This study was supported by a grant from the Murdoch Children's Research Institute (Royal Children's Hospital, Melbourne, Australia). The World Health Organization Collaborating Centre for Reference and Research on Influenza is supported by the Australian Government, Department of Health and Ageing.
Community-acquired respiratory viruses such as influenza A and influenza B have long been recognized as important causes of morbidity in patients after solid organ transplantation with predictable complications, such as viral pneumonitis, secondary bacterial pneumonia, and acute graft rejection.1, 2
A rapidly spreading novel strain of the H1N1 influenza virus [pandemic H1N1 2009 influenza A (pH1N1/09)] was identified in April 2009, leading to the announcement of a pandemic by the World Health Organization on June 11, 2009. Initial epidemiological work identified both children and immunosuppressed patients as being particularly susceptible, and associated morbidity and mortality rates were higher in comparison with those for seasonal influenza.3 Against this background, a monovalent, nonadjuvanted pH1N1/09 vaccine (Panvax H1N1, CSL Biotherapies, Parkville, Australia) was introduced in September 2009 in Australia.
Excellent seroconversion rates (>90%) after a single dose of the vaccine in healthy subjects3, 4 translated into national and international recommendations for a single-dose regimen for individuals 10 years old or older (including immunosuppressed patients).5, 6 Although recent epidemiological studies have been able to better describe the risk and burden of pH1N1/09 disease among solid organ transplant recipients,7-11 no data, to the best of our knowledge, are currently available on the effectiveness of this single-dose vaccination approach in this high-risk group.
The aim of this study was to assess the immunogenicity of a single dose of the pH1N1/09 vaccine in a cohort of pediatric liver transplant recipients ≥10 years of age.
CI, confidence interval; GMT, geometric mean titer; HAI, hemagglutination inhibition; NA, not applicable; pH1N1/09, pandemic H1N1 2009 influenza A.
PATIENTS AND METHODS
A purified, inactivated, monovalent, split-virion, and nonadjuvanted pH1N1/09 vaccine (Panvax H1N1) was available at the end of the 2009 influenza season in Australia; it was licensed in September 2009. The vaccine contained 15 μg of hemagglutinin per 0.5-mL dose of A/California/7/2009 (H1N1), the virus strain recommended by the World Health Organization for the pH1N1/09 vaccine.
All liver transplant recipients from the Royal Children's Hospital Liver Transplant Program (Melbourne, Australia) who were 10 years old or older were considered for recruitment. Informed consent was obtained from the parents or guardians of all subjects and from participants (if applicable). Exclusion criteria included the following: anaphylactic hypersensitivity to egg protein, neomycin, polymyxin B sulfate, or any of the constituents of the vaccine; previous laboratory-confirmed pH1N1/09 infection; and liver transplantation less than 3 months prior to the time of recruitment. Minor respiratory symptoms during the winter of 2009 before the study that did not lead to physician contact were not documented during recruitment. Ethics approval for the study protocol was obtained from the human research ethics committee of the Royal Children's Hospital.
The study was carried out as a prospective, observational cohort study between November 2009 and October 2010. All study participants were vaccinated in accordance with the recommendations of the Department of Health and Ageing and the Australian Technical Advisory Group on Immunisation.12 The participants received a single 0.5-mL injection of Panvax H1N1 in the deltoid muscle at baseline. Venous blood samples were taken during routine visits to the liver transplant outpatient clinic before vaccination and at least 6 weeks thereafter. The schedule of postvaccination sampling was determined by the timing of the next routine visit to the outpatient department. An upper time limit was not included in the study protocol. A full blood count, comprising white cell and lymphocyte counts, liver function tests, and measurements of trough levels of immunosuppressive medications such as tacrolimus and cyclosporine, were performed for all samples. Specimens for antibody testing were separated and stored at −20°C until they were processed further. Clinical data were documented (eg, age, gender, pretransplant diagnosis, time since transplantation, immunosuppressive regimen, and episodes of rejection until the end of the study period).
Following antibody analysis towards the end of the intended study, those patients with a lack of seropositivity (seroprotection) after initial vaccination were offered a second dose of the vaccine with a further assessment of the antibody response 6 weeks later.
Antibody Testing and Immunogenicity
Batched serum samples, frozen on dry ice, were sent to the World Health Organization Collaborating Centre for Reference and Research on Influenza (Victorian Infectious Diseases Reference Laboratory, Melbourne, Australia) for antibody testing. The humoral immune response was evaluated with a hemagglutination inhibition (HAI) assay for the A/California/7/2009 (H1N1) virus. The HAI assays were performed for all prevaccination and postvaccination samples with a previously described method.13, 14 Titers were expressed as reciprocals of the highest dilution of serum that prevented hemagglutination. Seroconversion was defined as a 4-fold increase in the antibody titer from the baseline value with a minimum titer of 40 in the postimmunization sample. Seropositivity was defined as an HAI titer of 40 and more. This putative threshold had previously been observed to correlate with 50% protection against seasonal influenza strains in challenge studies.15 Seropositivity was, therefore, used synonymously with seroprotection.
Outcome Measures and Analysis
The main immunological endpoint was the percentage of vaccine recipients who fulfilled the criteria of seroconversion and seropositivity after 1 dose of Panvax H1N1. Seroconversion and seropositivity rates of those receiving a second dose of the vaccine were included in the analysis after the intended study. The results were compared to the outcomes of a pH1N1/09 vaccination study previously published by Zhu et al.3 Their study involved a subgroup of healthy adolescent children who were vaccinated with 2 doses of a nonadjuvanted pH1N1/09 vaccine (15 μg) 14 days apart; the seroconversion and seroprotection rates were 94.3% and 97.1%, respectively, after the first dose and 98% and 100%, respectively, after the second vaccine dose (see the supporting information for the article by Zhu et al.3).
After antibody analysis, geometric mean titers (GMTs) with 95% confidence intervals (CIs) were calculated and compared for the subgroups of baseline-seropositive patients and baseline-seronegative patients. GMT differences were considered significant in the setting of nonoverlapping 95% CIs.
The group of patients displaying seroconversion after the first dose of the vaccine was compared with the nonconverters with respect to possible differences in clinical and laboratory parameters using Student's t test.
Twenty-eight pediatric liver transplant recipients older than 10 years were considered for participation in the study. Seven patients had to be excluded: 4 failed to give consent, 1 suffered from an anaphylactic egg protein allergy, and 2 had already received the pH1N1/09 vaccine before the study. Twenty-one patients completed the first vaccination between October 2009 and May 2010 and were included in the final analysis; 11 of these patients were male. The median time between vaccination and postvaccination blood sampling was 10.0 weeks (range = 6.0-18.9 weeks). The median age at the first vaccination was 12.8 years (range = 10.1-18.5 years), and the median time since transplantation was 6.9 years (range = 0.4-11.5 years). The indications for transplantation included extrahepatic biliary atresia (n = 10), congenital metabolic diseases (n = 6), autoimmune liver disease (n = 1), drug-induced liver failure (n = 2), and idiopathic acute liver failure (n = 2). Immunosuppressive medications included the following: tacrolimus monotherapy (n = 16); cyclosporine A monotherapy (n = 2); a combination of tacrolimus and azathioprine (n = 1); a combination of tacrolimus, azathioprine, and steroids (n = 1); and a combination of cyclosporine and mycophenolate mofetil (n = 1).
After the first vaccination, 13 of 21 patients (62%) fulfilled the criteria of seroconversion, and 14 of 21 patients (66.6%) those of seropositivity (Fig. 1). At baseline, 7 of 21 patients (33.4%) were already seropositive. Six of 7 baseline-seropositive patients (86%) seroconverted, whereas 7 of 14 baseline-seronegative patients (50%) seroconverted. The individual HAI titer changes of the seroconverters and nonconverters after the first dose of the vaccine are summarized in Fig. 2.
Of the 8 patients who did not seroconvert after the first dose of the vaccine, those without baseline seroprotection (n = 7) were offered a second vaccine dose (median time between vaccinations = 208 days, range = 146-238 days). Six agreed to proceed, and 1 did not attend further vaccinations. Four of the 6 patients (67%) seroconverted after a second dose of the vaccine. Hence, the cumulative seroconversion rate after both doses of the vaccine was 89.5% (ie, 17 of 19 patients; the patient who was baseline-seropositive and did not seroconvert and the nonconverter who did not attend the second vaccination were excluded). The cumulative seroprotection rate was 90% (18 of 20 patients). The seroconversion and seroprotection results for our cohort are summarized in Fig. 3 and are compared with the results for a subgroup of healthy adolescent children who were vaccinated with 2 doses of a comparable pH1N1/09 vaccine (see the supporting information for the article by Zhu et al.3).
Antibody titers and GMTs are summarized in Table 1. The GMTs after the first dose of the vaccine were considerably higher in the baseline-seropositive group versus the baseline-seronegative group. A significant difference in the GMT with nonoverlapping CIs was found in the baseline-seronegative group after vaccination, whereas no significant difference was seen in the GMT of the baseline-seropositive patients after vaccination.
Table 1. HAI Assay Results at the Baseline and After the First (n = 21) and Second Vaccinations (n = 6) for Baseline-Seropositive and Baseline-Seronegative Participants
pH1N1/09 Sampling Time
HAI Titer ≥ 40, n/N (%)
HAI Titer ≥ 160, n/N (%)
GMT (95% CI)
HAI Titer ≥ 40, n/N (%)
HAI Titer ≥ 160, n/N (%)
GMT (95% CI)
6.1 (−6.2 to 18.4)
≥6 weeks after first dose
≥6 weeks after second dose
In comparison with the nonconverters, the patients with successful seroconversion displayed no differences with respect to age, gender, immunosuppressive regimen, calcineurin inhibitor levels, lymphocyte counts, or the time between vaccination and postvaccination blood sampling (Table 2). Patients who underwent transplantation more recently were less likely to mount a successful immune response to the vaccine (P = 0.03, univariate analysis). The distribution of positive and negative seroresponses with respect to the time since transplantation is displayed in Fig. 4. There was no apparent time point post-transplantation when seroconversion was more likely to occur.
Table 2. Comparison of the Clinical and Laboratory Characteristics of Seroconverters and Nonconverters After 1 Dose of the pH1N1/09 Vaccine
Seroconverters (n = 13)
Nonconverters (n = 8)
Age (years), median (range)
Time since transplant (years), median (range)
Time between vaccination and postdose titer (weeks), median (range)
1 (plus tacrolimus and azathioprine)
1 (plus cyclosporine)
Tacrolimus level (μg/L), median (range)
Cyclosporine level (μg/L), median (range)
Lymphocyte count (×109/L), median (range)
No episodes of rejection were recorded during a median follow-up of 11 months (range = 8-13 months).
Since the declaration of the H1N1 pandemic by the World Health Organization, data on the clinical course and disease impact have been steadily increasing. The infection has been reported to be most common in the younger population, and particularly severe disease has been found during pregnancy and in traditional high-risk groups such as immunosuppressed patients.16, 17 Studies investigating the role of pH1N1/09 infection in transplant patients have been limited to retrospective and observational case series.7, 18, 19 The range of symptoms is comparable to the symptoms of seasonal influenza (from a self-limiting upper respiratory disease to respiratory failure). Morbidity and mortality rates in this group of patients are substantial, and they have high rates of hospitalization and admission to intensive care.7, 17, 19-21 Infection-associated episodes of cellular rejection have been described.2, 8, 22 Therefore, vaccination against pH1N1/09 appears to be essential and is strongly recommended as part of the pretransplant workup and posttransplant care.23, 24 However, little is known about the efficacy of the available pH1N1/09 vaccines, and all national and international vaccination guidelines are based on immunogenicity data emanating from healthy study participants.
This study provides for the first time data on the immunogenicity of a pH1N1/09 vaccine in a small cohort of pediatric liver transplant patients. Here, a single dose of the vaccine given to liver transplant patients who were 10 years old or older failed to achieve the seroconversion and seroprotection rates found in their healthy, nontransplant counterparts. These results follow the outcomes of recent studies addressing the immunogenicity of seasonal and pH1N1/09 influenza vaccines in solid organ22, 25 and stem cell transplant recipients.26
In contrast to previous experience27, 28 and the currently recommended single-dose approach for older children,5, 12for nearly 20% of our patients, a second dose of the vaccine was necessary to achieve seroconversion. Vaccination immunogenicity, measured by antibody or T cell responses to mitogens, appeared to be generally reduced in comparison with healthy controls.29
Because the generation of an immune response to vaccination with viral components is dependent on T and B cell function,30 the efficacy of vaccination in transplant patients likely depends on the level of immunosuppression. High doses of steroids and the use of T and B cell–depleting agents have been reported as important factors in blunting an immune response to vaccination.22 Our study, limited by small numbers, fails to provide this link because there were no significant differences in the immunosuppressive regimens and calcineurin inhibitor levels between the seroconverters and nonconverters (Table 2).
In agreement with previous transplant vaccination studies,26, 27 time since transplantation positively correlated with seroconversion in our patient cohort. However, the high percentage of preexisting seropositivity within the seroconversion group (further discussed later) may have acted as a possible confounder: preexisting memory rather than the time since transplantation may have been the more important contributor to seroconversion.
The high rate of preexisting baseline-seroprotective titers in the seroconverted patient group is an important observation. Although pH1N1/09 infection rates reached a climax in Victoria, Australia between May and July 2009, the H1N1/09 vaccine was not licensed until early September 2009. The first patient of this study was recruited and vaccinated in October 2009, 3 months after the peak pandemic activity. Thus, the high rate of baseline seropositivity most likely represents exposure to pH1N1/09 before vaccination with subsequent mild or subclinical infections not leading to pH1N1 testing. This corresponds well with the results from seroprevalence studies in school-age children, which indicated high rates of oligo- and asymptomatic infections during the winter of 2009.31 Because the seroconversion rate was higher in the baseline-seropositive group than in the baseline-seronegative group (86% versus 50%; Fig. 1), vaccination seemed to act as a booster in this group of patients enhancing the primed immune response. Hence, the overall seroconversion rate of 62% documented in our study is most likely an overestimate for immunosuppressed transplant patients. Nonetheless, because baseline seropositivity correlates with seroconversion, it is conceivable that in a nonpandemic situation, adolescents with preexisting immunity due to pH1N1 infection or vaccination in the previous year could potentially generate higher seroconversion rates to one dose of seasonal vaccine than documented in our study and thus not requiring a booster.
In accordance with previous work on influenza vaccination in solid organ transplantation,27, 28 we were unable to identify an increased rate of hepatocellular rejection during the study period.
The most important limitation of this study is the small number of participants. In addition, like the results of most other vaccine immunogenicity studies, the results of this study are restricted by the fact that the efficacy of the vaccine was assessed by the determination of surrogate markers of immunoprotection (seroconversion and seropositivity) rather than by the determination of its ability to prevent laboratory-confirmed influenza infections. It is conceivable that immunosuppressed patients might require higher antibody titers than the putative seroprotective level of 40 in order to be clinically protected.
In summary, we observed a modest rate of seroconversion after a single dose of the pH1N1/09 vaccine in a small cohort of pediatric liver transplant patients. A second dose of the vaccine is necessary to achieve an immune response comparable to that of healthy controls, at least in a pandemic setting. The actual seroconversion rates to the pH1N1/09 antigen in a nonpandemic scenario still need to be investigated. Further research into new vaccine strategies in order to improve and monitor vaccine responses in immunosuppressed patient groups is urgently needed.