Prevention of bacterial infection, which is a leading cause of morbidity in patients with rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE), is a priority. However, the safety and immunogenicity of the pneumococcal vaccine in such patients remain controversial. We evaluated the currently available pneumococcal vaccine in patients with RA or SLE. Pneumococcal vaccination was not associated with an appreciable deterioration in any clinical or laboratory measure of disease activity in either group. One month after vaccination, patients in both groups had significant increases in geometric mean concentrations of pneumococcal polysaccharide–specific IgG to all 7 serotypes tested, as did control subjects. However, 14 (33.3%) of 42 patients with RA and 5 (20.8%) of 24 patients with SLE responded either to none or to only 1 of the 7 polysaccharides. Pneumococcal vaccination is generally safe and immunogenic in patients with RA or SLE, but a subset of patients may remain unprotected by the currently available vaccine.
Both rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) are associated with 2- to 4-fold excess mortality [1–3]. Bacterial infection is the leading cause of mortality in patients with RA or SLE; pneumonia causes 15%–25% of deaths [1, 2, 4–7]. It is currently recommended that pneumococcal polysaccharide vaccine be administered to persons >65 years of age and to patients with chronic illness who have increased risk for pneumococcal disease [8–10]. Although RA and SLE patients are considered potential candidates for pneumococcal vaccination, it has not been uniformly recommended because of concerns about the safety and efficacy of the pneumococcal vaccine in these patients [4, 11–13]. Therefore, we evaluated whether the currently available pneumococcal vaccine is immunogenic in patients with RA and SLE and whether it induces acute worsening of disease activity in these patients.
Subjects. Forty-two consecutive patients who fulfilled the American College of Rheumatology criteria for RA [14], 24 consecutive patients who fulfilled the American College of Rheumatology revised classification criteria for SLE [15], and 20 healthy control subjects matched for age and institution participated in this study. Exclusion criteria were pregnancy, a history of past vaccination allergy, and previous pneumococcal vaccination. Each patient or subject received an intradeltoid injection with 0.5 mL of pneumococcal vaccine (Pneumovax; Merck), which is a 23-valent preparation containing 25 µg of each of the following capsular polysaccharides (Danish nomenclature): types 1–5, 6B, 7F, 8, 9N, 9V, 10A, 11A, 12F, 14, 15B, 17F, 18C, 19A, 19F, 20, 22F, 23F, and 33F.
Clinical assessment. Before vaccination, a complete history was obtained, a physical examination was done, and the subject's medical records were reviewed. For patients with RA, clinical assessment before and 2 months after vaccination included the following: duration of morning stiffness (in minutes); evaluation of daytime and nocturnal pain, which was done by use of a visual analogue scale of 10 cm on which 10 represented extreme pain and 0 represented no pain; and counting of the number of tender and swollen joints. For patients with SLE, disease activity was scored before and after vaccination by use of the Systemic Lupus Erythematosus Disease Activity Index [16].
Appropriate informed consent was obtained from all patients, and the clinical research was conducted in accordance with guidelines for human experimentation specified by the Tel Aviv Sourasky Medical Center and the Minneapolis Veterans Affairs Medical Center.
Laboratory assessment of disease activity. Routine laboratory tests done before and 1–2 months after vaccination included the following: complete blood cell counts, a serum chemistry panel, urinalysis, Westergren erythrocyte sedimentation rate, and C-reactive protein level. For SLE patients, levels of C3 and C4 complement, immunoglobulins, and antibodies to double-stranded DNA were determined.
Capsule pneumococcal polysaccharide–specific IgG measurement. Levels of pneumococcal capsule–specific IgG were measured by use of ELISA, as described elsewhere [17]. A reference pooled human serum was standardized by comparison with the international standard serum 89-SF (gift of Carl Frasch, US Food and Drug Administration, Bethesda, MD), for which the concentrations of vaccine type–specific IgG antibody have been established for 15 of the 23 vaccine serotypes [18, 19]. Concentrations of specific IgG antibody in serum 89-SF for serotypes without established standard values were determined by use of the cross-standardization method [18] and results were as follows: for pneumococcal polysaccharide type 8, 5.74 µg/mL; and for pneumococcal polysaccharide type 9N, 7.90 µg/mL.
Purified capsular pneumococcal polysaccharides (0.5 µg in 100 µL of PBS per well) were bound to 96-well polystyrene microtiter plates (Nunc Maxisorp; PGC Scientifics), by overnight incubation at 4°C. All serum samples and standard samples (10 µg/mL of serum) were preadsorbed with cell wall polysaccharide (Statens Seruminstitut) at room temperature for 30 min. Serum samples were added to plates at an initial dilution of 1 : 200 in 1% bovine serum albumin with 0.05% (vol/vol) Tween 20 in PBS. IgG that reacted with pneumococcal polysaccharide was detected with use of affinity-purified horseradish peroxidase–conjugated goat antihuman IgG (Jackson Laboratories). Each plate included a positive laboratory reference sample and a control serum sample that contained a known level of IgG to the relevant pneumococcal polysaccharide to assess the coefficient of variation of measurement between plates; for plates with coefficients of variation of >12%, the analysis was repeated. Results are reported in micrograms per milliliter, on the basis of comparison with the international standard serum 89-SF.
Statistical analysis. Differences between the geometric mean concentration of capsule-specific IgG in prevaccination and postvaccination serum samples were calculated by repeated measures of variance. Continuous variables were compared between groups by one-way analysis of variance, and proportions were compared by χ2 analysis. Associations between clinical predictors and vaccine response were assessed by logistic regression. Statistical analysis was performed with SPSS for Windows, version 6.1 (SPSS), and all P values are 2-tailed.
Characteristics of patients and control subjects. All groups were predominantly female (table 1). SLE patients were significantly younger than RA patients and control subjects (P = .001) and had had their disease for a significantly shorter time (P = .014) than had RA patients. Significantly more RA patients were treated with nonsteroidal anti-inflammatory drugs and methotrexate, and significantly more SLE patients were receiving hydroxychloroquine (table 1). Dosages of prednisone and methotrexate did not differ significantly between RA and SLE patients who were being treated with these immunosuppressive agents, and only small numbers of patients were receiving aurothioglucose, sulfasalazine, minocycline, or cyclophosphamide. Most of the RA and SLE patients had mild to moderate disease activity before vaccination (table 2).
Clinical characteristics of patients with systemic lupus erythematosus (SLE) or rheumatoid arthritis (RA) and control subjects who underwent pneumococcal vaccination.
Effects of pneumococcal vaccination on measures of disease activity in patients with rheumatoid arthritis or systemic lupus erythematosus.
Effects of pneumococcal vaccination on disease activity. Pneumococcal vaccination was not associated with a significant worsening of any clinical or laboratory parameter of disease activity (table 2). In fact, RA patients had significantly fewer tender joints 2 months after vaccination. Similarly, SLE patients did not experience any significant changes in clinical or laboratory measures of disease activity after vaccination (table 2). Minor adverse acute effects of pneumococcal vaccination were observed in 2 patients. In 1 patient with RA, transient diffuse musculoskeletal pain developed the day after vaccination, and 1 patient with SLE had an episode of clinical pleuritic pain that persisted for 5 days after vaccination. Thus, pneumococcal vaccination appeared to be safe and well-tolerated in both RA and SLE patients.
Immunogenicity of pneumococcal vaccine. Prevaccination antibody levels to the 7 pneumococcal serotypes tested did not differ significantly between groups (table 3). However, a greater proportion of RA patients than of SLE patients or control subjects had prevaccination specific antibody levels below the presumptively protective level of 1 µg/mL for 6 of 7 serotypes.
Prevaccination levels of pneumococcal capsular polysaccharide-specific antibody in patients with systemic lupus erythematosus (SLE) or rheumatoid arthritis (RA) and in control subjects.
One month after vaccination, both the RA and the SLE groups had significant increases in geometric mean concentrations of capsule pneumococcal polysaccharide–specific antibody and in the mean fold-increase in antibody levels to all 7 serotypes, compared with prevaccination levels (table 4). Immune responses at 1 month did not differ significantly between the RA and SLE groups and were similar to those seen in the group of control subjects, with the exception of antibody responses to types 8 and 2 in the SLE patients (table 4). Of the serotypes tested, serotype 14 appeared to be the most immunogenic and serotype 4 the least immunogenic in all groups.
Increase in levels of pneumococcal capsular polysaccharide-specific antibody 1 month after vaccination in patients with systemic lupus erythematosus (SLE) or rheumatoid arthritis (RA) and in control subjects.
Individual responses of RA and SLE patients to pneumococcal vaccination. Although the RA and SLE patients responded to pneumococcal vaccination, the 23-valent vaccine did not appear to be uniformly immunogenic in these patients. Whether an immune response was defined as a 2-fold increase in antibody levels or as an absolute change in specific antibody of 1 µg/mL, according to the different serotypes, only 35%–71% of RA patients and 36%–86% of SLE patients responded to pneumococcal vaccination within 1 month (table 5). A greater proportion of control subjects responded to pneumococcal vaccination, although the differences reached statistical significance only for serotype 7F.
Proportion of patients with systemic lupus erythematosus (SLE) or rheumatoid arthritis (RA) who responded to pneumococcal vaccination.
Furthermore, a substantial proportion of RA and SLE patients had an extremely poor response to pneumococcal vaccination. Whether an immune response was defined as 2-fold increase in antibody levels or as an absolute change in specific antibody of 1 µg/mL at 1 month after vaccination, 14 (33.3%) of 42 RA patients and 5 (20.8%) of 24 SLE patients responded to either none or only 1 of the 7 pneumococcal polysaccharides tested. In contrast, none of the control subjects failed to respond to pneumococcal vaccination (P = .004).
Because RA and SLE patients with such poor responses to pneumococcal vaccination presumably remain at risk of invasive pneumococcal infection, despite having been vaccinated, we attempted to identify clinical or laboratory parameters that were determined before vaccination that might help clinicians predict such poor response. However, demographic parameters, laboratory tests, and measures of disease activity (whether at baseline or after vaccination) did not significantly predict poor vaccine response in either group. In addition, poor vaccine response was not significantly associated with levels of vitamin B12 or folate or use of any immunosuppressive agents, including prednisone, methotrexate, hydroxychloroquine, or azathioprine.
We have shown that vaccination of SLE and RA patients with the available 23-valent pneumococcal vaccine was safe and that, 1 month after vaccination, the vaccine induced adequate mean antibody responses in both groups, which were similar to the response in a group of control subjects. However, a substantial proportion of both RA and SLE patients had impaired antibody responses after pneumococcal vaccination.
Although pneumonia is a leading cause of death in both SLE and RA patients, pneumococcal vaccination is not routinely recommended for immunization of patients with a connective-tissue disease [14, 20]. Concerns regarding the safety and efficacy of immunizing persons who have connective-tissue disease have persisted for >50 years [21]. This intuitive reluctance is mainly based on the fact that microbial infection has frequently been hypothesized to trigger autoantibody production by mechanisms such as molecular mimicry or superantigen stimulation. In theory, such a mechanism could also trigger antibody production following deliberate vaccination with foreign antigens. Indeed, onset of a connective-tissue disease after vaccination has been described in several case reports. Onset of SLE has been described after vaccination against tetanus [22, 23], hepatitis B [22–25], and other diseases [22]. Similarly, cases of RA after administration of tetanus toxoid and recombinant hepatitis B vaccine have been reported [25, 26]. However, these sporadic reports should not preclude routine vaccination of patients with SLE and RA. Vaccination of SLE patients with influenza vaccine [27, 28] and recombinant hepatitis B vaccine [29] seems to be well tolerated. In a recent study involving 73 SLE patients receiving multiple vaccines simultaneously, including pneumococcal vaccine, 6 patients had a mild increase in SLE disease activity, but an equal number of patients showed clinical improvement after vaccination [30]. Several studies concerning the former 14-valent pneumococcal vaccine in SLE patients have demonstrated its safety [11, 31]. The present study, which included 24 SLE patients vaccinated with the 23-valent pneumococcal vaccine, reaffirms the safety of vaccination for such patients, at least in the short term. Similarly, in our group of RA patients, pneumococcal vaccination did not exacerbate disease activity. Although 1 patient experienced transient musculoskeletal pains after vaccination, the group of RA patients had improved in most measures of disease activity 1 month after vaccination.
Overall, the RA and SLE groups had significant increases in mean antibody responses at 1 month after pneumococcal vaccination. Although our study had 35%–99% power to detect a 50% difference in immune responses, the proportions of RA and SLE patients with significant antibody responses to pneumococcal vaccine were similar to the proportion of control subjects. However, in contrast to control subjects, a substantial subset of RA and SLE patients did not have significant responses to pneumococcal vaccination. Among RA patients, one-third responded to 0 or only 1 of the 7 serotypes tested. None of the clinical and laboratory measures evaluated appeared to predict poor response to pneumococcal vaccination in RA patients. In contrast, O'Dell et al. [32] reported preliminary findings that immune responses to pneumococcal vaccination were significantly decreased among RA patients who were being treated with methotrexate. Age, sex, duration of disease, measures of disease activity, and the use of other immunosuppressive agents, including corticosteroids, azathioprine, sulfasalazine, and antimalarial agents, were not significantly associated with the antibody response. However, the mean dose of corticosteroids for the RA patients was <10 mg/day, and only 1 patient was treated with a more potent immunosuppressive drug (cyclophosphamide). Thus, the effects of high-dose corticosteroids or potent immunosuppressive drugs on the ability of RA patients to respond to pneumococcal vaccination remains to be determined.
The immunogenicity of vaccines in patients with SLE has long been a matter of controversy. Titers of antibody to influenza virus after vaccination in SLE patients with serologically active disease have been reported to be significantly lower than those in other SLE patients and in healthy control subjects [28]. In 38 patients with SLE, pneumococcal vaccination induced mean antibody levels at 1 month and 1 year that were significantly lower than those in healthy control subjects [11]. In contrast, several studies have shown that mean antibody levels in patients with SLE are similar to those in control subjects after pneumococcal vaccination, although responses of individual patients varied significantly [13, 31, 33]. In agreement with these findings, we found that SLE patients, as a group, had a good response to the pneumococcal vaccine, but 20% of SLE patients had a poor response. Again, none of the clinical or laboratory measures predicted lack of response in SLE patients.
In conclusion, pneumococcal vaccination of patients with SLE or RA was safe and immunogenic in the majority of the patients; however, an appreciable subgroup of patients had depressed responses to the current vaccine. Large-scale clinical trials are now needed to determine the efficacy of the current vaccine to prevent pneumococcal infection in patients with RA or SLE. In addition, further studies are required to determine whether the new conjugate pneumococcal vaccine, or other vaccines under development, are more uniformly immunogenic and effective in these patients.
Financial support: National Institutes of Health grant AI-02440 (to J.B.R.).
IDSA Members: For your free access to this journal, log in via the IDSA members area.
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy