equal groups. Children of the first group received vaccine, whereas those in the other received an indistinguishable

Transcription

1 INFECTION AND IMMNITY, June 1984, p /84/6734-6$2./ Copyright 1984, American Society for Microbiology Vol. 44, No. 3 Recombinant Cold-Adapted Attenuated Influenza A Vaccines for se in Children: Reactogenicity and Antigenic Activity of Cold- Adapted Recombinants and Analysis of Isolates from the Vaccinees G. I. ALEXANDROVA,' F. I. POLEZHAEV,1 G. N. BDILOVSKY,' L. M. GARMASHOVA,' N. A. TOPRIA,1 A. Y. EGOROV,' Y. R. ROMEJKO-GRKO,' T. A. KOVAL,' K. V. LISOVSKAYA,2 A. I. KLIMOV,2 AND Y. Z. GHENDON2* Research Institute for Experimental Medicine, Academy of Medical Sciences of the SSR, Leningrad,1 and Moscow Research Institute for Viral Preparations, Moscow,2 SSR Received 8 February 1983/Accepted 1 February 1984 Reactogenicity and antigenic activity of recombipants obtained by crossing cold-adapted donor of attenuation A/Leningrad/134/47/57 with wild-type influenza virus strains A/Leningrad/322/79(HlN1) and A/ Bangkok/1/79(H3N2) were studied. The recombinants were areactogenic when administered as an intranasal spray to children aged 3 to 15, including those who lacked or had only low titers of pre-existing antihemagglutinin and anti-neuraminidase antibody in their blood. After two administrations of vaccines at a 3- week interval, both strains induced antibody in 75 to 95% of the children. On coinfection of chicken embryos with both recombinants, only weak interference was observed. Administration to children of the bivalent vaccine containing HlNl and H3N2 recombinants induced efficient production of antibody to Hi and H3 hemagglutinins and Ni and N2 neuraminidases without adverse reactions. The recombinants studied were genetically stable as judged by retention of the temperature-sensitive phenotypes and a lack of reversion of the genes carrying temperature-sensitive mutations in all of the reisolates from vaccinated children. In an accompanying paper (5), we presented data on obtaining a cold-adapted donor of attenuation and its coldadapted recombinants and on molecular genetic analysis of these strains. This paper presents data on evaluation of reactogenicity and antigenicity of recombinant vaccines for children aged 3 to 15 and analysis of isolates from the vaccinees. MATERIALS AND METHODS The viruses and preparation of the recombinants have been described previously (5). Vaccine and placebo. Experimental vaccines were prepared with the cold-adapted recombinants in accordance with requirements currently existing in the SSR. Monovalent vaccines contained % egg infective doses (EID5) of infectious recombinant virus per ml. The vaccines were diluted 1:2 before administration. Bivalent vaccine was prepared by mixing two undiluted monovalent vaccines just before administration. Placebo consisted of normal allantoic fluid of uninfected chicken embryos diluted 1:2 before administration. Vaccination of humans. Vaccine was administered in.25- ml volumes (17 EID5 of the virus) intranasally into each nasal turbinate with a sprayer. Revaccination was carried out 21 days later by the same method. Evaluation of reactogenicity of vaccines. Evaluation of reactogenicity of the cold-adapted recombinant vaccines was carried out in stages. First, healthy young adults aged 18 to 25 years were vaccinated. Then, schoolchildren aged 12 to 15 and 7 to 11 years and children 3 to 6 years of age were vaccinated. Children were immunized in organized communities (schools and preschools) in February-March It should be noted that no influenza morbidity was registered during that period in the city where the vaccinations were * Corresponding author. 734 carried out, and no influenza viruses were isolated. Children and adults were carefully examined before vaccination to select healthy individuals, who were then divided into two equal groups. Children of the first group received vaccine, whereas those in the other received an indistinguishable placebo. They were observed for 5 days after the first vaccination for evaluation of their systemic and local reactions and temperature measurement. Evaluation of antigenic activity. Paired sera were taken from vaccinated children before vaccination, 21 days after the first vaccination, and 21 days after the second vaccination to determine anti-hemagglutinin (anti-ha) titers in a hemagglutination test modified by Alexandrova (1). To remove nonspecific inhibitors, sera were heated at 58 C for 3 min; to determine antibody levels to H3 hemagglutinin, a variant of AlBangkok/1/79(H3N2) influenza virus resistant to nonspecific inhibitors of blood serum was used as a test virus. Anti-neuraminidase (anti-na) antibody levels were determined in an elution-inhibition test (2, 1). The latter test was performed with recombinants R-11 (Heql-NA/BrazilIll/ 78) and R-17 (Heql-NA/Bangkok/1/79) as antigens. These recombinants were provided by N. E. Gorev (All-nion Influenza Institute, Leningrad, SSR). It should be noted that the variant of the method used by us for determining anti-na antibody correlated well with a conventional one (1). Recombination test. The recombination test with reisolates fromn vaccinated children has been described previously (4, 6) and in the accompanying paper (5). RESLTS Reactogenicity of monovalent vaccines prepared from recombinants 47/25/1(HlN1) and 47/7/2(H3N2). Only 1 of 122 (.8%) young adults (including 65 persons with anti-ha titers of <16) vaccinated once with the 47/25/1(HlN1) strain developed fever (-37.6, which was transient (Table 1). Administration of the 47/7/2(H3N2) vaccine to 138 young

2 VOL. 44, 1984 LD-ADAPTED INFLENZA A VACCINES FOR SE IN CHILDREN. II 735 TABLE 1. Frequency of febrile reactions in adults and children to administration of the recombinants 47/25/1(HlN1) and 47/7/2(H3N2)' Individuals with anti-ha level of Age of No. (%) with -16 Recombinant vaccinees Prepn No. fever (37.6- No. (%) with (yr) 38.5 No. fever ( /25/1(HlN1) 18-2 Vaccine (.8) 65 () Placebo 13 1 (.8) 7 1 (1.4) 7-15 Vaccine (.7) (1.) Placebo 46 () 183 () 3-6 Vaccine (1.1) (1.5) Placebo (2.5) 47 1 (1.1) 47/7/2(H3N2) 18-2 Vaccine 138 () 82 () Placebo 13 1 (.8) 84 1 (1.2) 7-15 Vaccine 37 1 (.3) (.6) Placebo 46 () 19 () 3-6 Vaccine (1.) (1.8) Placebo (2.5) 35 1 (2.9) a The vaccine (at a dose of 17 EID5 of the virus) or placebo was administered once to adults and children. Temperature was measured for 5 days. adults (including 82 with anti-ha titers of.16) caused no postvaccination fever. Because of these observations with young adults, trials were extended to schoolchildren aged 7 to 15 years, and later to children 3 to 6 years of age. Single administration of the 47/25/1(HlN1) strain to 192 children aged 7 to 15 and to 136 children aged 3 to 6 with low initial anti-ha antibody levels caused transient fever (.37.6 in two persons in each group (1. and 1.5%, respectively). Low reactogenicity was also observed in children vaccinated with the 47/7/2(H3N2) recombinant (Table 1). Reactogenicity after the vaccination was lower when calculated for the total number of persons under observation, irrespective of their initial prevaccination blood levels of anti-ha antibody. Both recombinants had low reactogenicity even for the most susceptible group of children, those who had both preexisting anti-ha and anti-na antibody at titers of s16 in their blood. The frequency of febrile reactions in this group was 1.1% for the 47/25/1(H1N1) recombinant (93 children) and % for the 47/7/2(H3N2) recombinant (77 children). None of the vaccinees developed severe postvaccination reactions with a temperature of >38.5 C or any changes in their general condition, catarrh, or other symptoms characteristic of acute respiratory diseases. TABLE 2. Evaluation of antigenicity of monovalent vaccines prepared from recombinants 47/25/1(HlNl) and 47/7/2(H3N2). In February 1981, we analyzed the susceptibility of children of different age to influenza virus by measuring the prevaccination prevalence of Hi and H3 serum antibody by using A/Brazil/11/78 and A/Bangkok/1/79 as antigens. Low levels of anti-ha antibody (s16) to H3 and Hi were detected in 52.5 and 55% of children aged 11 to 15, in 63.1 and 63.9% of children 7 to 1 years of age, and in 64.5 and 86.% of children aged 3 to 6 years, respectively. These data indicated that there were a considerable number of children potentially susceptible to infection with Hi and H3 influenza virus. Antigenicity of the recombinant vaccines was estimated as a fourfold or greater rise of anti-ha and anti-na antibody titers in sera taken from children after the first or the second vaccinations. Among individuals with low initial antibody levels (s16), 94.1 and 86.1% of children aged 3 to 6 and 7 to 15 vaccinated with the recombinant 47/25/1(HlN1), as well as 8.9 and 74.6% of those vaccinated with the recombinant 47/7/2(H3N2), developed a fourfold or higher rise in titers of anti-ha antibody, respectively (Table 2). A fourfold or higher rise in anti-na antibody was observed in 49.3 to 6.4% of the vaccinees. The high antigenicity of the recombinants tested was confirmed by high geometric mean titers of Antigenic activity of two-dose and one-dose immunizations of recombinant vaccine strainsa No. with % With a No. with % With a Antibody Age of initial fourfold Geometric initial fourfold Vaccination studied vaccinees Vaccine antibody antibody mean anti- antibody antibody (yr) titers of rise body titer titers of rs to 64 rs % Two doses Anti-HA /25/1(HlN1) : n/2(H3N2) : /25/1(HlN1) : /7/2(H3N2) : Anti-NA /25/1(HlNl) : n/2(H3N2) : One dose Anti-HA /25/1(HlNl) : /7/2(H3N2) : a Children vaccinated twice were vaccinated intranasally with 17 EID5 of the virus twice at an interval of 3 weeks. Blood samples were taken before immunization, 3 weeks after the first immunization, and 3 weeks after the second immunization. Seroconversion rates were evaluated in children with low (<16) and moderate (1:32 to 1:64) levels of anti-ha and anti-na antibody. Geometric mean antibody titers and antibody rises were determined in seronegative persons.

3 736 ALEXANDROVA ET AL. INFECT. IMMN. antibody after immunization, reaching 1:42 to 1:6 for anti- HA antibody and 1:28 to 1:42 for anti-na antibody. Among children with prevaccination antibody titers of 32 to 64, a fourfold or greater rise in titers of anti-ha antibody was observed in 12.1 to 36.5% of the vaccinees, and a rise in titers of anti-na antibody was detected in 19. to 31.6% of the vaccinees. We also tested the antigenicity of a single dose of vaccine. The data presented in Table 2 show that the single dose induced a fourfold or greater rise in antibody titers in 61 to 65% of the vaccinees. Geometric mean antibody titers were two times lower after the single dose of vaccine than after two doses (1:27 to 1:36 as compared to 1:42 to 1:6). se of a bivalent vaccine containing recombinants 47/7/2(H3N2) and 47/25/1(HlN1). Because recombinants 47/ 25/1 and 47/7/2 contain five common genes coding for nonglycosylated proteins (5) and differ from each other only in genes coding for hemagglutinin, neuraminidase, and the PB2 protein (gene 1), we supposed that simultaneous infection of these recombinants might be possible with only slight interference. This would make it possible to carry out simultaneous vaccination of children with both strains. To study such a possibility, we first examined reproduction of these recombinants after coinfection of chicken embryos. The result obtained indicated that reproduction of the A/Leningrad/322/79 strain was suppressed sharply during cocultivation of two wild-type strains [A/Leningrad/322/ 79(HlN1) and A/Bangkok/1/79(H3N2)]. Reproduction of the 47/25/1 strain was somewhat decreased during cocultivation of the cold-adapted recombinants 47/25/1(HlN1) and 47/7/2(H3N2), although to a lesser extent than during cocultivation of wild-type HlNl and H3N2 strains. After the experiments in chicken embryos showed only a low level of interference between the recombinants, we evaluated reactogenicity and antigenicity of a bivalent vaccine. This involved simultaneous administration of the recombinants to children 7 to 15 years of age. Reactogenicity of the recombinants administered either simultaneously or separately was equally low (Table 3). Other symptoms of influenza illness were also not produced by any of the vaccines. Studies of antigenicity of the recombinants administered simultaneously showed (Table 4) that the bivalent vaccine is just as potent as each of the recombinants applied separately with respect to the number of individuals with a fourfold and higher antibody rise and the geometric mean antibody titers after vaccination. Application of the bivalent vaccine consisting of recombinants 47/25/1(HlN1) and 47/7/2(N3N2) induced development of both anti-ha and anti-na antibody to both components of the preparation. The presence of seroconversion in individuals with initial antibody titers of 32 to 64 was additional proof for antigenicity of the bivalent vaccine. Studies of genetic stability of the recombinants and isolates. To study the genetic stability of the recombinants, we passaged them five times in chicken embryos at 32 C, after which we compared the temperature-sensitive (ts) phenotype of the passaged and initial viruses. The passaged variants failed to reproduce at 4 C and retained the ts phenotype of the initial recombinants (not shown). Next we studied the properties of isolates obtained from nasal swabs of children 48 h after vaccination with the coldadapted recombinant strains and passaged once in chicken embryos at 32 C. All isolates obtained from 23 children 3 to 15 years of age entirely retained the ts phenotype of the vaccines administered to children (Table 5). On the assumption that the recombinants used may have had ts mutations in several genes so that alterations in one of them during reproduction in the nasopharynx of the vaccinated children might have no effect on a ts phenotype of the reisolated virus, we decided to study the genetic properties of the isolates in a recombination analysis with the reference fowl plague virus ts mutants. All 14 isolates obtained from children 48 h after vaccination with recombinant 47/25/1 and all 9 isolates obtained from children 48 h after vaccination with recombinant 47/7/2 and passaged once in chicken embryos at 32 C behaved in these recombination experiments similarly to the vaccine recombinants administered to the children, i.e., they retained ts mutations in all the genes in which they were present in the initial recombinant strains used as monovalent vaccines (Table 5). DISCSSION Evaluation of reactogenicity of the cold-adapted recombinants showed that they caused transient fever in less than 1% of the vaccinees. No alterations in the general condition TABLE 3. Frequency of febrile reactions in children 3 to 15 years of age to administration of a bivalent vaccine from recombinants 47/25/1 (H1N1) and 47/7/2(H3N2)' Individuals with initial Anti-HA levels of No. with fever. 16 Vaccine Prepn No. 37.6C (%) No. with fever No. 37.6C (%) Bivalent [47/25/1(HlN1) + Vaccine (.9) 161b 2 (1.2) 47/7/2(H3N2)] 122c 2 (1.6) Placebo 72 1 (1.4) 41" 1 (2.4) 32C 1 (3.1) Monovalent Vaccine 96 1 (1.) 65 1 (1.5) [47/25/1(HlN1)] Placebo 3 () 17 () Monovalent Vaccine 89 () 59 () [47/7/2(H3N2)] Placebo 45 () 23 () a Children were vaccinated intranasally with monovalent or bivalent vaccines from the cold-adapted recombinants. When the bivalent vaccine was used, both monovalent vaccines were mixed in equal volumes just before vaccination, and 17 EID5 of each virus was administered intranasally. When a monovalent vaccine was used, the preparation was diluted 1:2 and administered at a dose of 17 EID5. Normal allantoic fluid was used as placebo. Temperature was measured for 5 days after the vaccination. b Individuals seronegative to 47/25/1(HlN1). c Individuals seronegative to 47/7/2(H3N2).

4 VOL. 44, 1984 LD-ADAPTED INFLENZA A VACCINES FOR SE IN CHILDREN TABLE 4. Antigenic activity of the bivalent vaccine from the recombinant vaccine strain on a two-dose vaccination of children of 3 to 15 years of agea Vaccine No. with with ~~~~~~~~~~~~~No. Initial % With a Geometric initial % With a Antibody Antigen intial fourfold mean Antibody rise antibody fourfold Vaccine studied Antigen antibody antibody antibody (no.)b titers of antibody ts6 rise titer 1:32 to rise -16 ~~~~~~~~~~~~~~~~~1:64 47/25/1(HlNl) + Anti-HA HlNl : /7/2(H3N2) H3N : Anti-NA HlNl : H3N : /25/1(HlN1) Anti-HA HlNl : Anti-NA : /7/2(H3N2) Anti-HA H3N : Anti-NA : a Children aged 3 to 17 were vaccinated twice intranasally with either monovalent vaccine from each recombinant or with bivalent vaccine containing 17 EID5 of each virus at an interval of 3 weeks. Three weeks after the second immunization, blood samples were taken, and anti- HA and anti-na antibody titers were determined. b Geometric mean antibody titer after immunization/geometric mean antibody titer before immunization. of the vaccinated children or any symptoms characteristic of acute respiratory diseases were observed. Cold-adapted recombinant vaccines obtained by using a cold-adapted A/Ann Arbor/6/6 strain as a donor of attenuation and which inherited five or six genes coding for nonglycosylated proteins from the cold-adapted donor also proved to be poorly reactogenic for adults with low initial antibody levels (3, 7-9), although they caused systemic afebrile reactions and catarrh symptoms in some volunteers (9). As judged by measurement of hemagglutinin and neuraminidase antibody response, the cold-adapted recombinant strains obtained by us had an adequate immunogenicity, even for children with low initial antibody titers. A two-dose immunization induced a fourfold or greater antibody rise in 75. to 95.% of the vaccinees; even after a one-dose immunization, a fourfold antibody rise was observed in over 6% of the vaccinated children. In addition, a fourfold or higher rise in anti-na antibody was observed in 5 to 7% of the vaccinees. A good antigenic response was achieved in recent investigations by Wright et al. (11) involving vaccination of 21 seronegative children with cold-adapted recombinant strains derived from the cold-adapted donor A/Ann Arbor/6/6. A fourfold increase in titers of anti-ha antibody was observed in the majority of children even after the single dose of vaccine. It should be emphasized that wild-type H3N2 and HlNl influenza virus strains used by us to obtain recombinants had had only a few passages in chicken embryos at 32 C after isolation. In addition, special variants which grew well at 4 C were selected and used in recombination experiments. Wide experience gained in the SSR in obtaining attenuated influenza virus strains used as live vaccines by passages in chicken embryos at the optimal temperature has shown that not less than 2 passages are required to obtain variants attenuated for humans. Lately, trials in volunteers of potentially reactogenic influenza virus strains have been prohibited in the SSR. Because of this we had no chance to evaluate reactogenicity for humans of the epidemic strains A/Leningrad/322/79(HlN1) and A/Bangkok/1/79(H3N2), which had had not more than six to eight passages in chicken embryos at optimal temperatures and might have retained the reactogenicity. Nevertheless we cannot rule out the possibility that the high degree of attenuation of the recombinants obtained may be due not only to the genes inherited from cold-adapted donor but also to ones derived from wildtype influenza virus. But it should be noted that in previous work (4) the transfer of genes coding for hemagglutinin and neuraminidase from the epidemic influenza virus strains whose reactogenicity was confirmed in volunteer trials did not change the degree of attenuation of recombinants which had inherited other genes from the cold-adapted parent. The recombinants obtained by us were not overattenuated and their antigenic activity is associated with the ability to reproduce in the vaccinee organism. This is indicated by the ability to isolate viruses possessing properties of vaccine strains from nasal swabs of vaccinated children. In addition, it is very unlikely that intranasal administration of a very small virus dose (ca. 3 CCA) might have induced such a marked antigenic response without virus reproduction, which we observed in our experiments. According to data in the literature, inactivated influenza vaccine intranasally administered to children, even at a dose of 4 to 8 CCA, induced a very moderate antigenic response (12). We have also shown that it is possible to simultaneously use two recombinant strains in a bivalent vaccine of HlNl and H3N2 serotype, and in this instance the antigenic response of vaccinated children proved to be similar to that after separate administration of each preparation. Lack or drastic decrease in interference between the recombinants under study was confirmed by coinfection of chicken embryos. Apparently, recombinant strains derived from the same donor of attenuation and for which the majority of genes coding for nonglycosylated proteins are identical are capable of reproducing in the same cell, having no interfering effect upon each other. However, one cannot exclude a possibility that this may be true only for this particular pair of recombinants. A problem of reversion and restoration of virulence of a vaccine strain is one of the important problems in using attenuated live influenza vaccines. Because of this, we paid particular attention in our experiments to the genetic stability of our cold-adapted recombinants during reproduction in children. Studies of 23 isolates showed that they did not change their ts characteristics as judged by analysis of replication at different temperatures or by recombination analysis with reference ts mutants. It must be noted that we studied the viruses isolated 48 h after vaccination. To date we have failed to isolate viruses later than 48 h. At present, investigations in this direction are in progress.

5 738 ALEXANDROVA ET AL. INFECT. IMMN. cl O z E..qe ON N- ON1 ON all.- N 'C Nx IC -N oc co z -o Cl N._ *) ' ) - ' (4- ) m r.. E r- A c._ to CZ. D p E N m c > ) CZ Cl NNNNNNNNN en ON ONal ON (ON 'n X en Xt 1n X Cl CfX... NNNNNN NNNNN=NNNNNNN Cl e~n ) r ON enrncc cl CN ON W) Ns N N N N N N N N N nqon e eq PON=m r- 'IC I ~CD r N- (ON ON _~ 'O n en O x~~ ^, ro o~~~ se ^\ bc \ X r O ioun Cl OCD Cl~ Cl r- Cl6 'i -:16~ V VVV ~~in QON NuN ON aon Nr.. a-.ic.nq NN b\s9#e b b oooocloooonooo x ~ m N N m ooon NV eq oo obb ro Or_ ooe o "IToo ot ON cl N ClClC C Cl! I"ClClCl N.. N. N. N.q N. oa. N. N. N. N 'i eno r6 r' N N N N N NE N N N N C~..l el..l..~.. C-:r-l o6n-4 N x r- Cl Cl Cl Cl Cl Cl Cl Cl Cl Cl NNNTh) ) n-4)r-- e N N N N N N N m o m 4-4 m o-4 V z Cl "it "it ' t ) w Cc L) I. c c ).toq.2 Cd D ) ^in 6 Z1 6 -r-:r- NNNNNNr-~~~~~~~~~~N- u Cd ), (4- ) -) E-

6 VOL. 44, 1984 LD-ADAPTED INFLENZA A VACCINES FOR SE IN CHILDREN. II 739 Recently, Wright et al. (11) carried out investigations involving vaccination of children with recombinant coldadapted vaccines obtained by recombination of the coldadapted donor A/Ann Arbor/6/6 with wild-type viruses A/Alaska/6/77(H3N2) and A/Hong Kong/123/77(H1N1). Analysis of isolates from the vaccinees has shown that the virus contained in swabs retained a ts phenotype; at the same time, efficiency of plaquing at 39 C increased after one passage of the isolates in MDCK cells at 32 C, indicating a certain genetic instability of the vaccines under study. It should be emphasized that in our experiments we examined isolates which had had one passage in chicken embryos at 32 C after isolation; however, all of them proved to be entirely stable. Attenuation of cold-adapted donors which had had multiple passages in chicken embryos might be due to ts, coldadapted, host-range, or other types of mutations. Possibly, multiple passages at a lowered temperature, as used to prepare the cold-adapted donor strain, might result in appearance of not only a single but several mutations in the same gene, and reversion (or suppression) of a single mutation would have no effect on the phenotype of the gene as a whole. The data presented in this report and in an accompanying paper (5) indicate that the cold-adapted ts A/Leningrad/134/ 47/57(H2N2) strain was a satisfactory donor of attenuation to obtain recombinant live vaccine strains for immunization of children. It should be noted in conclusion that in autumn of 1983 we vaccinated nearly 2, children 3 to 15 years of age with the cold-adapted recombinants 47/7/2(H3N2) and 47/25/1 (HlNl) and observed no adverse reactions, although the epidemiological activity was high (manuscript in preparation). ACKNOWLEDGMENTS We thank A. Kendal for helpful criticism during the preparation of manuscript, as well as V. Kandaurov and M. Botin for assistance in vaccination. LITERATRE CITED 1. Alexandrova, G. I Analysis of the structure of inhibitorsensitive and inhibitor-resistant influenza virus strains of A2 type. Acta Virol. 6: Appleyard, G., and Y. Oram The assay of influenza antineuraminidase activity by an elution inhibition technique. J. Gen. Virol. 34: Cate, T. R. and R. B. Couch Live influenza A/Victoria/ 75(H3N2) virus vaccines: reactogenicity, immunogenicity, and protection against wild-type virus challenge. Infect. Immun. 38: Ghendon, Y. Z., A. I. Klimov, G. I. Alexandrova, and F. I. Polezhaev Analysis of genome composition and reactogenicity of recombinants of cold-adapted and virulent virus strains. J. Gen. Virol. 53: Ghendon, Y. Z., F. I. Polezhaev, K. V. Lisovskaya, T. E. Medvedeva, G. I. Alexandrova, and A. I. Klimov Recombinant cold-adapted attenuated influenza A vaccines for use in children: molecular genetic analysis of the cold-adapted donor and recombinants. Infect. Immun. 44: Ghenkina, D. B., and Y. Z. Ghendon Recombination and complementation between orthomyxoviruses under conditions of abortive infection. Acta Virol. 29: Moritz, A. J., C. Kunk, H. Hofman, E. Liehl, P. Reeve, and H. F. Maassab Studies with a cold-recombinant A/Victoria/3n5(H3N2) virus. II. Evaluation in adult volunteers. J. Infect. Dis. 142: Murphy, B. R., R. Chanock, M. L. Clements, W. C. Anthony, A. T. Sear, L. A. Cisneros, M. B. Reunels, E. H. Miller, R. E. Black, M. N. Levine, R. F. Botts, K. G. Douglas, H. F. Maassab, N. J. Cox, and A. P. Kendal Evaluation of A/AlaskaI6/ 77(H3N2) cold-adapted recombinant viruses derived from A/Ann Arbor/6/6 cold-adapted donor virus in adult seronegative volunteers, Infect. Immun. 32: Reeve, P., B. Gerendas, A. Moritz, B. Liehl, C. Kunz, H. Hofmann, and H. F. Maassab Studies in man with coldrecombinant influenza virus (HlNl) live vaccines. J. Med. Virol. 6: Topuriya, N. V Application of a modified technique of inhibition of elution of erythrocytes for detecting anti-neuraminidase antibody in influenza infection. Izv. Akad. Nauk Gruz. SSR Ser. Biol. 5: (In Russian.) 11. Wright, P. F., N. Okabe, K. T. McKee, H. F. Maassab, and D. T. Karson Cold-adapted recombinant influenze A virus vaccines in seronegative young children. J. Infect. Dis. 166: Yamison, R., B. Hilman, and D. Kelly Immunization of children with chronic lung diseases with inactivated bivalent influenza vaccines. South. Med. J. 65: