Serological responses and results of natural infectious challenge of recipients of zonal ultracentrifuged influenza A2/Aichi/2/68 vaccine*

Bull. Org. mond. Sante 11971, 45, 767-771 Bull. Wld Hlth Org. Serological Responses and Results of Natural Infectious Challenge of Recipients of Zonal Ultracentrifuged Influenza A2/Aichi/2/68 Vaccine* VERNON KNIGHT,' ROBERT B. COUCH,' R. GORDON DOUGLAS 2 & NICOLA M. TAURASO 3 In an epidemic of Hong Kong influenza in a penal institution, acute respiratory illness occurred in 5.6 % of 196 inmates given single subcutaneous doses of 332, 535, or 1 265 CCA of zonal centrifuged, inactivated vaccine and in 11.7 % of 755 nonvaccinated inmates (P <0.03). The frequency of illness among inmates who received single doses of 54 or 137 CCA of vaccine was not significantly different from that among the nonimmunized inmates (X2=0.93, not significant). The vaccine was prepared with the A2/Aichi/2/68 strain of influenza and administered 7 weeks before the outbreak. Virus shedding was significantly lower in vaccinated inmates with respiratory disease (11.1 %) than in non- vaccinated inmates (73.7%, P<0.04). Serological studies showed an increasing mean titre and increasing frequency of 4-fold or greater haemagglutination inhibition and neutralizing antibody responses with increasing dosage of vaccine. In November and December 1968, volunteers at a state correctional institution near Houston, Texas, were immunized with a monovalent vaccine, pre- pared by zonal ultracentrifugation, and containing influenza A2/Aichi/2/68. In January 1969, a natural outbreak of influenza caused by the Hong Kong variant occurred at the institution. The present report describes the serological responses to graded doses of the vaccine given subcutaneously and the prophylactic effect of the immunization during the epidemic. MATERIALS AND METHODS The investigation was made at the Ramsey I farm unit of the Texas Department of Corrections. During January 1969 there were 1 066 inmates at the unit. * Supported by funds from the Infectious Disease Branch contract No. HEW PH-43-68-963 and by grant No. RR- 00350, General Clinical Research Centers, National Institutes of Health. 1 Departments of Microbiology and Medicine, Baylor College of Medicine, Houston, Tex., USA. 2 Departments of Medicine and Microbiology, University of Rochester School of Medicine and Dentistry, Rochester, N.Y., USA. 3 Division of Biologics Standards, National Institutes of Health, Bethesda, Md., USA. Vaccine was administered to inmates who volun- teered to receive it, under regulations of the State Department of Corrections and of Baylor College of Medicine.4 Excluded from consideration in this study are 21 men who received live influenza virus chal- lenge before the epidemic and who as a result devel- oped a high titre of antibody to the Hong Kong strain. None of them developed influenza during the epidemics. Details of this study are described else- where (Kasel & Couch, 1969). In addition, 20 men were given 1 265 CCA units of the Aichi vaccine intranasally. A high proportion of these volunteers showed 4-fold increases in serum antibody to the Aichi strain, and two men developed afebrile illnesses during the epidemic. These data are not included in the report. The epidemic of Hong Kong influenza first ap- peared in the last few days of December 1968 at a smaller prison unit, Ramsey II, located two miles away. No volunteers from Ramsey II participated in the live virus experiments. Altogether 34% of the 276 inmates at Ramsey II developed febrile respira- 4Volunteers were enlisted without coercion and their written informed consent was obtained after a verbal explana- tion of the study. The procedure used is described in: Knight, V. (1964) Progr. med. Virol., 6, 1-26. 2768 - 767 - V. KNIGHT AND OTHERS tory illnesses during the epidemic, and the Hong Kong variant was isolated from 23 of 31 patients (74.2 %). Vaccine Formol-inactivated monovalent influenza vaccine containing the A2/Aichi/2/68 (Hong Kong variant) strain was supplied by Eli Lilly and Company. Five pools of vaccine containing graded doses of vaccine were made by making serial 2-fold dilutions of a single lot of vaccine concentrate prepared by conti- nuous-flow zonal ultracentrifugation (Reimer et al., 1967). The antigen content of the final vaccine pools was determined by the chicken cell agglutination (CCA) test, which was performed in the laboratory of the Division of Biologics Standards, National Institutes of Health, according to the methods of Miller & Stanley (1944), as described in a National Institutes of Health (1946) memorandum. The NIH CCA-10 reference vaccine containing 400 CCA units per ml was included in all tests, and the corrected CCA value of each vaccine lot represents an average of five determinations. Monovalent influenza B vac- cine containing 640 CCA units/ml of the B/Mass/3/66 strain was given to 60 men, and saline was given to 180 men at the time of administration of the A2 vac- cine. The numbers of volunteers from the 10 resi- dence wings in the prison were different, but vaccine and saline were administered to inmates in each wing roughly in the same proportions as they were admin- istered in the prison as a whole. Virus isolation and identification Rhesus monkey kidney cell cultures were inocu- lated with throat swab specimens. Cultures that sub- sequently demonstrated haemadsorption of guinea- pig erythrocytes were tested with chicken immune sera prepared against influenza A2/Aichi/2/68, A2/Taiwan/1/64, A2/Japan/170/62, and B/Singa- pore/3/64. Neutralization occurred only with the Aichi antiserum. Details of the method have been published previously (Kasel et al., 1965). Neutraliza- tion and haemagglutination inhibition (HI) tests for serum antibody were performed by standard proce- dures (Couch et al., 1971). Case finding Daily sick call was held by medical personnel of each prison unit. Patients with acute respiratory disease, whose illness appeared severe enough to warrant excuse from work, were held for examina- tion by the investigators. Temperatures were taken at each visit to the dispensary, usually once or twice daily during illness. Samples for culturing the in- fluenza virus were taken from approximately 1 in 3 of the men with respiratory disease. Neither the prison medical staff nor the investigators knew which inmates had received influenza A2 vaccine, influenza B vaccine, or saline, or which inmates had not been injected. Symptoms commonly consisted of sudden onset of illness with myalgia and prostration, head- ache, sore throat, cough, and fever. A detailed de- scription of some of these cases is included in a previous report of the effect of amantadine on in- fluenza (Knight et al., 1970). RESULTS Serological responses to zonal-ultracentrifuged vaccine Table 1 shows the frequency of 4-fold or greater serum antibody responses and the mean titres of groups of subjects tested by the HI test. There was an increase in mean antibody titre and in the frequency of rises with increasing dosage of vaccine. The data are similar for the neutralizing antibody response among smaller groups of men who were selected as being free of antibody before vaccination. Epidemic of respiratory disease at Ramsey I Table 2 shows the time of occurrence of cases of acute respiratory illness among immunized men in the period 1-15 January 1969, subdivided according to high dose (332-1265 CCA) and low dose (54-137 CCA) regimens. In the periods before and after that shown, respiratory illness was infrequent. The fre- quency of illness among all immunized men (8.5 %) was not significantly less than that among nonim- munized men (11.7%). However, there was signifi- cantly less illness (5.6%) among men who received the three higher doses of vaccine than among non- immunized men (I1.70%) (X2=5.46, 1 d.f., P<0.03). The difference between those given the two lower doses of vaccine (16.2% of illness) and nonim- munized men (11.7%) was not significant. When only febrile cases (>990F->37.20C) were consi- dered (Table 3), the difference between the occur- rence of illness among all immunized and non- immunized men was significant (X2 = 4.46, 1 d.f., P<0.05), and the differences between febrile men given high doses of vaccine and febrile men given low doses of vaccine or no vaccine were of even greater significance. Virus shedding (Table 2) was significantly less frequent among immunized inmates with respiratory illness (11 .1 %) than among nonimmunized inmates 768 ZONAL ULTRACENTRIFUGED INFLUENZA A2/AICHI/2/68 VACCINE 769 Table 1. Haemagglutination inhibition and neutralizing antibody responses to vaccination with zonal centrifuged influenza A2/Aichi/2/68 (Hong Kong) vaccine. The inoculations were carried out on 7 November 1968 Haemagglutination inhibition antibody Neutralizing antibody (4 weeks) aSubcutaneousI(4 weeks) a dose(CCA units) No. subjects . 4-fold Mean titre No. subjects . 4-fold Mean titre(CCA_____units)_____ tested increase |_Mean_titre_ tested increase |_Mean_titre 1265 50 80% 1:36.2 15 100% 1: 59.7 535 58 76% 1: 29.0 14 93 % 1: 24.7 332 50 60 % 1: 21.3 15 93 % 1: 22.4 137 51 46% 1:13.8 15 80% 1:16 54 - - - 15 80 % 1:12.7 Controls b 47 8 % 1: 5.3 1 5 7% <1: 2 a Before vaccination the haemagglutination inhibition test mean titre was 1: 4.2 and the neutralizing mean titre was <1: 2. b Received 641 CCA units of influenza B/Mass/3/66. Table 2. Acute respiratory illness and influenza virus isolations in immunized and nonimmunized inmates Immunized Nonimmunized Date of onset No. of cases No. of cases No. No. in January Total no. receiving receiving of positive Total no. of positive1970 of cno. doses of Idoses of virus of cases virusof cases 54-137 332-1265 isolations/ isolations/ CCA units CCA units no. tested no. tested 1 4 1 3 -6 1/1 2 0 - - - 12 3/6 3 3 2 1 0/2 14 5/7 4 0 - - - 7 5/6 5 2 1 1 0/1 4 - 6 1 - 1 3 1/1 7 2 1 1 - 13 5/7 8 1 1 - 0/1 9 5/5 9 3 2 1 - 8 2/3 10 2 1 1 0/1 3 - 11 0 - - - 1 0/1 12 3 2 1 0/2 2 1/1 13 2 1 1 1/2 2 - 14 0 ---1- 15 0 ---3- total 23 12 11 1/9 88 28/38 no. at risk 270 74 196 - 755 - percentage 8.5 16.2 5.6 11.1 11.7 73.7 6 V. KNIGHT AND OTHERS Table 3. Occurrence of febrile and afebrile respiratory illness in relation to vaccine dose Acute respiratory illness a Subcutaneous dose No. (1-15 January, 1969) (7 Nov. 1968) vaccinated (CCA units) Febrile Afebrile (> 990 F) (< 990 F) 1265 78 3 1 535 59 1 3 332 59 2 1 137 60 7 2 54 14 2 1 total no. immunized 270 15 8 no vaccine 755 76 12 a Statistical analysis: No. of febrile immunized (1265, 535, and 332 CCA) versus no. of febrile immunized (137 and 54 CCA): 6 out of 196 versus 9 out of 74 ( x2, 6.834; 1 d.f.; P<0.01). No. of febrile immunized (1265, 535, and 332 CCA) versus no. of febrile nonimmunized: 6 out of 196 versus 76 out of 755 ( x2, 8.82; 1 d.f.; P<0.01). No. of febrile immunized (137 and 54 CCA) versus no. of febrile nonimmunized: 9 out of 74 versus 76 out of 755 ( x2, 0.134; 1 d.f.; P>0.50). (73.7%) (P <0.04, Fisher's exact test). The number of cultures taken was too few for valid testing of the two subgroups of vaccine dosage. Illness in residence wings Immunized men constituted from 10% to about 50% of the population of the 10 residence wings of the prison. Illness occurred in immunized men in 9 of the 10 wings in proportions ranging from 3.4% to 27.6%. Illness in nonimmunized men ranged from 3.3 % to 30.1 % in the 10 wings. There was no significant correlation in the frequency of illness between immunized and nonimmunized men in the various residence wings. Virus isolation from subjects without respiratory ill- ness Cultures for influenza virus were also made during the epidemic on 22 immunized men who reported to sick call with minor complaints not involving the respiratory tract. None of these was positive for influenza. In addition, cultures for influenza virus were made on 24 nonimmunized men who reported to sick call with minor complaints not involving the respiratory tract, and 4 were positive. These 4 were not included in the study because of the absence of respiratory symptoms. DISCUSSION From the data presented it is apparent that vacci- nation resulted in substantial serum antibody res- ponses that were dose-related. The high frequency of responses and the substantial titres achieved with the three higher doses of vaccine were consistent with a potential protective effect. The principal finding of this study is that, of five potency levels of single doses of vaccine given 7 weeks before an epidemic of influenza due to the Hong Kong variant, the three higher levels were associated with a significant reduction in respiratory illness during the epidemic in comparison with the frequency of such illness among a large group ofnonimmunized inmates. If only the patients in whom fever was observed are considered, there was significantly less illness in all five vaccine dosage groups in compari- son with nonimmunized men. Virus shedding in all vaccinated inmates with respiratory disease was sig- nificantly reduced. These results are considered to represent a protective effect of vaccination. It was noted that the concurrent epidemic in a neighbouring institution only 2 miles away involved 34% of inmates, and outbreaks of this severity had occurred in the institution where the trial was carried out in previous years. This suggests that vaccination, by inducing partial herd immunity, may have reduced the overall severity of the epidemic at Ramsey I. It is conceivable, therefore, that immunization was even more beneficial than the data suggest. The epidemic occurred only 7 weeks after immunization, at a time when the protective effect of immunization was pro- bably at its maximum, and it is not certain how long this level of protection would persist. In the present study, influenza virus was isolated from 4 of 24 nonimmunized inmates who came to the dispensary with complaints not involving the res- piratory tract. One immunized patient with afebrile respiratory disease also yielded influenza virus. In contrast, no virus was isolated from 22 immunized afebrile men with complaints not related to the respiratory tract. While there may be different explanations for these findings, we believe that they represent asymptomatic infections similar to those that have occurred, in our experience, in volunteers inoculated with live influenza virus. Many previous studies have attempted to relate the efficacy of vaccines to dose, type of vaccine, or 770 ZONAL ULTRACENTRIFUGED INFLUENZA A2/AICHI/2/68 VACCINE 771 mode of administration; but in view of the possibly uncertain content of antigen in many such vaccines, the different studies are probably not comparable. Recent research conducted in the Division of Bio- logics Standards (NIH) laboratories resulted both in the standardization of the CCA test for measuring vaccine antigen content and in the preparation of stable reference vaccines (Tauraso et al., 1969). With the advent of the Hong Kong influenza epidemic in the autumn of 1968, these improvements were ap- plied to the certification of the potency of the mono- valent A2/Aichi/2/68 influenza vaccine used in the present study. This standardization of potency of the antigen will be available in the future and should provide a better basis of comparison of the effective- ness of vaccines. ACKNOWLEDGEMENTS Assistance in statistical analysis was provided by Mrs E. Osborne. The authors thank Dr George Beto, Commissioner of Corrections, State of Texas, and Warden T. M. Hutto and his staff, especially Captain Clive Aldridge and Captain H. D. Cottrell, for their co-operation and for providing facilities for this study. RE'SUME REPONSES SEROLOGIQUES ET REACTIONS A UNE INFECTION NATURELLE CHEZ DES SUJETS IMMUNISES PAR UN VACCIN ANTIGRIPPAL A2/AICHI/2/68 PURIFIt PAR ULTRACENTRIFUGATION DE ZONE Au cours d'une epidemie de grippe de Hong Kong survenue dans un etablissement penitentiaire du Texas en 1969, on a enregistre 11 cas d'affection aigue de l'appareil respiratoire parmi 196 detenus immunises 7 semaines auparavant par un vaccin A2/Aichi/2/68 administre en une dose unique sous-cutanee de 332 a 1265 unites CCA (agglutination des hematies de poulet). Le taux de morbidite de 5,6% parmi ces detenus differait significativement (P<0,03) de celui observe parmi leurs 755 compagnons non vaccines chez lesquels on a releve 88 cas d'affection respiratoire (11,7 %). En revanche, chez des detenus immunises par des doses plus faibles (54 a 137 unites CCA), la morbidite a ete sensiblement la meme que chez les sujets non vaccines. On a constate une elimi- nation du virus chez 11,1 % des malades vaccines et 73,7 % des malades non immunises, la difference etant significative (P<0,04). Les cas de fievre ont e significa- tivement moins nombreux (6 sur 196) chez les sujets vaccines a forte dose que chez les sujets ayant requ une dose faible (9 sur 74) et chez les sujets non vaccines (76 sur 755). Les examens serologiques ont montre que la hausse des titres moyens d'anticorps inhibant l'hemagglutination (IH) ou neutralisants (N) etait fonction de la dose vaccinale. II en etait de meme en ce qui regarde les hausses de titres de 4 fois et plus qui ont e observees chez 80% (titres IH) et 100% (titres N) des sujets vaccines par une dose forte. Dans un etablissement penitentiaire voisin, en I'absence de toute vaccination, une epidemie de grippe du meme type a provoque une morbidite de 34%. Il apparait que la vaccination, en induisant une immunite partielle de groupe, a pour effet de reduire la frequence de la maladie egalement chez les sujets non vaccines. REFERENCES Couch, R. B., Douglas, R. G., Fedson, D. S. & Kasel, J. A. (1971) J. infect. Dis., 124, 473-480 Kasel, J. A. & Couch, R. B. (1969) Bull. Wld Hlth Org., 41, 447-452 Kasel, J. A., Waddell, G. H., Alford, R. H., Sigel, M. M. & Knight, V. (1965) Nature (Lond.), 206, 41-43 Knight, V., Fedson, D., Baldini, J., Douglas, R. G. & Couch, R. B. (1970) Infection and Immunity, 1, 200-204 Miller, G. L. & Stanley, W. M. (1944) J. exp. Med., 79, 185-195 National Institutes of Health (1946) Titration of chicken red cell agglutination (CCA) value, Bethesda, Md., Division of Biologics Standards, National Institutes of Health Reimer, C. B., Baker, R. S., van Frank, R. M., Newlin, T. E., Cline, G. B. & Anderson, N. G. (1967) J. Virol., 1, 1207-1216 Tauraso, N. M., O'Brien, T. C. & Seligmann, E. G., Jr (1969) Buill. Wld Hlth Org., 41, 497-506