Cellular and humoral immunity elicited by influenza vaccines in pediatric hematopoietic-stem cell transplantation

Human Immunology 73 (2012) 884–890

Contents lists available at SciVerse ScienceDirect

www.ashi-hla.org

journal homepage: www.elsevier.com/locate/humimm

Cellular and humoral immunity elicited by influenza vaccines in pediatric hematopoietic-stem cell transplantation Valérie Guérin-El Khourouj a, Marie Duchamp a, Anne Krivine b, Béatrice Pédron a, Marie Ouachée-Chardin c, Karima Yakouben c, Marie-Louise Frémond c, André Baruchel c, Jean-Hugues Dalle c, Ghislaine Sterkers a,⇑ a b c

Laboratory of Immunology, Robert Debré Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Paris 7 Denis Diderot University, Paris, France Department of Virology, Cochin Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), France Department of Haematology, Robert Debré Hospital, Assistance Publique-Hôpitaux de Paris (AP-HP), Paris 7 Denis Diderot University, Paris, France

a r t i c l e

i n f o

Article history: Received 27 March 2012 Accepted 11 July 2012 Available online 20 July 2012

a b s t r a c t Immunity induced by influenza vaccines following hematopoietic stem-cell transplantation (HSCT) is poorly understood. Here, 14 pediatric recipients (mean age: 6 years) received H1N1 (n = 9) or H1N1/H3N2 (n = 5) vaccines at a median of 5.7 months post-HSCT (HLA-identical related bone-marrow graft: 10/14). Fourteen clinically-matched non-vaccinated recipients were included as controls. Cellular response to vaccination was assessed by a T-cell proliferation assay. Humoral response was assessed by H1N1-specific antibody titration. IL2 and IFNc responses to influenza were also evaluated by an intracellular cytokine accumulation method for some of the recipients. Higher proliferative responses to H1N1 (p = 0.0001) and higher H1N1-specific antibody titers (p < 0.02) were observed in vaccines opposed to non-vaccinated recipients. In some cases, proliferative responses to H1N1 developed while at the same time antibody titers did not reach protective (P1:40) levels. Most recipients vaccinated with only the H1N1 strain had proliferative responses to both H1N1 and H3N2 (median stimulation index H1N1: 96, H3N2: 126 in responders). Finally, IL2 responses predominated over IFNc responses (p < 0.02) to influenza viruses in responders. In conclusion, H1N1 vaccination induced substantial cell-mediated immunity, and to a lesser extent, humoral immunity at early times post-HSCT. H1N1/H3N2 T-cell cross-reactivity and protective (IL2) rather than effector (IFNc) cytokinic profiles were elicited. Ó 2012 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved.

1. Introduction The outbreak of H1N1 swine-origin influenza A virus in April 2009 [1,2] raised a new threat to public health because most humans under 60 years of age lack antibodies against the virus [3]. Hematopoietic-stem cell transplant (HSCT) recipients have an increased risk for more severe infections [4–6]. Annual influenza immunization is therefore recommended for these patients starting at six months after HSCT [7–11]. In the case of a pandemic, Abbreviations: HSCT, hematopoietic-stem cell transplantation; GVHD, graft versus host disease; TIV, trivalent inactivated seasonal vaccine; PBMC, peripheral blood mononuclear cells; AdV, adenovirus; Cpm, counts per minute; SI, stimulation index; HA, hemagglutinin; NA, neuraminidase. ⇑ Corresponding author. Address: Hôpital Robert Debré-Laboratoire d’Immunologie, 48 Boulevard Sérurier, 75019 Paris, France. Tel.: +33 1 40 03 53 05; fax: +33 1 40 03 47 76. E-mail address: [email protected] (G. Sterkers).

the recommendation is to vaccinate recipients earlier, from the third month post-transplantation, although vaccine efficacy is expected to be less at that time. The aim of this study was the evaluation of times to transplantation compatible with vaccine immunogenicity in transplanted children during the 2009–2010 pandemic. The immune response to influenza consists of both antibody and cell-mediated responses [12,13]. All studies reported so far agree on the indication of low antibody responses to influenza vaccines in HSCT recipients, especially at early times post-transplantation [7,11,14–16]. Studies on both cellular and humoral responses to influenza vaccination in HSCT settings are scarce and performed in limited series of patients. Furthermore, data from these studies are controversial. Two studies, from the same center and performed in 14 adult HSCT recipients, reported relatively poor antibody and IFNc responses to influenza vaccines [11]. Optimal in vitro proliferative responses to influenza was however observed

0198-8859/$36.00 - see front matter Ó 2012 American Society for Histocompatibility and Immunogenetics. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.humimm.2012.07.039

885

V. Guérin-El Khourouj et al. / Human Immunology 73 (2012) 884–890

in a very limited series of four pediatric recipients [17]. In this latter study, effective cellular responses contrasted with defective antibody responses. In the present study, we assessed the immunogenicity of the 2009 pandemic (pdm) H1N1 vaccines in 14 pediatric recipients. Both humoral (H1N1-specific antibody titration) and cellular immune responses to the vaccines were evaluated. CD8+ T-cells form a part of the immune defence against influenza viruses [18,19]. Yet, certain types of vaccines, such as preparations from inactivated viruses, that are recommended for immune-compromised individuals, induce CD4+ T-cell responses but marginal if any CD8+ T-cell responses. We therefore focused on the CD4+ T-cell subset to assess cellular immune responses to influenza vaccine in this study. The sensitive 3H-Thymidine assay was used to assess immunogenicity. Cytokine (IL2 and/or IFNc) responses to influenza viruses also addressed the quality of CD4 Tcell responses in a portion of the recipients. Finally, the potential induction by influenza vaccines of T-cell cross-reactivity to distinct strains was evaluated by the use of A/PR8 (H1N1) and A/x31 (H3N2) viruses for in vitro stimulation. 2. Material and methods 2.1. Populations In this case series study, 42 consecutive patients transplanted from January 2009 to January 2010 were initially considered. In the absence of clear recommendations, the decision whether to vaccinate or not the recipients at early times post-HSCT was up to the choice of the physicians who took care of the patients. Ongoing graft-versus host disease (GVHD) and/or symptomatic infection at time of H1N1 vaccine availability (December 2009) was however a contraindication for all physicians and informed consent of the parents or the guardians of the index cases was required. Fourteen recipients were excluded because of early death or relapse (n = 11), vaccination with the trivalent inactivated seasonal (TIV) but not with the pandemic H1N1 vaccine (n = 1) or for technical reasons (unavailability of blood samples: n = 2). Among the 28 remaining patients, 14 had been H1N1 vaccinated and 14 had not. Seven H1N1-vaccinated healthy individuals from the medical staff of the transplant center were included as controls. The study was approved by the local ethics committee of Robert Debre Hospital (Paris, France).

PandemrixÒ formulation was adjuvanted (adjuvant AS03). The PanenzaÒ formulation was not. The 2009–2010 TIV VaxigripÒ vaccine (Sanofi Pasteur) contained the A/Brisbane 59/07 (H1N1), A/ Brisbane 10/07 (H3N2), B/Brisbane 60/08 (B) strains. All vaccines were inactivated. The vaccinated recipients were immunized (0.5 ml per dose) in the deltoid muscle from December 2009 to March 2010. The median lag time from HSCT to vaccination was 171 days (76–330 days). The seven healthy controls from the medical staff received one dose of PandemrixÒ (n = 5) or one dose of PanenzaÒ (n = 2) followed by one dose of VaxigripÒ at four weeks intervals during the same period (December 2009–February 2010). 2.3. Blood sampling For ethical reasons (no blood selectively drawn for the study) and/or technical reasons (low CD4 counts) data on cellular immunity to influenza was not available before vaccination for most patients. Residual cells from blood samples drawn on ACD anticoagulant at months 6 and/or 12 post-HSCT and dedicated to routine care (immune recovery evaluation) were used for post-vaccination cellular immunity investigations. Influenza-specific antibodies were titrated before and after vaccination in residual sera dedicated to immunoglobulin dosage for routine care. 2.4. Viral antigens used for in vitro T-cell stimulation Influenza prototypes from the collection of the National Influenza Reference Center (Lyon, France) were used. Influenza A/Puerto-Rico/8/34 (A/PR8) is a H1N1 virus known to induce CD4+ T-cell cross-reactivity between all H1N1 strains [20,21]. Influenza A/x31 is a reassortant virus which inherits six genome segments from A/ PR8 and two from a virus of the H3N2 Hong-Kong epidemic strain, these segments encoding the hemagglutinin and the neuraminidase glycoproteins [22]. These strains were selected for their capacity to reveal cross-reactivity. As previously described, viruses were grown in the allantoic cavity of embryonated chicken eggs, purified by sucrose gradient centrifugation and UV inactivated [20]. Titers were determined by hemaglutinin (HA) assays and were expressed as HA units/ml. Viruses were stored at 80 °C until use. Inactivated adenovirus (Virion, Zurich, Switzerland) was used as a control antigen.

2.2. Vaccines 2.5. T-cell proliferation assays The influenza A H1N1 2009 PanenzaÒ (Sanofi-Pasteur, Lyon, France) and PandemrixÒ (Glaxo Smith Kine, Marly Le Roi, France) vaccines contained the A/California/7/2009-like strain. The

T-cell proliferation assays were performed as previously described [23]. In brief, peripheral blood mononuclear cells (PBMCs)

Table 1 Immunogenicity according to vaccine protocols. Vaccination protocols

a b c

Vaccine immunogenicity

First dose

Second dose

Third dose

Numbers of recipients

Number of recipients with cellular immunity to H1N1 among vaccines

Number of recipients with protective humoral immunity to H1N1 among evaluable vaccines

PandemrixÒa PandemrixÒ PandemrixÒ PandemrixÒ Panenzab PanenzaÒ

— PandemrixÒ VaxigripÒc PandemrixvÒ PanenzaÒ PanenzaÒ

— — — VaxigripÒ — VaxigripÒ

n=1 n=2 n=2 n=1 n=6 n=2

0/1 1/2 1/2 1/1 6/6 2/2

ND ND 0/2 1/1 3/4 1/1

PandemrixÒ from Glaxo Smith Kline: adjuvanted pandemic (2009) H1N1 monovalent inactivated vaccine. PanenzaÒ from Sanofi-Pasteur: non-adjuvanted pandemic (2009) H1N1 monovalent inactivated vaccine. VaxigripÒ from Sanofi-Pasteur: seasonal (2009–2010): H3N2 (A/Brisbane/10/2007) + H1N1 (A/Brisbane/59/07) + B (B/Brisbane/60/08) trivalent inactivated vaccine.

886

V. Guérin-El Khourouj et al. / Human Immunology 73 (2012) 884–890

Table 2 Main patient characteristics. Characteristics

H1N1-vaccinated HSCT (n = 14)

Mean age, years, 6y 4m (5m 16y) months (range) Male/female, n 9/5 Underlying disease, n Acute leukemia 9 1 JMMLa Aplastic anemia 1 Others 3b Conditioning regimen Myeloablative, n (%) 13 (93%) TBI 4 (28%) ATG 5 (36%) HLA donor matching, n Bone marrow: 6/6 HLA-matched 10 related P9/10 HLA-matched 1 unrelated Cord-blood: P4/6 HLA matched 3 unrelated Acute GVHD (preceding vaccination period) Maximum grade 0–2 11 Maximum grade 3–4 3 Chronic GVHD (at time of vaccination period) Limited 0/14 Extensive 0/14 Documented H1N1 0 infection

Non H1N1-vaccinated HSCT (n = 14) 5y 9m (9m

15y)

9/5 9 3 1 1c 13 (93%) 3 (21%) 7 (50%)

4 7

3

9 5 3/14 0/14 1

TBI: total body irradiation; ATG: antithymoglobulin. a JMML: juvenil myelomonocytic leukemia. b Sickle-cell anaemia (n = 2), Hurler (n = 1). c Myeloproliferative syndrome (n = 1).

described [25]. PBMCs were cultured at 106/ml for 18 h in culture medium supplemented with recombinant IL2 (10 IU/mL; Roche diagnostics, Meylan, France), 2 lg/mL of anti-CD28 antibody (Beckman Coulter, Marseille, France) and 2 lg/mL of anti-CD49d antibodies (BD biosciences, Le Pont de Claix, France). Optimal amounts of H1N1, H3N2 or AdV antigens were used for virusspecific stimulations [20,24]. Antigens were omitted in negative controls. Cultures containing 0.2 lg/mL of CD3 antibodies (Janssen-Cilag, Issy Les Moulineaux, France) were used as positive controls. Brefeldin A (10 lg/ml, Sigma, Saint-Quentin-Fallavier, France) was added after 1 h. After fixation and permeabilisation (IntraStain Kit, Dako, Cytomation, Glostrups, Denmark), cells were labeled with CD3-FITC (Beckman-Coulter), CD4-PerCP, IFNc-PE and IL2-APC (BD Biosciences). Cells were analyzed on a Navios instrument (Beckman-Coulter, Villepinte, France) using CXP analysis software (Beckman-Coulter). Files were gated on small CD3+/CD4+ lymphocytes. A minimum of 100,000 events in the lymphocyte gate was acquired. The percentage of cells secreting IFNc and/or IL2 was calculated as the percentage of cells secreting cytokines in stimulated cultures minus the percentage of unstimulated cells secreting cytokines (background). Results were considered positive when P0.03% and P3-fold the background. 2.8. Statistics The Fisher exact test was used to compare percentages. The Mann Withney test was used to denote differences between groups. The rank test of Wilcoxon was used to compare paired samples. A p value 60.05 was considered significant.

3. Results were cultured for 6 days with predetermined optimal concentrations of H1N1, H3N2 (5HA units/ml) or AdV antigen [20,24]. Unstimulated cells were used as negative controls. Cells stimulated by phytohemagglutinin were used as positive controls. [3H]-Thymidine was added for the last 18 h of culture. Results are given as counts per minute (cpm) and stimulation index (SI) (SI: cpm in stimulated cultures divided by cpm in unstimulated cultures). Based on data found in vaccinated controls, SI P 9 associated with cpm P 5000 was considered as a positive response to influenza. SI P 5 associated with cpm P 3000 was previously determined as a positive response to AdV [24]. 2.6. Serum antibody titers Sera were tested against the pandemic 2009 H1N1 strain. Antibody titers were determined by a standard inhibition of hemagglutination method using human 0Rh-negative red blood cells and PanenzaÒ vaccine as antigen. All sera were treated with receptordestroying enzyme and heat-inactivated at 56 °C for 30 min to remove non-specific inhibitors. Twofold dilutions of sera beginning at 1:10 were tested against four hemagglutinin units of antigen. The titer of HAI antibodies was defined as the reciprocal of the highest serum dilution that completely inhibited hemagglutination. Sera whose titers were <10 were assigned a titer of five for calculation purposes. Seroprotection was defined as specific antibody titer P1:40 and seroconversion as a fourfold titer increase. 2.7. Enumeration of antigen-specific cytokine secreting cells IFNc and/or IL2-secreting cells were enumerated in a flow cytometry-intracellular cytokine staining assay as previously

3.1. Patients The protocols of vaccination (vaccine composition and number of doses) were up to the choice of the physician who took care of the patients. They are detailed in Table 1. Eight recipients received one or two doses of H1N1 non-adjuvanted vaccine (PanenzaÒ) either alone (n = 6) or followed by one dose of the trivalent seasonal vaccine (TIV, VaxigripÒ, n = 2). Six recipients received one or two doses of the adjuvanted H1N1 vaccine (PandemrixÒ) either alone (n = 3) or followed by one dose of the TIV (VaxigripÒ, n = 3). The median lag time from HSCT to vaccination was 171 days (76–336) in the 14 vaccinated recipients. The main clinical characteristics of the study populations are given in Table 2. Note that the vaccinated and non-vaccinated populations were matched for gender, age, underlying disease and conditioning. The non-vaccinated group were however more likely to receive a grant from unrelated donors and to present with cGVHD during the vaccination period. No severe adverse reactions to the vaccines were reported. Especially, there was no worsening of GVHD during the first month following vaccination. The only recipient with documented H1N1 influenza infection (H1N1-PCR positivity) was not vaccinated. 3.2. Proliferative responses to H1N1 and to H3N2 The lower level of proliferative response to H1N1 was SI = 9 in vaccinated healthy individuals used as controls. SI P 9 was therefore used to denote a positive response in patients. Four recipients were tested before vaccination. Proliferative responses to H1N1 were at background, in all four as expected from prior myeloablative conditioning (median SI = 1.5; range: 1–4).

V. Guérin-El Khourouj et al. / Human Immunology 73 (2012) 884–890

Stimulation index

A

1000

p = 0.0001

p = NS

p = 0.0009

115

100

108 52

42 10

2 1

1

H1N1(1) (n = 14) (n= 13)

H3N2 (n = 11) (n= 12)

AdV (n= 13) (n= 14)

B H1N1 specific antibody titers

10000

p < 0,02

* 1000

3.3. Antibody response to H1N1

Vaccinated

% secreting cells

% of CD4 secreting cytokines

the non-vaccinated, non H1N1-infected recipients (median SI = 1, range: 1–40). In more detail, SI values over the cut-off of positivity were observed in 11/14 vaccinated but 1/13 non-vaccinated and non-infected recipients (Fig. 1A). Of the recipients immunized with the TIV vaccine (H1N1 + H3N2) 3/3 responded to both H1N1 and H3N2 strains. Seven recipients vaccinated against the only pandemic H1N1 strain responded to H1N1. 5/7 (70%) also responded with similar intensity levels to H3N2. (H1N1 median SI = 96, range: 17–257; H3N2 median SI = 126, range: 14–225). The non-vaccinated recipient with positive H1N1 PCR in nasal fluids 27 weeks before immunological investigations also reacted to both H1N1 (SI = 61) and H3N2 (SI = 43). Adenovirus (AdV) was previously shown to immunize 60–70% of pediatric recipients within the first six months post-HSCT [24]. Proliferative reactivity to AdV was therefore used to assess the overall immune status of recipients. Similar intensity levels of proliferative responses to AdV were observed in vaccinated and nonvaccinated responders (median SI = 108 and 52, respectively).

100

≤ 10

C

887

1

0.1

Non-vaccinated Non-infected

p = 0.02

0.08

0.02 0.01 0.01

H1N1

H3N2

Fig. 1. Immune responses to influenza vaccine. (A) Proliferative responses: Results are expressed as individual values of stimulation index (SI). PBMC from vaccinated (empty symbols) and non-vaccinated (full symbols) recipients were cultured with H1N1 (triangles), H3N2 (circles) or AdV (squares) antigens. The median values are indicated (—) for each group (median of all values for influenza stimulation and median of positive values for AdV stimulation). The horizontal dashed lines indicate the threshold of positivity for H1N1 and H3N2 (SI: 9, that is the lower value found in healthy vaccinated controls) and the threshold of positivity for AdV (SI: 5, that discriminates immune from non-immune individuals). p values (Wilcoxon test) <0.05 were considered significant. The non-vaccinated but H1N1-infected recipient was excluded from analysis. (B) Anti-H1N1 antibody responses: Individual antibody titers at a median time of 55 days post-vaccination (range 40–111) in eight recipients vaccinated at a median time of 165 days post-HSCT are reported. Individual values in four non-vaccinated recipients following the vaccination period are also reported. ⁄Pre-vaccination antibody titer was 1:80 in this recipient. Wilcoxon test was used to make comparisons. (C) Cytokine responses to H1N1 and H3N2: An intracytoplasmic-cytokine assay was used to denote the percentages of CD4+ T-cells among total CD4+ T-cells secreting cytokines (IL2+/IFNc + IL2+/ IFNc+ + IL2 /IFNc+). Stimulation used H1N1 (triangle) or H3N2 (circle) virus strains. Empty symbols represent vaccinated recipients. Full symbols represent nonvaccinated recipients (the one H1N1 infected was excluded). Background (percentage of positive cells in non-stimulated cultures) was deduced. The horizontal dashed line indicates the threshold of positivity (0.03%). Wilcoxon test was used to make comparisons.

The median lag time from HSCT to T-cell proliferation assay was 335 days (range: 211–517) in vaccinated recipients and 335 days (range: 188–380) in non-vaccinated recipients. Higher (p = 0.0001) intensity levels of proliferative responses were observed in vaccines (median SI = 42; range: 2–257) opposed to

Eight recipients vaccinated at a median of 107 days post-HSCT (range: 92–326) had anti-H1N1 antibody titers available after vaccination (median time post-vaccination 49 days; range 41–281). H1N1 antibodies were undetectable after vaccination in one vaccine who did not develop a proliferative response to H1N1. Of the seven recipients who developed proliferative responses to H1N1 after vaccination, five had titer levels P1:40, of whom one seroconverted (titers: 1:80 and 1:2560 before and after vaccination, respectively). Low antibody titers (1:10 and 1:20) contrasted with optimal T-cell proliferative responses to H1N1 (SI = 53 and 96) in the remaining two. As shown in Fig. 1B, H1N1 antibody levels were significantly lower (p < 0.02) in the four non-vaccinated recipients we analyzed (geometric mean 1260 and <10 in vaccinated and non-vaccinated recipients, respectively). Finally, anti-H1N1 antibody titer in the H1N1-infected recipient was 1:80 at 69 days, following infection. 3.4. Cytokine responses to H1N1 and to H3N2 Cytokine responses to H1N1 and to H3N2 viruses were evaluated in seven vaccines who showed proliferative responses to the two viral strains. Five non-vaccinated and non-H1N1 infected recipients and the seven healthy controls were also evaluated for comparison. As expected, the inactivated viruses used for in vitro stimulation mostly elicited responses by CD4+ T-cells (not shown). Higher (p = 0.02) percentages of H1N1-specific CD4+ T-cell secreting cytokines were observed in vaccines (median: 0.08%; range: 0.02–0.24%) opposed to the non-vaccinated recipients (Fig. 1C). Two recipients vaccinated against the only H1N1 strain and also the H1N1-infected non-vaccinated recipient evidenced cytokine responses to both H1N1 and H3N2. Fig. 2 illustrates the typical profiles of the cytokinic responses to H1N1, H3N2 and AdV in (i) one representative non-vaccinated recipient, (ii) one representative recipient vaccinated with the only PanenzaÒ (H1N1) vaccine, (iii) in the H1N1-infected recipient and (iv) in one representative vaccinated healthy control. A predominant IL2+/IFNc response to influenza viruses was observed in the vaccinated recipients. Overall, total IL2 response to influenza viruses (IL2+/IFNc + IL2+/IFNc+; median: 0.11%; range: 0.05–0.49) was higher (p = 0.014) than total IFNc response (IL2+/IFNc+ + IL2 / IFNc+; median: 0.03%; range: 0.01–0.09). In contrast, in the only H1N1-infected recipient, and in healthy controls did a polyfunctional (IL2+/IFNc+) response to influenza viruses predominated

888

V. Guérin-El Khourouj et al. / Human Immunology 73 (2012) 884–890

IL2

U ti Unstimulated l t d 0.01 %

0.00 %

AdV 0.06 %

H1N1 0.19 %

0.01 %

H3N2 0.00 %

0.01 %

0.00 %

A: Non vaccinated patient 0.01 0 01 %

0.12 0 12 %

0.01 %

0.00 0 00 %

IL2

IFN-γ IFN

0.01 %

B: H1N1 vaccinated patient

0.04 %

0.03 %

0.50 %

0.11 %

0.11 %

0.10 %

0.08 %

0.02 %

0.22 %

0.07 %

0.05 %

I IL2

IFN-γ

0.01 %

C: H1N1 influenzainfluenza infected patient

0.02 %

0.08 %

0.01 %

0.19 %

0.08 %

0.11 %

0.03 %

0.02 %

0.04 %

0.08 %

0.02 %

IL2 2

IFN-γ

D: Healthy vaccinated subject

0.06 %

0.01 %

0.08 %

0.03 %

0.24 %

0.06 %

0.05 %

0.03 %

0.01 %

0.08 %

0.12 %

0.04 %

IFN-γγ Fig. 2. Profiles of cytokines. Cytokines responses in unstimulated cultures and after simultaneous stimulations with AdV, H1N1 or H3N2 are shown as dot blots. Four representative subjects were used: A = one representative non-vaccinated HSCT recipient, B = one HSCT recipient vaccinated with the only H1N1 pandemic vaccine, C = the H1N1 influenza-infected and non-vaccinated recipient and D = one representative healthy control vaccinated with the H1N1 pandemic vaccine and TIV.

and the total number of IFNc secreting cells was over (p = 0.022) the total number of IL2 secreting cells in healthy controls.

4. Discussion According to myeloablative conditioning, which would be expected to abrogate confounding pre-existing cellular immunity, proliferative responses to influenza were at background in all recipients evaluable before vaccination. At a median time of 335 days post-HSCT, the levels of T-cell proliferative responses to influenza were significantly higher in vaccinees than in non-vaccinated recipients. Similar proliferative responses to AdV in vaccinees and non-vaccinated recipients linked these differences to vaccination and not to the overall immune status. Eight out of 10 recipients who received HLA-identical grafts from a sibling and three out of three recipients who received P4/6 HLA-matched cord blood units developed T-cell immunity to H1N1 (from the 3rd and 4th month post-HSCT, respectively).

These results support good vaccine immunogenicity in these two kinds of grafts. Only one patient had unrelated HSCT. No patient had ongoing GVHD at vaccination. Further research is needed to determine appropriate conditions to obtain the best response to the vaccination in these more vulnerable groups. Note also that 6/6 recipients vaccinated with two doses of the non-adjuvanted H1N1 vaccine proliferated to H1N1 (Table 1). Higher (p = 0.0007) median intensity levels were observed in these six patients than in the 13 non-vaccinated and non-infected patients (median SI: 82.5; range 27–257 and 1; range 1–40, respectively). This result supports good immunogenicity, at least for this protocol. The emergence of a novel influenza virus, against which individuals had little or no pre-existing immunity, provided the opportunity to assess the humoral response to the 2009 pandemic H1N1 vaccines. A poor rate (20–30%) of humoral responses to influenza vaccines has been reported in adults [11,14,26–28]. In our study 5/8 (61%) of recipients who were examined for antibody responses had a post-vaccination titer to H1N1 generally accepted as being

V. Guérin-El Khourouj et al. / Human Immunology 73 (2012) 884–890

protective (1:40). Both young age (associated with a better shortterm immune reconstitution) and GVHD exclusion might have contributed to the apparent good humoral response in our series. In line with our observation, Yalcin et al. reported seroconversion for H1N1 in 10/14 (71.4%) of recipients less than 9 years of age following vaccination with a trivalent inactivated vaccine (TIV) [29]. It is notable that low levels of antibody contrasted with optimal T-cell proliferative responses to H1N1 in two patients. In contrast to this observation, Haining et al. showed CD4+ T-cell responses, but no increase in serum antibody levels, in four children following TIV vaccination 4–22 months after allogeneic HSCT [17]. How an influenza-specific T-cell response, in the absence of influenza-specific antibodies, might be able to modify the course of the infection remains an unanswered question. Influenza A viruses are subdivided into subtypes based on their surface antigens: hemagglutinin (HA) or neuraminidase (NA). It is widely recognised that antibody responses to influenza vaccination or infection are directed mostly toward variable epitopes from viral surface-exposed proteins [30]. Current vaccines are developed to generate neutralizing antibodies that target the HA and NA of influenza strains now in circulation. Recent comparative effectiveness and efficacy research increasingly indicates however that humoral immunity alone may be insufficient to defeat the influenza virus [31]. In natural infection, epitopes recognized by T-cells are far more conserved among different influenza strains [32]. T-cells with broad cross-reactivity may contribute in preventing heterosubtypic infection [13,33]. As far as we know, only one study has suggested that H1N1 vaccination could induce CD4 T-cell reactivity to conserved epitopes shared by several H1N1 strains [34]. Here, we show a broad H1N1/H3N2 cross-reactivity following H1N1 vaccination. Altogether these two studies indicate that, as in natural immunization, CD4+ T-cell response to influenza vaccines can target conservative influenza epitopes. Continuing antigenic drift allows influenza viruses to escape antibody mediated recognition, and as a consequence, the vaccine currently in use needs to be altered annually. As suggested by Assarson et al., highly conserved epitopes recognized by T-cells may represent an alternative approach for the generation of more universal influenza virus vaccines [33]. In this study we also aimed to clarify the nature of the immune response to the vaccine. It is now clear that the magnitude of a T-cell response as measured by a single parameter does not reflect its full functional potential. Most studies on cellular immune responses to influenza vaccines focused on IFNc responses and more occasionally on Th1/Th2 balance [35,36], though there is compelling evidence that IL2 is a crucial factor in defining a protective T-cell response against viruses. We showed that IL2 responses were clearly elicited by H1N1 vaccines in transplanted children, in fact they even predominated over IFNc responses. In contrast, predominant IFNc and primarily IL2+/IFNc+ responses were observed in the only HSCT recipient with documented natural infection and also in healthy vaccinated adults in whom prior natural immunization likely contributed to the response. These results are in line with the paradigm that vaccinations of naive individuals with inactivated organisms in general induce a CD4+ T-cell memory response biased toward the production of IL2, whereas responses primed by infection with live viruses induce effector (IFNc+/IL2 ) and polyfunctional (IL2+/IFNc+) responses biased toward IFNc secretion [37]. How the distinct patterns observed in natural immunization, opposed to vaccination, influence protection remains to be determined. The strength of the study was inclusion of the largest pediatric series analyzed so far. Also, it analyzed both the cellular and humoral arms of immune defences to influenza vaccines. Finally, it took advantage of early vaccination recommendations in pandemic conditions to determine vaccine immunogenicity from

889

the 3rd month post-HSCT. Limitations were the absence of prevaccine studies in a portion of the recipients and heterogeneity of vaccination protocols. A new prospective design would be difficult to envision outside of a pandemic situation. This study: (i) provides further evidence that inactivated influenza vaccines are safe and immunogenic from the third month following genoidentical HSCT in children, (ii) demonstrates for the first time the immunogenicity of influenza vaccines from the fourth month following cord blood HSCT, (iii) the finding of CD4+ T-cell cross-reactivity to H1N1 and H3N2 antigens following vaccination with an H1N1 strain underlines the potential use of conserved peptides for influenza vaccination in HSCT settings, and finally (iv) our study indicates that various immune profiles can be elicited by inactivated influenza vaccines. Machado et al. reported influenza infection in 10.5% of vaccinated HSCT recipients [38]. Further vaccine efficacy trials could answer how various profiles could influence protection against influenza infection. Conflict of interest The authors declare that they have no conflict of interest in publishing this article. Acknowledgments We acknowledge the technical expertise of Guylaine Boiry, Anne-Marie Courchinoux, Priscilia Egremonte, Elodie Geneletti and Judith Tholle. We acknowledge the nurses and staff of the Department of Haematology without whom this research would not have been possible. Finally, we greatly thank Sophie CaillatZucman for revising and Céline Neto for preparing the manuscript. This work was partly supported by Assistance Publique, Hôpitaux de Paris. References [1] Dawood FS, Jain S, Finelli L, Shaw MW, Lindstrom S, Garten RJ, Gubareva LV, Xu X, Bridges CB, Uyeki TM. Emergence of a novel swine-origin influenza A (H1N1) virus in humans. N Engl J Med 2009;360:2605–15. [2] Fraser C, Donnelly CA, Cauchemez S, Hanage WP, Van Kerkhove MD, Hollingsworth TD, Griffin J, Baggaley RF, Jenkins HE, Lyons EJ, et al. Pandemic potential of a strain of influenza A (H1N1): early findings. Science 2009;324:1557–61. [3] Hancock K, Veguilla V, Lu X, Zhong W, Butler EN, Sun H, Liu F, Dong L, DeVos JR, Gargiullo PM, et al. Cross-reactive antibody responses to the 2009 pandemic H1N1 influenza virus. N Engl J Med 2009;361:1945–52. [4] Ljungman P, Ward KN, Crooks BN, Parker A, Martino R, Shaw PJ, Brinch L, Brune M, De La Camara R, Dekker A, et al. Respiratory virus infections after stem cell transplantation: a prospective study from the Infectious Diseases Working Party of the European Group for Blood and Marrow Transplantation. Bone Marrow Transplant 2001;28:479–84. [5] Raboni SM, Nogueira MB, Tsuchiya LR, Takahashi GA, Pereira LA, Pasquini R, Siqueira MM. Respiratory tract viral infections in bone marrow transplant patients. Transplantation 2003;76:142–6. [6] Kharfan-Dabaja MA, Velez A, Richards K, Greene JN, Field T, Sandin R. Influenza A/pandemic 2009/H1N1 in the setting of allogeneic hematopoietic cell transplantation: a potentially catastrophic problem in a vulnerable population. Int J Hematol 2011;91:124–7. [7] Engelhard D, Nagler A, Hardan I, Morag A, Aker M, Baciu H, Strauss N, Parag G, Naparstek E, Ravid Z, et al. Antibody response to a two-dose regimen of influenza vaccine in allogeneic T cell-depleted and autologous BMT recipients. Bone Marrow Transplant 1993;11:1–5. [8] Ljungman P, Engelhard D, de la Camara R, Einsele H, Locasciulli A, Martino R, Ribaud P, Ward K, Cordonnier C. Vaccination of stem cell transplant recipients: recommendations of the Infectious Diseases Working Party of the EBMT. Bone Marrow Transplant 2005;35:737–46. [9] Ljungman P, Nahi H, Linde A. Vaccination of patients with haematological malignancies with one or two doses of influenza vaccine: a randomised study. Br J Haematol 2005;130:96–8. [10] Dykewicz CA. Summary of the Guidelines for Preventing Opportunistic Infections among Hematopoietic Stem Cell Transplant Recipients. Clin Infect Dis 2001;33:139–44. [11] Avetisyan G, Aschan J, Hassan M, Ljungman P. Evaluation of immune responses to seasonal influenza vaccination in healthy volunteers and in patients after stem cell transplantation. Transplantation 2008;86:257–63.

890

V. Guérin-El Khourouj et al. / Human Immunology 73 (2012) 884–890

[12] Couch RB, Kasel JA. Immunity to influenza in man. Annu Rev Microbiol 1983;37:529–49. [13] Deng Y, Jing Y, Campbell AE, Gravenstein S. Age-related impaired type 1 T cell responses to influenza: reduced activation ex vivo, decreased expansion in CTL culture in vitro, and blunted response to influenza vaccination in vivo in the elderly. J Immunol 2004;172:3437–46. [14] Pauksen K, Linde A, Hammarstrom V, Sjolin J, Carneskog J, Jonsson G, Oberga G, Engelmann H, Ljungman P. Granulocyte-macrophage colony-stimulating factor as immunomodulating factor together with influenza vaccination in stem cell transplant patients. Clin Infect Dis 2000;30:342–8. [15] Ljungman P, Avetisyan G. Influenza vaccination in hematopoietic SCT recipients. Bone Marrow Transplant 2008;42:637–41. [16] Gandhi MK, Egner W, Sizer L, Inman I, Zambon M, Craig JI, Marcus RE. Antibody responses to vaccinations given within the first two years after transplant are similar between autologous peripheral blood stem cell and bone marrow transplant recipients. Bone marrow transplant 2001;28:775–81. [17] Haining WN, Evans JW, Seth NP, Callaway GD, Wucherpfennig KW, Nadler LM, Guinan EC. Measuring T cell immunity to influenza vaccination in children after haemopoietic stem cell transplantation. Br J Haematol 2004;127:322–5. [18] Estcourt MJ, Letourneau S, McMichael AJ, Hanke T. Vaccine route, dose and type of delivery vector determine patterns of primary CD8+ T cell responses. Eur J Immunol 2005;35:2532–40. [19] Moulton VR, Bushar ND, Leeser DB, Patke DS, Farber DL. Divergent generation of heterogeneous memory CD4 T cells. J Immunol 2006;177:869–76. [20] Sterkers G, Michon J, Henin Y, Gomard E, Hannoun C, Levy JP. Fine specificity analysis of human influenza-specific cloned cell lines. Cell Immunol 1985;94:394–405. [21] Lamb JR, Woody JN, Hartzman RJ, Eckels DD. In vitro influenza virus-specific antibody production in man: antigen-specific and HLA-restricted induction of helper activity mediated by cloned human T lymphocytes. J Immunol 1982;129:1465–70. [22] Baez M, Palese P, Kilbourne ED. Gene composition of high-yielding influenza vaccine strains obtained by recombination. J Infect Dis 1980;141:362–5. [23] Guerin V, Dalle JH, Pedron B, Ouachee-Chardin M, Yakouben K, Baruchel A, Sterkers G. Cellular immune parameters associated with spontaneous control of CMV in children who underwent transplantation. Bone Marrow Transplant 2010;45:442–9. [24] Guerin-El Khourouj V, Dalle JH, Pedron B, Yakouben K, Bensoussan D, Cordeiro DJ, Peltier L, Ouachee-Chardin M, Baruchel A, Sterkers G. Quantitative and qualitative CD4 T cell immune responses related to adenovirus DNAemia in hematopoietic stem cell transplantation. Biol Blood Marrow Transplant 2010;17:476–85. [25] Pedron B, Guerin V, Cordeiro DJ, Masmoudi S, Dalle JH, Sterkers G. Development of cytomegalovirus and adenovirus-specific memory CD4 Tcell functions from birth to adulthood. Pediatr Res 2010;69:106–11. [26] Mohty B, Bel M, Vukicevic M, Nagy M, Levrat E, Meier S, Grillet S, Combescure C, Kaiser L, Chalandon Y, et al. Graft-versus-host disease is the major

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34]

[35]

[36]

[37]

[38]

determinant of humoral responses to the AS03-adjuvanted influenza A/09/ H1N1 vaccine in allogeneic hematopoietic stem cell transplant recipients. Haematologica 2011;96:896–904. Nagler A, Naparstek E, Engelhard D. Combined therapy of granulocyte transfusions, intravenous opsonins and antibiotics for gram-negative pneumonia in neutropenic cancer patients. Acta Haematol 1994;91:42–5. Issa NC, Marty FM, Gagne LS, Koo S, Verrill KA, Alyea EP, Cutler CS, Koreth J, Armand P, Ho VT, et al. Seroprotective titers against 2009 H1N1 influenza A virus after vaccination in allogeneic hematopoietic stem cell transplantation recipients. Biol Blood Marrow Transplant 2011;17:434–8. Yalcin SS, Kondolot M, Albayrak N, Altas AB, Karacan Y, Kuskonmaz B, Aksu S, Cetin M, Goker H, Yurdakok K, Uckan D. Serological response to influenza vaccine after hematopoetic stem cell transplantation. Ann Hematol 2011;89:913–8. Bui HH, Peters B, Assarsson E, Mbawuike I, Sette A. Ab and T cell epitopes of influenza A virus, knowledge and opportunities. Proc Natl Acad Sci U S A 2007;104:246–51. Xing Z, Cardona CJ. Preexisting immunity to pandemic (H1N1) 2009. Emerg Infect Dis 2009;15:1847–9. Assarsson E, Bui HH, Sidney J, Zhang Q, Glenn J, Oseroff C, Mbawuike IN, Alexander J, Newman MJ, Grey H, Sette A. Immunomic analysis of the repertoire of T-cell specificities for influenza A virus in humans. J Virol 2008;82:12241–51. McElhaney JE, Xie D, Hager WD, Barry MB, Wang Y, Kleppinger A, Ewen C, Kane KP, Bleackley RC. T cell responses are better correlates of vaccine protection in the elderly. J Immunol 2006;176:6333–9. Iorio AM, Bistoni O, Galdiero M, Lepri E, Camilloni B, Russano AM, Neri M, Basileo M, Spinozzi F. Influenza viruses and cross-reactivity in healthy adults: humoral and cellular immunity induced by seasonal 2007/2008 influenza vaccination against vaccine antigens and 2009 A(H1N1) pandemic influenza virus. Vaccine 2012;30:1617–23. Guthrie T, Hobbs CG, Davenport V, Horton RE, Heyderman RS, Williams NA. Parenteral influenza vaccination influences mucosal and systemic T cellmediated immunity in healthy adults. J Infect Dis 2004;190:1927–35. Skowronski DM, Lu H, Warrington R, Hegele RG, De Serres G, HayGlass K, Stark D, White R, Macnabb J, Li Y, et al. Does antigen-specific cytokine response correlate with the experience of oculorespiratory syndrome after influenza vaccine? J Infect Dis 2003;187:495–9. Divekar AA, Zaiss DM, Lee FE, Liu D, Topham DJ, Sijts AJ, Mosmann TR. Protein vaccines induce uncommitted IL-2-secreting human and mouse CD4 T cells, whereas infections induce more IFN-gamma-secreting cells. J Immunol 2006;176:1465–73. Machado CM, Cardoso MR, da Rocha IF, Boas LS, Dulley FL, Pannuti CS. The benefit of influenza vaccination after bone marrow transplantation. Bone marrow transplant 2005;36:897–900.