Heterosubtypic immunity, or immunity induced against one influenza A subtype that is protective against other influenza A subtypes, is of great interest in the study of influenza. Theoretically, heterosubtypic immunity has the potential to provide an individual with protection against future pandemic influenza strains and could be the basis for broadly protective vaccines. In animals, heterosubtypic immunity has been demonstrated after infection with wild-type virus and administration of some types of influenza vaccines.1 Recent studies in mice, that have demonstrated heterosubtypic immunity, have led some to question the value of vaccinating children against influenza on an annual basis.2
Although multiple studies have described heterosubtypic immunity in animal models,1 there are limited data regarding protective heterosubtypic immunity in human beings. Two studies suggested that during the 1957 A/H2N2 pandemic, adults infected with A/H1N1 in prior years appeared to be protected against A/H2N23,4; however, this protective effect was not observed in children.3 In 1977–1978 when epidemics of A/H3N2 and A/H1N1 occurred concurrently or within a few months, Japanese school children with recent H3N2 infection appeared to be protected against subsequent H1N1 infection.5 Lastly, in pediatric clinical studies of cold-adapted vaccine viruses (N = 77), pre-existing immunity against 1 subtype (H1N1 or H3N2) had no demonstrable effect on the replication of or immune responses to viruses of the opposite subtype.6
There is a significant need for additional data regarding heterosubtypic immunity in human beings, particularly in children. With the cocirculation of 2 influenza A subtypes in recent years, data from multiyear placebo-controlled studies of influenza vaccines can provide an additional opportunity to look for evidence of heterosubtypic protection.
Four multiyear, culture-confirmed efficacy studies that compared live attenuated influenza vaccine with placebo in children 6 to 71 months of age were identified for analysis.7–10 All studies were conducted in children without a history of previous influenza vaccination at the time of enrollment. Additionally, all studies used consistent criteria for obtaining a nasal swab for viral culture. Starting from 11 days after vaccination, cultures were collected from any child having: (1) at least one of the following symptoms: fever, wheezing, shortness of breath, pulmonary congestion, pneumonia, or acute otitis media (suspected or diagnosed), or (2) at least 2 of the following symptoms concurrently: rhinorrhea, sore throat, cough, muscle aches, chills, headache, irritability, decreased activity, or vomiting. In all studies, samples were collected within 4 days of symptom onset; for each study, 64% to 81% and 80% to 89% of samples were collected within 2 and 3 days of symptom onset, respectively. For all studies, central laboratories evaluated nasal swabs for the presence of influenza virus and for subtype and serotype identification by antigenic methods.
For each study season, the dominant circulating types/subtypes were identified. For each study and for each dominant type/subtype, subjects who received placebo in both years of the study were separated into 2 cohorts based on whether the subject experienced culture-confirmed influenza illness in year 1. The first cohort consisted of subjects with confirmed illness due to the specific subtype, and the second cohort consisted of subjects with no evidence of any influenza illness caused by any type/subtype. Pooled analyses were conducted to examine the cumulative data across all studies. For each cohort as well as the pooled analyses, the incidence of influenza illness by subtype during year 2 was calculated and compared. Relative risk (RR) and 95% confidence intervals (CIs) were calculated.
Three studies permitted an analysis of the protective effect of influenza A (H1N1 or H3N2) illness in year 1 against influenza A illness of the alternate subtype in year 2 (heterosubtypic protection). There was no evidence of heterosubtypic protection in any study comparison or in the pooled analysis; in the pooled analysis, 12.7% of subjects with influenza A in year 1 versus 13.1% of those without influenza A in year 1 developed alternate subtype illness in year 2, for an RR of 0.97 (95% CI: 0.60, 1.52). In 1 study comparison, a statistically significant increased risk of subsequent heterosubtypic illness was observed (RR, 3.24 [95% CI: 1.33, 7.73]). Two studies permitted an analysis of the protective effect of influenza A/H1N1 illness in year 1 against influenza B illness in year 2 (heterotypic protection). There was no evidence of heterotypic protection in any of the study comparisons or the pooled analysis (RR, 1.5 [95% CI: 0.66, 3.27]). Two studies permitted an analysis of the protective effect of influenza A illness in year 1 against influenza A illness of the same subtype in year 2 (homosubtypic protection). In the first study, the A/H1N1 strains that circulated in years 1 and 2 were characterized as A/New Caledonia/20/99-like.8 In the second study, the A/H3N2 strains that circulated in year 1 were characterized as A/Wuhan/395/95-like; however, in year 2, antigenically distinct A/Sydney/05/97-like strains predominated.9 Results from both the study comparisons were consistent with homosubtypic protection (RR, 0.0 [95% CI: 0.00, 9.66] and RR, 0.13 [95% CI: 0.01, 0.72], respectively). The pooled analysis demonstrated an RR of 0.15 (95% CI: 0.01, 0.82). All results for comparisons evaluating heterosubtypic, heterotypic, and homosubtypic immunity are included in Table, Supplemental Digital Content 1, http://links.lww.com/INF/A871.
In this analysis of the 2-year incidence of culture-confirmed influenza illness in unvaccinated young children, there was no evidence of heterosubtypic protection. The single observation of an increased risk of heterosubtypic illness in year 2 is likely due to chance or the presence of underlying differences in the 2 compared cohorts. The lack of demonstrable heterosubtypic protection is consistent with previous observations that children do not develop lasting, clinically significant heterosubtypic protection after wild-type influenza illness.3 However, the current and prior analyses cannot rule out the fact that short-term, protective heterosubtypic immunity may occur, as demonstrated in Japanese schoolchildren in 1977–1978 when A/H3N2 and A/H1N1 circulated in the same influenza season.5 Short-term cross-subtype immunity is also consistent with the epidemiologic observation that A/H3N2 and A/H1N1 rarely codominate during a specific influenza season. Together with data suggesting that adults can develop protective heterosubtypic immunity in the years following influenza illness,3,4 the results of this analysis support the conclusion that it is only after multiple wild-type influenza illnesses acquired through childhood, adolescence, and early adulthood that humans might develop lasting, clinically significant heterosubtypic immunity. This analysis also demonstrates how data from other multiyear, placebo-controlled studies of influenza illness in children or adults could be examined for evidence of heterosubtypic immunity, assuming that the study population does not change between influenza seasons.
As expected based on the lack of common antigens for influenza A and B, there was no evidence that influenza A illness provided heterotypic protection against subsequent influenza B illness. However, there was evidence that wild-type influenza illness provided homosubtypic protection through the following year. Homosubtypic protection has been shown to persist for many years after wild-type influenza illness for antigenically related strains, as most recently demonstrated with the emergence of the novel pandemic H1N1 strain. In the current studies, as previously described, the level of homosubtypic protection observed against a novel, antigenically distinct A/H3N2 strain closely matched with the level of protection induced by annual vaccination with live attenuated influenza vaccine.9
Limitations to this analysis are driven largely by the available data. Studies were limited to a 2-year period and could only evaluate the effect of a single influenza exposure. Despite the large size of the studies analyzed, there were relatively few children with influenza illness caused by a single subtype in year 1 that could be followed through year 2 for subsequent influenza illness. As a result, the failure to detect heterosubtypic protection cannot rule out a moderate or low level of protection. With a combined cohort of 1286 unvaccinated recipients followed for more than 2 years, the current 3-study pooled analysis had 80% power to detect a risk reduction of 60%, 59% power to detect a risk reduction of 50%, and only 37%, 20%, and 10% power to detect risk reductions of 40%, 30%, and 20%, respectively. To achieve 80% power to detect a 40% risk reduction, a cohort of approximately 2500 to 4300 individuals would have been required, assuming an annual influenza A subtype illness rate of 10% to 13%.
In addition to the limitations related to study size and duration, the compared cohorts were not randomized, and differences in various underlying factors could have affected a cohort's risk of influenza in year 2. It is also possible that influenza infection in either year could have occurred without culture-confirmed illness. Although a sensitive definition of influenza illness was used in the studies, asymptomatic infections would have occurred in some subjects and there may have been symptomatic illness for which a proper culture was not obtained. Lastly, severity of illness in the cohorts could not be extensively compared. Heterosubtypic immunity in children could exist and could help limit the severity of influenza illness while failing to protect against influenza infection. The end point used in the studies required the presence of respiratory illness. As a result, the current analysis could have detected an effect on illness severity with regard to the incidence of respiratory symptoms, but it was not possible to evaluate the length of illness or the incidence of more severe illness such as lower respiratory tract disease.
The current results support previous observations that children do not develop lasting, clinically significant heterosubtypic protection after influenza illness. However, a moderate-to-low level of heterosubtypic protection could not be ruled out. These results help to advance our understanding of human anti-influenza immunity. Data from other multiyear, placebo-controlled studies of influenza illness in children or adults should be examined for evidence of heterosubtypic immunity.
Editorial assistance in the form of copyediting and preparation of the manuscript for submission was provided by Susan E. DeRocco, PhD, John E. Fincke, PhD, and Gerard P. Johnson, PhD, of Complete Healthcare Communications, Inc. (Chadds Ford, PA).
1. Grebe KM, Yewdell JW, Bennink JR. Heterosubtypic
immunity to influenza
A virus: where do we stand? Microbes Infect
2. Bodewes R, Kreijtz JH, Baas C, et al. Vaccination against human influenza
A/H3N2 virus prevents the induction of heterosubtypic
immunity against lethal infection with avian influenza
A/H5N1 virus. PLoS One
3. Epstein SL. Prior H1N1 influenza
infection and susceptibility of Cleveland Family Study participants during the H2N2 pandemic of 1957: an experiment of nature. J Infect Dis
4. Slepushkin AN. The effect of a previous attack of A1 influenza
on susceptibility to A2 virus during the 1957 outbreak. Bull World Health Organ
5. Sonoguchi T, Naito H, Hara M, et al. Cross-subtype protection in humans during sequential, overlapping, and/or concurrent epidemics caused by H3N2 and H1N1 influenza
viruses. J Infect Dis
6. Steinhoff MC, Fries LF, Karron RA, et al. Effect of heterosubtypic
immunity on infection with attenuated influenza
A virus vaccines in young children
. J Clin Microbiol
7. Vesikari T, Fleming DM, Aristegui JF, et al. Safety, efficacy, and effectiveness of cold-adapted influenza
vaccine-trivalent against community-acquired, culture-confirmed influenza
in young children
attending day care. Pediatrics
8. Tam JS, Capeding MR, Lum LC, et al. Efficacy and safety of a live attenuated, cold-adapted influenza
vaccine, trivalent against culture-confirmed influenza
in young children
in Asia. Pediatr Infect Dis J
9. Belshe RB, Gruber WC, Mendelman PM, 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
10. Bracco Neto H, Farhat CK, Tregnaghi MW, et al. Efficacy and safety of 1 and 2 doses of live attenuated influenza
vaccine in vaccine-naive children
. Pediatr Infect Dis J