Abstract

(See the editorial commentary by Halasa, on pages 1471–4.)

Background. The safety and immunogenicity of live, attenuated influenza vaccine (LAIV) has not been compared to that of the standard trivalent inactivated vaccine (TIV) in children with cancer.

Methods. Randomized study of LAIV versus TIV in children with cancer, age 2–21 years, vaccinated according to recommendations based on age and prior vaccination. Data on reactogenicity and other adverse events and blood and nasal swab samples were obtained following vaccination.

Results. Fifty-five eligible subjects (mean age, 10.4 years) received vaccine (28 LAIV/27 TIV). Both vaccines were well tolerated. Rhinorrhea reported within 10 days of vaccination was similar in both groups (36% LAIV vs 33% TIV, P > .999). Ten LAIV recipients shed virus; the latest viral shedding was detected 7 days after vaccination. Immunogenicity data were available for 52 subjects, or 26 in each group. TIV induced significantly higher postvaccination geometric mean titers against influenza A viruses (P < .001), greater seroprotection against influenza A/H1N1 (P = .01), and greater seroconversion against A/H3N2 (P = .004), compared with LAIV. No differences after vaccination were observed against influenza B viruses.

Conclusions. As expected, serum antibody response against influenza A strains were greater with TIV than with LAIV in children with cancer. Both vaccines were well tolerated, and prolonged viral shedding after LAIV was not detected.

Clinical Trials Registration. NCT00906750.

Influenza infection is a major cause of lower respiratory tract illness in children [1] and has greater potential to cause adverse outcomes in children with underlying medical conditions [2]. Pediatric solid organ transplant recipients are at risk for influenza-related complications, including pneumonia, sepsis, CNS disease, acute graft rejection, and death [3]. Pediatric patients with cancer have been reported to acquire influenza more frequently [4] and to have prolonged symptoms, compared with healthy controls [5], resulting in hospitalization, administration of antibiotics, and interruptions of chemotherapy [5]. Severe and fatal complications have also been reported [6]. For these reasons, the American Academy of Pediatrics and the Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention endorse the annual administration of the trivalent inactivated influenza vaccine (TIV) to immunocompromised children, including those who are undergoing treatment for malignancy [7]. However, there are conflicting data concerning the immunogenicity of influenza vaccines in children with cancer [8–11]. Discrepancies in reported immune responses may result from differences in patient characteristics, including age, underlying malignancies, and chemotherapeutic regimens, or methods of vaccination, including type of vaccine, dose, and timing of administration in relation to chemotherapy.

Live attenuated, cold-adapted, trivalent influenza-virus vaccine (LAIV) has been shown to be efficacious and safe in eligible adults and children 2–49 years of age [12, 13]. Moreover, LAIV has the potential in children to stimulate greater local anti-influenza immune responses and greater reductions in influenza illness, including against antigenically drifted strains not included in the vaccine [14], compared with the TIV. In addition, several studies have shown that the LAIV protective effect can extend beyond the year of administration [15–18]. LAIV has been shown to induce T-cell responses as well as humoral immunity [19]. In addition, LAIV has the benefit of having a needle-free delivery system, which may make this vaccine more desirable for use in patients with low platelet counts. Thus, we reasoned that LAIV may offer several potential advantages over TIV in vaccination of children with cancer. At present, LAIV is not licensed by the Food and Drug Administration for administration to immunocompromised individuals.

In a prior multicenter, randomized, double-blinded trial of LAIV (n = 10) versus placebo (n = 10) in pediatric cancer patients conducted during the 2004–2007 influenza seasons, LAIV was demonstrated to be well tolerated with no treatment-related serious adverse events. In that study, prolonged shedding of vaccine virus did not occur [20].

In the present study, we compared the safety and serologic immune response of LAIV to TIV in mild to moderately immunocompromised children with cancer and assessed the incidence and duration of viral replication following vaccination with LAIV.

MATERIAL AND METHODS

Mild to moderately immunocompromised children with cancer who had stable disease or were in remission, were receiving or had received chemotherapy and/or radiotherapy within the past 3 months, and were aged 2–21 years were eligible for enrollment. Subjects were excluded if they had an absolute neutrophil count (ANC) ≤500 cells/mm3 within 72 hours prior to study entry, had previous receipt of any current year influenza vaccine, had received immunoglobulin within 90 days, had close contact with another severely immunocompromised patient, were recipients of hematopoietic stem cell transplant, or met other safety exclusion criteria unrelated to immune status that were listed in the package inserts of either vaccine.

Stratified by age and influenza vaccination history, study participants were randomly assigned (1:1, LAIV to TIV) in blocks of 4 using a centrally managed computer-generated randomization schedule. Children received 2 doses of vaccine, 28–42 days apart, if they were <9 years of age and had a history of less than 2 prior influenza vaccinations. Children who received influenza vaccination in previous seasons or who were ≥9 years of age received 1 dose of vaccine. The study was approved by the St. Jude Children's Research Hospital (SJCRH) Institutional Review Board. Influenza activity was reported nationally in September 2008 and locally in October 2008 [21].

Study Vaccines

LAIV (FluMist; MedImmune), intranasal 0.2 mL (0.1 mL per nostril), or TIV (Fluzone; Aventis Pasteur), intramuscular 0.25 mL for children below 36 months of age or 0.5 mL for older children, were used. Standard vaccines for the 2008–2009 influenza season, including A/Brisbane/59/2007 (H1N1)–like virus, A/Brisbane/10/2007 (H3N2)–like virus and B/Florida/4/2006-like virus, were used.

Safety Evaluation

Information was collected about adverse events, including fever, rhinorrhea/nasal congestion, sore throat, cough, vomiting, headache, muscle aches, pain, chills, tiredness, and irritability in a diary card for 28 consecutive days after each vaccination. For subjects receiving TIV, the presence or absence of redness, swelling, and/or pain around the injection site was also recorded. All subjects were contacted (either via telephone or clinic visit) on days 3–5 and days 7–10 after each vaccination to assess safety. In the LAIV group, nasal swabs were obtained at days 3–5, 7–10, and 28–42 following each vaccination for viral culture. Families of subjects not scheduled to return to clinic were instructed on how to obtain a nasal swab and return it via overnight shipment using appropriate transport media. All participants were seen in clinic 28–42 days after each vaccination. Adverse events were assessed for severity using National Cancer Institute Common Toxicity Criteria and relationship to vaccination. Reactogenicity events were defined as those occurring within 28 days of vaccination and expected based on vaccine package inserts. Serious adverse events were collected for 180 days. If a subject developed influenza-like illness within the first 28 days after each vaccination, nasal swabs were collected and tested (by culture, PCR, and antigen detection using direct immunofluorescence) for influenza.

Immunogenicity Evaluation

Whole blood samples were obtained from all subjects prior to vaccination and 28–42 days after each dose of vaccine; serum was stored at –20°C until the time of analysis. The primary response was the detection of influenza-specific antibodies for the 3 influenza strains included in the vaccine, as measured by hemagglutination inhibition (HAI) as previously described [22]. Humoral responses were also assessed by microneutralization assay, and influenza-specific serum antibody isotype concentrations were determined using ELISA as described [22]. Total IgG, IgM, and IgG subclasses IgG1, IgG2, IgG3, and IgG4 were independently assayed. Results were expressed as geometric mean titers (GMTs) with 95% confidence intervals (CIs); seroprotection rates, defined as the percentage of subjects achieving an HAI titer ≥40; and seroconversion rates, defined as the percentage of subjects achieving at least a 4-fold increase in HAI titers from a seropositive prevaccination titer (≥10) or a rise from <10 to ≥40 in those who were seronegative. Serum immunogenicity analysis was also performed against mismatched B (B/Bris/60/08) to assess the immune response induced against vaccine mismatched viruses. Complete blood cell count and differential were obtained prior to vaccination. ELISPOT analysis (IFNγ) of PBMCs was done as previously described [23] using whole inactivated viruses corresponding to each vaccine antigen.

Madin-Darby canine kidney cells were inoculated with nasal swab specimens and the quantity of viral shedding was determined by assessment of the 50% tissue culture infectious dose (TCID50).

Immunogenicity and virology studies were performed at SJCRH by laboratory technicians not blinded to treatment assignment.

Statistical Analyses

A sample size of 60 participants, 30 in each group, was chosen based on estimates of enrollment. Estimating a seroresponse with TIV of 35.7%, we would have power of 80% (P = .05) to detect only a large difference (35%) between the 2 groups. The Fisher exact test was used to compare the seroconversion/seroprotection rates and other categorical variables between 2 vaccine groups. For comparison of ANC, absolute lymphocyte count (ALC), serum IgA and IgG concentrations, 2 groups were established according to the median value among the 55 subjects at baseline and compared using the Wilcoxon rank sum test. When appropriate [24], multiple logistic regression models were applied to examine the association between seroprotection and seroconversion of a particular strain and vaccine type (LAIV vs TIV) after adjusting for the following covariates: age, ANC, ALC, serum IgA and IgG concentrations, and cancer diagnosis (hematological diseases vs solid tumor). Interaction terms between vaccine type and other covariates were also considered in the logistic models. The nominal level for statistical significance was an alpha of .05. SAS version 9.2 (SAS Institute) and StatXact (Cytel Corporation) Windows version 8 were used for statistical analysis.

RESULTS

Subject Characteristics

Fifty-six subjects were randomized (14 October through 31 December 2008) at a ratio of 1:1 to receive either LAIV or TIV. One subject in the LAIV group was found to have a previous diagnosis of asthma after enrollment and was excluded before receiving study vaccine. Among the remaining 55 subjects, 28 received LAIV, while 27 received TIV; 48 received 1 dose, and 7 received 2 doses. Demographic data are summarized in Table 1; there were no significant differences between subjects in the 2 vaccine groups.

Table 1.

Demographic Characteristics of Subjects Receiving Either Live-Attenuated Influenza Virus Vaccine (LAIV) or Trivalent Influenza Virus (TIV)

Characteristic LAIV (n = 28) TIV (n = 27) 
Age   
    Mean, years (range) 10.4 (2.1–20) 10.4 (2.1–21) 
    <9 years 13 (46) 14 (52) 
    ≥9 years 15 (54) 13 (48) 
Male sex 15 (54) 15 (56) 
Race   
    White, non-Hispanic 19 (68) 20 (74) 
    Black 8 (29) 7 (26) 
    Asian/White 1 (3) 
Underlying cancer   
    Hematological malignancy 13 (46) 12 (44) 
    Solid tumor 15 (54) 15 (56) 
Previous seasonal influenza vaccine 23 (82) 18 (67) 
Mean ANC count, cells/mm3  
    Hematologic malignancy 1.9 2.8 
    Solid tumor 3.6 2.4 
Mean ALC count, cells/mm3  
    Hematologic malignancy 1.0 1.0 
    Solid tumor 1.3 1.1 
Mean total serum immunoglobulin concentration, mg/dLa,b 802.1 825 
Characteristic LAIV (n = 28) TIV (n = 27) 
Age   
    Mean, years (range) 10.4 (2.1–20) 10.4 (2.1–21) 
    <9 years 13 (46) 14 (52) 
    ≥9 years 15 (54) 13 (48) 
Male sex 15 (54) 15 (56) 
Race   
    White, non-Hispanic 19 (68) 20 (74) 
    Black 8 (29) 7 (26) 
    Asian/White 1 (3) 
Underlying cancer   
    Hematological malignancy 13 (46) 12 (44) 
    Solid tumor 15 (54) 15 (56) 
Previous seasonal influenza vaccine 23 (82) 18 (67) 
Mean ANC count, cells/mm3  
    Hematologic malignancy 1.9 2.8 
    Solid tumor 3.6 2.4 
Mean ALC count, cells/mm3  
    Hematologic malignancy 1.0 1.0 
    Solid tumor 1.3 1.1 
Mean total serum immunoglobulin concentration, mg/dLa,b 802.1 825 

Data are no. (%) of subjects, unless otherwise indicated.

Abbreviations: ANC, absolute neutrophil count; ALC, absolute lymphocyte count; ANC, absolute neutrophil count.

a

At the time of the first vaccination.

b

Concentration includes IgA, IgM, and IgG.

Safety Evaluation

Reactogenicity events after 10 and 28 days were similar in both groups, except for injection site reactions in the TIV group (18.5% overall) (Table 2). All reactogenicity events were grade 1 and 2. None of the subjects were removed from the study because of vaccine-related adverse events.

Table 2.

Summary of Reactogenicity Events Within 28 Days Following Dose 1 or 2 of Live-Attenuated Influenza Virus Vaccine (LAIV) or Trivalent Influenza Virus (TIV)

 After First Dose
 
After Second Dose
 
 Day 0–10
 
Day 0–28
 
Day 0–10
 
Day 0–28
 
Reactogenicity events LAIV (n = 28) TIV (n = 27) LAIV (n = 28) TIV (n = 27) LAIV (n = 3) TIV (n = 4) LAIV (n = 3) TIV (n = 4) 
Fevera 0 (0) 2 (7.4) 1 (3.6) 4 (14.8) 0 (0) 1 (25) 0 (0) 1 (25) 
Rhinorrhea 10 (35.7) 9 (33.3) 12 (42.9) 9 (33.3) 1 (33.3) 1 (25) 2 (66.7) 1 (25) 
Sore throat 3 (10.7) 3 (11.1) 6 (21.4) 3 (11.1) 0 (0) 0 (0) 0 (0) 0 (0) 
Cough 2 (7.1) 5 (18.5) 3 (10.7) 5 (18.5) 1 (33.3) 0 (25) 2 (66.7) 0 (0) 
Vomiting 5 (17.8) 5 (18.5) 8 (28.6) 6 (22.2) 1 (33.3) 1 (25) 1 (33.3) 1 (25) 
Headache 4 (14.3) 4 (14.8) 5 (17.8) 4 (14.8) 0 (0) 0 (0) 1 (33.3) 0 (0) 
Muscles aches 0 (0) 2 (7.4) 0 (0) 3 (11.1) 0 (0) 0 (0) 0 (0) 0 (0) 
Tiredness 4 (14.3) 5 (18.5) 6 (21.4) 5 (18.5) 0 (0) 1 (25) 0 (0) 1 (25) 
Irritability 3 (10.7) 0 (0) 3 (10.7) 1 (3.7) 0 (0) 0 (0) 0 (0) 0 (0) 
Chills 3 (10.7) 2 (7.4) 3 (10.7) 2 (7.4) 0 (0) 0 (0) 0 (0) 0 (0) 
Injection local reactionb NA 4 (14.8) NA 5 (18.5) NA 0 (0) NA 0 (0) 
 After First Dose
 
After Second Dose
 
 Day 0–10
 
Day 0–28
 
Day 0–10
 
Day 0–28
 
Reactogenicity events LAIV (n = 28) TIV (n = 27) LAIV (n = 28) TIV (n = 27) LAIV (n = 3) TIV (n = 4) LAIV (n = 3) TIV (n = 4) 
Fevera 0 (0) 2 (7.4) 1 (3.6) 4 (14.8) 0 (0) 1 (25) 0 (0) 1 (25) 
Rhinorrhea 10 (35.7) 9 (33.3) 12 (42.9) 9 (33.3) 1 (33.3) 1 (25) 2 (66.7) 1 (25) 
Sore throat 3 (10.7) 3 (11.1) 6 (21.4) 3 (11.1) 0 (0) 0 (0) 0 (0) 0 (0) 
Cough 2 (7.1) 5 (18.5) 3 (10.7) 5 (18.5) 1 (33.3) 0 (25) 2 (66.7) 0 (0) 
Vomiting 5 (17.8) 5 (18.5) 8 (28.6) 6 (22.2) 1 (33.3) 1 (25) 1 (33.3) 1 (25) 
Headache 4 (14.3) 4 (14.8) 5 (17.8) 4 (14.8) 0 (0) 0 (0) 1 (33.3) 0 (0) 
Muscles aches 0 (0) 2 (7.4) 0 (0) 3 (11.1) 0 (0) 0 (0) 0 (0) 0 (0) 
Tiredness 4 (14.3) 5 (18.5) 6 (21.4) 5 (18.5) 0 (0) 1 (25) 0 (0) 1 (25) 
Irritability 3 (10.7) 0 (0) 3 (10.7) 1 (3.7) 0 (0) 0 (0) 0 (0) 0 (0) 
Chills 3 (10.7) 2 (7.4) 3 (10.7) 2 (7.4) 0 (0) 0 (0) 0 (0) 0 (0) 
Injection local reactionb NA 4 (14.8) NA 5 (18.5) NA 0 (0) NA 0 (0) 

Data are no. (%) of reactogenicity events.

Abbreviation: NA, not applicable.

a

Defined as ≥100°C (orally) or ≥99.6°C (axillary).

b

Defined as tenderness, erythema, or induration at injection site.

Two serious adverse events were considered possibly related to the vaccine. An 11-year-old girl developed fever, cough, rhinorrhea, myalgia, and mild hypotension requiring hospitalization 12 days after LAIV receipt. Viral culture of a nasal swab had isolated the vaccine strain of influenza A virus on day 3 after vaccination but not on day 7. A nasal wash collected on day 13 at the time of onset of the symptoms was negative for influenza by DFA and culture, but influenza A virus was detectable by PCR. Insufficient RNA quantities prohibited further characterization. She was treated with antimicrobials and oseltamivir but remained febrile after 5 days of therapy. Multiple pulmonary nodules consistent with fungal infection were identified by computerized chest tomography on day 9 of hospitalization, and antifungal therapy was begun; fever resolved, and she was discharged from the hospital on day 11. In another case, a 2-year-old boy with a diagnosis of glioma developed afebrile seizure-like activity within 30 minutes after receiving TIV. His seizure was attributed to mild hyponatremia, although serum sodium level was normal, and anemia. He was admitted to the hospital and discharged the next day without further episodes of seizure. This subject had received TIV in the previous year without complications.

Immunogenicity

Data were available for 52 subjects, 26 in the LAIV group and 26 in the TIV group. Three subjects, 2 in the LAIV group and 1 in the TIV group, were excluded from the immunogenicity analysis because of events occurring after vaccination but before collection of immunogenicity samples. In the LAIV group, 1 subject had been receiving intravenous immunoglobulin and was inadvertently enrolled, and 1 subject developed relapsed malignancy between the first and second vaccinations. In the TIV group, 1 subject, who was asymptomatic, had a positive PCR for influenza A virus of unknown subtype 28 days after the first vaccine.

Pre- and postvaccination GMTs, seroprotection rates, and seroconversion rates are shown in Table 3. The postvaccination GMTs against the influenza A virus subtype H3N2 and H1N1 strains were significantly higher after TIV receipt than after LAIV receipt (P < .001), but titers for B/Flor/5/06 or B/Bris/60/08 were not (P = .41 and P = .15, respectively). Both vaccines elicited a comparably high seroprotection rate for A/H3N2, but TIV elicited a significantly higher seroprotection rate for A/H1N1, compared with LAIV (73.0% vs 34.6%; P = .01). The high seroprotection responses to A/H3N2 seen in both groups and to A/H1N1 in the TIV group are partially due to the high proportion of subjects with HAI baseline titers ≥40. Seroprotection responses to matched and mismatched influenza B strains remained low after vaccination in both groups. Seroconversion rates were higher for TIV, compared with LAIV, for A/H3N2 (46.1% vs 7.6%; P < .004). Seroconversion rates were low for A/H1N1 and both influenza B strains with each vaccine group.

Table 3.

Hemagglutinition Inhibition Assay (HAI) Immunogenicity Data After Completion of Vaccination With Live-Attenuated Influenza Virus Vaccine (LAIV) or Trivalent Influenza Virus (TIV)

Influenza Virus Strain, Variable LAIV (n = 26) TIV (n = 26) P
A (H3N2)    
    Pre-GMT (95% CI) 80 (5–1244) 126 (5–2920) .30 
    Post-GMT (95% CI) 82 (7–976) 228 (18–6286) <.001 
    Preseroprotection, %b 84 80.7 >.999 
    Postseroprotection, %b 80.7 92.3 .41 
    Postseroconversion, %c 7.6 46.1 <.004 
A (H1N1)    
    Pre-GMT (95% CI) 24 (3–216) 38 (3–456) .19 
    Post-GMT (95% CI) 17 (4–80) 89 (6–1336) <.001 
    Preseroprotection, %b 34.6 53.8 .26 
    Postseroprotection, %b 34.6 73.0 .01 
    Postseroconversion, %c 7.6 26.9 .13 
B (Flor) (Matched)    
    Pre-GMT (95% CI) 11 (2–66) 13 (1–131) .51 
    Post-GMT (95% CI) 14 (3–68) 21 (2–274) .41 
    Preseroprotection, %b 15.3 23 .73 
    Postseroprotection, %b 19.2 30.7 .52 
    Postseroconversion, %c 3.8 11.5 .60 
B (Bris)    
    Pre-GMT (Mismatched) (95% CI) 20 (3–153) 12 (2–81) .05 
    Post-GMT (95% CI) 8 (2–29) 11 (2–53) .15 
    Preseroprotection, %b 30.7 19.2 .52 
    Postseroprotection, %b 3.8 15.3 .34 
    Postseroconversion, %c 3.8 >.999 
Influenza Virus Strain, Variable LAIV (n = 26) TIV (n = 26) P
A (H3N2)    
    Pre-GMT (95% CI) 80 (5–1244) 126 (5–2920) .30 
    Post-GMT (95% CI) 82 (7–976) 228 (18–6286) <.001 
    Preseroprotection, %b 84 80.7 >.999 
    Postseroprotection, %b 80.7 92.3 .41 
    Postseroconversion, %c 7.6 46.1 <.004 
A (H1N1)    
    Pre-GMT (95% CI) 24 (3–216) 38 (3–456) .19 
    Post-GMT (95% CI) 17 (4–80) 89 (6–1336) <.001 
    Preseroprotection, %b 34.6 53.8 .26 
    Postseroprotection, %b 34.6 73.0 .01 
    Postseroconversion, %c 7.6 26.9 .13 
B (Flor) (Matched)    
    Pre-GMT (95% CI) 11 (2–66) 13 (1–131) .51 
    Post-GMT (95% CI) 14 (3–68) 21 (2–274) .41 
    Preseroprotection, %b 15.3 23 .73 
    Postseroprotection, %b 19.2 30.7 .52 
    Postseroconversion, %c 3.8 11.5 .60 
B (Bris)    
    Pre-GMT (Mismatched) (95% CI) 20 (3–153) 12 (2–81) .05 
    Post-GMT (95% CI) 8 (2–29) 11 (2–53) .15 
    Preseroprotection, %b 30.7 19.2 .52 
    Postseroprotection, %b 3.8 15.3 .34 
    Postseroconversion, %c 3.8 >.999 

Abbreviations: CI, confidence interval; GMT, geometric mean titer.

a

Based on the Wilcoxon rank sum test or the Fisher exact test.

b

Percentage of subjects achieving an HAI titer ≥40.

c

Percentage of subjects achieving a 4-fold rise in the prevaccination titer or a rise from <10 to ≥40 in those who were seronegative prevaccination titer.

Microneutralization titers were generally higher for all antigens than were HAI titers (Figure 1). IgG1 was the predominant IgG subclass induced by both vaccines (data not shown).

Figure 1.

Hemagglutination inhibition (HAI) and microneutralization (MN) responses following receipt of live attenuated influenza-virus vaccine (LAIV) and trivalent inactivated influenza vaccine (TIV) by children with cancer. HAI (A, C, and E) and MN (B, D, and F) titers after LAIV (n = 26) and TIV (n = 26) are shown pre- and postvaccination. The horizontal line indicates the mean titer.

Figure 1.

Hemagglutination inhibition (HAI) and microneutralization (MN) responses following receipt of live attenuated influenza-virus vaccine (LAIV) and trivalent inactivated influenza vaccine (TIV) by children with cancer. HAI (A, C, and E) and MN (B, D, and F) titers after LAIV (n = 26) and TIV (n = 26) are shown pre- and postvaccination. The horizontal line indicates the mean titer.

Analysis of cellular immunity by ELISPOT demonstrated extreme variability in responses, with values equally likely to decrease or increase from baseline (Figure 2). No differences in the direction or magnitude of these changes were observed in comparing LAIV to TIV or between the 3 antigens (representative data shown for A/H1N1 only).

Figure 2.

T-cell responses by ELISPOT to live attenuated influenza-virus vaccine (LAIV) and trivalent inactivated influenza vaccine (TIV) in children with cancer. ELISPOT analyses (IFNγ spot-forming units [SFU] per million PBMCs) before (day 0) and after (day 28–42) vaccination with LAIV (n = 26) or TIV (n = 26). The horizontal line indicates the limit of the assay; no spots were detectable for samples below this line.

Figure 2.

T-cell responses by ELISPOT to live attenuated influenza-virus vaccine (LAIV) and trivalent inactivated influenza vaccine (TIV) in children with cancer. ELISPOT analyses (IFNγ spot-forming units [SFU] per million PBMCs) before (day 0) and after (day 28–42) vaccination with LAIV (n = 26) or TIV (n = 26). The horizontal line indicates the limit of the assay; no spots were detectable for samples below this line.

The data were further analyzed for the effect of several covariates on response to vaccination. Logistic regression model revealed no interaction effect between vaccine type and any of the covariates. After control for vaccine type, younger subjects were more likely to seroconvert to A/H1N1, compared with older subjects (P= .05; data not shown), and older subjects were more likely to have seroprotection against A/H3N2 (P= .04; data not shown). When age was dichotomized (<9 vs ≥9 years), improvement in seroconversion to A/H1N1 in younger subjects was no longer observed. In addition, subjects with higher serum IgG concentration at baseline were more likely to reach seroprotection against A/H1N1 (P = .05; data not shown), compared with subjects with lower serum IgG concentrations at baseline.

Although this study was not designed to assess the clinical efficacy of any of the studied vaccines in terms of the prevention of influenza infection, 3 participants had documented laboratory-confirmed influenza A infection. In one case (in the TIV group), infection occurred 28 days after first and before second vaccination, and in the other 2 cases (one in each group), infection developed more than 60 days after vaccination. Of the 2 who had completed the vaccination series, neither had a postvaccination titer predictive of seroprotection to any of the 3 influenza strains contained in the vaccine.

Viral Shedding

Viral shedding was assessed as described in all 28 LAIV recipients and documented in 10 (35.7%). Six subjects shed virus on the day 3–5 assessment only; 1 shed virus on the day 7–10 assessment only, and 3 subjects shed virus at both the day 3–5 and day 7–10 assessments. The peak titer of vaccine virus shedding was ≤103 TCID50/mL in all subjects, and no viral shedding was detected past the day 7–10 visit.

DISCUSSION

In our study of children with cancer, higher postvaccination GMT was observed in the TIV group with both A/H3N2 and A/H1N1 viruses when compared with LAIV using HAI. Greater seroprotection (postvaccination HAI titer ≥40) was observed with A/H1N1 and greater seroconversion (at least a 4-fold increase from prevaccination titer or a rise from <10 to ≥40) with A/H3N2 in the TIV group as well. High concentrations of preexisting antibody to A/H3N2 may explain the lack of difference in seroprotection in the 2 groups.

Overall, studies with LAIV have demonstrated that the proportion of individuals experiencing at least a 4-fold rise in the serum HAI titers is often correlated with protective efficacy [15, 25]. However, the absence of serologic responses following LAIV administration may not correlate with lack of protection, as high levels of efficacy and effectiveness occur with low serum HAI responses elicited by LAIV [13, 15, 25–27]. A previous study comparing TIV and LAIV in healthy adults during an influenza season when an antigenically drifted type A/H3N2 influenza predominated revealed lower antibody responses with LAIV, compared with TIV, for all 3 influenza subtypes using HAI. However, LAIV still appeared to be protective, particularly against type A influenza, and the estimation of relative efficacy did not indicate a significant advantage of TIV over LAIV [27]. This result may suggest that the use of antibody titers to confirm protection against influenza may lead to overestimation of the efficacy of TIV and underestimation of LAIV efficacy [27]. Additionally, an earlier trial found an association between serum HAI titers and protection but found that other factors contributed to the protection of vaccinated subjects [26]. In that study, some vaccinated children had neither protective serum HAI titers nor nasal wash IgA, and it was postulated that alternative immune mechanisms, such as cellular immunity, may have contributed to protection by LAIV [28]. One recent study in young children demonstrated a correlation between IFNγ ELISPOT responses induced by LAIV and protection from culture-confirmed influenza illness [29]. In our study, we saw variable T-cell responses as measured by ELISPOT both within participants and between groups. This likely reflects the effect of ongoing immunosuppression through chemotherapy or other immunosuppressive drugs that affect the function and number of immunocompetent cells, including lymphocytes, limiting IFNγ production [30].

Response to the influenza B vaccine component appeared poorer than to influenza A virus strains, with GMT following vaccination remaining below the level normally considered protective in both vaccine groups. Influenza B virus circulated only briefly in 2006 and was almost absent from circulating influenza strains in 2007, resulting in low levels of immunity in this population, as reflected in prevaccination titers [31, 32]. The low prevaccination influenza B virus titers may have influenced the postvaccination titers in the TIV recipients. Some studies using TIV in immunocompromised patients have found different immunogenicity rates to different vaccine strains, with a lower rate of immunogenicity for B type strain using the same or different vaccine strains [33, 34]. The immune response to different vaccine strains seems to be more balanced in immunocompetent populations [34, 35], although healthy children receiving TIV frequently have very poor immune response to influenza B virus [36].

Some previous reports claimed that children with solid tumors produced higher postvaccination HAI titers, compared with those with leukemia [8, 37], but this was not observed in our vaccine groups. Similarly, previous studies using TIV in children with cancer showed an inverse correlation between the rates of seroconversion and total numbers of circulating lymphocytes or neutrophils [37, 38], although other studies were not able to confirm this finding [8]. We did not find any correlation between prevaccination ALC or ANC and immune response to any of the studied vaccines (data not shown), but this may be related to our small sample size.

Logistic regression analysis revealed a significant association between baseline serum IgG concentrations and seroprotection against A/H1N1, with patients with higher serum IgG concentrations being more likely to respond to vaccination, compared with patients with lower serum IgG concentrations, in both vaccine groups. This association was not statistically significant for A/H3N2 and B strains. In a previous study among patients with cancer vaccinated with TIV, a lower protective immune response to A/H1N1 and influenza B virus was observed for patients with low baseline serum IgG concentrations, compared with those with high serum IgG concentrations [28].

Two serious adverse events were judged to be possibly related to vaccination. One subject had a seizure within minutes after receiving TIV. Although previous reports have described seizures after administration of TIV in children [39, 40], no relationship could be established in this case. Another subject developed influenza-like symptoms and a nasal wash positive for influenza A virus by PCR but negative by DFA and culture at day 13 after LAIV. The genotype of the virus (vaccine strain vs wild-type strain) could not be determined because of insufficient quantity of RNA in the sample. Wild-type virus was not circulating in the community at that time, so it is likely that this finding represents residual vaccine virus present after the day 7–10 visit. The clinical impression at the time was probable fungal pneumonia.

The safety profiles of LAIV and TIV were similar to those previously reported [13, 41]. Although previously observed in individuals receiving LAIV [42], we did not find more rhinorrhea or fever in the LAIV recipients, compared with the TIV group.

In our study, the last day of viral shedding among LAIV recipients was 7 days after vaccination. This finding is similar to previous reports [43], although shedding has been reported up to 21 days after LAIV in healthy children <36 months of age [44]. It has been reported that shedding of vaccine virus is reduced in those who have been vaccinated previously with TIV [43]; thus, it was not surprising that less shedding occurred in our cohort of LAIV recipients, who had high rates of previous TIV vaccination.

Our study has several limitations. A small sample size resulted in greater variability in laboratory assessments and limited the power of our conclusions. Furthermore, the subjects had many different malignant conditions with heterogeneous treatments. We were also only able to report antibody response rather than relative efficacy for preventing influenza. We attempted to evaluate influenza-specific mucosal IgA antibody response, but yield was not adequate.

The results of this study indicate that both TIV and LAIV were safe in mild to moderately immunosuppressed children with cancer. In addition, the theoretical concerns about transmission from vaccinated individuals to immunocompromised children are unwarranted. Since the vaccine is safe to administer to this patient group, inadvertent transmission is unlikely to cause significant adverse effects. In this study, standard TIV induced a greater serum antibody response than LAIV against influenza A subtypes in an immunosuppressed population. It is unclear at this time whether the relatively poor humoral responses from LAIV translate to poor efficacy. However, the finding that cellular immune responses were extremely variable because of ongoing immune suppression suggests that other correlates of immunity to LAIV cannot be assumed in this patient population.

Notes

Acknowledgments.

 We thank the physicians, nurses, and parents of participating children who assisted with this study, and key laboratory personnel, Susannah Keck and Margaret Griffith, for excellent technical support. Study personnel contributing to this study include Pamela Finnie and Donna Nance. We also thank Chong Wang, for her assistance with statistical analysis, and Sarah Heston and Ramin B. Tabrizi, for helping with database information.

Financial support.

 This work was supported by the American Lebanese Syrian Associated Charities; and by a grant from MedImmune (to S. C.) for conduct and analysis of the study. MedImmune was not involved in the conception, design, conduct, or interpretation of this study.

Potential conflicts of interest.

 S. C. received a MedImmune research grant for the conduct and analysis of the clinical trial, and J. A. M. has served as a paid consultant to GlaxoSmithKline, Novartis, and Pfizer. All other authors report no potential conflicts.

All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References

1.
Glezen
WP
Paredes
A
Taber
LH
Influenza in children. Relationship to other respiratory agents
JAMA
 , 
1980
, vol. 
243
 (pg. 
1345
-
9
)
2.
Neuzil
KM
Wright
PF
Mitchel
EF
Jr
Griffin
MR
The burden of influenza illness in children with asthma and other chronic medical conditions
J Pediatr
 , 
2000
, vol. 
137
 (pg. 
856
-
64
)
3.
Apalsch
AM
Green
M
Ledesma-Medina
J
Nour
B
Wald
ER
Parainfluenza and influenza virus infections in pediatric organ transplant recipients
Clin Infect Dis
 , 
1995
, vol. 
20
 (pg. 
394
-
9
)
4.
Kempe
A
Hall
CB
MacDonald
NE
, et al.  . 
Influenza in children with cancer
J Pediatr
 , 
1989
, vol. 
115
 (pg. 
33
-
9
)
5.
Feldman
S
Webster
RG
Sugg
M
Influenza in children and young adults with cancer: 20 cases
Cancer
 , 
1977
, vol. 
39
 (pg. 
350
-
3
)
6.
Potter
MN
Foot
AB
Oakhill
A
Influenza A and the virus associated haemophagocytic syndrome: cluster of three cases in children with acute leukaemia
J Clin Pathol
 , 
1991
, vol. 
44
 (pg. 
297
-
9
)
7.
Harper
SA
Fukuda
K
Uyeki
TM
Cox
NJ
Bridges
CB
Prevention and control of influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP)
MMWR Recomm Rep
 , 
2004
, vol. 
53
 (pg. 
1
-
40
)
8.
Chisholm
JC
Devine
T
Charlett
A
Pinkerton
CR
Zambon
M
Response to influenza immunisation during treatment for cancer
Arch Dis Child
 , 
2001
, vol. 
84
 (pg. 
496
-
500
)
9.
Porter
CC
Edwards
KM
Zhu
Y
Frangoul
H
Immune responses to influenza immunization in children receiving maintenance chemotherapy for acute lymphoblastic leukemia
Pediatr Blood Cancer
 , 
2004
, vol. 
42
 (pg. 
36
-
40
)
10.
Blumberg
EA
Albano
C
Pruett
T
, et al.  . 
The immunogenicity of influenza virus vaccine in solid organ transplant recipients
Clin Infect Dis
 , 
1996
, vol. 
22
 (pg. 
295
-
302
)
11.
Stiver
HG
Weinerman
BH
Impaired serum antibody response to inactivated influenza A and B vaccine in cancer patients
Can Med Assoc J
 , 
1978
, vol. 
119
 (pg. 
733
-
5
738
12.
Nichol
KL
Mendelman
PM
Mallon
KP
, et al.  . 
Effectiveness of live, attenuated intranasal influenza virus vaccine in healthy, working adults: a randomized controlled trial
JAMA
 , 
1999
, vol. 
282
 (pg. 
137
-
44
)
13.
Belshe
RB
Mendelman
PM
Treanor
J
, et al.  . 
The efficacy of live attenuated, cold-adapted, trivalent, intranasal influenzavirus vaccine in children
N Engl J Med
 , 
1998
, vol. 
338
 (pg. 
1405
-
12
)
14.
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
 , 
2000
, vol. 
136
 (pg. 
168
-
75
)
15.
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
 , 
2007
, vol. 
26
 (pg. 
619
-
28
)
16.
Gaglani
MJ
Piedra
PA
Herschler
GB
, et al.  . 
Direct and total effectiveness of the intranasal, live-attenuated, trivalent cold-adapted influenza virus vaccine against the 2000–2001 influenza A(H1N1) and B epidemic in healthy children
Arch Pediatr Adolesc Med
 , 
2004
, vol. 
158
 (pg. 
65
-
73
)
17.
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
 , 
2009
, vol. 
28
 (pg. 
365
-
71
)
18.
Ambrose
CS
Yi
T
Walker
RE
Connor
EM
Duration of protection provided by live attenuated influenza vaccine in children
Pediatr Infect Dis J
 , 
2008
, vol. 
27
 (pg. 
744
-
8
)
19.
He
XS
Holmes
TH
Zhang
C
, et al.  . 
Cellular immune responses in children and adults receiving inactivated or live attenuated influenza vaccines
J Virol
 , 
2006
, vol. 
80
 (pg. 
11756
-
66
)
20.
Halasa
N
Englund
JA
Nachman
S
, et al.  . 
Safety of live attenuated influenza vaccine in mild to moderately immunocompromised children with cancer
Vaccine
 , 
2011
 
29:4110--5
21.
Centers for Disease Control and Prevention (CDC)
Update: influenza activity—United States, September 28, 2008–January 31, 2009
MMWR Morb Mortal Wkly Rep
 , 
2009
, vol. 
58
 (pg. 
115
-
9
)
22.
McCullers
JA
Van De Velde
LA
Allison
KJ
Branum
KC
Webby
RJ
Flynn
PM
Recipients of vaccine against the 1976 “swine flu” have enhanced neutralization responses to the 2009 novel H1N1 influenza virus
Clin Infect Dis
 , 
2010
, vol. 
50
 (pg. 
1487
-
92
)
23.
Andreansky
SS
Stambas
J
Thomas
PG
Xie
W
Webby
RJ
Doherty
PC
Consequences of immunodominant epitope deletion for minor influenza virus-specific CD8+-T-cell responses
J Virol
 , 
2005
, vol. 
79
 (pg. 
4329
-
39
)
24.
Peduzzi
P
Concato
J
Kemper
E
Holford
TR
Feinstein
AR
A simulation study of the number of events per variable in logistic regression analysis
J Clin Epidemiol
 , 
1996
, vol. 
49
 (pg. 
1373
-
9
)
25.
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
 , 
2006
, vol. 
118
 (pg. 
2298
-
312
)
26.
Belshe
RB
Gruber
WC
Mendelman
PM
, et al.  . 
Correlates of immune protection induced by live, attenuated, cold-adapted, trivalent, intranasal influenza virus vaccine
J Infect Dis
 , 
2000
, vol. 
181
 (pg. 
1133
-
7
)
27.
Ohmit
SE
Victor
JC
Rotthoff
JR
, et al.  . 
Prevention of antigenically drifted influenza by inactivated and live attenuated vaccines
N Engl J Med
 , 
2006
, vol. 
355
 (pg. 
2513
-
22
)
28.
Thomas
PG
Keating
R
Hulse-Post
DJ
Doherty
PC
Cell-mediated protection in influenza infection
Emerg Infect Dis
 , 
2006
, vol. 
12
 (pg. 
48
-
54
)
29.
Forrest
BD
Pride
MW
Dunning
AJ
, et al.  . 
Correlation of cellular immune responses with protection against culture-confirmed influenza virus in young children
Clin Vaccine Immunol
 , 
2008
, vol. 
15
 (pg. 
1042
-
53
)
30.
Sato
S
Nagai
H
The clinical application of quantiferon TB-2G: its usefulness and limitations
Kekkaku
 , 
2011
, vol. 
86
 (pg. 
101
-
12
)
31.
Centers for Disease Control and Prevention. Update: influenza activity–United States and worldwide, 2005–06 season, and composition of the 2006–07 influenza vaccine
MMWR Morb Mortal Wkly Rep
 , 
2006
, vol. 
55
 (pg. 
648
-
53
)
32.
Centers for Disease Control and Prevention. Update: influenza activity–United States, September 30, 2007–February 9, 2008
MMWR Morb Mortal Wkly Rep
 , 
2008
, vol. 
57
 (pg. 
179
-
83
)
33.
Sanchez-Fructuoso
AI
Prats
D
Naranjo
P
, et al.  . 
Influenza virus immunization effectivity in kidney transplant patients subjected to two different triple-drug therapy immunosuppression protocols: mycophenolate versus azathioprine
Transplantation
 , 
2000
, vol. 
69
 (pg. 
436
-
9
)
34.
Smith
KG
Isbel
NM
Catton
MG
Leydon
JA
Becker
GJ
Walker
RG
Suppression of the humoral immune response by mycophenolate mofetil
Nephrol Dial Transplant
 , 
1998
, vol. 
13
 (pg. 
160
-
4
)
35.
Kroon
FP
van Dissel
JT
de Jong
JC
Zwinderman
K
van Furth
R
Antibody response after influenza vaccination in HIV-infected individuals: a consecutive 3-year study
Vaccine
 , 
2000
, vol. 
18
 (pg. 
3040
-
9
)
36.
Neuzil
KM
Jackson
LA
Nelson
J
, et al.  . 
Immunogenicity and reactogenicity of 1 versus 2 doses of trivalent inactivated influenza vaccine in vaccine-naive 5–8-year-old children
J Infect Dis
 , 
2006
, vol. 
194
 (pg. 
1032
-
9
)
37.
Gross
PA
Lee
H
Wolff
JA
Hall
CB
Minnefore
AB
Lazicki
ME
Influenza immunization in immunosuppressed children
J Pediatr
 , 
1978
, vol. 
92
 (pg. 
30
-
5
)
38.
Matsuzaki
A
Suminoe
A
Koga
Y
Kinukawa
N
Kusuhara
K
Hara
T
Immune response after influenza vaccination in children with cancer
Pediatr Blood Cancer
 , 
2005
, vol. 
45
 (pg. 
831
-
7
)
39.
McMahon
AW
Iskander
JK
Haber
P
Braun
MM
Ball
R
Inactivated influenza vaccine (IIV) in children <2 years of age: examination of selected adverse events reported to the Vaccine Adverse Event Reporting System (VAERS) after thimerosal-free or thimerosal-containing vaccine
Vaccine
 , 
2008
, vol. 
26
 (pg. 
427
-
9
)
40.
Kelly
H
Carcione
D
Dowse
G
Effler
P
Quantifying benefits and risks of vaccinating Australian children aged six months to four years with trivalent inactivated seasonal influenza vaccine in 2010
Euro Surveill
 , 
2010
, vol. 
15
  
:19661
41.
Belshe
RB
Edwards
KM
Vesikari
T
, et al.  . 
Live attenuated versus inactivated influenza vaccine in infants and young children
N Engl J Med
 , 
2007
, vol. 
356
 (pg. 
685
-
96
)
42.
Neuzil
KM
Dupont
WD
Wright
PF
Edwards
KM
Efficacy of inactivated and cold-adapted vaccines against influenza A infection, 1985 to 1990: the pediatric experience
Pediatr Infect Dis J
 , 
2001
, vol. 
20
 (pg. 
733
-
40
)
43.
Block
SL
Yogev
R
Hayden
FG
Ambrose
CS
Zeng
W
Walker
RE
Shedding and immunogenicity of live attenuated influenza vaccine virus in subjects 5–49 years of age
Vaccine
 , 
2008
, vol. 
26
 (pg. 
4940
-
6
)
44.
Vesikari
T
Karvonen
A
Korhonen
T
, et al.  . 
A randomized, double-blind study of the safety, transmissibility and phenotypic and genotypic stability of cold-adapted influenza virus vaccine
Pediatr Infect Dis J
 , 
2006
, vol. 
25
 (pg. 
590
-
5
)

Author notes

Presented in part: 48th Annual Meeting of the Infectious Disease Society of America, Vancouver, Canada, 21--24 October 2010.
a
Present affiliation: Department of Pediatrics, University of South Florida, Tampa, Florida (S. C.); Division of Pediatrics, The University of Texas MD Anderson Cancer Center, Houston, Texas (N. C. D.); Department of Pediatrics, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts (C. R.-G.).