Antibody response is reduced following vaccination with 7-valent conjugate pneumococcal vaccine in adult methotrexate-treated patients with established arthritis, but not those treated with tumor necrosis factor inhibitors†
ClinicalTrials.gov identifier: NCT00828997; EudraCT database no. EU 2007-006539-29.
To study the influence of antiinflammatory treatments, including methotrexate (MTX) and tumor necrosis factor (TNF) inhibitors, on antibody response following vaccination using a 7-valent conjugate pneumococcal vaccine in adult patients with established arthritis.
Patients with rheumatoid arthritis (RA) or spondylarthropathy (SpA) (including psoriatic arthritis) were vaccinated (n = 505). All patients were stratified into 6 prespecified groups based on diagnosis and treatment (RA patients receiving MTX, RA patients receiving anti-TNF agents and MTX, RA patients receiving TNF inhibitors as monotherapy, SpA patients receiving anti-TNF agents and MTX, SpA patients receiving TNF inhibitors as monotherapy, and SpA patients receiving nonsteroidal antiinflammatory drugs [NSAIDs] and/or analgesics). SpA patients receiving only NSAIDs/analgesics served as a control group. All patients received 1 dose (0.5 ml) of vaccine intramuscularly. Levels of IgG antibodies against 23F and 6B serotypes were measured at vaccination and at 4–6 weeks following vaccination, using standardized enzyme-linked immunosorbent assays.
Positive antibody response was defined as an antibody response ratio (ARR) (i.e., ratio of post- to prevaccination antibody levels) of ≥2. The ARR differed significantly between the groups. A better ARR was seen among patients in the control group compared to those in groups treated with MTX or MTX in combination with TNF inhibitors. Among patients treated with TNF inhibitors as monotherapy, ARRs for both serotypes were lower numerically, but were not significantly different, compared to those in controls. Ongoing MTX treatment was predictive of reduced response (odds ratio 0.41 [95% confidence interval 0.24–0.68], P = 0.001). Higher age was associated with impaired positive antibody response. Concomitant prednisolone treatment elicited better positive antibody response in patients with RA.
Treatment with MTX and higher age were predictive of an impaired antibody response to the 7-valent conjugate pneumococcal vaccine in this cohort of patients with chronic arthritis. TNF inhibitors did not significantly affect antibody responses.
Patients with rheumatoid arthritis (RA) and spondylarthropathies (SpA), including psoriatic arthritis, receiving traditional disease-modifying antirheumatic drugs (DMARDs), corticosteroids, and/or biologic agents have an increased risk of vaccine-preventable diseases, including those caused by Streptococcus pneumoniae (1–9). A standard 23-valent polysaccharide vaccine is recommended by the Centers for Disease Control and Prevention Advisory Committee on Immunization Practices, the European Centre for Disease Prevention and Control, the World Health Organization, and the Swedish National Board of Health and Welfare for all individuals ages 65 and older as well as for those receiving immunosuppressive treatment (10, 11). Despite these recommendations, vaccination coverage among patients with chronic arthritis is lower than in the general population (12). Possible explanations include concern about vaccination safety, but also its immunogenicity and efficacy.
We have previously reported the serologic response following pneumococcal vaccination using standard 23-valent polysaccharide vaccine in patients with RA who have been treated with different antiinflammatory medications (13). Serologic response was significantly lower in RA patients receiving methotrexate (MTX) compared to those receiving tumor necrosis factor (TNF) inhibitors or to healthy controls. Furthermore, <50% of patients developed positive antibody response, defined as a ≥2-fold increase in antibody levels. In contrast, when using polypeptide immunization (influenza vaccine) we found that >90% of the patients developed protective levels of antibodies, regardless of treatment (14).
The effectiveness of standard 23-valent polysaccharide vaccine in preventing invasive pneumococcal diseases in adults has been the subject of debate (15, 16). This vaccine contains only polysaccharide antigens and stimulates antibody response in a T cell–independent manner. To enhance antibody response, a novel vaccine that includes 7 pneumococcal serotypes (4, 6B, 9V, 14, 18C, 19F, and 23F) conjugated to a mutant diphtheria toxin (CRM197) as a carrier protein has been developed (17). Since antigen-presenting B cells may take up the glycoconjugate and present the peptides to T helper cells, this probably results in T cell–dependent stimulation of B cells and their maturation into plasma cells and memory cells (18, 19). We hypothesized that, in arthritis patients treated with potent immunomodulating drugs, a protein–polysaccharide challenge would elicit better antibody response compared to a polysaccharide challenge only. Our hypothesis is supported by findings of superior antibody response to pneumococcal conjugate vaccine compared to those of polysaccharide vaccine in patients with chronic obstructive pulmonary disease (20). Corresponding studies involving patients with rheumatic diseases are lacking.
The objective of the current study was to examine the impact of different immunosuppressive treatments, including MTX and TNF inhibitors, on antibody response following 7-valent conjugate pneumococcal vaccination in patients with established arthritis. Serotypes 23F and 6B were identified among the 8 most common serotypes of invasive S pneumoniae isolates reported to the Swedish Institute for Infectious Disease Control during 1988–1998; for this reason, we chose these serotypes to study antibody response (21). We compared antibody responses in different treatment groups to those in a control group consisting of patients with SpA not receiving immunosuppressive treatment. In addition, we studied the tolerability of this vaccine and possible influence of the vaccination on existing rheumatic disease.
PATIENTS AND METHODS
Adult patients who had established arthritis and were regularly followed up at the Department of Rheumatology in Lund and Malmö at Skåne University Hospital were eligible for this study. Patients in the RA group had to fulfill the American College of Rheumatology 1987 RA classification criteria (22) and patients in the SpA group had to have been diagnosed as having ankylosing spondylitis (AS), an undifferentiated spondylarthritis (including those related to inflammatory bowel disease), or psoriatic arthritis. Patients were stratified into 6 prespecified groups based on diagnosis and treatment, as follows: 1) RA patients receiving MTX (with or without other DMARDs) (group 1); 2) RA patients receiving anti-TNF agents and MTX (with or without other DMARDs) (group 2); 3) RA patients receiving TNF inhibitors as monotherapy (group 3); 4) SpA patients receiving anti-TNF agents in combination with MTX (group 4); 5) SpA patients receiving anti-TNF agents as monotherapy (group 5); and 6) SpA patients receiving nonsteroidal antiinflammatory drugs (NSAIDs) and/or analgesics (control group) (group 6). Patients who had changes in antirheumatic treatment within 4 weeks before and 6 weeks after vaccination were not included in the study. Other exclusion criteria included previous vaccination with standard 23-valent pneumococcal polysaccharide within 5 years, history of allergic reaction at previous vaccinations, pregnancy, or ongoing infection.
Consecutive patients fulfilling inclusion criteria were invited to participate in the study. Oral and written information was provided to all subjects who were invited to participate, and written informed consent was obtained from each participant before enrollment.
All participants received a single 0.5-ml dose of the 7-valent conjugate pneumococcal vaccine administrated intramuscularly by a rheumatology nurse. At the time of vaccination, a clinical examination was performed by a rheumatologist and data were collected on disease and treatment characteristics, previous vaccinations, and comorbidity, using a structured followup program developed in southern Sweden (23). All patients were encouraged to monitor and report possible adverse or unexpected effects of the vaccination, as well as changes in rheumatic disease. Adverse events (AEs) and adverse drug reactions (ADRs) were recorded according to the Guideline for Good Clinical Practice and Clinical Safety Data Management (24).
Measurement of IgG specific for S pneumoniae capsular polysaccharides.
Serum samples were collected immediately prior to vaccination and after 4–6 weeks. Human IgG antibodies specific for S pneumoniae capsular polysaccharides 23F and 6B were quantified using enzyme-linked immunosorbent assays (ELISAs) meeting World Health Organization standards, as previously described (13, 25). Briefly, ELISA plates were coated with the polysaccharides 23F or 6B. Dilutions of human sera absorbed with pneumococcal cell wall polysaccharide were then added to the ELISA plates. A reference serum was included on all plates. The serotype-specific antibodies for 23F and 6B were detected using alkaline phosphatase–conjugated goat anti-human IgG (γ-chain specific) F(ab′)2 fragments, followed by addition of the substrate p-nitrophenyl phosphate. Thus, the influence of rheumatoid factor (RF) in the measurement of antibodies could be avoided. The optical density at 405 nm was measured with an ELISA plate reader. The optical density of the colored end product is proportional to the amount of specific antibodies present in the serum. The lower limit of detection was 0.02 mg/liter for serotype 6B and 0.01 for 23F.
At our laboratory, antibody levels of ≥1 mg/liter are generally considered protective. However, as a protective level of serum antibody has not been strictly defined and may differ among serotypes, these values were not used to identify individuals with probable or possible protective levels of antibodies. In order to study the impact of different antirheumatic treatments, we used the antibody response ratio (ARR) (i.e., the ratio of post- to prevaccination antibody levels). A positive antibody response was defined as an ARR of ≥2.
Powered calculation revealed that a population of ∼500 patients was needed in order to detect a 20% difference between the different treatment groups with 80% power. Differences between treatment groups were analyzed using the chi-square test and the Mann-Whitney U test when appropriate. Geometric mean antibody levels were calculated from log-transformed values, and differences were compared by paired-samples t-test. Possible predictors of positive antibody response were analyzed using univariate and multivariate logistic regression models.
Patient and treatment characteristics.
The study population comprised 505 adult patients with established inflammatory rheumatic disease (253 patients with RA, 121 patients with psoriatic arthritis, 78 patients with ankylosing spondylitis, 53 patients with another form of SpA) who were receiving different antiinflammatory treatments. Patient and disease characteristics in the prespecified treatment groups are shown in Table 1. The groups of RA patients receiving MTX had a shorter disease duration, a lower score on the Health Assessment Questionnaire (HAQ) (26), and a higher number of swollen and tender joints compared to the group of RA patients not treated with MTX. The group of SpA patients receiving anti-TNF agents in combination with MTX had a higher HAQ score, a higher 28-joint Disease Activity Score (DAS28) (27), and a greater proportion of women and smokers compared to other groups of patients with SpA (Table 1).
Table 1. Demographic and disease characteristics at the time of vaccination*
Group 1 (n = 85)
Group 2 (n = 89)
Group 3 (n = 79)
P between groups 1–3
Group 4 (n = 83)
Group 5 (n = 83)
Group 6 (n = 86)
P between groups 4–6
Differences between treatment groups were calculated separately for rheumatoid arthritis (RA) and spondylarthropathy (SpA). P values were determined by Mann-Whitney U test or chi-square test. Group 1 = RA patients receiving methotrexate (MTX) (with or without other disease-modifying antirheumatic drugs [DMARDs]); group 2 = RA patients receiving anti–tumor necrosis factor (anti-TNF) agents and MTX (with or without other DMARDs); group 3 = RA patients receiving anti-TNF agents as monotherapy; group 4 = SpA patients receiving anti-TNF agents in combination with MTX; group 5 = SpA patients receiving anti-TNF agents as monotherapy; group 6 = SpA patients receiving nonsteroidal antiinflammatory drugs and/or analgesics (control group). NS = not significant; HAQ = Health Assessment Questionnaire; DAS28 = Disease Activity Score in 28 joints; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; TJC = tender joint count; SJC = swollen joint count; RF = rheumatoid factor; anti-CCP = anti–cyclic citrullinated peptide.
Sex, % female
Age, mean ± SD (range) years
61.5 ± 14 (24–85)
60.1 ± 10 (31–81)
59.8 ± 14 (24–86)
50.4 ± 11 (25–70)
49.2 ± 12 (26–76)
51.6 ± 12 (22–75)
Disease duration, mean ± SD (range) years
11.4 ± 10 (0–40)
16.2 ± 12 (1–48)
20.6 ± 11 (1–48)
<0.001 1–3; 0.004 1–2; 0.007 2–3
12.7 ± 9 (0–40)
15.8 ± 11 (1–44)
12.7 ± 12 (0–45)
HAQ score, mean ± SD
0.7 ± 0.6
0.9 ± 0.7
1.2 ± 0.7
<0.001 1–3; 0.056 1–2; 0.001 2–3
0.7 ± 0.5
0.5 ± 0.5
0.5 ± 0.5
0.019 4–5; 0.008 4–6
DAS28, mean ± SD
3.7 ± 1.2
3.4 ± 1.2
3.9 ± 1.1
3.1 ± 1.2
2.5 ± 12
3.0 ± 1.1
0.002 4–5; 0.006 5–6
Ongoing prednisolone treatment, %
Prednisolone dosage, mean mg/week
0.019 1–2; 0.038 2–3
CRP, mean mg/liter
0.019 1–2; 0.009 2–3
<0.001 5–6; 0.001 4–6
ESR, mean mm/hour
MTX, mean mg/week
TJC, mean from 28-joint count (68-joint count)
SJC, mean from 28-joint count (66-joint count)
0.013 1–2; 0.015 2–3
Current smoker, %
0.055 4–5; 0.039 4–6
RF titer, mean gm/liter
RF positive, %
Anti-CCP positive, %
Previous pneumococcal vaccination, %
When all RA patients (n = 253) were compared to all SpA patients (n = 252) significant differences in several baseline and disease characteristics were found. As expected, RA patients were older (P < 0.001), had a longer mean disease duration (P = 0.025), had a higher HAQ score (P < 0.001), and had a higher DAS28 (P < 0.001). A larger proportion of RA patients were female (P < 0.001), were treated with prednisolone (at higher mean dosage) (P < 0.001), and had previously been exposed to pneumococcal vaccine (P < 0.001), while a larger proportion of SpA patients were being treated with concomitant NSAIDs (P = 0.016).
Antibody levels and ARRs.
Prevaccination-specific IgG levels were similar in all groups for the same serotype, despite a higher proportion of RA patients who had previously been vaccinated compared to SpA patients. Postvaccination serotype-specific IgG increased significantly for both serotypes in all groups compared to baseline (Table 2). The ARR differed among the groups. Figures 1A and B show box plots of the ARRs for 6B and 23F according to treatment group. After adjustment for age and sex, the ARR was higher in controls (patients with SpA treated with NSAIDs and/or analgesics) compared to groups of patients treated with MTX (P = 0.046 for 6B and P = 0.002 for 23F) or MTX combined with TNF inhibitors (P < 0.002 for 6B and P < 0.001 for 23F). Among all patients receiving TNF inhibitors as monotherapy, ARRs for both serotypes were somewhat lower numerically but were not significantly different, compared to those in controls. Furthermore, in RA or SpA patients receiving TNF inhibitors in combination with MTX, ARRs were not significantly different from those in RA patients receiving MTX. Among patients treated with MTX, significantly lower ARRs were found for both serotypes compared to those not treated with MTX (P < 0.001 in both). ARRs did not differ between patients treated with TNF inhibitors compared to those not receiving these drugs for either serotype tested.
Table 2. Prevaccination and postvaccination antibody concentrations*
No. (%) of patients with ≥2-fold increase in prevaccination antibody levels for both serotypes
No. (%) of patients with ≥4-fold increase in prevaccination antibody levels for both serotypes
A better antibody response was seen in patients with SpA compared to patients with RA. Additionally, men had better antibody response compared to women. While higher age at vaccination was associated with a lower ARR, no other demographic characteristics or disease activity at vaccination, measured by the DAS28, HAQ score, erythrocyte sedimentation rate, presence of anti–cyclic citrullinated protein antibodies, or RF status, significantly influenced the ARR.
Positive antibody response.
Overall, the pattern of positive antibody response mirrored the ARR in the different treatment groups. The proportions of patients with positive antibody response for each serotype alone and for both serotypes are shown in Figure 2. The proportions of positive antibody responses for each serotype alone and for both serotypes together were significantly lower among the groups of patients receiving MTX, using the univariate logistic regression model, while anti-TNF treatment was not significantly associated with the frequency of positive antibody response. Patients exhibiting positive antibody response were significantly younger (P = 0.017, P = 0.001, and P = 0.002 for serotypes 23F, 6B, and both 23F and 6B, respectively).
Pneumococcal vaccination status and baseline antibody levels.
At vaccination, a large proportion of patients included in this study had putative protective levels of ≥1 mg/liter for each serotype (43.8% for serotype 23F and 67.1% for serotype 6B). A significant proportion of these patients were nonresponders for both serotypes (P = 0.002 for 23F and P = 0.013 for 6B) (univariate logistic regression model). The proportion of all patients with protective antibody levels postvaccination was 73.5% for serotype 23F and 80.2% for serotype 6B.
Altogether, 45 patients (9%) had received a standard 23-valent pneumococcal vaccine >5 years before enrollment in the study. A significantly larger proportion of RA patients in all treatment groups had previously been vaccinated (10.6%, 18.0%, and 16.5% in groups 1, 2, and 3, respectively) compared to SpA patients (2.4%, 4.8% and 1.2%, in groups 4, 5 and 6, respectively) (Table 1). Previously vaccinated patients had lower median pre- and postvaccination antibody levels for both serotypes tested, and their ARRs for the 6B serotype were significantly lower (P = 0.019) and showed a trend toward impaired ARRs for the 23F serotype (P = 0.052 by Mann-Whitney U test).
Possible predictive factors of positive antibody response for both serotypes were analyzed using univariate regression, as well as multivariate logistic regression with adjustment for age, sex, ongoing MTX treatment, ongoing anti-TNF treatment, ongoing prednisolone treatment, patient global assessment on a visual analog scale, and prevaccination antibody levels for 23F and 6B. In a primary regression analysis of all patients included in the study, only higher age and ongoing MTX treatment were identified as predictors of impaired antibody response for both serotypes (Table 3). Neither RA diagnosis nor ongoing anti-TNF treatment had a significant impact on antibody response.
Table 3. Predictors of positive antibody response against Streptococcus pneumoniae serotypes 23F and 6B following vaccination with 7-valent conjugate pneumococcal vaccine, in multivariate regression models*
OR (95% CI)
Adjusted for age, sex, ongoing methotrexate (MTX) treatment (yes or no), ongoing anti–tumor necrosis factor treatment, ongoing prednisolone treatment, patient global assessment on a visual analog scale, and prevaccination antibody levels for 23F and 6B. OR = odds ratio; 95% CI = 95% confidence interval; RA = rheumatoid arthritis; SpA = spondylarthropathy.
Ongoing treatment with MTX
Ongoing treatment with MTX
Ongoing treatment with MTX
Subanalyses including only patients with the same disease showed that, in RA patients, higher age and MTX treatment were predictors of impaired antibody response both in univariate and multivariate regression analysis. Concomitant prednisolone treatment was associated with better antibody response in the univariate analysis (odds ratio 1.93 [95% confidence interval 1.1–3.5], P = 0.030) but this became nonsignificant in the adjusted multivariate analysis (P = 0.09). Prednisolone dose did not significantly influence antibody response. In SpA patients, only concomitant MTX treatment was predictive of an impaired antibody response for both serotypes in the multiple logistic regression analysis (Table 3).
ADRs and AEs.
All ADRs and AEs were recorded according to the Guideline for Good Clinical Practice and Clinical Safety Data Management. The majority of side effects of 7-valent conjugate vaccine were mild and transitory, and the vaccine was well tolerated. No deaths were reported within 2 months following vaccination. Two cases of serious AEs were reported. One patient developed pneumonia requiring inpatient hospitalization, and chest radiography showed a lobar pneumonia but no pathogenic bacteria were found in blood cultures. The patient recovered completely after treatment with antibiotics (penicillin G). The second AE was a deep infection of the right forefoot requiring inpatient hospitalization, orthopedic surgery, and treatment with intravenous antibiotics. The most frequently reported AEs were tenderness or pain at the injection site (n = 52) lasting an average of 2–3 days, slightly increased body temperature (n = 37) lasting several days, headache, muscle and joint pain, and influenza-like symptoms. In total, 34 patients reported changes in existing rheumatic disease, the vast majority of which were a transitory worsening of joint pain lasting, on average, a week after vaccination. In 1 patient, worsening of existing arthritis was observed by a rheumatologist. No AEs involving the onset of new rheumatic or other autoimmune diseases were observed.
This is the first report concerning antibody response following pneumococcal vaccination using conjugate vaccine in immunosuppressed adult patients with established arthritis. Our primary finding is that MTX-treated arthritis patients have a reduced antibody response after pneumococcal vaccination with 7-valent conjugate vaccine. A considerable proportion of patients with RA and SpA (82% and 74%, respectively) who were being treated with MTX did not exhibit a 2-fold increase in antibody levels for the 2 serotypes tested. These findings confirm our own previous findings, as well as findings by others, in patients with RA (13, 28) and psoriatic arthritis (29) after 23-valent pneumococcal vaccination. Two studies that did not demonstrate an effect of MTX on antibody response had considerably less power (30, 31).
Regardless of diagnosis, patients receiving anti-TNF treatment as monotherapy had numerically, but not statistically significant, lower antibody response for both serotypes tested, compared to controls. Furthermore, patients receiving TNF inhibitors in combination with MTX had similar antibody responses to those of patients treated with MTX only. Taken together, these findings suggest that TNF inhibitors alone or added to MTX do not significantly impair the antibody response after pneumococcal vaccination. Our results are consistent with previously published experimental data demonstrating that anti-TNF treatment does not suppress the T cell responses to subsequent challenge with peptide antigens, but rather enhances these responses and brings them toward normal responses in patients with RA (32).
The impact of anti-TNF treatments on antibody response has been investigated by others with somewhat conflicting results. Corresponding to our previous report (13) and the present study, addition of etanercept, infliximab, or adalimumab to MTX did not affect the antibody response following pneumococcal vaccination (29, 33, 34). In another study, anti-TNF treatment was not associated with impaired antibody response, but a large proportion of patients did not respond to pneumococcal vaccination (31).
The mechanisms by which MTX or anti-TNF treatments affect the antibody response are not known, but our data suggest that they are similar for different rheumatic diseases. Although patients with SpA in our study had somewhat better response compared to corresponding RA treatment groups, the pattern of antibody response was similar. The effects of MTX in vivo may be mediated by increasing apoptosis of T cells as well as alteration of cytokine expression and humoral response (35).
One unexpected result seen in this study was that RA patients receiving concomitant prednisolone had better antibody response in the univariate analysis compared to those not treated with corticosteroids. The positive effect of oral corticosteroids on antibody response after pneumococcal vaccination in patients with RA has previously been reported (33), and in another study, high-dose glucocorticoid therapy in children with nephritic syndrome did not inhibit antibody response (36). The exact pathophysiologic mechanism for the effect of corticosteroids on antibody response is not known, but in the present study, mean prednisolone dosage was low (Table 1).
Higher age was predictive of an impaired antibody response for both serotypes. This is in accordance with several previous reports showing that elderly adults have lower antibody levels following vaccination (33, 37).
We also examined the influence of previous vaccination with standard 23-valent polysaccharide vaccine on antibody response. In spite of a 5-year pneumococcal vaccine–free window, previous vaccination was associated with both lower prevaccination antibody levels and lower antibody response. These data are consistent with results from the only study examining the persistence of antibody following pneumococcal vaccination in patients with inflammatory rheumatic disease. In that study, antibody levels were below those considered to be protective 3 years after pneumococcal vaccination (38). It may be speculated that immunosuppressive treatments, such as MTX, can reduce the ability to produce specific antibodies following polysaccharide–polypeptide challenge, and may hamper the capability of memory B cells to respond to subsequent vaccination. Our findings are somewhat in contrast to recently reported findings of higher antibody levels in previously vaccinated patients 5 years after vaccination, compared to vaccine-naive persons (39).
In general, a lower antibody response was elicited in RA patients compared to that in patients with SpA, as confirmed in a univariate regression model. Several confounding factors may have contributed to these results. Patients with RA were older, were all treated with immunosuppressive drugs, and were more frequently previously vaccinated with standard 23-valent pneumococcal vaccine; all of these confounders can probably explain why this difference was not seen after adjustment in the multivariate model. The influence of an underlying disease-related immunologic effect on antibody response following vaccination challenge cannot be ruled out, despite the fact that prevaccination antibody levels in our study did not differ significantly between different treatment groups or between RA and SpA patients.
It has previously been reported that smoking was associated with increased antibody prevaccination levels, at least in men (40). However, the proportion of smokers was similar among all treatment groups.
There are limitations to this study. Patients with RA who were not receiving immunosuppressive treatment were not included. Because early initiation of DMARD treatment for RA is now common, it was not feasible to recruit a sufficient number of non-DMARD–treated patients in our observational setting. Additionally, the inclusion of patients whose RA is at a very early stage and who have never received DMARDs would not be generalizable to the majority of arthritis patients in the observed setting. We used SpA patients who had never received DMARD therapy as controls for practical reasons, since the recruitment of population-based controls would encompass several ethical and logistic/administrative problems regarding administration of a vaccine not licensed for healthy adults younger than 65 years old (as is the case with the other vaccine serotypes).
We chose to measure the antibody response to 2 serotypes in this study. Since the main objective was to investigate the effect of different treatments on antibody response, we presume that underlying mechanisms responsible for the effect of antirheumatic drugs would influence the response to all serotypes in a similar way regardless of serotype. There were also logistical reasons for this choice.
In this study, we examined antibody response following vaccination, and the study was not designed to assess the effectiveness of pneumococcal conjugate vaccine on clinical outcomes, such as hospitalization for pneumococcal disease. Although the correlation between positive antibody response and protection against the disease was expected, we cannot claim that vaccination prevents pneumococcal disease in patients with rheumatic disease treated with immunosuppressive agents who do have a positive antibody response. Antibody response is generally considered as a surrogate marker for clinical efficacy, and controlled trials examining the effectiveness of different vaccines in preventing invasive pneumococcal disease in adult immunocompromised patients are critically needed (41).
Heptavalent conjugate vaccine was well tolerated and, with the exception of a few cases of time-limited worsening of arthritis symptoms, did not influence the course of rheumatic disease. Likewise, no cases of new-onset autoimmune disease were reported. Our results are consistent with those of a recent study showing no association between vaccination and RA (42).
In conclusion, in patients with RA and those with SpA, MTX reduces antibody response after vaccination with pneumococcal conjugate vaccine, but anti-TNF treatment does not. Pneumococcal vaccination, therefore, should preferably be given before the initiation of MTX treatment in patients with established arthritis. However, this study does not address whether repeated vaccinations can boost antibody response as in children. The possible benefit of pneumococcal vaccination in the prevention of invasive pneumococcal diseases is such that patients with established arthritis should be encouraged to be vaccinated, regardless of their treatment regimen.
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. Kapetanovic 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. Kapetanovic, Roseman, Jönsson, Truedsson, Saxne, Geborek.
Acquisition of data. Kapetanovic, Roseman, Jönsson, Truedsson, Saxne, Geborek.
Analysis and interpretation of data. Kapetanovic, Roseman, Jönsson, Truedsson, Saxne, Geborek.
The authors wish to thank the late research nurse Lotta Larsson for her professional and meticulous monitoring of this study. We thank nurses Elna Haglund, Eva-Karin Kristoffersson, Helén Axelsson, and Käthe Nillson for their help with vaccination, collecting the blood samples, and carrying out the study, as well as Peter Kapral, Maria Jacobsson, Ingrid Moberg, Ingrid Bondesson, Ingrid Hermansson, and Eva Hommerberg at the Skånes Universitetssjukhus Lund and Malmö Department of Rheumatology for centrifuging and storing blood samples, and Ingrid Mattsson-Geborek for her skillful help with figures. We thank all patients for their participation in the study and all colleagues for their cooperation and support during the study. We also thank Professor John Isaacs for valuable scientific advice.