Influenza is an acute viral respiratory illness that causes annual epidemics resulting in substantial morbidity in all age groups, accounting for up to 25% of all respiratory illnesses in preschool populations during epidemic periods. 1 In these young children seasonal rates of influenza-associated hospitalization are 4- to 20-fold higher than those seen among children >5 years of age. 2, 3 Although the currently available inactivated influenza vaccine is recommended for high risk adult and pediatric populations, its use in these groups is disappointing in part because of the variable age-dependent protective efficacy and poor acceptance of annual administration by injection. 4 This is especially true for children <9 years of age who require two injections for primary immunization. Poor vaccine uptake among young children may also impact on community prevention programs because children enhance ongoing viral transmission in communities.
Live influenza vaccines have been developed by cocultivation of “cold-adapted” viral strains with wild-type strains to create attenuated reassortant vaccine for intranasal administration. These vaccines contain six attenuating genes of the cold-adapted strain, each gene coding for neuraminidase and hemagglutinin surface proteins from wild-type strains. 5, 6 They are administered intranasally and result in local replication of influenza virus, simulating natural infection. Such vaccines provide an advantage over inactivated vaccines by virtue of (1) ease of administration, (2) induction of mucosal as well as systemic immune responses 7 and (3) enhanced acceptance in children.
Trivalent formulations of intranasally administered, cold-adapted influenza vaccine (CAIV) have been shown to be safe and immunogenic among young children and adults. 8–10 Recently CAIV was shown to be efficacious in preventing culture-confirmed influenza among toddlers. 11, 12 We conducted a prospective, randomized, placebo-controlled study to directly compare, in the same population, the safety and immunogenicity of the CAIV lot used in that efficacy trial with 3 new consistency lots.
We enrolled 500 healthy children 12 to 36 months old from the Kaiser Permanente, Southern California Region outpatient pediatric clinic population. Subjects were randomly assigned to 1 of 5 study groups and received 1 of the following: 1 of the 3 consistency lots of CAIV (Group 1, 2, 3); a lot of CAIV used in a completed efficacy trial 11 (Group 4); or placebo (Group 5). Randomization lists were computer-generated and parents/guardians, subjects and investigators were blinded to the treatment group to which participants were assigned.
Children were not enrolled in this trial if any of the following were present: (1) history of hypersensitivity to eggs or egg protein; (2) significant underlying chronic illness for which the inactivated influenza vaccines are recommended; (3) immunodeficiency disease or immunosuppressive therapy in the participant or household member; (4) acute febrile illness within 7 days or upper respiratory illness within 3 days of vaccination; (5) prior receipt of CAIV or inactivated influenza vaccine; (6) administration of an investigational drug within 1 month of vaccination in this study; (7) administration of any live virus vaccine within 1 month, or expected to receive another live virus vaccine within 1 month of vaccination in this study; (8) administration of any inactivated vaccine within 2 weeks, or expected to receive another inactivated vaccine within 2 weeks of vaccination in this study; (9) history of wheezing or bronchodilator medication use within 2 weeks before vaccination; (10) receipt of any blood product within 3 months before vaccination or expected receipt within the study duration; (11) expected administration of any nasal medications during the first 10 days after vaccination; (12) no telephone in the household.
Each child received two doses of intranasal study vaccine, ∼60 days apart. Before vaccination the medical history, review of systems and a brief physical examination were performed. After vaccination all subjects were observed at the clinical study site for at least 15 min. Families were given a digital thermometer and asked to record on a diary card the temperature of the subject and the occurrence of any symptoms listed on the diary card (including fever, lethargy, irritability, runny nose/nasal congestion, sore throat, cough, headache, muscle aches, chills, vomiting), beginning on the evening of the day of each vaccination and for 10 days thereafter. All other adverse events and medications taken were also recorded on the diary card. Study personnel called the home on Day 1 and Day 10 after vaccination to encourage compliance. At each scheduled visit the research nurse interviewed the parents and reviewed the child’s medical record for information about interval illnesses, clinic visits and hospitalizations. A clinical safety monitor (a pediatric infectious disease specialist) had independent access to the randomization code and continuously monitored vaccine safety during the study.
Four lots of intranasal trivalent CAIV were used. The strains selected for the three consistency lots were those recommended by the Food and Drug Administration for the licensed influenza vaccine for that season (1997 to 1998). The efficacy lot contained strains used in the licensed vaccine during the season in which the efficacy trial was conducted (1996 to 1997). Groups 1, 2 and 3 received vaccine containing ∼107.0 50% tissue culture-infective doses of each A/Shenzhen/227/95 (H1N1)-, A/Wuhan/359/95(H3N2)- and B/Harbin/7/94-like viral strains. Group 4 received the same lot used in a multicenter efficacy trial 11 containing ∼106.7 50% tissue culture-infective doses of A/Texas/36/91-like (H1N1), A/Wuhan/359/95(H3N2)-like and B/Harbin/7/94-like viral strains. CAIV is delivered in egg allantoic fluid containing sucrose-phosphate-glutamate. Each dose is 0.5 ml by intranasal spray (0.25 ml/nostril). Group 5 received placebo as a 0.5-ml intranasal spray (0.25 ml/nostril) of egg allantoic fluid containing sucrose-phosphate-glutamate. Study vaccines were given with the child sitting on an adult’s lap, with the head stabilized, and kept in an upright position for at least 30 s after vaccine administration.
Blood samples were obtained just before the first dose and ∼4 weeks after the second dose of study vaccine. For ∼20 participants in each group an additional sample was obtained just before the second dose. All sera were stored at −20°C before testing. Sera were tested in a blinded manner for strain-specific serum hemagglutination inhibition (HAI) antibody titers against type A (H3N2 or H1N1) or type B as appropriate. Sera from each participant were tested simultaneously.
For each consistency lot group and for all consistency lot groups combined, the distribution of demographic characteristics, frequency of adverse events, post-Dose 1 and post-Dose 2 strain-specific HAI geometric mean titer (GMT) and seroconversion rates, with their 95% confidence intervals, were determined. Comparisons were made to the efficacy lot and placebo groups using analysis of variance, Fisher’s tests or chi square tests, with a significance level of 0.05 with adjustment for multiple comparisons. Primary immunogenicity endpoints included post-Dose 2 strain-specific GMT regardless of baseline serostatus and post-Dose 2 strain-specific seroconversion. Seronegative was defined as having a serum HAI titer of ≤1/4. Seroconversion was defined as ≥4-fold rise in antibody titer.
The study was designed to have at least 95% probability to detect lot-to-lot differences if, in fact, the lots differ by at least 4-fold in the fold change for the GMT. Conversely there is high power to rule out 4-fold or greater differences in GMT fold changes if in fact the lots are consistent. Comparisons of GMTs between lots were done by constructing 95% confidence intervals for the ratio of GMT fold change (post-vaccination/baseline) between the lots. Confidence intervals were constructed with the standard error from the analysis of variance model. The post-Dose 2 GMTs were expected to have 95% confidence ranges for all strains not exceeding ±25% of the GMT for single groups and not exceeding ±13% for the combined three consistency lots. The study was expected to provide 95% assurance that true seroconversion rates for all strains and all lots are within ±12% of measured rates.
Informed consent was obtained from the families of all children who participated in this study. We performed this study in concordance with US Department of Health and Human Services Guidelines and that of our institutions for conduct of clinical trials.
Five hundred children were enrolled in the study, and 474 received 2 doses of study vaccine. The ethnic, racial, age and gender distributions of children in each group were not significantly different. Overall 53% were male and ∼48% were Caucasian, 35% Latino and 12% black, representing the approximate distributions found in the Southern California patient population from which these children were enrolled.
Each consistency lot (Groups 1, 2 and 3) and the efficacy lot (Group 4) of CAIV was generally safe and well-tolerated. There were no significant group differences between consistency lots in the proportion of children with fever or local or systemic reactions after vaccination (Table 1). Runny nose/nasal congestion was the most commonly reported adverse event. After the first dose children who received any individual CAIV lot were significantly more likely to report runny nose/nasal congestion than those who received placebo (nasal congestion, 63 to 68% CAIV lots vs. 49% in placebo, P < 0.05). Children had nasal congestion for a median of 3.5 days. After the second dose the general likelihood of any reactogenicity event was diminished compared with the first dose, and the group differences in the proportion of children with nasal congestion were no longer statistically significant. Consistency lot recipients were also significantly more likely than placebo recipients to report headache after the first dose and chills after the second dose, but the percentage of children with these symptoms was low (6 to 11% CAIV lots vs. 2% placebo, 2 to 8% CAIV lots vs. 0% placebo, respectively). No serious vaccine-related adverse events occurred in any child who received CAIV.
Tables 2 and 3 list strain-specific GMTs and fold changes after each dose for each study group. Of the three strains (A/H3N2, A/H1N1 and B) post-Dose 2 GMTs were generally highest for H3N2 strains and lowest for H1N1 strains (Table 2). H3N2 strains also had the most robust fold change responses (Table 3). Groups 1, 2 and 3 (consistency lots) were a priori considered equivalent if, when comparing lot-to-lot fold change ratios after the second dose (e.g. lot1/lot2), all 95% confidence intervals fell within the interval bounded by 0.25 and 4.0. For all viral strains individual lot-to-lot ratios ranged from 0.70 to 2.12, and the extreme limits of the confidence intervals ranged from 0.48 to 3.11; thus the immunogenicity of the consistency lots were not statistically different.
For the A/H3N2 and B strains Groups 1, 2 and 3 had similar distributions of antibody level after the second dose without significant difference between lots. Ninety-seven and 84% had HAI titers ≥1/32 against A/H3N2 and B strains, respectively. In Group 4 (efficacy trial lot) 100% of children had HAI titers ≥1/32 against A/H3N2. For the B strain 72% children in Group 4 had HAI titers ≥1/32 (P < 0.02 compared with the B strain in the consistency lots combined).
The response to the H1N1 strain was less robust and the consistency lots performed better than the efficacy vaccine; 62% of the children who received a consistency lot vaccine had post-Dose 2 antibody titers ≥1/32 compared with 27% of the efficacy lot group (P < 0.01). Although the primary immunogenicity evaluations showed similar immune responses between lots, the post-Dose 2 H1N1 GMT and fold change responses were slightly less in Group 3 than those of Groups 1 and 2 (Tables 2, 3 and 4 ). All, however, were significantly greater than that seen with placebo and with Group 4 (efficacy trial lot).
Generally after two doses seroconversion rates in Groups 1, 2 and 3 were high, ranging from 64 to 88% in all children and 79 to 100% among those initially seronegative (Table 4). We evaluated the impact of only 1 dose of vaccine in a subset of 18 to 20 children in each group. A second dose of vaccine did not elicit significantly higher GMTs or seroconversion rates for the A/H3N2 strain for any CAIV lot (Tables 2 and 4). A second dose for the H1N1 strain did result in a significantly higher GMT compared with the first dose (P < 0.02) and a higher seroconversion proportion (51% to 77%, P < 0.02). For the B strain in the consistency lots only, GMT was significantly increased after a second dose (P < 0.001).
There were differences in immunogenicity noted between those children seronegative at enrollment compared with the group as whole. At enrollment ∼85% of all study children were seronegative to the H1N1 strain, ∼60% to the H3N2 strain and 50% to the B strain. Seronegative children had generally lower post-Dose 2 GMTs but higher levels of seroconversion and higher fold changes in antibody titers after the second dose (Tables 2, 3 and 4). This effect was most pronounced for the H3N2 and B strains. Too few seronegative children were included in the post-Dose 1 cohort to make meaningful comparisons of GMT or seroconversion to the whole cohort as a group.
Because different strains of A/H1N1 were used in Groups 1, 2 and 3 (A/Shenzhen) compared with the Group 4 efficacy lot (A/Texas), we evaluated immune responses using the appropriate homologous and heterologous H1N1 antigen for each lot. The A/Shenzhen-containing H1N1 vaccine was generally more immunogenic and induced a greater degree of cross-reactivity with a heterologous strain than the A/Texas H1N1-containing vaccine. That is, among children vaccinated with A/Shenzhen H1N1, 80% seroconverted after two doses using the Shenzhen antigen compared with 49% using the A/Texas antigen. Conversely among those who received the A/Texas H1N1 vaccine, 37% seroconverted after two doses of the A/Texas antigen compared with 15% receiving the Shenzhen antigen.
Age at initial vaccination, time between vaccinations (28 to 41 days vs. 42 to 60 days; data not shown), and gender did not substantially affect immune responses.
We prospectively evaluated in young children the safety and immunogenicity of three consistency lots of CAIV compared with a lot previously reported to provide substantial clinical protection in an efficacy trial 11 and to placebo. We found that for each of the three vaccine strains (A/H3N2, A/H1N1 and B), all three CAIV consistency lots showed very similar safety and immunogenicity profiles. Overall CAIV was safe, well-tolerated and free of serious adverse events. The immune responses of the consistency lots were robust and comparable with or better than the lot used in an earlier efficacy trial.
By all measures of immune response (serum HAI antibody GMT, post-Dose 2 fold change and seroconversion), the consistency lots performed well individually and as a group were as good as or better than the efficacy trial lot. After two doses of a CAIV consistency lot, no fewer than 79% of initially seronegative children seroconverted, regardless of virus strain, with robust GMT responses. It is generally believed the serum HAI antibody titer is correlated with clinical protection, with antibody GMTs similar to values noted in this study. 13 Direct and reliable extrapolation of our data as a predictor of clinical protection is not possible, however, because other measures of immunity may play a role, including mucosal IgA 7, 14 or other uncharacterized cell-mediated immune mechanisms. In addition the dynamics of protection may vary with the type of vaccine administered (inactivated vs. live 14) and virus subtype. 15 We believe, however, that these CAIV lots would be clinically protective given the magnitude of the immune responses compared with the efficacy trial lot which has a known high level of protective efficacy. 11
Although immune responses between lots were similar, we noted certain trends regarding the H3N2 immune response. Post-Dose 2 GMTs to H3N2 were generally more robust than H1N1 and B antibody titers in all CAIV groups. This phenomenon has been seen with some formulations of CAIV, 16, 17 whereas H1N1 responses were superior to H3N2 with others. 18, 19 Differential influenza type A subtype responses may result from varied surface protein expression, preexisting antibody titers or other biologic characteristics of the vaccine. It is unlikely that concurrent, undetected wild-type influenza infections contributed to measured immune responses because this study was performed in the summer months during which ongoing surveillance for influenza infections did not reveal cases in our study population from which these children were enrolled. We also found that for each consistency lot, the H3N2 antibody levels did not appreciably change after a second dose of CAIV. The efficacy trial showed that both the one dose and two dose regimens of CAIV were efficacious against naturally acquired H3N2 and B disease. 11 The true impact of the magnitude of strain-specific antibody titers and the number of doses required for protection will require prospective evaluations during annual influenza epidemics.
The efficacy lot (which contained A/Wuhan/H3N2) used in this study was previously shown to cross-react with and provide significant clinical protection against a drifted strain (A/Sydney/H3N2). 12 Immunologic cross-reactivity, however, is unpredictable and has been well-documented. 20 In this study the CAIV consistency lots included a different H1N1 strain than the efficacy lot, and we were able to test for heterologous immunity. Children who received the A/Shenzhen vaccine strain displayed higher levels of cross-reactivity to the A/Texas strain than vice versa. In both comparisons, however, cross-reactivity was noted in <50% of children.
We found that all lots of CAIV were well-tolerated and easy to administer. As noted in other studies 11, 18 the most common side effect was runny nose/nasal congestion of short duration and not associated with other systemic complaints. Because it is administered intranasally, CAIV vaccination is free of local reactions associated with intramuscular injection of the inactivated influenza vaccine. In addition intranasal administration eliminates the problem of repeated intramuscular dosing necessary for children <9 years of age receiving influenza vaccine for the first time. The safety profile of this vaccine and its ease of administration will likely enhance its acceptability to health care providers, children and their families.
Influenza remains an important cause of morbidity in young children and vaccination remains the most effective and rapid means for its prevention. Several studies have shown that CAIV is safe, immunogenic and efficacious. In addition its favorable cost effectiveness compared with other vaccines routinely used in children has been noted. 21 Our study has shown that there are no important safety or immunologic differences among three consistency lots of CAIV. Further studies should focus on evaluation of this vaccine among those at particular risk of influenza-associated morbidity such as older adults and those with chronic high risk medical conditions, in all age groups.
PM and IC are employees of Aviron. T his study was funded by Aviron.
We thank members of the UCLA Center for Vaccine Research including Mellie Badar, Cathy Martinez, Pat Wilson, Jane Solomon, Elvira Soto, Carol Farnworth, Kim McMullin, Gloria Leon, Charlotte Gardea, Jeffree Sirowy, Candy Byers, Pam Hastings, Cathy Skokan, Pat Chatfield, Pam Johnson and Deanna Bundy without whom this study would not have been completed.
1. Glezen WP. Serious morbidity and mortality associated with influenza
epidemics. J Epidemiol 1982; 4: 25–44.
2. Neuzil KM, Mellen BG, Wright PF, Mitchel EF, Griffen MR. The effect of influenza
on hospitalization, outpatient visits, and courses of antibiotic in children. N Engl J Med 2000; 342: 225–31.
3. Izurieta HS, Thompson WW, Kramarz P, et al. Influenza
and the rates of hospitalization for respiratory disease among infants and young children. N Engl J Med 2000; 342: 232–9.
4. CDC. Prevention and control of influenza
: recommendations of the Advisory Committee on Immunization
Practices (ACIP). MMWR 2001; 50. (RR-4)
5. Maassab HF. Biologic and immunologic characteristics of cold-adapted influenza
virus. J Immunol 1969; 102: 728–32.
6. Cox NJ, Maassab HF, Kendal AP. 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 degrees between recent H23N2 and H1N1 epidemic strains and cold-adapted A/Ann Arbor/6/60. Virology 1979; 97: 190–4.
7. Boyce TG, Gruber WC, Coleman-Dockery SD, et al. Mucosal immune response to trivalent live attenuated intranasal influenza
vaccine in children. Vaccine 1999; 18: 82–8.
8. Swierkosz EM, Newman FK, Anderson EL, Nugent SL, Mills GB, Belshe RB. Multidose, live attenuated, cold-recombinant, trivalent influenza
vaccine in infants and young children. J Infect Dis 1994; 169: 1121–4.
9. Atmar RL, Keitel WA, Cate TR, Quarles JM, Couch RB. Comparison of trivalent cold-adapted recombinant (CR) influenza
virus vaccine with monovalent CR vaccines in healthy unselected adults. J Infect Dis 1995; 172: 253–7.
10. Gruber WC, Kirschner K, Tollefson S, et al. Comparison of monovalent and trivalent live attenuated influenza
vaccines in young children. J Infect Dis 1993; 168: 53–60.
11. Belshe RB, Mendelman P, Treanor J, et al. The efficacy of live attenuated, cold-adapted, trivalent, intranasal influenza
virus vaccine in children. N Engl J Med 1998; 338: 1408–12.
12. Belshe RB, Gruber WC, Mendelman P, et al. Efficacy of vaccination with live attenuated, cold-adapted, trivalent, intranasal influenza
virus vaccine against a variant (A/Sydney) not contained in the vaccine. J Pediatr 2000; 136: 168–75.
13. Belshe RB, Gruber WC, Mendelman P, et al. Correlates of immune protection induced by live, attenuated, cold-adapted, trivalent, intranasal influenza
virus vaccine. J Infect Dis 2000; 181: 1133–7.
14. Clements ML, Betts RF, Tierney EL, Murphy BR. Serum and nasal wash antibodies associated with resistance to experimental challenge with influenza
A wild-type virus. J Clin Microbiol 1986; 24: 157–60.
15. Fox JP, Cooney MK, Hall CE, Foy HM. Influenzavirus infections in Seattle families, 1975–1979. Am J Epidemiol 1982; 116: 228–42.
16. Gruber WC, Belshe RB, King JC, et al. Evaluation of live attenuated influenza
vaccine in children 6–18 months of age: safety
, and efficacy. J Infect Dis 1996; 173: 1313–9.
17. Swierkosz EM, Newman FK, Anderson EL, et al. Multidose, live attenuated, cold-recombinant, trivalent influenza
vaccine in infants and young children. J Infect Dis 1994; 169: 1121–4.
18. Edwards KM, DuPont WD, Westrich MK, Plummer WD, Palmer PS, Wright PF. A randomized controlled trial of cold-adapted and inactivated vaccines for the prevention of influenza
A disease. J Infect Dis 1994; 169: 68–76.
19. Sears SD, Clements ML, Betts RF, et al. Comparison of live, attenuated H1N1 and H3N2 cold-adapted and avian-human influenza
A reassortant viruses and inactivated virus vaccine in adults. J Infect Dis 1988; 158: 1209–19.
20. Kilbourne ED, Arden NH. Inactivated influenza
vaccines. In: Plotkin SA, Orenstein WA, eds. Vaccines. 3rd ed. Philadelphia: Saunders, 1999: 531–51.7
21. Luce BR, Zangwill KM, Palmer CS, et al. Cost-effectiveness analysis of an intranasal influenza
vaccine for the prevention of influenza
in healthy children. Pediatrics (in press).