had clinical andlor serological signs of infections with wild measles virus during the 8 to 9 years post-booster period. 10 out of the 11 children with infections had non-hemagglutinating-inhibiting (HI) hemolysis-inhibiting (HLI) antibodies demonstrable In their sera after removal of HI antibodies by absorption with TE antigen. In contrast 13 out of 16 vaccinees without detectable signs of infection lacked non-HI HLI antibodies. 10 out of these 13 children were vaccinated with further attenuated Live measles virus. There were no clinical reactions to vaccination. 4 vaccinees with low pre-vaccination HI antibody titers showed signifkant rises of antibody titers including non-m HLI antibodies. In the remaining children no take of the Live vaccine could be demonstrated. Thus HI antibodies of a certain minimal concentration can block the replication of vaccine virus even in the absence of non-HI HLI antibodies. However, since it will be diffkult to establish these conditions by use of available inactivated vaccines it is recommended that future vaccine products should include both major virus envelope surface components, the hemagglutinin and the hemolysin. KEY WORDS: Measles vaccine, inactivated, immunity
In two small scale field trials the immunizing capacity of two different types of inactivated measles virus vaccines were studied (8-14). Formalin inactivated crude whole virus and purified hemagglutinin from Tween 80 and ether (TE) treated material were used both for primary immunization with three rnontly doses and for boostering 17 to 23 months after the primary immunization. High titers of hemagglutinating-inhibiting (HI) and neutralizing antibodies were induced in many vaccinees and appeared to protect against clinically overt
-
Supported by grants from the Swedish Medical Research Council (Project no. 16X- 116)
infections upon exposure to “wild” measles virus. However, in vaccinees with readily demonstrable, but relatively lower titers of antibodies, infections were encountered. These ranged from subclinical infections to cases of regular and in some cases even atypical measles infections. The atypical reactions included cases of pneumonia interpreted to represent Arthus reactions occurring in the respiratory tract. Similar atypical reactions have been encountered in other studies of killed measles vaccines ( 2 . 3 ) . In recent studies it has been found that the measles virus envelope contains two major Acta Pzdiatr S c a d 65
172
E. Norrby and R. Lagercrantz
components available at the surface of virions. Antibodies against these components can be separately identified in the form of HI and non-HI hemolysing-inhibiting (HLI) antibodies (7). The latter antibodies can be demonstrated after removal of HI antibodies by absorption with TE treated material. Both formalin and T E treated materials have been found to induce HI but not non-HI HLI antibodies when used for hyperimmunization of rabbits (7) and for vaccination of humans. In a collaborative study (15) sera collected from children immunized with TE treated or live measles vaccine or a combination of both products were examined for the presence of HI and non-HI HLI antibodies. The further attenuated live vaccine caused a production of antibodies against both types of surface envelope components, but the inactivated vaccine only induced HI antibodies. Further it was found that when the live vaccine was administered after a preimmunization with inactivated vaccine no non-HI HLI antibody response was evoked. In a separate study of sera collected during a field trial arranged by the Medical Research Council in England (4) it was found that immunization with formalin inactivated vaccine followed by live measles vaccine also induced HI, but not non-HI HLI antibodies ( 5 ) . In the present study children included in the small scale field trials with inactivated measles vaccines initiated in 1961 and 1967, (8-14) were reexamined. The occurrence of serological and possibly clinically demonstrable infections since the last follow-up analysis in 1967 was analysed and sera collected were used for determination of both HI and non-HI HLI antibodies. Children with detectable HI but no non-HI HLI antibodies were revaccinated with live measles vaccine and their antibody response and clinical reactions were studied. MATERIALS AND METHODS Sfudy populution. The schedule for two m a l l scale field trials with inactivated measles vaccine were previously described in detail (cf. ref. 14). The present follow-up Acta Pzdiatr Scand 65
study was made five years after the last examination of vaccinees (14), i.e. 8-9 years after the last injection of vaccine. Mainly children who at the last time of sampling were found not to have responded either clinically or serologically to exposure to wild virus were included in the study. 1 1 children from vaccine trial I and 16 children from vaccine trial I1 were available for interviewing and collection of a blood sample. Venous bleedings were made and no anticoagulant was added. Sera were collected after incubation of the blood samples for some hours at room temperature and at +4" over night. A selected group comprising 10 children from both trials who did not have detectable non-HI HLI antibodies were vaccinated with further attenuated live measles vaccine (Lirurgen, Dow Chemical Co., Indianapolis, Indiana, USA). Before vaccination a venous blood sample was collected and additional samples were taken one and 4 weeks after vaccination. Serological analyses. The techniques for determination of hemagglutinin-inhibition (HI), hemolysis-inhibition (HLI) and nucleocapsid complement fixing (CF) antibodies were previously described in detail (6). HLI antibodies were determined both before and after removal of HI antibodies by absorption with TE treated concentrated extracellular measles virus material. Only after removal of HI antibodies can HLI antibodies reacting with envelope components separate from the hemagglutinin be determined (7, 15). Sera with an HI antibody titer of 160 or less were absorbed to a final dilution of 1 : 8, whereas sera with a higher HI antibody titer were absorbed to a dilution of I : 10.
RESULTS HI antibody titers in serir~nsumnples collected 3 to 4 and 8 to 9 years ufter a foirrtii dose of I-wecine. Indicutions of measles tiirus infections. Serum samples were collected 8 to 9 years after booster vaccination from 1 1 children in vaccine trial I and 16 children in vaccine trial 11. 3 to 4 years post-booster samples were available from 9 and 14 vaccinees in each group, respectively. The titers of HI antibodies in the different late post-booster samples are shown in Fig. 1. In most vaccinees there was a 4 to %fold reduction in antibody titers during the five-year interval between collection of samples. However, four vaccinees from trial I and three from trial I1 showed significantly increased titers of HI antibodies indicating infections with wild measles virus. Altogether 7 out of I 1 vaccinees in group I showed signs of infections (Table I ) . Besides the four vaccinees with increased titers of HI antibodies (one of which had clinical signs of
Inactivated measles vaccine. VZZZ
I280
-
,'
173
0
/ /
a
320
80
,/
am.
.
00
/ /
..
.
0
. O
/ /
/
O .
o
0
/
-
/
/
00
/ /
/
0
0
/ 20
-
420
-
/
/
0
/
/
I
I
Fig. f. Correlation between serum HI antibody titers 4 ( 3 ) years and 9 (8) years after a booster injection to children in field trials I (circles) and I1 (squares). A filled symbol denotes clinical and/or serological signs of infection with wild measles virus.
infection), there were two children who had had serologically verified infections (one case with clinical symptoms) just before the 3 to 4 years postbooster sampling (14) and still another child with clinical signs interpreted as measles in connection with a known exposure to wild virus inbetween the two times of sampling. One of the two cases of both clinically and serologically verified disease had a some-
what aberrant appearance. There was a high fever and an atypical exanthema beginning on the distal extremities. However, there was no detectable pneumonia. In field trial I1 4 out of 16 children were interpreted to have had infections with wild measles virus. Two of these had mild clinical symptoms, whereas the other two cases only showed a rise in antibody titers.
Table I. Occurrence of children with and M-ithout infections during 3 and 9 years after a booster injection of inactivated vaccine
Appearance of non-HI HLI antibodies in vaccinees with and M-iithout signs of infection uith Mild measles r*irus. The titer of non-HI HLI antibodies in serum samples was determined after removal of HI antibodies by absorption with TE treated measles antigen. Sera from 10 out of the 11 children with signs of infection contained non-HI HLI antibodies. The exceptional case of infection was identifiable both by clinical signs and by a rise in HI antibody titers from 40 to 160. However, no non-HI HLI antibodies were detectable a t a dilution of 1 : 8. The situation in children without signs of infection was completely different (Table I ). No non-HI HLI antibodies were detectable in 3 out of 4 children in trial I and in 10 out of 12 children in trial 11. Eight out of the 13 vaccinees devoid of non-HI HLI antibodies were known to have been exposed to wild
Presence of detectable non-HI HLI antibodies in childre; without signs of infections
Total number of children Children with signs of infection ( a ) clinical aad serological ( 6 ) only clinical ( c ) only serological Children without signs of infection ( a ) without demonstrable non-HI HLI antibodies ( 6 ) with demonstrable non-HI HLI antibodies
Vaccine trial I
Vaccine trial I1
I1
16
7 2 I 4 (2)
4 2 2 (2)
4
12
3(2)"
IO(6)
20)
' Figures within parenthesis denote number of children with known exposure to wild measles virus.
Acto Pzediatr Scond 65
174
E. Norrby and R . Lagercrantz
measles virus. The three vaccinees who had non-HI HLI antibodies demonstrable in their sera, all had been exposed to cases of measles but had shown neither serological nor clinical signs of infection. Nucleocapsid CF antibodies were not detected in a 1 : 5 dilution of sera from children without signs of infection, except in three cases. Two of these cases also had demonstrable non-HI HLI antibodies. In contrast nucleocapsid CF antibodies were present in sera from 6 out of the 7 children with signs of infection in vaccination group I and 2 out of 4 of the corresponding children in group 11. Effect of immunization withfiuther attenuated live measles \vaccine in children Mith HI but no non-HI HLJ antibodies. 10 out of 13 children without demonstrable non-HI HLI antibodies were revaccinated with further attenuated measles vaccine. In no case did this vaccination induce either local or general clinical reactions. The effect on antibody titers are shown in Table 2. Only titers in the four week postvaccination serum samples have been included in the table since no changes in titers were detectable in one week postvaccination samples. Serological reactions to vaccination were found to depend on the titer levels of prevaccination antibodies. In 4 vaccinees with a preimmunization antibody titer of 10 to 40, significant (4-fold or higher) rises in HI, nonHI HLI and nucleocapsid CF antibodies occurred in 3,3 and 2 children, respectively. One vaccinee (R. B.) showed a rise in HI antibody titer from 40 to 80 but no appearance of nonHI HLI and npcleocapsid CF antibodies. 4 additional vaccihees had prevaccination HI antibody titers of 160 to 320. There was no appearance of detectable non-HI HLI antibodies in any of these children after vaccination. Three of these vaccinees had detectable nucleocapsid CF antibody titers which did not change except in one case, who displayed a titer rise from 4 to 8 paralleled by a rise in HI antibodies from 160 to 320. The tenth child selected for vaccination with live vaccine Acta Pdediatr Scand 65
Table 2. Effect of revaccination with further attenuated live measles %saccineof children previously immunized ir’ith only inactivated measles vaccine Serum titers of
Vaccinee“
M. S .
Time of samplingb Pre post
NucleoNon-HI capsid HI anti- HLI anti- CF antibodies bodies bodies 10
MEASLES VACCINATION VIII. The Occurrence of Antibodies against Virus Envelope Components after Imnzunization vtlith Inactil-luted Vaccine. Effects of Rel-vaccination btith Live Measles Vaccine ERLING NORRBY and RUTGER LAGERCRANTZ From the Department of Virology, Karolinska lnsritutet und rhe Paediatric Clinic, Karolinska Sjukhuset, Stockholm, Snaeden
ABSTRACT. Norrby, E. and Lagercrantz, R. (Department of Virology, Karolinska Institutet and the Paediatric Clinic, Kardinska Sjukhuset, Stockholm, Sweden). Measles vaccination VIII. The occurrence of antibodies against virus envelope components after immunization with inactivated vaccine. Effects of revaccination with live measles vaccine. Acta Paediatr Scand 65: 171, 1976.-Children immunized with 4 doses of formalin-inactivated vaccine and/or purified hemagglutinin prepared from Tween 80-ether (TE) treated material were subjected to a follow-up 8-9 years after the last dose of vaccine. 11 out of 27 children had clinical andlor serological signs of infections with wild measles virus during the 8 to 9 years post-booster period. 10 out of the 11 children with infections had non-hemagglutinating-inhibiting (HI) hemolysis-inhibiting (HLI) antibodies demonstrable In their sera after removal of HI antibodies by absorption with TE antigen. In contrast 13 out of 16 vaccinees without detectable signs of infection lacked non-HI HLI antibodies. 10 out of these 13 children were vaccinated with further attenuated Live measles virus. There were no clinical reactions to vaccination. 4 vaccinees with low pre-vaccination HI antibody titers showed signifkant rises of antibody titers including non-m HLI antibodies. In the remaining children no take of the Live vaccine could be demonstrated. Thus HI antibodies of a certain minimal concentration can block the replication of vaccine virus even in the absence of non-HI HLI antibodies. However, since it will be diffkult to establish these conditions by use of available inactivated vaccines it is recommended that future vaccine products should include both major virus envelope surface components, the hemagglutinin and the hemolysin. KEY WORDS: Measles vaccine, inactivated, immunity
In two small scale field trials the immunizing capacity of two different types of inactivated measles virus vaccines were studied (8-14). Formalin inactivated crude whole virus and purified hemagglutinin from Tween 80 and ether (TE) treated material were used both for primary immunization with three rnontly doses and for boostering 17 to 23 months after the primary immunization. High titers of hemagglutinating-inhibiting (HI) and neutralizing antibodies were induced in many vaccinees and appeared to protect against clinically overt
-
Supported by grants from the Swedish Medical Research Council (Project no. 16X- 116)
infections upon exposure to “wild” measles virus. However, in vaccinees with readily demonstrable, but relatively lower titers of antibodies, infections were encountered. These ranged from subclinical infections to cases of regular and in some cases even atypical measles infections. The atypical reactions included cases of pneumonia interpreted to represent Arthus reactions occurring in the respiratory tract. Similar atypical reactions have been encountered in other studies of killed measles vaccines ( 2 . 3 ) . In recent studies it has been found that the measles virus envelope contains two major Acta Pzdiatr S c a d 65
172
E. Norrby and R. Lagercrantz
components available at the surface of virions. Antibodies against these components can be separately identified in the form of HI and non-HI hemolysing-inhibiting (HLI) antibodies (7). The latter antibodies can be demonstrated after removal of HI antibodies by absorption with TE treated material. Both formalin and T E treated materials have been found to induce HI but not non-HI HLI antibodies when used for hyperimmunization of rabbits (7) and for vaccination of humans. In a collaborative study (15) sera collected from children immunized with TE treated or live measles vaccine or a combination of both products were examined for the presence of HI and non-HI HLI antibodies. The further attenuated live vaccine caused a production of antibodies against both types of surface envelope components, but the inactivated vaccine only induced HI antibodies. Further it was found that when the live vaccine was administered after a preimmunization with inactivated vaccine no non-HI HLI antibody response was evoked. In a separate study of sera collected during a field trial arranged by the Medical Research Council in England (4) it was found that immunization with formalin inactivated vaccine followed by live measles vaccine also induced HI, but not non-HI HLI antibodies ( 5 ) . In the present study children included in the small scale field trials with inactivated measles vaccines initiated in 1961 and 1967, (8-14) were reexamined. The occurrence of serological and possibly clinically demonstrable infections since the last follow-up analysis in 1967 was analysed and sera collected were used for determination of both HI and non-HI HLI antibodies. Children with detectable HI but no non-HI HLI antibodies were revaccinated with live measles vaccine and their antibody response and clinical reactions were studied. MATERIALS AND METHODS Sfudy populution. The schedule for two m a l l scale field trials with inactivated measles vaccine were previously described in detail (cf. ref. 14). The present follow-up Acta Pzdiatr Scand 65
study was made five years after the last examination of vaccinees (14), i.e. 8-9 years after the last injection of vaccine. Mainly children who at the last time of sampling were found not to have responded either clinically or serologically to exposure to wild virus were included in the study. 1 1 children from vaccine trial I and 16 children from vaccine trial I1 were available for interviewing and collection of a blood sample. Venous bleedings were made and no anticoagulant was added. Sera were collected after incubation of the blood samples for some hours at room temperature and at +4" over night. A selected group comprising 10 children from both trials who did not have detectable non-HI HLI antibodies were vaccinated with further attenuated live measles vaccine (Lirurgen, Dow Chemical Co., Indianapolis, Indiana, USA). Before vaccination a venous blood sample was collected and additional samples were taken one and 4 weeks after vaccination. Serological analyses. The techniques for determination of hemagglutinin-inhibition (HI), hemolysis-inhibition (HLI) and nucleocapsid complement fixing (CF) antibodies were previously described in detail (6). HLI antibodies were determined both before and after removal of HI antibodies by absorption with TE treated concentrated extracellular measles virus material. Only after removal of HI antibodies can HLI antibodies reacting with envelope components separate from the hemagglutinin be determined (7, 15). Sera with an HI antibody titer of 160 or less were absorbed to a final dilution of 1 : 8, whereas sera with a higher HI antibody titer were absorbed to a dilution of I : 10.
RESULTS HI antibody titers in serir~nsumnples collected 3 to 4 and 8 to 9 years ufter a foirrtii dose of I-wecine. Indicutions of measles tiirus infections. Serum samples were collected 8 to 9 years after booster vaccination from 1 1 children in vaccine trial I and 16 children in vaccine trial 11. 3 to 4 years post-booster samples were available from 9 and 14 vaccinees in each group, respectively. The titers of HI antibodies in the different late post-booster samples are shown in Fig. 1. In most vaccinees there was a 4 to %fold reduction in antibody titers during the five-year interval between collection of samples. However, four vaccinees from trial I and three from trial I1 showed significantly increased titers of HI antibodies indicating infections with wild measles virus. Altogether 7 out of I 1 vaccinees in group I showed signs of infections (Table I ) . Besides the four vaccinees with increased titers of HI antibodies (one of which had clinical signs of
Inactivated measles vaccine. VZZZ
I280
-
,'
173
0
/ /
a
320
80
,/
am.
.
00
/ /
..
.
0
. O
/ /
/
O .
o
0
/
-
/
/
00
/ /
/
0
0
/ 20
-
420
-
/
/
0
/
/
I
I
Fig. f. Correlation between serum HI antibody titers 4 ( 3 ) years and 9 (8) years after a booster injection to children in field trials I (circles) and I1 (squares). A filled symbol denotes clinical and/or serological signs of infection with wild measles virus.
infection), there were two children who had had serologically verified infections (one case with clinical symptoms) just before the 3 to 4 years postbooster sampling (14) and still another child with clinical signs interpreted as measles in connection with a known exposure to wild virus inbetween the two times of sampling. One of the two cases of both clinically and serologically verified disease had a some-
what aberrant appearance. There was a high fever and an atypical exanthema beginning on the distal extremities. However, there was no detectable pneumonia. In field trial I1 4 out of 16 children were interpreted to have had infections with wild measles virus. Two of these had mild clinical symptoms, whereas the other two cases only showed a rise in antibody titers.
Table I. Occurrence of children with and M-ithout infections during 3 and 9 years after a booster injection of inactivated vaccine
Appearance of non-HI HLI antibodies in vaccinees with and M-iithout signs of infection uith Mild measles r*irus. The titer of non-HI HLI antibodies in serum samples was determined after removal of HI antibodies by absorption with TE treated measles antigen. Sera from 10 out of the 11 children with signs of infection contained non-HI HLI antibodies. The exceptional case of infection was identifiable both by clinical signs and by a rise in HI antibody titers from 40 to 160. However, no non-HI HLI antibodies were detectable a t a dilution of 1 : 8. The situation in children without signs of infection was completely different (Table I ). No non-HI HLI antibodies were detectable in 3 out of 4 children in trial I and in 10 out of 12 children in trial 11. Eight out of the 13 vaccinees devoid of non-HI HLI antibodies were known to have been exposed to wild
Presence of detectable non-HI HLI antibodies in childre; without signs of infections
Total number of children Children with signs of infection ( a ) clinical aad serological ( 6 ) only clinical ( c ) only serological Children without signs of infection ( a ) without demonstrable non-HI HLI antibodies ( 6 ) with demonstrable non-HI HLI antibodies
Vaccine trial I
Vaccine trial I1
I1
16
7 2 I 4 (2)
4 2 2 (2)
4
12
3(2)"
IO(6)
20)
' Figures within parenthesis denote number of children with known exposure to wild measles virus.
Acto Pzediatr Scond 65
174
E. Norrby and R . Lagercrantz
measles virus. The three vaccinees who had non-HI HLI antibodies demonstrable in their sera, all had been exposed to cases of measles but had shown neither serological nor clinical signs of infection. Nucleocapsid CF antibodies were not detected in a 1 : 5 dilution of sera from children without signs of infection, except in three cases. Two of these cases also had demonstrable non-HI HLI antibodies. In contrast nucleocapsid CF antibodies were present in sera from 6 out of the 7 children with signs of infection in vaccination group I and 2 out of 4 of the corresponding children in group 11. Effect of immunization withfiuther attenuated live measles \vaccine in children Mith HI but no non-HI HLJ antibodies. 10 out of 13 children without demonstrable non-HI HLI antibodies were revaccinated with further attenuated measles vaccine. In no case did this vaccination induce either local or general clinical reactions. The effect on antibody titers are shown in Table 2. Only titers in the four week postvaccination serum samples have been included in the table since no changes in titers were detectable in one week postvaccination samples. Serological reactions to vaccination were found to depend on the titer levels of prevaccination antibodies. In 4 vaccinees with a preimmunization antibody titer of 10 to 40, significant (4-fold or higher) rises in HI, nonHI HLI and nucleocapsid CF antibodies occurred in 3,3 and 2 children, respectively. One vaccinee (R. B.) showed a rise in HI antibody titer from 40 to 80 but no appearance of nonHI HLI and npcleocapsid CF antibodies. 4 additional vaccihees had prevaccination HI antibody titers of 160 to 320. There was no appearance of detectable non-HI HLI antibodies in any of these children after vaccination. Three of these vaccinees had detectable nucleocapsid CF antibody titers which did not change except in one case, who displayed a titer rise from 4 to 8 paralleled by a rise in HI antibodies from 160 to 320. The tenth child selected for vaccination with live vaccine Acta Pdediatr Scand 65
Table 2. Effect of revaccination with further attenuated live measles %saccineof children previously immunized ir’ith only inactivated measles vaccine Serum titers of
Vaccinee“
M. S .
Time of samplingb Pre post
NucleoNon-HI capsid HI anti- HLI anti- CF antibodies bodies bodies 10
