Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study

G Model MEDCLI-4783; No. of Pages 7

ARTICLE IN PRESS Med Clin (Barc). 2019;xxx(xx):xxx–xxx

www.elsevier.es/medicinaclinica

Original article

Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study Patricia Richi a,∗ , María Dolores Martín b , María Teresa Navío c , Laura González-Hombrado d , Marina Salido e , Jesús Llorente f , Israel Thuissard-Vasallo g , Patricia Alcocer h , Carmen María Saa-Requejo i , Ana Jiménez-Diaz a , Laura Cebrián c , Leticia Lojo c , Marta García-Castro d , David Sanz-Rosa g , Patricia Castro e , Sandra Fernández-Rodríguez j , María José Martínez de Aramayona j , Martina Steiner a , Tatiana Cobo a , Cristina García-Fernández i , a ˜ Mónica Fernández-Castro a , Óscar Illera a , Ricardo Valverde k , Santiago Munoz-Fernández a

Rheumatology Department, Infanta Sofía University Hospital, Paseo de Europa 34, San Sebastián de los Reyes, 28702 Madrid, Spain Bactereology Department, BR Salud Laboratories, Infanta Sofía University Hospital, Paseo de Europa 34, San Sebastián de los Reyes, 28702 Madrid, Spain c Rheumatology Department, Infanta Leonor University Hospital, Gran Vía del Este 80, 28031 Madrid, Spain d Rheumatology Department, Tajo University Hospital, Av. Amazonas Central s/n, Aranjuez, 28300 Madrid, Spain e Rheumatology Department, Infanta Cristina University Hospital, Av 9 de Junio 2, Parla, 28981 Madrid, Spain f Pharmacy Department, Infanta Sofía University Hospital, Paseo de Europa 34, San Sebastián de los Reyes, 28702 Madrid, Spain g School of Doctoral Studies & Research, Europea University, Calle Tajo s/n, Villaviciosa de Odón, 28670 Madrid, Spain h Rheumatology Department, HM Hospital, Avenida de Manoteras n◦ 10, 28050 Madrid, Spain i Preventive Medicine Department, Infanta Sofía University Hospital, Paseo de Europa 34, San Sebastián de los Reyes, 28702 Madrid, Spain j Occupational Medicine Department, Infanta Sofía University Hospital, Paseo de Europa 34, San Sebastián de los Reyes, 28702 Madrid, Spain k Dermatology Department, Infanta Sofía University Hospital, Paseo de Europa 34, San Sebastián de los Reyes, 28702 Madrid, Spain b

a r t i c l e

i n f o

Article history: Received 27 August 2018 Accepted 7 February 2019 Available online xxx Keywords: Influenza vaccine Autoimmune inflammatory diseases Biological therapy Anti-TNF Rituximab

a b s t r a c t Background and objectives: Influenza vaccine is recommended for patients with autoimmune inflammatory rheumatic diseases who receive biological therapy. To evaluate if biological therapy impairs immunization after seasonal influenza vaccine. Material and methods: Patients with inflammatory arthopathies, psoriasis, inflammatory bowel disease or connective tissue diseases who were receiving or were going to initiate biological therapy were included and vaccinated during 2014–2015 influenza season. ELISA was used to measure influenza antigen A and B antibodies, before and after vaccination. Demographic parameters, diagnosis and kind of treatment were recorded and their influence on the final serological status against influenza was studied. Results: 253 subjects were analyzed. After vaccination, 77% of participants presented detectable antibodies against antigen A and 50.6% of them had detectable antibodies against antigen B. Final seropositivity rate against antigen B antibodies increased from baseline (50.6% vs 43.5%, p < 0.001). Anti-TNF drugs were associated with better response and rituximab with the worst (79.2% vs 55.0% for final seropositivity against antigen A, p = 0.020). Vaccine response in the rituximab group tended to improve when the interval between the drug administration and the vaccination was at least 12 weeks (seropositivity rate 80.0% in those with the longer interval vs 25.0% in the other group, p=0.054). Conclusions: Among the patients on biological therapy vaccinated against influenza, anti-TNF therapy was identified as a predictive factor of final seropositivity. Rituximab presented a lower rate of final seropositivity, which could be increased with an accurate administration schedule. ˜ S.L.U. All rights reserved. © 2019 Elsevier Espana,

Abbreviations: AIIRD, autoimmune inflammatory rheumatic diseases; VE, vaccines effectiveness; Pso, psoriasis; WHO, World Health Organization; Ab, antibodies; Ag, antigens; AU, Absorbance Units. Abreviaturas: AIIRD, enfermedades reumáticas inflamatorias autoinmunes; VE, efectividad vacunal; Pso, psoriasis; WHO, organización Mundial de la Salud; Ab, anticuerpos; Ag, antígenos; AU, unidades de absorbancia. ∗ Corresponding author. E-mail addresses: [email protected], [email protected] (P. Richi). https://doi.org/10.1016/j.medcli.2019.02.003 ˜ S.L.U. All rights reserved. 0025-7753/© 2019 Elsevier Espana,

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003

G Model MEDCLI-4783; No. of Pages 7

ARTICLE IN PRESS P. Richi et al. / Med Clin (Barc). 2019;xxx(xx):xxx–xxx

2

Respuesta a la vacuna contra la gripe en pacientes que reciben tratamientos biológicos: resultados del estudio de cohortes RIER r e s u m e n Palabras clave: Vacuna contra la gripe Enfermedades inflamatorias autoinmunes Terapia biológica Anti-TNF Rituximab

Antecedentes y objetivos: La vacunación antigripal está recomendada en pacientes con enfermedades autoinmunes sistémicas que reciben tratamientos biológicos. Evaluar si la terapia biológica puede perjudicar la inmunización después de la administración de la vacuna contra la gripe estacional. Material y métodos: Los pacientes con artropatías inflamatorias, psoriasis, enfermedad inflamatoria intestinal o enfermedades del tejido conectivo, que estaban en tratamiento o que iban a iniciar tratamiento con terapia biológica, fueron incluidos en el estudio y vacunados durante la temporada de influenza 20142015. Se utilizó ELISA para medir los anticuerpos contra los antígenosA y B de la gripe, antes y después de la vacunación. Se registraron los datos demográficos, diagnósticos y el tipo de tratamiento y se estudió su influencia sobre el estado serológico final contra la influenza. Resultados: Se analizaron 253 sujetos. Después de la vacunación, el 77% de los participantes presentaron anticuerpos detectables contra el antígeno A y el 50,6% de ellos tenían anticuerpos detectables contra el antígeno B. La tasa de seropositividad final de anticuerpos contra el antígeno B aumentó desde los valores basales (50,6% frente a 43,5%, p < 0,001). Los fármacos anti-TNF se asociaron con la mejor respuesta y rituximab con la peor (79,2% vs. 55,0% para la seropositividad final contra el antígeno A, p = 0,020). La respuesta a la vacuna en el grupo de rituximab tuvo tendencia a mejorar cuando el intervalo entre la administración del fármaco y la vacunación fue por lo menos de 12 semanas (tasa de seropositividad del 80,0% en aquellos con el intervalo más largo frente al 25% en el otro grupo, p=0.054). Conclusiones: Entre los pacientes en terapia biológica vacunados contra la influenza, la terapia anti-TNF se identificó como un factor predictivo de la seropositividad final. Rituximab presentó una tasa más baja de seropositividad final, que podría aumentarse con un programa de administración preciso. ˜ S.L.U. Todos los derechos reservados. © 2019 Elsevier Espana,

Introduction Hu man Influenza (flu) is a contagious respiratory illness that can cause mild to severe disease. Serious outcomes of flu infection can result in hospitalization or death. People with autoimmune inflammatory rheumatic diseases (AIIRD), especially those on biological therapy have a greater risk of complications,1 this being the reason why seasonal flu vaccination is recommended for these patients.2 Nevertheless, vaccines effectiveness (VE) in these patients is not clear as different studies have shown inconsistent results. Some, report high rates of responders3 or no differences with healthy controls4 and others find worse responses.5 It seems that VE depends not only on the disease but also on the treatment used.6 The aim of our study was to determine whether immunological response to flu vaccination in people with AIIRD is influenced by the biological therapy and if the effect depends on the type of biological DMARD (disease modifying anti-rheumatic drug) used in routine clinical practice. We also wanted to know if other parameters such as demographic characteristics, diagnoses and concomitant treatment with glucocorticoids and/or synthetic DMARDs, impact on the vaccine effectiveness. Finally, we investigated the compliance to the vaccination recommendations in this group of patients.

Patients and methods A non-interventional, multicentre, cohort study was designed. Patients attended in Rheumatology, Dermatology or Gastroenterology services of the four participant hospitals (Infanta Sofia University Hospital, Infanta Leonor University Hospital, El Tajo University Hospital and Infanta Cristina University Hospital, Comunidad Autónoma de Madrid, Spain) that fulfilled inclusion criteria were asked to enrol in the study. The recruitment period started with the beginning of the flu vaccination campaign in October 2014. Follow-up period went on until the final serological test of the latest patient included was performed (February, 2015).

Subjects older than 18, suffering from an AIIRD such as rheumatoid arthritis (RA), psoriatic arthritis(PsA), spondyloarthritis (SpA), undifferentiated arthritis (UA) or connective tissue diseases (CTD), psoriasis (Pso) or inflammatory bowel disease (IBD) and followed up in the hospitals outclinics, were included. Furthermore, patients had to be on or were going to initiate (naïve subjects) a biological DMARD therapy with an anti-tumour necrosis factor (anti-TNF␣) such as infliximab, adalimumab, etanercept, golimumab or certolizumab or other biological agent like, rituximab, tocilizumab, abatacept or anakinra. The choice of the biological agent depended on the patients’ characteristics and clinical parameters. Patients who changed their biological DMARD between the basal and the final serological test were excluded from the analysis. Number of biological agents used and disease and treatment duration were documented. Data about concomitant treatments with synthetic DMARDs (Metotrexate [MTX], leflunomide, sulfasalazine, hidroxychloroquine, azatioprine or ciclophosphamide) and corticosteroids was recorded. Patients reported if they had been vaccinated or not with the influenza vaccine during the previous influenza season (2013–2014) before entering the study. Subjects were excluded if they had already received the 2014–2015 influenza vaccine. All data was collected during the first study visit, before the vaccine was administered. All participants received seasonal inactivated trivalent influenza vaccine (TIV) recommended by the World Health Organization (WHO) in the 2014–2015 northern hemisphere influenza season, that contained: an A/California/7/2009 (H1N1)pdm09-like virus, a A/Texas/50/2012 (H3N2)-like virus and a B/Massachusetts/2/2012like virus.7 Naïve patients were vaccinated before their first dose of biological agent was administered. Basal antibody (Ab) titres against A and B antigens (Ag) were measured just before vaccination. The final Ab determinations were done at least 4 weeks after the vaccine administration. Serum samples obtained from all participants, were aliquoted, stored at −20 ◦ C and tested with NovagnostTM Influenza A IgG

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003

G Model MEDCLI-4783; No. of Pages 7

ARTICLE IN PRESS P. Richi et al. / Med Clin (Barc). 2019;xxx(xx):xxx–xxx

Statistics For the descriptive analysis, the absolute (n) and relative (%) frequencies were used to express the qualitative variables. Mean ± standard deviation (SD) or median [interquartile range; IQR] were used to express the quantitative variables according the gaussian behaviour. The total of cases per the variable to study was used to perform the calculations. To test the statistically significant differences of the vaccine response in the different scenarios either the Chi-square test or Fisher’s exact test was performed for qualitative variables. Besides, either Student T test or U-Mann Whitney was used for quantitative variables according to the normality test. Serologic status before and after vaccination was assessed by McNemar test, for the overall study population and for each variable of interest (diagnosis, being naive or not for biological treatment, having received one or more biological agents, concomitant treatment with a DMARD agent and the use of glucocortcoids). Possible predictors of positive antibody response were analyzed using univariate and multivariate logistic regression models. All variables were included in the univariate analysis. Only those univariate variables with significance level of p < 0.100 were used for the multivariate logistic regression model. An enter method was employed. When the p-value was inferior to the alpha error (5%), a statistical significance was considered. The data analysis was performed with IBM SPSS statistics version 21.0 (IBM Corp; USA). Results Two hundred and eighty subjects were included in the study. Four of them stopped the biological DMARD during the study period, so they were excluded from the analysis. A further 23 did not attend the study schedule and therefore there were no Influenza A/B IgG data available. So, finally the 253 patients that completed the study, with a follow up duration of 6.7 ± 2.5 weeks, were analyzed. Demographic data is shown in Table 1.

Table 1 Demographic data and disease duration of the 253 patients analyzed. Total n = 253 Sex, n (%) Male Female Age, years Disease duration (years)

103 (40.7) 150 (59.3) 49 ± 13 5.0 [9.0]

100 90

Ag B Ab

*

Ag A Ab

80

% of patieents

(Siemens Healthcare Diagnostics), enzyme immunoassay for semiquantitative detection of specific IgG antibodies (Ab) of Influenza A virus in human serum. NovagnostTM Influenza B IgG was used for the detection of specific IgG Ab of Influenza B. The semiquantitative immunoenzymatic determination of IgG-class Ab of Influenza A/B virus is based on the Enzyme-linked Immunosorbent Assay (ELISA) technique. The amount of anti-Influenza A/B virus IgG Ab in the sample was measured by an ELISA microwell plate reader at 450 nm. In accordance with the provider’s instructions cut-off value was defined as 10 Absorbance Units (AU). Samples were considered positive if the absorbance value was at least 15% over the cut-off: ≥11.5 AU and negative if the absorbance value was lower than 15% below the cut-off: <8.5 AU. Samples with an absorbance value of up to 15% above or below the cut-off should not be considered as clearly positive or negative (grey zone). Nonetheless, to avoid bias that could lead to better results we considered negative any value under 11.5 AU. Final seropositivity rates against Ag A and Ag B after vaccination were calculated. Moreover, we investigated stronger immunological responses consisting in double or a fourfold rise of basal titres. Differences in seropositivity rates depending on the lapse of time between the vaccination and the final Ab determination, were also studied. All participants provided signed written informed consent. All procedures and methods were approved by the local ethics committees.

3

70 60 50 40 30 20 10 0 Anti-TNF

Rituximab

Tocilizumab

Fig. 1. Percentage of patients with final detectable Ab depending on the biological agent: *p = 0.024.

Recorded data about patients’ adherence to vaccination recommendations during 2013–2014 was as follows. Nineteen patients did not remember if they had been vaccinated against influenza. From the 234 subjects who reported their vaccination status during the previous influenza season, 131 patients (56.0%) received 2013–2014 vaccine. Patients ≥60 years showed a trending towards significance higher rate of vaccination during the previous influenza season compared to patients <60 years (64.0 vs 48.7%, p = 0.050). The highest percentage of vaccines during the previous campaign was reached by the patients on rituximab or certolizumab (76.5 and 70.0%, respectively) while patients on golimumab showed a lower rate when compared with the rest of biological DMARDs (30.4% of patients, p = 0.009). Antibodies to Antigen A We found baseline detectable Influenza A Ab in 192 (75.9%) of the patients and in 194 (76.7%) of them after vaccination. Twentyone (34.4%) out of 61 the basal seronegative subjects became seropositive. In contrast, 19 (9.8%) out of 192 seropositive patients at baseline, turned to seronegative. Basal Ab titres ≥11.5 AU were independent from the previous season’s vaccination (70.9% of unvaccinated patients vs 80.2% of those vaccinated, p = 0.099). Table 2 shows basal and final seropositive patients depending on the diagnosis, type of biological agent and concomitant treatment. Final seropositivity was related to basal similar condition (90.1% of basal seropositive patients vs 34.4% of basal seronegative ones, p < 0.001) and to younger age (48.5 ± 13.2 vs 53.0 ± 11.7 years, p = 0.017), but not to gender, diagnosis, disease and biological therapy duration, previous season vaccination, being naïve for or having been treated with just one or more biological DMARDs and receiving concomitant treatment with glucocorticoids or synthetic DMARDs. We compared final seropositivity rates between patients on different biological agents and found out Rituximab presented lower rates compared to patients on anti-TNF. Results are shown in Fig. 1. Patients on rituximab with an interval ≥12 weeks between vaccination and the drug administration showed a tendency towards better responses (final seropositivity in 80% of subjects with a lapse longer than 12 weeks between the rituximab infusion and the

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003

G Model MEDCLI-4783; No. of Pages 7

ARTICLE IN PRESS P. Richi et al. / Med Clin (Barc). 2019;xxx(xx):xxx–xxx

4

Table 2 Comparison between basal and final seropositivity against Ag A and Ag B. Data shows n (%). Basal Ab Ag A ≥11.5 Diagnosis RA (n = 90) SpA (n = 87) PsoA (n = 42) Pso (n = 13) IBD (n = 8) CTD (n = 7) UA (n = 6)

Final Ab Ag A ≥11.5

p value

Basal Ab Ag B ≥11.5

Final Ab Ag B ≥11.5

p value

69 (74.4) 67 (77.0) 34 (81.0) 9 (69.2) 7 (87.5) 4 (57.1) 4 (66.7)

64 (71.1) 72 (82.8) 33 (78.6) 10 (76.9) 7 (87.5) 4 (57.1) 4 (66.7)

0.648 0.267 1.000 1.000 1.000 1.000 1.000

37 (41.1) 40 (46.0) 22 (55.4) 5 (38.5) 2 (25.0) 3 (42.9) 1 (16.7)

50 (55.6) 45 (51.7) 22 (55.4) 3 (23.1) 4 (50.0) 2 (28.6) 2 (33.3)

0.031 0.486 1.000 0.625 0.500 1.000 1.000

Biological treatment Naive (n = 20) Not naive (n = 233)

17 (85.0) 175 (75.1)

15 (80.0) 178 (76.4)

1.000 0.728

7 (35.0) 103 (44.2)

15 (75.0) 113 (48.5)

0.008 0.302

Number of biological agents received Treatment with one biological agent (n = 168) Treatment with >1 biological agent (n = 67)

127 (74.6) 50 (74.6)

131 (78.0) 49 (73.1)

0.523 1.000

84 (50.0) 34 (50.8)

69 (41.1) 35 (52.2)

0.072 1.000

Kind of biological agent Anti TNF (n = 215) Infliximab (n = 31) Adalimumab (n = 77) Etanercept (n = 71) Certolizumab (n = 12) Golimumab (n = 24) Rituximab (n = 20) Tocilizumab (n = 13)

165 (76.7) 24 (77.4) 54 (70.1) 59 (83.1) 11 (91.7) 17 (70.8) 13 (65.0) 10 (76.9)

170 (79.1) 26 (83.9) 60 (77.9) 56 (78.9) 9 (75.0) 19 (79.2) 11 (55.0) 9 (69.2)

0.472 0.625 0.146 0.508 0.500 0.625 0.625 1.000

92 (42.8) 18 (58.1) 30 (39.0) 29 (40.9) 6 (50.0) 9 (37.5) 10 (50.0) 6 (46.2)

113 (52.6) 18 (58.1) 33 (42.9) 41 (57.8) 8 (66.7) 13 (54.2) 8 (40.0) 4 (30.8)

0.023 1.000 0.690 0.045 0.625 0.289 0.625 0.500

DMARDs Not DMARD (n = 132) DMARD (n = 121)

95 (72.0) 97 (80.2)

97 (73.5) 97 (80.2)

0.824 1.000

57 (43.2) 53 (43.8)

63 (47.7) 65 (53.7)

0.440 0.090

Kind of DMARD MTX (n = 83) Leflunomide (n = 25)

64 (77.2) 20 (80.0)

65 (78.3) 21 (84.0)

1.000 1.000

35 (42.2) 11 (44.0)

42 (50.6) 15 (60.0)

0.248 0.344

157 (76.2) 35 (74.5) 29 (74.4) 6 (75.0)

162 (78.6) 32 (68.1) 27 (69.2) 5 (62.5)

0.441 0.581 0.754 1.000

88 (42.7) 22 (46.8) 18 (46.2) 4 (50.0)

103 (50.0) 25 (53.2) 20 (51.3) 5 (62.5)

0.092 0.607 0.774 1.000

Glucocorticoids Not glucorticoids (n = 206) Glucocorticoids (n = 47) Dairy prednisone dose ≤7.5 mg (n = 39) Dairy prednisone dose >7.5 mg (n = 8)

Changes in the proportion of basal and final seropositive patients depending on the diagnosis, biological treatment, kind of biological agent and concomitant treatments were assessed using McNemar test. p values in bold highlight significant difference before and after vaccination in each category. RA: rheumatoid arthritis; SpA: spondyloarthritis; PsA: psoriatic arthritis; Pso: psoriasis; IBD: inflammatory bowel disease; CTD: connective tissue diseases; UA: undifferentiated arthritis; DMARD: disease modifying anti-rheumatic drug.

vaccine administration, vs 25% in patients with interval shorter than 12 weeks, p = 0.054). Final blood samples were taken before the 10th week after immunization in all patients except in 31, who presented longer interval (11.0[3.0] weeks) between vaccination and the final blood test. In this small group, we found detectable Ab in most patients (87%), with no differences with those with the shorter period. We found similar results for intervals longer than 12 weeks (13.0[2.0] weeks). There was a group of patients who showed a stronger response. Fourteen (5.5%) doubled Ab titres, 7 of them suffered spondyloarthritis and 12 were on anti-TNF therapy. No patient on rituximab doubled baseline titres. Only 2 patients (0.8%) reached a fourfold rise. Having a powerful response did not depend on age, gender, type of disease or concomitant treatments. The univariate regression analysis identified treatment with an Anti-TNF agent and basal Ag A Ab ≥11.5 as predictive factors of final seropositivity and older age and treatment with Rituximab as predictors of final seronegativity. In the multivariate analysis only Anti-TNF treatment and serological basal status remained as predictors of final Ag A Ab ≥11.5, Fig. 2.

AgA > 11.5

Anti TNF Basal Ab AgA ≥ 11.5

0

10

20

30

OR

CI (95%)

p value

3.030

1.207-7.609

0.018

18.660

8.955-38.885

<0.001

40

Fig. 2. Multivariate logistic regression for the variables associated to Influenza A Ab titres ≥11.5 AU.

Antibodies to Antigen B Detectable Influenza B Ab was found in 110 patients (43.5%) before vaccination and in 128 patients (50.6%) after it (p < 0.001). Fifty-one (35.6%) patients out of 143 basal seronegative subjects reached seropositivity and 33 (30.0%) seropositive patients at baseline, became seronegative. As basal A titres, basal B Ab titres ≥11.5 AU were independent from the 2013 to 2014 previous vaccination (40.8% of non vaccinated patients vs 47.3% of those vaccinated, p = 0.317).

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003

G Model

ARTICLE IN PRESS

MEDCLI-4783; No. of Pages 7

P. Richi et al. / Med Clin (Barc). 2019;xxx(xx):xxx–xxx AgB > 11.5

OR

CI (95%)

p value

Psorias is

0.214

0.049 - 0.926

0.039

Basal Ab AgB ≥ 11.5

5.512

2.987 - 10.170

0.001

0

5

10

Fig. 3. Multivariate logistic regression for the variables associated to Influenza B Ab titres ≥11.5 AU.

Rheumatoid arthritis patients, naïve ones, subjects receiving an anti-TNF agent and those on etanercept increased the rate of seropositive patients from basal to final status (p < 0.050). Results are shown in Table 2. Age, gender, disease duration, biological therapy survival, number of previous biological agents used and concomitant treatment with synthetic DMARDs or glucocorticoids did not affect the final serological status. There was a higher proportion of naïve subjects with final detectable B Ab than non-naïve patients (75.0% of naïve patients vs 57.1% of non-naïve ones p = 0.023). We found no significant differences between immunological responses depending on the biological DMARD used. Rituximab did not show better outcomes when increasing the interval between the vaccination and the administration of rituximab over to 12 weeks. Seropositive rates did not decrease when the interval between the vaccination and the final Ab determination increased. Furthermore, when we compared the proportion of final seropositive patients depending on the duration of this lapse, we found it was higher in the group with an interval longer than 10 weeks (71.0% vs 48.2%, p = 0.018). This behaviour remained for 12 week periods. We found a small group of patients that presented a stronger response, consisting in at least double basal titres (n = 15). They were younger than those who did not double (39.1 ± 15.6 years vs 50.2 ± 12.5 years respectively, p = 0.001). Fourteen were on antiTNF inhibitors therapy and none of them on rituximab treatment. Only 2 patients (0.8%) presented a 4 fold raise. Both univariate and multivariate regression analysis identified basal Ag B Ab ≥11.5 as predictive factor of final seropositivity and Psoriasis as predictor of final seronegativity, Fig. 3. Discussion Our study presents the results of influenza vaccination in 253 patients with AIIRD, Psoriasis or Inflammatory bowel diseases, treated with biological therapy, with or without concomitant treatment with synthetic DMARDs and/or glucocorticoids. We found a significant increase of the rate of seropositive patients for Ag B Ab after vaccination during 2014–2015 campaign. However, we did not find the same response for Ag A Ab during the same season probably due to the high percentage of baseline seropositive Ab levels (75.9%) possibly pointing to prior exposure to the virus by direct contamination but not to 2013–2014 vaccination influence. We did not investigate the exposure to other flu seasons’ vaccines, but detectable titres had been described 22 months after vaccination in patients with inflammatory arthritis on synthetic and/or biological DMARDs.8 On the other hand, in our study baseline seropositivity against Ag A was considerably higher than the one reported by others, like Polachek et al. who found Ab against Ag A basal seropositive rates between 24.0 and 40.9% in patients with PsA or Pso respectively.9 We used ELISA technique, that has already been used to evaluate the immunological response to attenuate cold-adapted influenza

5

vaccines in healthy adults10 as well as to measure serological changes after influenza infection. Li et al. studied the Ab responses to A(H1N1) pdm09, in flu infected healthy adults, during 2009 influenza pandemic, measured by ELISA and compared with those obtained by HIA and/or microneutralization. ELISA showed less sensitivity than the other methods.11 This fact contributes to explain the low proportion of patients that became seropositive after vaccination. We found out that patients on biological therapy with final detectable Ab against Ag A were younger than those with no response. Also, patients with stronger immune response against Ag B were younger compared to those whose final positive titres did not double the basal ones - though this was a small group of 15 patients so these results should be interpreted with caution-. Lack of protection following vaccination has been described in the elderly due to the immunosenescence process.12 Although all participants were younger than 65 years, these findings might reflect less vigorous immune responses as age increases. Influenza vaccine effectiveness in patients on anti-TNF treatment has been assessed in different studies with diverse methods. Burmester et al. found influenza infection-related adverse effects occurred in 5% of vaccinated patients compared with 14% of patients non vaccinated in a study that included 15,132 patients with RA exposed to adalimumab from 28 RA clinical trials.13 Kivitz et al. used a hemagglutination inhibition antibody assay (HIA) to evaluate post-vaccination seroprotection rates in patients on certolizumab-pegol and found no differences with the results obtained in patients on placebo.14 Kapetanovic et al., also used HIA to investigate the impact of different anti-rheumatic treatments on antibody response to pH1N1 vaccination in patients with RA and SpA. RA patients on rituximab had significantly worse results and SpA on anti-TNF monotherapy significantly better response rates compared to other treatment groups.8 Although we used different methods to evaluate VE we found similar results with Anti-TNF predicting final detectable Ab against Ag A and Rituximab as predictor factor of impaired response. Final seropositivity status was less common in patients on rituximab, but better responses were achieved when the interval between the influenza vaccination and rituximab administration was over 12 weeks, concurring with other reports already published.15 In fact, the ideal timing for administering inactivated vaccines to patients treated with rituximab, has been defined as 4 weeks before or 6 months after the drug infusion.16 Different studies have shown no significant relation between VE and synthetic DMARDs treatment, mainly MTX and glucocorticoids,3,4,17,18 while other reports have found an impaired response to vaccination in patients treated with MTX, probably related to the unselective inhibition of cell proliferation caused by MTX.19 Our results reveal similar findings to former studies, with no decreased response due to synthetic DMARDs treatment in patients on biological therapy. Regarding glucocorticoids, although we did not demonstrate worse responses associated with this treatment, it is worth highlighting the importance of achieving a good control of disease activity that allows glucocorticoids withdrawal, as its combination with anti-TNF-␣ has been associated with higher risk of serious infections.20,21 In our study, psoriasis was a predictor factor of final seronegativity. We have not found comparable results in previous studies. Nevertheless, this result has to be interpreted cautiously due to the small sample size (13 patients). The response to the influenza vaccine is achieved approximately 2 weeks after vaccination, reaching a peak between the 4th and 6th

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003

G Model MEDCLI-4783; No. of Pages 7 6

ARTICLE IN PRESS P. Richi et al. / Med Clin (Barc). 2019;xxx(xx):xxx–xxx

weeks. The Ab titres decrease over time, so Advisory Committees have been concerned about the life of seroprotective Ab. Populations that could present more rapid declines, such as people over 60, have been analyzed, showing seroprotective rates over 16 weeks after vaccination.22 We wanted to know if patients on biological therapies continued presenting detectable Ab several months after immunization, as the last 30 years the peak month of flu activity has been predominately February23 and annual vaccination campaigns begin at the end of October. In our study there were no differences between the amount of patients with detectable A Ab before or after 10 weeks of vaccination. Similar results were found when this period was increased until 12 weeks, but in a very small sample (12 subjects, 92.37% seropositive). For B Ab there was a significantly higher proportion of seropositive patients in the groups with longer intervals. It may reflect a late seroconversion that could help to protect individuals against influenza B activity that frequently reaches its peak in the later influenza season.24,25 We investigated the adherence to vaccination recommendations and found that almost half of patients had not received the influenza vaccine during the previous flu season, while in other studies unvaccinated patients reached 60–80%.26,27 In patients on biological treatment, aged over 60 tended to be associated with higher probability of immunization as has been reported by other authors.28 Consequently, age continues to be recognized as a risk factor for influenza infection and its complications while synthetic and biological DMARDs are not. On the other hand, even though rituximab has been related to worse responses,29 it was found to be the biological agent associated with the highest rate of previous vaccination, reflecting probably an increasing awareness of the risk of infection related to B-cell depletion therapies between clinicians. This risk is also associated with the rest of immunosuppressive treatments so a great effort must be made in terms of implementation of vaccination recommendations for immunosuppressed subjects. Patient’s lack of knowledge has been identified as one of the principle reasons for not receiving vaccines.26 Providing information to patients and physicians is an efficient way to increase vaccination rates.30 Our study has some limitations. The first one was assessing VE. Organizations like Centers of Disease Control and Prevention (CDC) and the U.S. Flu VE Network have been working together since the 2003–2004 flu season, using observational studies to estimate how effective the flu vaccine is. These investigations compare the ratio of vaccination among outpatients with acute respiratory illness and laboratory-confirmed by RT-PCR influenza infection to the ratio of vaccination among outpatients with acute respiratory illness who test negative for influenza infection. We were not able to carry out an observational study as a huge sample of subjects is needed. Therefore we decided to evaluate the immune response to vaccination assessing the presence or absence of detectable titres after immunization, instead of the rate of infection in those vaccinated vs those unvaccinated. On the other hand, we found a group of patients in each category that turned to negative after vaccination. As mentioned above, ELISA results lower than 11.5 but higher than 8.5 AU, are considered by the technique supplier as “grey zone”, but we recorded them as negative to avoid false positive records. Patients with Ab titres over but near 11.5 AU, could change easily its final category to negative, just with a decrease of 0.2 AU. This could explain the change from positivity to negativity present in some patients and contributes to decrease the amount of final seropositive subjects. The sample involved different diagnosis, some of them with a reduced number of patients (6 with UA, 13 with Pso, 7 with CTD and 8 with IBD). This could be another limitation of the study but we decided to include all data, as it was a representation of the habitual population on biological DMARs, attended in our clinics.

Despite the relative small number of patients vaccinated in several of the biological therapy groups, significant differences in the response to vaccination were found between patients with Rituximab and those in the anti-TNF group. A larger number of patients should be included in the study in order to test significance in the differences in the vaccination response for the rest of the biological therapies. In conclusion, our study shows that basal seroposivity against Influenza and Anti-TNF agents are predictive factors of final seropositive status between patients on biological therapy. Furthermore, Rituximab is the biological DMARD associated with worse results, but better responses can be achieved with an accurate schedule that combines rituximab doses, influenza vaccine and adequate time intervals between them. Fundings This study were funded by Sociedad de Reumatología Comunidad de Madrid (SORCOM)/MSD and Pfizer. The funding sources had no role in the study design, collection, analysis or interpretation of the data, writing of the manuscript, or in the decision to submit the manuscript for publication. Conflict of interest Patricia Richi declares she has received research grants from Sociedad de Reumatología Comunidad de Madrid (SORCOM)/MSD and Pfizer. ˜ Santiago Munoz-Fernández declares he has received research grants from Sociedad de Reumatología Comunidad de Madrid (SORCOM)/MSD and Pfizer. The rest of authors declare that they have no conflict of interest. References 1. Blumentals WA, Arreglado A, Napalkov P, Toovey S. Rheumatoid arthritis and the incidence of influenza and influenza-related complications: a retrospective cohort study. BMC Musculoskelet Disord. 2012;13:158–67, http://dx.doi.org/10.1186/1471-2474-13-158. 2. van Assen S, Agmon-Levin N, Elkayam O, Cervera R, Doran MF, Dougados M, et al. EULAR recommendations for vaccination in adult patients with autoimmune inflammatory rheumatic diseases. Ann Rheum Dis. 2011;70:414–22, http://dx.doi.org/10.1016/j.autrev.2010.12.003. 3. Kostianovsky A, Charles P, Alves JF, Goulet M, Pagnoux C, Le Guern V, et al. Immunogenicity and safety of seasonal and 2009 pandemic A/H1N1 influenza vaccines for patients with autoimmune diseases: a prospective, monocentre trial on 199 patients. Clin Exp Rheumatol. 2012;30 Suppl. 70:S83–9. 4. Kubota T, Nii T, Nanki T, Kohsaka H, Harigai M, Komano Y, et al. Anti-tumor necrosis factor therapy does not diminish the immune response to influenza vaccine in Japanese patients with rheumatoid arthritis. Mod Rheumatol. 2007;17:531–3, http://dx.doi.org/10.1007/s10165-007-0632-5. 5. Saad CG, Borba EF, Aikawa NE, Silva CA, Pereira RM, Calich AL, et al. Immunogenicity and safety of the 2009 non-adjuvanted influenza A/H1N1 vaccine in a large cohort of autoimmune rheumatic diseases. Ann Rheum Dis. 2011;70:1068–73, http://dx.doi.org/10.1136/ard.2011.150250. 6. Adler S, Krivine A, Weix J, Rozenberg F, Launay O, Huesler J, et al. Protective effect of A/H1N1 vaccination in immune-mediated disease–a prospectively controlled vaccination study. Rheumatology (Oxford). 2012;51:695–700, http://dx.doi.org/10.1093/rheumatology/ker389. 7. World Health Organization [Internet]. Recommended composition of influenza virus vaccines for use in the 2014–2015 northern hemisphere influenza season; [about 2 screens]. Genève: World Health Organization; 2014. Available from: http://www.who.int/influenza/vaccines/virus/recommendations/2014 15 north/en/ [accessed 12.01.17]. 8. Kapetanovic MC, Kristensen LE, Saxne T, Aktas T, Mörner A, Geborek P. Impact of anti-rheumatic treatment on immunogenicity of pandemic H1N1 influenza vaccine in patients with arthritis. Arthritis Res Ther. 2014;16:R2, http://dx.doi.org/10.1186/ar4427. 9. Polachek A, Korobko U, Mader-Balakirski N, Arad U, Levartovsky D, Kaufman I, et al. Immunogenecity and safety of vaccination against seasonal 2012 influenza virus among patients with psoriatic arthritis and psoriasis. Clin Exp Rheumatol. 2015;33:181–6. 10. Karron RA, Talaat K, Luke C, Callahan K, Thumar B, Dilorenzo S, et al. Evaluation of two live attenuated cold-adapted H5N1 influenza virus vaccines in

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003

G Model MEDCLI-4783; No. of Pages 7

ARTICLE IN PRESS P. Richi et al. / Med Clin (Barc). 2019;xxx(xx):xxx–xxx

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

healthy adults. Vaccine. 2009;27:4953–60, http://dx.doi.org/10.1016/j.vaccine. 2009.05.099. Li ZN, Lin SC, Carney PJ, Li J, Liu F, Lu X, et al. IgM, IgG, and IgA antibody responses to influenza A (H1N1)pdm09 hemagglutinin in infected persons during the first wave of the 2009 pandemic in the United States. Clin Vaccine Immunol. 2014;21:1054–60, http://dx.doi.org/10.1128/CVI. 00129-14. Goodwin K, Viboud C, Simonsen L. Antibody response to influenza vaccination in the elderly: a quantitative review. Vaccine. 2006;24:1159–69. Burmester GR, Landewé R, Genovese MC, Friedman AW, Pfeifer ND, Varothai NA, et al. Adalimumab long-term safety: infections, vaccination response and pregnancy outcomes in patients with rheumatoid arthritis. Ann Rheum Dis. 2017;76:414–7, http://dx.doi.org/10.1136/annrheumdis-2016-209322. Kivitz AJ, Schechtman J, Texter M, Fichtner A, de Longueville M, Chartash EK. Vaccine responses in patients with rheumatoid arthritis treated with certolizumab pegol: results from a single-blind randomized phase IV trial. J Rheumatol. 2014;41:648–57, http://dx.doi.org/10.3899/jrheum.130945. Gabay C, Bel M, Combescure C, Ribi C, Meier S, Posfay-Barbe K, et al. Impact of synthetic and biologic disease-modifying antirheumatic drugs on antibody responses to the AS03-adjuvanted pandemic influenza vaccine: a prospective, open-label, parallel-cohort, single-center study. Arthritis Rheum. 2011;63:1486–96, http://dx.doi.org/10.1002/art.30325. Pham T, Fautrel B, Gottenberg JE, Goupille P, Hachulla E, Masson C, et al. Rituximab (MabThera) therapy and safety management. Clinical tool guide. Joint Bone Spine. 2008;75 Suppl. 1:S1–99, http://dx.doi.org/10.1016/S1297-319X(08)73620-0. Fomin I. Vaccination against influenza in rheumatoid arthritis: the effect of disease modifying drugs, including TNF blockers. Ann Rheum Dis. 2006;65:191–4, http://dx.doi.org/10.1136/ard.2005.036434. Kobashigawa T, Nakajima A, Taniguchi A, Inoue E, Tanaka E, Momohara S, et al. Vaccination against seasonal influenza is effective in Japanese patients with rheumatoid arthritis enrolled in a large observational cohort. Scand J Rheumatol. 2013;42:445–50, http://dx.doi.org/10.3109/03009742.2013.788733. Kapetanovic MC, Saxne T, Sjöholm A, Truedsson L, Jönsson G, Geborek P. Influence of methotrexate, TNF blockers and prednisolone on antibody responses to pneumococcal polysaccharide vaccine in patients with rheumatoid arthritis. Rheumatology (Oxford). 2006;45:106–11, http://dx.doi.org/10.1093/rheumatology/kei193. ˜ T, Descalzo MÁ, Loza-Santamaría E, Carmona L, Munoz-Fernández ˜ Cobo-Ibánez S. Serious infections in patients with rheumatoid arthritis and other immunemediated connective tissue diseases exposed to anti-TNF or rituximab: data

21.

22.

23.

24.

25.

26.

27. 28.

29.

30.

7

from the Spanish registry BIOBADASER 2.0. Rheumatol Int. 2014;34:953–61, http://dx.doi.org/10.1007/s00296-014-2945-y. Germano V, Cattaruzza MS, Osborn J, Tarantino A, Di Rosa R, Salemi S, et al. Infection risk in rheumatoid arthritis and spondyloarthropathy patients under treatment with DMARDs, corticosteroids and TNF-␣ antagonists. J Transl Med. 2014;12:77, http://dx.doi.org/10.1186/1479-5876-12-77. Skowronski DM, Tweed SA, De Serres G. Rapid decline of influenza vaccineinduced antibody in the elderly: is it real, or is it relevant? J Infect Dis. 2008;197:490–502, http://dx.doi.org/10.1086/524146. Centers for Disease Control and Prevention. [Internet]. Atlanta. Centers for Disease Control and Prevention; c 1990–2018 [accessed 12.01.17]. Influenza (Flu). The Flu Season; [about 3 screens]. Available from: https://www.cdc.gov/flu/about/season/flu-season.htm. Centers for Disease Control and Prevention. [Internet] Atlanta. Centers for Disease Control and Prevention; c 2014 [accessed 12.01.17]. What You Should Know for the 2014–2015 Influenza Season. [about 9 screens]. Available from: http://www.cdc.gov/flu/pastseasons/1415season.htm. Ministerio de Ciencia Innovación y Universidades. [Internet]. Madrid. Instituto de Salud Carlos III; 2015 [accessed 12.01.17]. Informe de Vig˜ ilancia de la Gripe en Espana. Temporada 2015-2016 (Desde la semana 40/2015 hasta la semana 20/2016); [about 2 screens]. Available from: http://www.isciii.es/ISCIII/es/contenidos/fd-servicios-cientificotecnicos/fdvigilancias-alertas/fd-enfermedades/gripe.shtml. Pradeep J, Watts R, Clunie G. Audit on the uptake of influenza and pneumococcal vaccination in patients with rheumatoid arthritis. Ann Rheum Dis. 2007;66:837–8, http://dx.doi.org/10.1136/ard.2006.060285. Gluck T. Vaccinate your immunocompromised patients! Rheumatology. 2006;45:9–10, http://dx.doi.org/10.1093/rheumatology/kei237. Krasselt M, Ivanov J-P, Baerwald C, Seifert O. Low vaccination rates among patients with rheumatoid arthritis in a German outpatient clinic. Rheumatol Int. 2017;37:229–37, http://dx.doi.org/10.1007/s00296-016-3608-y. Hua C, Barnetche T, Combe B, Morel J. Effect of methotrexate, anti-tumor necrosis factor ␣, and rituximab on the immune response to influenza and pneumococcal vaccines in patients with rheumatoid arthritis: a systematic review and metaanalysis. Arthritis Care Res (Hoboken). http://dx.doi.org/10.1002/acr.22246. Lanternier F, Henegar C, Mouthon L, Blanche P, Guillevin L, Launay O. Low influenza-vaccination rate among adults receiving immunosuppressive therapy for systemic inflammatory disease. Ann Rheum Dis. 2008;67:1047, http://dx.doi.org/10.1136/ard.2007.081703.

Please cite this article in press as: Richi P, et al. Antibody responses to influenza vaccine in patients on biological therapy: Results of RIER cohort study. Med Clin (Barc). 2019. https://doi.org/10.1016/j.medcli.2019.02.003