Autoantibody response to adjuvant and nonadjuvant H1N1 vaccination in systemic lupus erythematosus

Authors

  • Murray B. Urowitz,

    1. University of Toronto Lupus Clinic and Centre for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Toronto, Ontario, Canada
    Search for more papers by this author
  • Anoja Anton,

    1. University of Toronto Lupus Clinic and Centre for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Toronto, Ontario, Canada
    Search for more papers by this author
  • Dominique Ibanez,

    1. University of Toronto Lupus Clinic and Centre for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Toronto, Ontario, Canada
    Search for more papers by this author
  • Dafna D. Gladman

    Corresponding author
    1. University of Toronto Lupus Clinic and Centre for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Toronto, Ontario, Canada
    • Toronto Western Hospital, 399 Bathurst Street, 1E-410B, Toronto Ontario, M5T 2S8, Canada
    Search for more papers by this author

Abstract

Objective

It has been reported that influenza vaccination increases autoantibody production and/or disease activity in a significant proportion of patients with systemic lupus erythematosus (SLE). During the recent H1N1 epidemic, we investigated whether the use of adjuvant- and nonadjuvant-containing H1N1 vaccine induced increased autoantibody production in patients with SLE.

Methods

Patients with SLE who received H1N1 vaccination and had a battery of 9 autoantibodies tested before and 1 and 3 months after vaccination were included. Antibodies tested included rheumatoid factor (nephelometry), antinuclear antibody (immunofluorescence), anti-DNA (Farr), anti-RNP, anti-Sm, anti-Ro, anti-La, anti–Scl-70, and anti–Jo-1 (enzyme-linked immunosorbent assay). Patients were evaluated by standard protocol, including items necessary to calculate the Systemic Lupus Erythematosus Disease Activity Index 2000 and the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index. Descriptive statistics and McNemar's test were performed to evaluate change in antibody positivity. Multivariate logistic regression was performed to adjust for repeated measures in the comparisons of autoantibodies over visits and vaccine types.

Results

One hundred three patients (94 women, 9 men) with a mean ± SD age at vaccination of 43.9 ± 15.2 years and a mean ± SD disease duration of 14.2 ± 11.0 years were included. Fifty-one patients received adjuvant and 52 received nonadjuvant vaccines. Antibody testing was performed a mean of 1.9 months prior to the vaccination. The first postvaccination sample was taken a mean of 1 month after vaccination and the second a mean of 3.5 months after vaccination. The percentage of patients with changes in antibodies following vaccination was not statistically significant for most antibodies. After adjusting for the number of tests performed, none of the associations was significant.

Conclusion

H1N1 vaccination (both adjuvant and nonadjuvant) did not increase the levels of autoantibodies in patients with SLE.

INTRODUCTION

Systemic lupus erythematosus (SLE) is characterized by B cell hyperreactivity and consequent increase of autoantibody production. Concurrently with active disease, SLE patients are more susceptible to infection. Therefore, it is important to ensure protection against viral and bacterial agents when available (1). However, immunization with a foreign protein might further enhance B cell hyperactivity with consequent production of potentially pathogenic autoantibodies.

In the fall of 2009/winter of 2010, there was a worldwide alert for H1N1 influenza infection. The recommendations of the World Health Organization, as well as local health authorities, were to vaccinate the population against this influenza strain. Vaccines, both those with adjuvant and those without adjuvant, were produced, with the latter reserved for pregnant women.

It has been reported that influenza vaccination increases autoantibody production and/or disease activity in a proportion of patients with SLE (2). An additional concern would be whether adjuvant-containing vaccinations might further enhance autoantibody production in these patients.

During the recent H1N1 epidemic, we investigated whether the use of adjuvant- and nonadjuvant-containing H1N1 vaccine induced increased autoantibody production in patients with SLE.

Significance & Innovations

  • Large cohort of patients with systemic lupus erythematosus receiving H1N1 vaccination.

  • Both adjuvant and nonadjuvant vaccines were tested.

  • No effect on disease activity was noted.

  • No significant effect on antibody production was noted.

PATIENTS AND METHODS

Patient selection.

Patients with SLE followed at the University of Toronto Lupus Clinic who received H1N1 vaccination (with or without adjuvant) and had a battery of 9 autoantibodies tested before and 1 and 3 months after vaccination were included.

Assessments.

Patients were evaluated according to the standard protocol at the lupus clinic. The protocol includes clinical and laboratory assessments that allow the calculation of disease activity according to the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K), a valid disease activity measure for SLE (3). The protocol also allows the calculation of accumulated damage according to the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI), a valid measure to assess damage in SLE (4).

Antibody measurements.

Antibodies tested included rheumatoid factor (RF; nephelometry), antinuclear antibody (immunofluorescence), anti-DNA (Farr), anti-RNP, anti-Sm, anti-Ro, anti-La, anti–Scl-70, and anti–Jo-1 (enzyme-linked immunosorbent assay).

Statistical analysis.

Descriptive statistics and McNemar's test were performed to evaluate change in antibody positivity. These were assessed in the adjuvant and the nonadjuvant groups separately. Multivariate logistic regression adjusting for repeated values was used to compare the levels of autoantibodies over visits and vaccine types.

RESULTS

One hundred three patients (94 women, 9 men) were included. The mean ± SD age at vaccination was 43.9 ± 15.2 years, the mean ± SD disease duration was 14.2 ± 11.0 years, the mean ± SD SLEDAI-2K score was 4.38 ± 4.28, and the mean ± SD SDI score was 1.26 ± 1.52. Sixty-four percent of the patients were taking steroids, 79% were receiving antimalarials, and 62% were receiving immunosuppressive agents. Fifty-one patients received adjuvant and 52 received nonadjuvant vaccines (Table 1).

Table 1. Characteristics of the patient population at the time of vaccination*
 Value
  • *

    SLEDAI-2K = Systemic Lupus Erythematosus Disease Activity Index 2000; SDI = Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index.

Female sex, no. (%)94 (91.3)
Age, mean ± SD years43.9 ± 15.2
Disease duration, mean ± SD years14.2 ± 11.0
Steroids, %63.7
Immunosuppressive agents, %61.8
Antimalarials, %79.4
SLEDAI-2K, mean ± SD4.38 ± 4.28
SDI, mean ± SD1.26 ± 1.52

Antibody testing was performed a mean ± SD of 1.9 ± 2.8 months prior to the vaccination. The first postvaccination sample was taken a mean ± SD of 1 ± 0.5 month after vaccination and the second sample was taken a mean ± SD of 3.5 ± 1.0 months after vaccination. Table 2 shows the frequency of antibody positivity at baseline and the 2 followup visits postvaccination for all patients in whom serum was available at each time point. As can be seen, only RF and anti-Ro showed a statistically significant change in frequency between visits. However, only RF and anti-DNA antibodies are reported numerically. The actual mean values did not change significantly (for RF: mean ± SD 15.1 ± 42.9 prevaccination visit, 18.2 ± 22.0 first postvaccination visit, and 19.6 ± 33.1 second postvaccination visit; for DNA: mean ± SD 17.2 ± 29.6 prevaccination visit, 16.3 ± 26.3 first postvaccination visit, and 15.5 ± 26.8 second postvaccination visit). Antinuclear antibodies are reported in titers and the median value remained unchanged. All other autoantibodies are reported as positive or negative. Comparing the results for vaccine type, we find that there is no difference between adjuvant and nonadjuvant vaccines.

Table 2. Autoantibody positivity at prevaccination and in the first and second followup visits in all patients*
 Prevaccination, no./total (%)First postvaccination, no./total (%)Second postvaccination, no./total (%)Comparing over the different visits, PComparing between vaccine type, P
  • *

    RF = rheumatoid factor; ANA = antinuclear antibody; N/A = not applicable.

  • Logistic regression adjusting for repeated measures.

  • Regression could not be performed due to the low event rate.

RF14/89 (15.7)17/68 (25.0)25/89 (28.1)0.010.91
DNA31/91 (34.1)23/68 (33.8)17/50 (34.0)0.560.57
La15/93 (16.1)9/74 (12.2)13/92 (14.1)0.450.79
RNP24/93 (25.8)20/74 (27.0)24/92 (26.1)0.600.18
Jo-14/93 (4.3)4/73 (5.5)2/90 (2.2)0.570.87
Positive ANA73/92 (79.4)58/71 (81.7)70/95 (73.7)0.300.11
Ro44/91 (48.4)29/74 (39.2)33/92 (35.9)0.040.34
Sm10/93 (10.8)7/74 (9.5)12/92 (13.0)0.130.17
Scl-701/93 (1.1)1/73 (1.4)0/90 (0)N/AN/A

The percentage of patients with changes in antibodies following vaccination was not statistically significant for most antibodies. For RF, 80% remained unchanged, whereas 15% who were negative became positive and 5% who were positive became negative (McNemar's P = 0.07). This was more pronounced in the nonadjuvant group than in the adjuvant group. In the nonadjuvant group, 88% remained unchanged, 12% converted to positive, and none converted to negative (P = 0.03), whereas in the adjuvant group, 71% remained unchanged, 18% converted to positive, and 12% converted to negative (P = 0.53).

For anti-Ro antibodies, 83% remained unchanged, whereas 3% converted to positive and 14% converted to negative (P = 0.03), but no difference was significant when looking at the vaccine type. No other antibodies changed significantly in either group. After adjusting for the number of tests performed, none of the associations remained significant.

Sixty-eight patients had SLEDAI-2K values prior to and following the second postvaccination visit. The mean ± SD SLEDAI-2K prevaccination was 4.22 ± 4.41, while the value postvaccination was 3.90 ± 4.06 (paired t-test P = 0.39). At the next followup clinic visit (mean ± SD 4.5 ± 1.7 months), 11.5% of the patients had a flare of their disease, with the SLEDAI-2K score increased by ≥4. However, in our database, 10.5% of the patients have a flare between 2 consecutive visits (P = 0.78).

DISCUSSION

Vaccines are an important preventive measure for patients with SLE because of their increased susceptibility to infection that results from the disease process or its treatment. On the other hand, patients with SLE are prone to autoantibody production, which may be aggravated by foreign protein injections. We have now shown that vaccination against H1N1 influenza, both with and without adjuvant, did not lead to an increase in autoantibody production. A study by Kanakoudi-Tsakalidou et al showed that in 70 children with chronic rheumatic diseases receiving influenza vaccine of whom only 11 had SLE, there was no increase in autoantibody production (5). Abu-Shakra et al showed that there was no increase in anti-DNA antibody production post–influenza virus vaccination in 24 patients with lupus. In their study, although there were increases in antibodies to RNP, SM, Ro, La, and anticardiolipin 6 weeks postvaccination, the increases were usually transient, and improved at 12 weeks (6). Mercado studied the effect of vaccination in 18 women with SLE and showed that anti-DNA antibodies did not change after vaccination (1). Del Porto et al studied 14 patients with SLE and showed no increases in antinuclear antibodies, anti–double-stranded DNA, extractable nuclear antigen, RF, anticardiolipin, and anti–β2-glycoprotein I antibody titers (7). Our study included more patients than those previously reported, and confirms that there is no increase in autoantibody production postvaccination for vaccines with and without adjuvant.

Abu-Shakra et al (8) compared 24 SLE patients receiving inactivated influenza virus vaccine with 24 patients who were not vaccinated, and each group was followed for 6 and 12 weeks. They demonstrated that the SLEDAI scores of cases and controls at each assessment were not statistically different, suggesting that the vaccination did not lead to an increase in disease activity. Mercado also showed that disease activity and autoantibody did not change after influenza vaccination (1). Therefore, H1N1 vaccination (both adjuvant and nonadjuvant) did not increase the levels of autoantibodies in patients with SLE or cause an increase in disease activity.

A weakness of our study is that we did not measure anti-H1N1 antibodies, and therefore could not confirm that the vaccination was effective. However, the strength of our study is the large number of patients in a single center with predefined outcome criteria that confirm the safety of vaccination in patients with SLE. In addition, we were able to confirm the safety of vaccines with and without adjuvant.

A recent Chinese study demonstrated the safety of H1N1 influenza vaccine in postmarketing surveillance (9). Therefore, the influenza vaccination is safe and should be offered to patients with SLE.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Gladman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Urowitz, Anton, Gladman.

Acquisition of data. Urowitz, Anton, Gladman.

Analysis and interpretation of data. Urowitz, Ibanez, Gladman.

Ancillary