Safety and immunogenicity of co-administered MF59-adjuvanted 2009 pandemic and plain 2009–10 seasonal influenza vaccines in rheumatoid arthritis patients on biologicals

Abstract

Rheumatoid arthritis (RA) patients under immunosuppressive therapy are particularly susceptible to infections, mainly of the respiratory tract, thus vaccination may represent a strategy to reduce their incidence in this vulnerable population. In the 2009–10 influenza season, the safety and immunogenicity of co-administered non-adjuvanted seasonal and MF59-adjuvanted pandemic influenza vaccines were evaluated in this study in 30 RA patients under therapy with anti-tumour necrosis factor (TNF)-α agents or Abatacept and in 13 healthy controls (HC). Patients and HC underwent clinical and laboratory evaluation before (T0), 1 (T1) and 6 months (T2) after vaccinations. No severe adverse reactions, but a significant increase in total mild side effects in patients versus HC were observed. Both influenza vaccines fulfilled the three criteria of the Committee for Proprietary Medicinal Products (CPMP). Seroconversion rate for any viral strain in patients and HC was, respectively, 68 versus 45 for H1-A/Brisbane/59/07, 72 versus 81 for H3-A/Brisbane/10/07, 68 versus 54 for B/Brisbane/60/08 and 81 versus 54 for A/California/7/2009. A slight increase in activated interferon (IFN)-γ-, TNF-α- or interleukin (IL)-17A-secreting T cells at T1 compared to T0, followed by a reduction at T2 in both patients and HC, was registered. In conclusion, simultaneous administration of adjuvanted pandemic and non-adjuvanted seasonal influenza vaccines is safe and highly immunogenic. The largely overlapping results between patients and HC, in terms of antibody response and cytokine-producing T cells, may represent further evidence for vaccine safety and immunogenicity in RA patients on biologicals.

Introduction

Patients with rheumatoid arthritis (RA) present an increased incidence of infections, including those affecting the respiratory tract, compared with age- and sex-matched subjects without RA 1. This susceptibility to infections is dependent upon the immune system impairment related to both the disease itself and the use of immunosuppressive drugs, including biological agents 2.

Influenza vaccination is considered to be the most effective measure for preventing influenza-related morbidity and mortality. However, despite several studies on its effect in RA patients suggest vaccine safety and satisfactory immunogenicity 3 and the World Health Assembly has recommended a target of 75% seasonal flu vaccine coverage in at-risk populations since 2010 4, the vaccination rate of immunosuppressed patients is generally low in clinical practice 5–8, probably for the fear of disease reactivation and reduced immune response.

In March 2009, clustered cases of respiratory infection in Mexico led to the discovery of a new influenza virus A H1N1 subtype. The virus has spread across the world rapidly, and the World Health Organization on June 2009 declared an influenza pandemic. The specific vaccine was available in Italy only in adjuvanted form, thus during the season 2009–10 a double (monovalent adjuvanted pandemic and trivalent non-adjuvanted seasonal) simultaneous vaccination was administered to 30 RA patients treated with immunosuppressive drugs, including anti-tumour necrosis factor (TNF)-α and the co-stimulation inhibitor, Abatacept. Safety and immunogenicity 1 and 6 months after vaccinations were evaluated. Moreover, in a patient subgroup, non-specific T cell responsiveness was also analysed.

Materials and methods

The study was approved by the ethical committee of the S. Andrea University Hospital. Thirty RA patients [mean age ± standard deviation (s.d.) 50 ± 10 years, 23 (77%) women], diagnosed on the basis of the 1987 American College of Rheumatology criteria 9, were enrolled from the out-patients attending the Clinical Immunology and Rheumatology clinic of the S. Andrea University Hospital in Rome during the 2009–10 influenza season. The patients were usually treated with low doses of corticosteroids (< 10 mg/daily of prednisone), disease-modifying anti-rheumatic drugs (DMARDs), mostly methotrexate (MTX) (10–15 mg/weekly) in combination with either TNF-α blockers [Infliximab (four patients), Etanercept (13 patients) and Adalimumab (seven patients) or the co-stimulus inhibitor Abatacept (six patients)]. Inclusion criteria were low–moderate disease activity score (DAS) < 3·7 and stable disease (not needing any increase of therapy for at least 6 months). Thirteen healthy controls (HC), matched for age and sex [41·8 ± 12 years, eight (62%) women], were also enrolled.

Six patients and three HC had already received influenza vaccination in the 2008–09 season (Table 1).

Table 1.

Clinical and demographic data in patients with rheumatoid arthritis (RA) and healthy controls (HC)

	RA	HC	P
Sex, female n (%)	23 (77)	8 (62)	n.s.
Age (years) mean ± s.d.	50 ± 10	41·8 ± 12	n.s.
Vaccination 2008–09 n (%)	6 (20)	3 (23)	n.s.
Biological therapy n (%)		n.a.
Etanercept	13 (43)
Adalimumab	7 (23)
Infliximab	4 (13)
Abatacept	6 (20)
DAS T0 mean ± s.d.	2·33 ± 0·8	n.a.

DAS = Disease Activity Score; n.a. = not applicable; n.s. = not significant; s.d. = standard deviation.

Patients underwent clinical and laboratory evaluation [specific anti-influenza antibodies, anti-nuclear antibodies (ANA), rheumatoid factor (RF) and peripheral blood mononuclear cell (PBMC) evaluation] before (T0), 1 (T1) and 6 (T2) months after vaccination. Blood samples were collected from HC at the same time.

After informed consent and in the absence of contraindications (referred allergy for egg or any vaccine component, acute infections, pregnancy, etc.) subjects were immunized by intramuscular route with 0·5 ml trivalent non-adjuvanted split influenza vaccine (Vaxigrip; Sanofi Pasteur MSD, Lyon, France) containing 15 μg for each viral strain (A/Brisbane/59/07 H1, A/Brisbane/10/07 H3 and B/Brisbane/60/08). Contemporaneously, but on a different arm, they received a single dose of the pandemic monovalent (A/California/7/2009) MF59-adjuvanted influenza vaccine (A[H1N1]pdm09, Focetria; Novartis Vaccines, Siena, Italy).

Safety

Safety has been monitored with:

1. DAS at T0, T1 and T2, to register possible vaccine-induced disease reactivation.

2. A diary card given to all patients, in order to register possible local and systemic adverse reactions.

3. A telephone interview 1 week after vaccination to all patients, asking for the possible appearance of a list of clinical systemic and/or local side effects including: shivering, fever (>37·5°C), headache, malaise, asthenia, arthralgia, myalgia, local pain, redness, induration or swelling.

4. Laboratory evaluation at T0, T1 and T2, including erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), blood cell count, RF and ANA.

Moreover, influenza-like-illness (ILI) episodes, characterized by ‘acute respiratory tract infections and fever > 38°C, accompanied by systemic and respiratory symptoms’ were also recorded in both patients and HC.

Laboratory evaluation

Specific anti-influenza antibodies

Sera were analysed by haemagglutination-inhibition (HAI) test, according to standard procedures 10. Briefly, sera were treated with receptor-destroying enzyme (RDE; Sigma-Aldrich, St Louis, MO, USA) overnight at 37°C and subsequently incubated at 56°C for 30 min. HAIs were performed in duplicate, using V-bottomed 96-well microtitre plates (Costar, Lowell, MA, USA). Twofold serial dilutions of each RDE-treated serum, starting from 1:10 dilution, were tested for their ability to inhibit the agglutination of 0·5% turkey erythrocytes by four haemagglutinating units of the seasonal A/Brisbane/59/07 (H1), A/Brisbane/10/07 (H3) and B/Brisbane/60/08 and pandemic A/California/7/2009 (H1) influenza viruses. HAI titres were recorded as the reciprocal of the maximum dilution that caused complete inhibition.

Geometric mean titres (GMTs), seroprotection rate (the percentage of vaccine recipients with a serum HAI titre of at least 1:40 after vaccination), seroconversion rate (the percentage of vaccine recipients with an increase in serum HAI titres of at least fourfold after vaccination) and seroconversion factor (the post-vaccination antibody titre divided by the prevaccination antibody titre) were calculated. A seroprotection rate exceeding 70% (60% in people aged > 60 years), a seroconversion rate exceeding 40% (30% in people aged > 60 years) and a seroconversion factor exceeding 2·5 (2·0 in people aged > 60 years) were considered as vaccine immunogenicity cut-off levels for adults aged 18–60 years, according to the guidelines of the Committee for Proprietary Medicinal Products (CPMP) 11. To fulfil the CPMP guidelines for the seasonal flu vaccination, each of the three vaccine antigens must meet at least one of the above criteria, while all criteria must be met for pandemic vaccination.

Autoantibodies

RA sample sera were evaluated for ANA by indirect immunofluorescence (Alifax Diagnostica S.p.A., Padova, Italy); RF was assessed by nephelometry (Aeroset; Abbott Diagnostics, Abbott Park, IL, USA), according to the manufacturer's instructions.

Peripheral blood mononuclear cells

PBMCs were isolated from heparinized whole blood by density gradient centrifugation using Ficoll-Hystopaque (Sigma-Aldrich) and washed three times in phosphate-buffered saline (PBS). Aliquots of 10⁷ cells were dispensed in cryovials and placed in a freezing device (Nalgene, Rochester, NJ, USA).

PBMCs from six patients and three matched controls at T0, T1 and T2 were simultaneously thawed and batch-processed. Cells were stimulated with 20 ng/ml phorbol 12-myristate 13-acetate (PMA) and 1 μg/ml ionomycin (Sigma-Aldrich) at 37°C for 5 h to detect the frequencies of interleukin (IL)-17, TNF-α and interferon (IFN)-γ-producing T cells. Brefeldin A (1·5 μg/ml; Sigma-Aldrich) was added to cultured PBMCs for 3 h. The stimulated PBMCs were washed in PBS and incubated for 30 min in the dark at 4°C with CD4-phycoerythrin (PE)-cyanin (Cy), CD8-allophycocyanin (APC)-Cy7 and CD69-peridinin chlorophyll (PerCP). Then, PBMCs were fixed in 4% formaldehyde, permeabilized with 0·1% saponin (Sigma-Aldrich), and stained with IFN-γ-fluorescein isothiocyanate (FITC), IL-17A-PE and TNF-α-APC. One hundred thousand events per sample were acquired using a fluorescence-activated cell sorter (FACS)Canto flow cytometer and analysis was performed with FACSDiva software (BD Biosciences, Mountain View, CA, USA).

Statistical analysis

All statistical analyses were performed using Stat Soft version 6·0 and the GraphPad Software, Inc. program. The seroconversion factor for the single subjects was derived in a logarithmic (base 10) scale to obtain a roughly normal distribution and its statistical significance between different groups was determined with Student's t-test for unpaired data. Statistical comparison of DAS was made by Mann–Whitney U- and Wilcoxon's tests for paired data. Ordinary and repeated measures of one-way analysis of variance (anova), with Bonferroni post-test correction, were performed for variance evaluation of phorbol myristate acetate (PMA)–ionomycin-stimulated cells. P-values < 0·05 were considered significant.

Results

Demographic and clinical data related to RA patients and HC are reported in Table 1. In particular, no significant differences were observed between patients and HC relative to age, sex and percentage of vaccinated subjects in the previous year. In vaccinated RA patients, no significant DAS modifications were observed at T1 and T2 (2·18 ± 0·8 and 2·24 ± 0·9, respectively) compared with T0 (2·33 ± 0·8). Moreover, no severe adverse reactions (life-threatening, requiring hospitalization and/or intravenous anti-infective treatment), but a significant increase in total mild systemic and local side effects in patients versus HC (P < 0·005), were detected (Table 2).

Table 2.

Systemic and local side effects in vaccinated rheumatoid arthritis (RA) patients and healthy controls (HC)

Side effects	RA	HC	P
Redness/swelling injection site	3	0	n.s.
Fever (> 37·5°C)	2	0	n.s.
Headache	3	0	n.s.
Asthenia/malaise	2	1	n.s.
Arthralgia/myalgia	4	1	n.s.
Chills	1	0	n.s.
Total events/n patients	15/30 (50%)	2/13 (7%)	< 0·005
n patients with events/n patients	8/30 (26%)	2/13 (7%)	n.s.

n.s. = not significant.

No statistically significant changes of the ANA titre, RF and inflammatory indices were registered.

Specific antibody responses against different vaccine antigens [A/Brisbane/59/07 H1, A/Brisbane/10/07 H3 (H1 and H3 had also been included in the 2008–09 influenza vaccine formulation), B/Brisbane/60/08 for the seasonal and A/California/7/2009 for the adjuvanted pandemic vaccine] were assessed. Pandemic and seasonal influenza vaccines met all three CPMP criteria, in both patients and HC (Table 3). In particular, a seroconversion rate exceeding 40% and a seroprotection rate higher than 70% at T1 have been observed in both patients and HC for all seasonal and pandemic antigens. At T2, the seroprotection rate was maintained only for seasonal vaccine (all antigens in HC and B/Brisbane/60/08 in patients). Finally the seroconversion factor exceeded 2·5 in both patients and HC for all seasonal and pandemic antigens.

Table 3.

Immunogenicity in patients with rheumatoid arthritis (RA) and healthy controls (HC)

	A/Brisbane/59/07 H1N1			A/Brisbane/10/07 H3N2			B/Brisbane/60/08			A/California/7/2009/H1N1
	T0	T1	T2	T0	T1	T2	T0	T1	T2	T0	T1	T2
Seroconversion rate (%)
RA		68			68			68			82
HC		55			64			55			55
Seroprotection rate (%)
RA	22	80	46	41	88	69	67	100	96	0	84	46
HC	54	100	93	23	83	71	85	91	100	0	66	36
Seroconversion factor
RA		4·1			6·4			4·9			8·5
HC		3·7			6·2			4·8			5·1

GMT values are reported in Table 4. Between T0 and T1 they increased significantly for all antigens in RA patients (P < 0·05), with a reduction at T2 (Fig. 1). As a consequence of this reduction, protection at T2 was not always maintained (Table 3). Moreover, the antibody response to the A vaccine antigens in the six RA patients and three HC already vaccinated for influenza in the previous year (in which the two A influenza antigens were the same as the 2009–10 vaccine) showed a non-statistically significant increase from T0 to T1, whereas such an increase was significant (P < 0·001) in the 24 RA patients and 10 HC not vaccinated previously. No difference was found in vaccine response between the 24 and the six patients on therapy with TNF-α blockers or the co-stimulus inhibitor Abatacept, respectively.

Table 4.

Geometric mean titre values of antibody response to seasonal and pandemic vaccines in patients with rheumatoid arthritis (RA) and healthy controls (HC)

RA	A/Brisbane/10/07 (H3)	A/Brisbane/59/07 (H1)	B/Brisbane/60/8	A/California/7/2009(H1pdm)
T0	11	22	45	8
T1	61	174	263	100
T2	31	57	148	33

HC	A/Brisbane/10/07 (H3)	A/Brisbane/59/07 (H1)	B/Brisbane/60/8	A/California/7/2009(H1pdm)
T0	32	15	68	7
T1	113	107	302	50
T2	93	72	195	24

Fig. 1.

Geometric mean titre values in patients with rheumatoid arthritis (RA) and healthy controls (HC). *T0 versus T1 P < 0·05; #T1 versus T2 P < 0·05.

The frequency of ILI episodes in vaccinated patients was two of 30 (7%) versus none in HC, the difference being non-significant.

A slight increase in activated cytokine-producing T cells was found at T1 compared to T0 followed by a reduction at T2 in both patients and HC. Mean values were not significantly higher in patients compared to HC at every time-point. In particular, the percentage of CD69⁺IFN-γ⁺ cells increased significantly at T1 in both patients and HC, whereas that of CD69⁺TNF-α⁺ was increased significantly at T1 in HC only (Fig. 2).

Fig. 2.

Percentages of activated cytokine-secreting cells at T0, T1 andT2 in patients with rheumatoid arthritis (RA) and healthy controls (HC). *P < 0·05.

Discussion

The current study shows that simultaneous administration of pandemic adjuvanted and seasonal non-adjuvanted influenza vaccines is safe, even in RA patients on biologicals. No disease reactivation and appearance/increase in titres of autoantibodies were registered as a consequence of vaccine administration. However, mild local and systemic adverse events were observed with a significantly higher frequency in RA patients than in HC. Regarding immunogenicity, these vaccines were able to stimulate an antibody protective response in RA patients with low–moderate stable disease under treatment with DMARDs (mainly MTX) and/or biological agents (TNF-α blockers and co-stimulation inhibitor) comparable to HC. Vaccination in both patients and HC increased the frequency of T cells able to secrete IFN-γ/TNF-α and IL-17A following in vitro stimulation. Finally, the scarcity of ILI episodes in patients (not confirmed, however, by virological assays), their absence in HC and the lack of significant difference between patients and HC, suggest a satisfactory level of vaccine efficacy.

The number of vaccine responders and the high levels of the humoral immune responses specific for the seasonal viral strains were unexpected, as in our previous experience on similar RA patient and control populations the response to the same non-adjuvanted vaccination resulted as protective, but only partially met the CPMP criteria 11. One possible reason for the overall increased efficacy of the vaccination in the tested population may be provided by inclusion in the flu pandemic vaccine formulation of the innovative adjuvant MF59. Although the non-adjuvanted flu seasonal vaccine was administered simultaneously in a different arm, the adjuvant effect of MF59 could have improved innate immune response and stimulated priming of T helper cells broadly reacting with shared epitopes among different viral haemagglutinins 13. Alternatively, it may be speculated that pandemic vaccine per se, independently of adjuvant, may have stimulated the appearance of a broadly neutralizing heterosubtypic antibody response as a consequence of preferential naïve rather than memory B cell recruitment 14. In this study, a slight increase of local and systemic adverse events was observed, mainly among patients, probably because the observed adverse events are associated with vaccine-induced local inflammation 15 that may mimic or intensify symptoms already present in RA patients. In this view, MF59 may also be responsible for the increased frequency of adverse events observed. In fact, adjuvanted vaccines, even though useful in immunosuppressed subjects, have seldom been used in patients with autoimmune diseases (no more than 20% 3), for the fear of disease reactivation. The data reported here are in agreement with those of Pellegrini et al., who analysed 64 clinical trials involving MF59-adjuvanted pandemic and seasonal influenza vaccines in different populations, finding that MF59-vaccinated subjects were at higher risk of solicited local or systemic adverse events within 3 days from vaccination. By contrast, the significantly increased relative risk for Bell's palsy, paraesthesia and inflammatory bowel diseases, reported by Bardage et al. in more than 1 million Swedish vulnerable people vaccinated with the similar squalene-containing As03-adjuvanted pandemic influenza vaccine, has not been observed in our study 16. Lack of underlying disease reactivation has already been observed 17,18 in patients with rheumatic diseases immunized with the As03-adjuvanted vaccine. Moreover, in the first cited study 16, safety was observed even after two adjuvanted vaccine administrations. Conversely, in RA patients, Adler et al. found a possible vaccination-induced disease reactivation in eight of 149 patients with immune-mediated diseases, including connective tissue disease, RA, vasculitis and spondiloarthropathies 19. The lack of post-vaccine autoantibody significant modifications is in agreement with Urowitz et al. on more than 100 systemic lupus erythematosus (SLE) patients treated with either adjuvanted or non-adjuvanted influenza pandemic vaccine 20.

The current study demonstrates that, even though patients and HC had no detectable antibody levels to pandemic influenza antigen before vaccination (confirming the low cross-reaction rate between pandemic and seasonal viral strains 21), specific post-vaccine antibodies were achieved at far higher levels and in a greater patient percentage than observed previously 12. Unexpectedly, the B viral strain, which had been the least immunogenic in the previous experience, has become the most immunogenic in this study. The loss of protective anti-A[H1N1]pdm09 antibody levels at 6 months in patients and HC, instead, may be linked probably to its new appearance, thus lacking in cohort effect. In other studies involving patients with rheumatic diseases treated only with pandemic and not seasonal vaccine, antibody levels comparable to those obtained in HC have never been achieved after a single vaccine dose 16–18. Surprisingly, we observed a specific antibody response to the pandemic one-dose vaccine higher in RA patients than in HC (for all three CPMP parameters), and an excellent response, also superior or comparable to HC towards the seasonal vaccine strains. These results are probably a consequence of a series of factors, including the MF59 adjuvant activity, the simultaneous pandemic and seasonal vaccine administration, as well as the previous natural and/or vaccine immunizations. The reciprocal seasonal and pandemic cross-reaction is also supported by data obtained in normal adult and elderly subjects. Gasparini et al. 22 observed a doubled antibody response to the seasonal H1N1 strain when pandemic and seasonal vaccines were co-administered compared with the response obtained with the seasonal vaccine alone. Cross-reaction and protection in a great percentage of subjects is present even among apparently highly divergent viral strains, such as A[H1N1]pdm09 and seasonal strains 23–25; these shared epitopes may be most probably recruited and stimulated with high efficiency by MF59. To the best of our knowledge, this is the first study analysing the specific immune response to two simultaneously administered pandemic and seasonal vaccines in inflammatory rheumatic diseases. However, the immune response to the same vaccine schedule has been analysed in type 1 diabetes and in healthy, including elderly, subjects 22,26–28. In particular, Zuccotti et al. found an overall 100% seroprotection rate at 1 month after vaccination, without significant differences between those who received either the one- or the two-dose schedule in 80 young type 1 diabetes patients 26.

The role of the adjuvant in stimulating a specific immune response is also witnessed indirectly by five studies reporting statistically significant reduced GMTs, seroconversion and/or seroprotection rates in RA patients (partially anti-TNF-α or Abatacept-treated) compared to HC after non-adjuvanted pandemic influenza vaccine administration 29–33. The aim of the adjuvant is to enhance immune response in a population expected to be mostly naïve to an antigen and to increase vaccine production by sparing antigen. In humans, the adjuvant-induced immune response results in higher antibody titres and high-quality long-term persisting antibodies. These results are probably related to the induction by MF59 of strong CD4⁺ T cell help and more germinal centre reactions 34. Khurana et al. demonstrated that MF59 induces the epitope spreading from HA2 to HA1, thus stimulating a broader antibody response 35, and increases the maturation of antibody affinity 36. In contrast, subjects vaccinated with non-adjuvanted or alum-adjuvanted vaccines present narrower recognition of the fragments of the HA2 stem region 36.

The seasonal vaccine administration 3 weeks before the pandemic has been observed to be associated with a lower response to the latter 27,37; these results have also been reported by Gabay et al. 16, independently of the time interval between seasonal and pandemic vaccine administrations. Thus, seasonal and pandemic vaccines should be administered together, or at least in close proximity, to obtain a better response 38.

The influence of previous vaccination with the same antigens on the immune response modulation was already observed 12,39–46 and it has been confirmed in this study. In the six previously vaccinated patients and three HC, a non-significant increased antibody response to A H1 and H3 (both included in the influenza vaccine formulation for 2008–09 and 2009–10 seasons) antigens versus a significant (P < 0·001) antibody rise in the previously non-vaccinated patients and HC has been observed. Previous immunizations, therefore, may negatively modulate the antibody response, as already reported in 1979 47 as the so-called ‘original antigenic sin’ phenomenon 48, and this behaviour may also be observed in the protection against influenza 49.

Cellular immune response plays a crucial role in clearing influenza infection; furthermore, in elderly people it correlates with protection more than serum antibody response 50. Recent studies have reported preserved cellular immune response to influenza vaccination in rituximab-treated RA patients 51, in Wegener's 52 and scleroderma patients 53, while a diminished specific cellular immune response has been observed in SLE 54. The cellular immune response to influenza vaccine has never been investigated in RA patients treated with anti-TNF-α or Abatacept. In the current study, stimulated CD69-positive T cells presented a similar profile between patients and HC, in relation to cytokine patterns of IFNγ-, TNF-α- and IL-17A-secreting cells, respectively, and with dynamics showing an increase 1 month after vaccine stimulation followed by a decrease at lower levels than baseline 6 months later, which is reminiscent of what has been already observed with T regulatory cells (T_regs) 12. Probably, the vaccine-stimulated specific immune response may induce a secondary regulatory response, mediated by T_regs. The increase in activated IFN-γ-producing T cells at T1 reflects the major implication of T helper type 1 (Th1)-like response to vaccination. The non-significant increase in activated TNF-α⁺ T cells in patients is due probably to their anti-TNF-α therapy, while IL-17A⁺ T cells do not seem to be particularly involved in influenza response. The early recruitment of antigen-specific CD4 and memory B lymphocytes has been reported recently by Faenzi et al. in healthy adult subjects immunized with MF59-adjuvanted pandemic influenza vaccine. The authors explain this phenomenon as being due to priming from cross-reactive seasonal vaccines, with the pandemic vaccine able to recruit not only naïve B cells but also seasonal-primed memory B cells 55, an observation already glimpsed in vitro for yearly repeated seasonal vaccine 56,57. In conclusion, the current study demonstrates the cutting-edge profile of safety and immunogenicity of simultaneously administered MF59-adjuvanted pandemic and non-adjuvanted seasonal influenza vaccines. The similarity between patients and HC, not only in antibody but also in IFN-γ-, TNF-α- or IL-17A-secreting T cell response, may be interpreted as further evidence for vaccine safety and immunogenicity in RA patients on biologicals. This study shows that the adjuvanted pandemic and non-adjuvanted seasonal influenza vaccines may be administered both safely and effectively in RA patients. This has important implications for public health, considering the low influenza vaccine coverage in this vulnerable patient population, far below the World Health Assembly recommended target 6–8, due mainly to scarce sensitivity or fear of disease reactivation by medical doctors 7.

Disclosure

All authors declare no financial conflicts of interest.