Infectivity and reactogenicity of reassortant cold-adapted influenza A/Korea/1/82 vaccines obtained from the USA and USSR

Bulletin of the World Health Organization, 65 (3): 295-301 (1987) ©) World Health Organization 1987 Infectivity and reactogenicity of reassortant cold-adapted influenza A/Korea/1/82 vaccines obtained from the USA and USSR K. G. NICHOLSON,' D. A. J. TYRRELL,2 J. S. OXFORD,3 J. WOOD,3 G. C. SCHILD,4 C. W. POTTER,5 R. JENNINGS,6 R. H. MICHAELS,7 & G. APPLEYARD8 The safety and immunogenicity of two live influenza A virus vaccine strains, the CR 59 and 17/25/1 cold-adapted (ca) reassortants, were evaluated in 170 healthy young adult volunteers. The vaccines were produced by recombining A/Korea/1/82 (H3N2) wild-type virus with either A/Ann Arbor/6/60 (H2N2) or A/Leningrad/134/17/57 (H2N2) ca donors of attenuation. Both vaccines were well tolerated in volunteers. The 17/25/1 strain, prepared from A/Leningrad, infected at least 70% of seronegative volunteers after the first dose and 84% after the second; the CR 59 strain infected 62% and 72% of volunteers after first and second doses, respectively. Among the vaccinees who were initially seropositive, 17/25/1 infected 66% after one dose and 85% after two, while CR 59 infected 62% and 71%, respectively. Despite differences in temperature sensitivity, genetic composition, and serological reactivity to monoclonal antibodies, both vaccines behaved almost identically in animal models and man. We conclude that both donors of attenuation may be of great potential value. Because the existing inactivated influenza vaccines are relatively ineffective, even when given annually (1), there is increasing interest in live attenuated ' Senior Lecturer, Infectious Diseases Unit, Department of Pharmacology and Therapeutics, Groby Road Hospital, Leicester LE3 9QE, England. Requests for reprints should be sent to this author. 2 Director, MRC Common Cold Unit, Salisbury, Wilts., England. 3Scientific Officer, National Institute for Biological Standards and Control, Hampstead, London, England. 4 Director, National Institute for Biological Standards and Control, Hampstead, London, England. 5 Professor, Department of Virology, University of Sheffield Medical School, Sheffield, England. 6 Senior Lecturer, Department of Virology, University of Sheffield Medical School, Sheffield, England. ' Visiting Scientist, Department of Virology, University of Sheffield Medical School, Sheffield, England. At present at The Children's Hospital of Pittsburgh, Pittsburgh, PA, USA. 8 Section Head, Virology, Wellcome Biotechnology Ltd, Becken- ham, Kent, England. vaccines administered by the intranasal route. In the USA and USSR an approach now being pursued for the rapid attenuation of new influenza A viruses involves the use of an empirically derived donor of attenuation which grows well at a temperature that is suboptimal for the replication of wild-type virus (2, 3). Reassortant viruses derived from co-infection with cold-adapted (ca) donor strains and virulent influenza A virus can be rapidly and reliably obtained and, when evaluated in adults and children, have been found to be safe, stable, and antigenic in both the USA and USSR (4). Although based on the same concepts, the methods used in the two countries show many differences in detail and it was uncertain whether the two ca vaccine strains really had similar properties of attenuation and ability to induce protective antibody when administered to man. We therefore prepared, under identical conditions, two influenza viruses from seed ca viruses provided 4781 -295- K. G. NICHOLSON ET AL. by Dr H. Maassab in the USA and Dr Y. Ghendon in the USSR. The two seed viruses were derived from the same "wild" human influenza parent, A/Korea/ 1/82, but different ca viruses were used as the donors of attenuation. The vaccines were administered to healthy young adults whose responses were evaluated clinically and by determining the anti-haemagglutinin antibody (HI) titres. In addition, we sought to compare the strains by laboratory procedures. METHODS The seed ca reassortant (CR 59) was produced by Dr H. Maassab by recombining A/Ann Arbor/6/60 (H2N2) ca virus with A/Korea/i /82 (H3N2) wild- type virus using methods previously described (5). The seed ca reassortant (17/25 / 1) was produced by Dr Y. Ghendon by recombining the ca donor of attenuation for adults, A/Leningrad/134/17/57 (H2N2), with A/Korea/ 1/82 (H3N2) wild-type virus that had twice been cloned by terminal dilution in embryonated hens' eggs. Vaccine preparation in specific pathogen-free eggs and quality control tests were carried out at the Wellcome Research Laboratories, Beckenham, Kent. Bulk CR 59 vaccine, prepared from the American seed, had a titre of i07.5 EID5o (50% egg infectious doses) per ml; bulk 17/25/1 vaccine, prepared from the Russian seed, had a titre of 107.8 EID5o/ml. Neither vaccine seed for this study underwent prior selection in preliminary human studies and both can be considered as random clones selected only on the basis of laboratory properties. The study protocols were approved by Northwick Park Hospital and Leicestershire Area Health Authority Ethical Committees. Informed consent was obtained for all human studies. In a preliminary study at the Common Cold Unit, Salisbury, 23 healthy adults between the ages of 18 and 40 years were randomly assigned to receive 107 5 EID5o of either CR 59 or 17/25/1 ca reassortant virus intra- nasally by drops. The vaccine was administered in 1-ml volumes to volunteers lying with their necks hyperextended; they remained supine for a minute after immunization and were asked to sniff rather than blow their nose for a further hour. Volunteers were examined daily for systemic and respiratory symptoms and signs and nasal washings were collected for three days after immunization. Paired sera were collected for determining pre- and post- immunization serum haemagglutination-inhibition (HI) antibody titres. A second trial in 147 healthy students was carried out when results from the first trial became available. Healthy volunteers between the ages of 18 and 40 years were randomly assigned to receive 107 5 EID5o of either CR 59 or 17/25/1 ca reassortant virus intranasally, as described above. Volunteers were given a standard diary card to complete daily for three weeks and venous blood was taken for serology. Four weeks later they returned for a second dose of the same vaccine as before and were bled again and given another diary card. Blood samples were obtained 5 weeks thereafter. Standard methods were used to measure the HI antibody in coded sera from volunteers. In the first trial, A/England/40/83 was used as antigen in the HI test because of its greater sensitivity in detecting serum antibody than that of the CR 59 and 17/25/1 reassortant vaccines and the parent A/Korea/1/82 strain. All four strains were used as antigen in the HI tests in the second trial, and here seronegative individuals were defined as having pre-immunization titres of < 1:10 against all four test antigens. Laboratory evidence of influenza A virus infection was defined either by isolation of influenza A virus (attempted in the first study only), or by a fourfold or greater rise in serum antibody (HI) against A/Korea/1/82, A/England/40/83, or the reassort- ant virus administered. This decision was based on the greater sensitivity of the A/England strain in the HI test compared with the parental wild-type virus and the homologous vaccine strain. The use of four test antigens also helped establish the comparability of the two experimental groups before immunization and, with antibody rises generally occurring with two or more antigens, it reduced the likelihood of spurious results. Dilutions of nasal washings for virus isolation were inoculated into 10-day embryonated eggs by the allantoic route. In certain experiments the canine kidney cell line (MDCK) was used to detect virus in nasal washings. After incubation at 36 °C for 72 hours, the allantoic fluids or tissue culture fluids were harvested and tested for virus haemagglutinin using 0.5% chick cells. Wild-type A/Korea/I /82 virus and the two vaccine strains, CR 59 and 17/25 / 1, were tested for virulence in two animal models. First, groups of four ferrets were each inoculated intranasally under ether anaesthesia with i05 EID5o of test virus in 0.5 ml of Hanks' saline. The temperature response, virus replication, and changes in nasal wash protein were assessed as described previously (6). Secondly, groups of newborn rats aged 24-48 hours were inoculated intranasally with 0.01 ml of CR 59, 17/25/1, or wild-type A/Korea/i /82 virus contain- ing 108 EIDso/ml. Forty-eight hours later, i05 cfu of a virulent strain of Haemophilus influenzae type b were inoculated into the anterior nares of each rat; the incidence of bacteraemia was determined 48 hours later, as described previously (7). The temperature-sensitivity of replication of the 296 REASSORTANT COLD-ADAPTED INFLUENZA VACCINES wild parent strain, A/Korea/ 1/82, the two reassort- ants CR 59 and 17/25/1, and one other parental virus (namely, A/Ann Arbor/6/60) were determined by comparing virus replication in MDCK cells incubated at temperatures in the range of 32-39 'C. Viruses were titrated in the presence of 4 Ag/ml of trypsin on microwell cultures of MDCK cells. Haemagglutinins of CR 59, 17/25/1 and the wild parent A/Korea/1/82 virus were compared using a panel of 27 monoclonal antibodies and 4 post- infection ferret sera raised against recent influenza A H3N2 strains including A/England/23/76, A/Texas/ 1/77, and A/Bangkok/i /79. Similarly, neuraminidase antigens of the two reassortants and the wild parent A/Korea/1/82 virus were compared in the lectin test (8) using 3 post-infection ferret sera and 3 monoclonal antibodies directed against the neuraminidase of recent isolates. Finally, the two reassortants CR 59 and 17/25/1, the wild parent strain A/Korea/ 1/82, and one ca donor of attenuation A/Ann Arbor/6/60 were all subjected to polyacrylamide gel analysis of virus polypeptides (9). RESULTS Properties of the vaccines Gene analysis. The CR 59 virus contained six genes that code for internal and non-structural proteins from the donor cold-adapted strain and the genes that code for the haemagglutinin and neuraminidase surface glycoproteins from the influenza A/Korea virus (Dr H. Maassab, personal communication). The 17/25/1 virus inherited genes 1, 2, 3 and 7 from the ca parent and genes 4, 6 and 8 from the A/Korea strain. Gene 5 could not be differentiated by means of cRNA/DNA hybridization, although since gene 5 of 17/25/1 has the same temperature-sensitive (ts) mutation as gene 5 of the ca parent, it is thought to be inherited from the ca parent. The 17/25/1 virus contains ts mutations in genes 1, 5 and 7 (Dr Y. Ghendon, personal communication). Preliminary biochemical analysis of the viruses confirmed that both the CR 59 and 17/25/1 viruses were re- assortants since, for example, neither virus possessed the nucleoprotein (NP) or matrix (M) protein of A/Korea/ 1/82. Virulence studies. Four separate experiments were carried out to determine the virulence of CR 59, 17/25/1, A/Korea/i/82 wild-type virus, and a control virulent virus, A/Finland/74 (H3N2), in an infant rat model of infection (7). H. influenzae bacteraemia occurred in 32 out of 35 (91%7o) rats following inoculation with A/Finland/74. In con- trast, bacteraemia developed in 6% of rats (3 of 51) inoculated with A/Korea/I /82, 3% (1 of 37) inocu- lated with CR 59, and 19% (7 of 37) with 17/25/1; there were no significant differences between the rates of bacteraemia for A/Korea/ I /82, CR 59, and 17/25/1 viruses. Inoculation of the viruses into ferrets showed similar results: both reassortant viruses and A/Korea/i/82 wild-type virus induced little or no febrile reaction and little change in nasal protein. All three viruses grew to equally high titre in ferret nasal washings and induced comparable titres of post- infection serum HI antibody. Temperature sensitivity. A total of four experiments showed the CR 59 reassortant to have marked ts properties, with a reduction in titre of at least 4.0 Iogio TCID5o (50% tissue culture infectious doses) per ml when titrated in MDCK cells at 39 'C compared to 32 'C. In contrast, the 17/25/1 vaccine virus showed a 2.25 log,o reduction in titre under comparable conditions; the parental wild virus, A/Korea/ I /82, showed no ts properties. Serological analysis of the CR 59 and 17/25/1 haemagglutinin. Serological analysis of the viruses using monoclonal antibodies to haemagglutinin (HA) and the HI test demonstrated that the HA of the CR 59 vaccine virus was antigenically similar to that of the A/Korea/I /82 virus. In contrast, the HA of the 17/25/1 vaccine virus showed a number of antigenic differences. Post-infection ferret sera also distinguished the two vaccine strains but to a much lesser extent than the monoclonal antibodies. One hundred human sera failed to detect any fourfold differences between 17/25/1 and CR 59; however, there was a statistically significant difference between the geometric mean reciprocal HI titres (GMT) against CR 59 and 17/25/1 (23.8 versus 28.8 respectively, P<0.005, paired t test). Serological analysis of the neuraminidase showed the CR 59 and 17/25/1 vaccine strains to be closely related to the neuraminidase of the parental A/Korea/1/82 virus. Concomitant with serological differences in the HA of the 17/25/i vaccine strain compared to CR 59 and A/Korea/1 /82, electrophoretic migration dif- ferences in HA were also detected. Infectivity of volunteers For evaluation of the results of both clinical trials, the volunteers were divided into two groups according to the vaccine received and further subdivided into categories according to their initial antibody status. Among the 13 persons given CR 59 in the preliminary study, 10 had pre-immunization titres to A/England/40/83 of < 1:40 and 5 (50%) of these had fourfold or greater rises in HI antibody after infection. Among the 10 volunteers given 297 K. G. NICHOLSON ET AL. Table 1. Antibody status of vaccinees before immunization No. with reciprocal antibody titre to: A/Korea/ 1 /82, A/Eng/40/83 A/Korea/1 /82 A/Eng/40/83 and vaccine strain Vaccine No. receiving vaccine <10 <40 <10 <40 <10 CR 59 63 45 56 36 44 29 17/25/1 58 32 53 28 35 20 17/25 /1 vaccine, 7 had initial titres of < 1:40 and 4 were infected by the first dose and 2 out of 5 (40%/) (57/) of these had serological evidence of infection. were infected by the second; 16 out of 19 (84%) None of six volunteers with pre-immunization titres recipients of two doses were infected overall: one of >1:40 was infected. person developed fourfold rises in antibody after Serological data are available from 121 out of 147 both the first and second doses. Of 13 seropositive vaccinees in the second clinical trial. There were no individuals with pre-immunization titres of < 1:40 significant differences in the initial distribution of against A/Korea/1/82, A/England/40/83, and antibody to either A/Korea/1/82 or A/England/ 17/25/1 viruses, 11(85%0) were infected by the first 40/83 between the two vaccine groups (Table 1) dose and 1 out of 2 was infected by the second; 12 out (P<0.05, X2 test). of 13 (920/o) individuals given both doses were Among the 20 recipients of 17/25/1 vaccine who infected overall. Among 25 volunteers with pre- were initially seronegative to A/Korea/1/82, immunization titres of > 1:40 against A/Korea/ A/England/40/83, and 17/25/1 viruses, 14 (70%o) 1/82, A/England/40/83, or 17/25/1 viruses, 14 Table 2. Hi antibody status before and after intranasal inoculation with 17/25/1 and CR 59 attenuated influenza vaccines Percentage infected after: Geometric mean reciprocal antibody titre Vaccine and initial Before After After HI-antibody status Hi test antigen 1 st dose 2nd dose 1 st dose 1 st dose 2nd dose 17/25/1 vaccine: Seronegative A/England/40/83 70 84 <10 69 135° A/Korea/1 /82 <10 17 19 Seropositive A/England/40/83 85 92 9 70 154 b (< 1:40) A/Korea/ 1 /82 14 29 54 Seropositive A/England/40/83 56 80 71 224 278 (> 1:40) A/Korea/1 /82 17 36 44c CR 59 vaccine: Seronegative A/England/40/83 62 72 <10 31 31 A/Korea/ 1 /82 <10 13 16 Seropositive A/England/40/83 71 83 14 76 76 b (< 1:40) A/Korea/ 1 /82 9 32 39 Seropositive A/England/40/83 55 63 106 276 413 (> 1:40) A/Korea/ 1 /82 16 61 82C P<0.005. bP<0.025. ¢ P<0.05. 298 REASSORTANT COLD-ADAPTED INFLUENZA VACCINES (56%) were infected by the first dose and 2 out of 6 were infected by the second; 16 out of 20 (80%) recipients of both doses were infected overall. The infection rate for all 33 persons, initially seronegative or having titres of < 1:40, was 76%o (25 of 33) after the first dose and 87.50o (28 of 32) after the second; for all 58 recipients of one and two doses of 17/25 / 1, the infection rates were 67%o (39 of 58) and 85% (44 of 52) respectively. The geometric mean antibody titres are presented in Table 2. Among the 29 recipients of CR 59 vaccine who were initially seronegative to A/Korea/ 1/82, A/England/40/83, and CR 59 viruses, 18 (627o) were infected by the first dose and 3 out of 11 (27 /o) were infected by the second; 21 out of 29 (72%o) recipients of two doses were infected overall. Of 14 seropositive individuals with pre-immunization titres of <1:40 against A/Korea/ 1/82, A/England/40/ 83, and CR 59 viruses, 10 (71%) were infected by the first dose and none of 2 was infected by the second; 10 out of 12 (837o) individuals given both doses were infected overall. Among 20 volunteers with pre- immunization titres of > 1:40 against A/Korea/ 1/82, A/England/40/83, or CR 59 viruses, 11 (5570o) were infected by the first dose and 1 out of 8 was infected by the second; 12 out of 19 (63%o) recipients of both doses were infected overall. The infection rate for all 43 persons initially seronegative or having titres of < 1:40 was 650o (28 of 43) after the first dose and 76%o (31 of 41) after the second; for all 63 recipients of one and two doses of CR 59, the infection rates were 62/o (39 of 63) and 72% (43 of 60) respectively. Although there were no significant differences between the infection rates for CR 59 and 17/25/1 viruses in any of the groups analysed, comparison of the antibody responses (Table 2) revealed signifi- cantly higher titres against A/England/40/83 after the second dose of 17/25/1, as compared with CR 59, both among seronegatives and seropositives with initial titres of < 1:40. Virus was isolated from only one of the 23 volunteers inoculated with either reassortant virus in the preliminary studies and this had the ca properties of the parental virus. No virus isolation procedures were carried out in the second study. Adverse effects on volunteers The frequency and severity of reactions to CR 59 and 17/25/1 vaccines were unaffected by the antibody status before inoculation. The majority of infected and non-infected persons (81-90%) were either asymptomatic or complained of no more than one local reaction (rhinorrhoea, cough, or sore throat) during the 5 days after immunization. Systemic complaints (headache, malaise, myalgia, or feeling feverish) together with two or more upper respiratory symptoms developed after 2% (3 / 135) of non-infecting immunizations, as compared with 12% (5/43) for the CR 59 infections (P<0.05) and 7% (3/43) for the 17/25/1 infections. Two volunteers took time off work during the 5-day period after immunization but in neither case was this attributable to the inoculation: one volunteer, infected with CR 59, had a diarrhoeal illness with no rhinorrhoea or sore throat; the other, infected with 17/25 / 1, was off work with otitis which developed immediately after the immunization. DISCUSSION The most impressive observation of this study was the remarkable similarity of the Russian and American ca reassortants in terms of infectivity, immunogenicity and reactogenicity. We found that a single inoculation with CR 59 or 17/25/1 virus was sufficient to infect at least 62%7o and 707o, respect- ively, of seronegative volunteers, 71% and 85% of subjects with pre-immunization titres of < 1:40, and 55% and 56%o of persons with initial titres > 1:40. Although infection of persons possessing HI anti- body is not without precedent, we cannot rule out the possibility that the antigenic mass of HA in the inoculating material caused a host immune response in persons with pre-existing antibody. Nevertheless, in practical terms a large percentage of the volunteers had, at the end of the study, increased levels of HI antibody which, assuming a correlation between antibody and protection, would be advantageous. The second inoculation with CR 59 virus infected an additional 10%o of volunteers overall and the 17/25/1 reassortant a further 18%, suggesting that second doses may be highly cost-effective. It is also of note that a second dose reinfected only 1 out of 87 vaccinees: although not a formal protection study, the results are comparable to the high levels of protection against illness and infection found in recent studies using similar ca reassortants and wild- type virus, with virus challenge occurring up to 3 months after immunization (10, 11). Both the ca reassortant virus vaccines used in the present study were well tolerated; 817o of infections with CR 59 and 88%1o of infections with 17/25 /1 were either asymptomatic or were associated with only one upper respiratory symptom. A total of 12%o ofCR 59 infections, 707o of 17/25/1 infections, and 207o of non-infective inoculations were followed by various combinations of systemic and upper respiratory symptoms; these were generally mild and of short duration and resulted in no time off work; their frequency and severity were uninfluenced by the levels of antibody prior to immunization and, in the 299 K. G. NICHOLSON ET AL. context of protection against influenza, were considered acceptable. The studies in the rat and ferret animal models confirmed that the two reassortants CR 59 and 17/25/1 were attenuated but failed to establish that the wild-type A/Korea virus was virulent. Although in this study we did not compare the reassortants with wild-type virus in man, colleagues in the United States inoculated 14 seronegative volunteers with A/Korea/ 1/82 wild-type virus and observed systemic reactions in 29% and rhinitis in 50% (Dr M. H. Snyder, personal communication). Since intranasal infection with 17/25/1 and CR 59 caused significantly less rhinitis (14%o and 160/0, respectively) and fewer systemic reactions than the wild-type virus, we conclude that the two reassortants were attenuated but acknowledge that the A/Korea/I /82 parent is only of moderate virulence. Serological analysis of the viruses using mono- clonal antibodies to HA and the HI test demonstrated that the HA of the CR 59 vaccine virus was antigenically similar to that of the A/Korea/i/82 virus, whereas the HA of the 17/25/1 virus showed a number of differences. Concomitant with serological differences in the HA of the 17/25/1 vaccine strain compared to CR 59 and A/Korea/i/82, electro- phoretic migration differences in HA were also detected. It must be assumed therefore that the A/Korea wild-type virus was a heterogenous mixture and that the laboratory manipulations in the USSR and USA resulted in the emergence of antigenically distinguishable viruses. The haemagglutinins of the Korea wild-type parent and two reassortants are currently being sequenced and the results will be published with other biochemical data. Previous studies show that the transfer of the six internal genes from the A/Ann Arbor/6/60 ca virus reproducibly confers properties of attenuation for man, genetic stability, lack of transmissibility, and the ability to induce resistance to challenge by wild- type virus. The Russian A/Leningrad/134/17/57 ca reassortants are said to behave identically to the A/Ann Arbor reassortants and have been given to many thousands of people in the USSR. Despite differences in temperature sensitivity and genetic composition, our findings indicate CR 59 and 17/25 /1 to be of equal efficacy in man. We conclude that intranasal immunization with reassortants from both donors of attenuation may be of great value and that comparative field trials should be conducted to determine their clinical efficacy. ACKNOWLEDGEMENTS We thank Dr J. J. Skehel and Dr A. Douglas from the National Institute for Medical Research, Mill Hill, and Dr P. Yates from the National Institute for Biological Standards and Control, Hampstead, for certain monoclonal antibodies; Dr J. Willman, Dr P. G. Higgins, Mrs M. Andrews and staff members from the MRC Common Cold Unit; Dr P. Farrow, Dr B. Young and Mrs B. Dunlop from Groby Road Hospital, Leicester; Dr C. Price and staff members from the Medical Centre, Loughborough University; Dr T. Bektimirov of WHO, Geneva, for help in obtaining seed viruses; Mr J. H. Bennett and Mr W. Matchett from Wellcome Biotechnology Ltd; members of Loughborough University Students' Union for assistance in the conduct of this study; and especially Dr H. Maassab, Dr Y. Ghendon and the volunteers for their contributions. RESUME INFECTIOSITt ET RtACTOGtNICITS DE VACCINS GRIPPAUX A/KOREA/ 1/82 REASSORTIS ET CRYO-ADAPTtS OBTENUS AUX tTATS-UNIS D'AMtRIQUE ET EN UNION SOVIETIQUE On a evalue chez 170 volontaires adultes, jeunes et en bonne sante, 1'innocuite et l'immunogenicite de deux souches de vaccins grippaux vivants, les vaccins cryo- adaptes (ca) et reassortis CR 59 et 17/25/1. Ces vaccins ont et obtenus en recombinant le virus de type sauvage A/Korea/l /82 (H3N2), avec des attenuateurs A/Ann Arbor/6/60 (H2N2), ou A/Leningrad/134/17/57 (H2N2) cryo-adaptes. Le virus CR 59 contenait six genes codant pour les proteines internes non structurales de la souche attenuatrice, cryo-adaptee, A/Ann Arbor/6/60 ainsi que les genes codant pour les glycoprotenes de surface (hemag- glutinine et neuraminidase) provenant du virus A/Korea. Le virus 17/25/1 a herite les genes 1, 2, 3 et 7 de la souche mere A/Leningrad/134/17/57 et les genes 4, 6 et 8 de la souche A/Korea, le gene 5 n'ayant pu etre differencie par hybri- dation ARN/ADN, encore que, puisque le gene 5 du virus 17/25/1 presente la meme mutation thermosensible que le gene 5 de la souche cryo-adaptee, on puisse legitimement penser qu'il provient de cette souche. On a determine la virulence du virus A/Korea/ 1/82 de 300 REASSORTANT COLD-ADAPTED INFLUENZA VACCINES 301 type sauvage et des deux souches variantes CR 59 et 17/25/1 chez le furet et le rat. Ces etudes ont confirme I'attenuation des deux virus reassortis, mais elles n'ont pu mettre en evidence la virulence du virus A/Korea de type sauvage. La souche reassortie CR 59 presentait des proprietes thermosensibles marquees, avec une reduction de titre d'au moins 4,0 log1o DICT5o (dose infectante 50%o en culture tissulaire), quand le titrage avait lieu sur cellules MDCK a 39 °C plutot qu'a 32 'C. Par opposition, le virus vaccinal 17/25/1 presentait une reduction de titre de 2,25 log1o dans des conditions comparables; le virus A/Korea/ 1/82 de type sauvage, provenant de la meme lignee, n'a montre aucune thermosensibilite. L'analyse serologique des virus a l'aide d'anticorps monoclonaux hemagglutinants et la reaction d'inhibition de l'hemagglutination ont revele que l'hemagglutinine du virus de la souche vaccinale CR 59 etait antigeniquement identique a celle du virus A/Korea/ 1/82, tandis que celle du virus vaccinal 17/25/1 presentait un certain nombre de differences antigeniques. En outre, parallelement a ces differences serologiques au niveau de l'hemagglutinine de la souche vaccinale 17/25/1 par rapport a CR 59 et A/Korea/ 1/82, on a observe des differences en ce qui concerne la migration electrophoretique. Les volontaires ont bien supporte les deux vaccins. Le virus 17/25/1 obtenu a partir de A/Leningrad a infecte au moins 70% des volontaires seronegatifs apres la premiee dose et 84% apres la seconde; une et deux doses du virus CR 59 ont infecte respectivement 62% et 72% des volontaires. Le virus 17/25/1 a infecte 66% des sujets vaccines, initialement seropositifs, apres une dose et 85% apres deux doses; de meme, le virus CR 59 a infecte respectivement 62% et 71% de ces sujets. Malgre des differences au niveau de la thermo- sensibifite, des composants genetiques et de la reactivite serologique aux anticorps monoclonaux, les deux vaccins se sont comportes pratiquement de la meme facon chez les modeles animaux et chez l'homme. Nous en concluons au grand interet potentiel de ces deux attenuateurs. REFERENCES 1. HOSKINS, T. W. ET AL. Assessment of inactivated influenza A vaccine after three outbreaks of influenza A at Christ's Hospital. Lancet, 1: 33-35 (1979). 2. ALEXANDROVA, G. I. & SMORODINTSEV, A. A. Obtaining an additionally attenuated vaccinating cryophilic influenza strain. Revue roumaine d'inframicrobiologie, 2: 179-186 (1965). 3. MAASSAB, H. F. Adaptation and growth characteristics of influenza virus at 25 'C. Nature, 213: 612-614 (1967). 4. KENDAL, A. S. ET AL. Development of cold-adapted recombinant live attenuated influenza vaccines in the USA and USSR. Antiviral research, 1: 339-365 (1981). 5. Cox, N. J. ET AL. Comparative studies of wild-type and cold-mutant (temperature sensitive) influenza viruses: nonrandom reassortment of genes during preparation of live virus vaccine candidates by recombination at 25 'C between recent H3N2 and HINI epidemic strains and cold-adapted Ann Arbor/6/60. Virology, 97: 190-194 (1979). 6. POTTER, C. W. ET AL. Immunity to influenza in ferrets; response to live and killed virus. British journal of experimental pathology, 53: 153-167 (1972). 7. JENNINGS, R. ET AL. The replication of type A influenza viruses in the infant rat: a marker for virus attenuation. Journal ofgeneral virology, 49: 343-354 (1980). 8. LUTHER, P. ET AL. An investigation of antigenic drift of neuraminidases of influenza A(HlNl) viruses. Journal of hygiene, 92: 223-229 (1984). 9. OXFORD, J. S. ET AL. Biochemical and antigenic analysis using monoclonal antibodies of a series of infuenza A(H3N2) and (HINI) virus reassortants. Vaccine, 4: 9-14 (1986). 10. CLEMENTS, M. L. ET AL. Advantage of live attenuated cold-adapted influenza A virus over inactivated vaccine for A/Washington/80 (H3N2) wild-type virus infection. Lancet, 1: 705-708 (1984). 11. BETTS, R. F. ET AL. Resistance to challenge with influenza A/Hong Kong/123/77 (HlNl) wild-type virus induced by live attenuated A/Hong Kong/ 123 /77 (H 1 N 1) cold-adapted reassortant virus. Journal of infectious diseases, 151: 744-745 (1985).