Infliximab, but not etanercept, induces IgM anti–double-stranded DNA autoantibodies as main antinuclear reactivity: Biologic and clinical implications in autoimmune arthritis

Authors


Abstract

Objective

To analyze the clinical and biologic correlates of autoantibody induction during longer-term tumor necrosis factor α (TNFα) blockade with either the monoclonal antibody infliximab or the soluble receptor etanercept.

Methods

Thirty-four patients with spondylarthropathy (SpA) and 59 patients with rheumatoid arthritis (RA) were treated with infliximab for 2 years. Additionally, 20 patients with SpA were treated with etanercept for 1 year. Sera were blindly analyzed for antinuclear antibodies (ANAs), anti–double-stranded DNA (anti-dsDNA) antibodies, anti–extractable nuclear antigen (anti-ENA) antibodies, and antihistone, antinucleosome, and anticardiolipin antibodies (aCL). The anti-dsDNA antibodies were isotyped.

Results

High numbers of infliximab-treated patients with SpA or RA had newly induced ANAs (61.8% and 40.7%, respectively) and anti-dsDNA antibodies (70.6% and 49.2%, respectively) after 1 year, but no further increase between year 1 and year 2 was observed. In contrast, induction of ANAs and anti-dsDNA antibodies was observed only occasionally in the etanercept-treated patients with SpA (10% of patients each). Isotyping revealed almost exclusively IgM or IgM/IgA anti-dsDNA antibodies, which disappeared upon interruption of treatment. Neither infliximab nor etanercept induced other lupus-related reactivities such as anti-ENA antibodies, antihistone antibodies, or antinucleosome antibodies, and no clinically relevant lupus-like symptoms were observed. Similarly, infliximab but not etanercept selectively increased IgM but not IgG aCL titers.

Conclusion

The prominent ANA and anti-dsDNA autoantibody response is not a pure class effect of TNFα blockers, is largely restricted to short-term IgM responses, and is not associated with other serologic or clinical signs of lupus. Similar findings with aCL suggest that modulation of humoral immunity may be a more general feature of infliximab treatment.

Since the first proof of efficacy of tumor necrosis factor α (TNFα) blockade, both the number of patients treated worldwide and the number of indications for treatment have grown steadily (1–3). Surprisingly, the profound immunomodulation induced by TNFα blockers is associated with a relatively low incidence of immune-related complications such as demyelinating disease and lupus-like syndromes (3–5). This contrasts sharply with the prominent induction of autoantibodies such as antinuclear antibodies (ANAs) and anti–double-stranded DNA (anti-dsDNA) antibodies by TNFα blockers (6–14). Although this phenomenon has been recognized for several years, the clinical and biologic correlates of this antibody induction in autoimmune arthritis are not yet fully understood. Recent studies showed that induced anti-dsDNA antibodies belong to the IgM and IgA subclasses but not to the IgG subclass (7, 8, 12). Moreover, it became clear that ANA and anti-dsDNA antibody induction was observed not only in patients with rheumatoid arthritis (RA) but also in patients with spondylarthropathy (SpA) and Crohn's disease (8, 12–17). However, large systematic studies comparing different TNFα blockers, different disease groups, different autoantibodies, and longer-term clinical and serologic outcomes are not yet available, which precluded better understanding of the clinical and biologic relevance of the changes in the autoantibody profiles that occur during TNFα blockade.

The aim of the present study was to assess the above-mentioned issues in further detail by systematically analyzing the serologic, biologic, and clinical parameters in a large cohort of patients who were treated with either the monoclonal anti-TNFα antibody infliximab or the soluble TNFα receptor etanercept for up to 2 years. Because baseline humoral autoimmunity is uncommon in SpA, we used this disease as our primary human model.

PATIENTS AND METHODS

Patients and samples.

As shown in Table 1, a total of 113 patients with either SpA or RA were included in the study after giving informed consent. This study was approved by the Ethics Committee of the University Hospital Ghent. All patients with SpA fulfilled the European Spondyloarthropathy Study Group classification criteria for SpA (18); all patients with RA fulfilled the American College of Rheumatology (ACR; formerly the American Rheumatism Association) classification criteria for RA (19). Cohort 1 comprised 34 patients with SpA who were treated with a loading dose of infliximab (5 mg/kg) at weeks 0, 2, and 6, followed by a dose of 10 mg/kg every 14 weeks until week 48. In the second year of treatment, patients in cohort 1 received 5 mg/kg of infliximab every 8 weeks until week 104. Cohort 2 comprised 59 patients with RA who received 3 mg/kg of infliximab at weeks 0, 2, and 6, and every 8 weeks thereafter for up to 2 years (week 102). All patients with RA received concomitant treatment with methotrexate (≥7.5 mg/week). In 2 of the patients with RA, infliximab treatment was stopped between year 1 and year 2 due to pneumonia (n = 1) or fever/esophagitis/cystitis (n = 1). Cohort 3 comprised 20 patients with SpA who were treated with etanercept (25 mg twice weekly) for a 1-year period.

Table 1. Baseline characteristics of patients who were treated with TNFα blockers*
CharacteristicCohort 1, SpA (n = 34)Cohort 2, RA (n = 59)Cohort 3, SpA (n = 20)
  • *

    Except where indicated otherwise, values are the number of patients. TNFα = tumor necrosis factor α; SpA = spondylarthropathy; RA = rheumatoid arthritis; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate; MTX = methotrexate; NSAIDs = nonsteroidal antiinflammatory drugs.

TNFα blockerInfliximabInfliximabEtanercept
Age, median (range) years47.5 (30–67)55 (31–77)37.5 (20–71)
No. men/no. women23/1120/3913/7
Disease duration, median (range) years7.5 (1–35)10 (1–39)10 (1–41)
SpA subtype   
 Psoriatic arthritis176
 Ankylosing spondylitis165
 Undifferentiated SpA19
CRP, median (range) mg/dl1.67 (0.1–7.42)1.77 (0.1–8.35)0.95 (0.1–12.8)
ESR, median (range) mm/hour18 (1–101)22 (2–70)16 (1–86)
Concomitant therapy   
 MTX (≥7.5 mg/week)0590
 Corticosteroids3330
 NSAIDs304119
Time points of serum collectionBaseline, week 48, week 104Baseline, week 46, week 102Baseline, week 48

Additionally, we analyzed a control group of 15 patients with systemic lupus erythematosus (SLE) fulfilling the ACR classification criteria for SLE (20), who were not treated with TNFα blockers. Finally, 7 other patients (5 with RA and 2 with SpA) who were treated with infliximab for at least 5 months, and in whom treatment had to be interrupted, were tested at baseline, at the moment that infliximab was stopped, and after withdrawal from treatment in order to analyze the persistence over time of the induced autoimmune profiles.

In order to rule out inter-test variability and exclude possible technical biases, all serologic analyses were performed in a single test run, and investigators were blinded to the diagnosis, treatment, and time point of treatment.

Detection of ANAs.

For reasons of sensitivity, serum was diluted 1:40 in phosphate buffered saline (PBS). Serum samples were tested for ANA reactivity using fixed HEp-2000 cells (Immunoconcepts, Sacramento, CA) and a fluorescein isothiocyanate (FITC)–labeled conjugate (anti-human IgG heavy and light chain specific; Immunoconcepts). The fluorescence intensity was scored semiquantitatively from 0 to 5+, relative to the intensity of a negative control and a positive (4+) control (21). A sample was considered positive for ANA if a score of at least 2+ was obtained, which corresponds with positivity using a serum dilution between 1:80 and 1:160 (data not shown). A significant increase in ANA intensity was defined as an increase of at least 2 points on a scale of 0 to 5+.

Detection of anti-dsDNA antibodies.

Serum samples diluted 1:20 in PBS were tested for anti-dsDNA antibodies, using Crithidia luciliae–coated slides (Immunoconcepts) as substrate and an FITC-labeled conjugate (anti-human IgG heavy and light chain specific; Immunoconcepts). Samples were scored as being positive or negative, according to comparison with a negative control and a positive control. Similarly, isotyping of the anti-dsDNA antibodies was performed by indirect immunofluorescence on C luciliae, using specific FITC-labeled conjugates against human IgG heavy (γ) chain (Production d'Anticorps Réactifs Immunologiques & Services, Compiègne, France), human IgM heavy (μ) chain (Dako, Glostrup, Denmark), and human IgA heavy (α) chain (Dako). The specificity of the conjugates was validated by the manufacturer as well as in our own enzyme-linked immunosorbent assay (ELISA) experiments, using purified human IgG, IgM, and IgA as substrate (data not shown). We and other investigators previously demonstrated that, in the context of TNFα blockade, indirect immunofluorescence on C luciliae is a more sensitive and reliable technique than ELISA for the detection of anti-dsDNA antibodies (8, 11).

Detection of anti–extractable nuclear antigen (ENA), antihistone, and antinucleosome antibodies.

Sera were analyzed for anti-ENA and antihistone antibodies by line immunoassay (LIA) (INNO-LIA ANA update K1090; Innogenetics, Zwijnaarde, Belgium). This multiparameter assay contains the following antigens: SmB, SmD, RNP 70, RNP-A, RNP-C, SSA/Ro 52, SSA/Ro 60, SSB/La, CENP-B, Scl-70, Jo-1, ribosomal P, and histones (H1, H2a, H2b, H3, and H4). For detection of antinucleosome antibodies, sera were analyzed by ELISA (Anti-Nucleo; GA Generic Assays, Dahlewitz, Germany). In a diagnostic setting, the cutoff for positivity is 50 units/ml. Both tests were performed according to the manufacturer's instructions.

Detection of lupus-like characteristics.

Because all of the patients were followed up in the context of clinical trials, lupus-related manifestations such as oral ulcers, serositis, neurologic disorder, and skin symptoms (rash, photosensitivity), as well as any other adverse events were reported systematically in the medical charts at every visit. Biologic evaluation included the peripheral blood cell count to test for hematologic disorders and urinalysis to test for proteinuria (i.e., kidney disease).

Detection of anticardiolipin antibodies (aCL).

IgG aCL were detected by an in-house ELISA coated with cardiolipin (Sigma, St. Louis, MO). Diluted sera (1:1,000) were tested using anti-human IgG Fc-specific conjugate coupled with alkaline phosphatase (Sigma) and p-nitrophenylphosphate substrate in diethanolamine buffer (Bio-Rad, Hercules, CA). For the detection of IgM aCL, we used a commercially available kit (Orgentec Diagnostika, Mainz, Germany). The ELISA was performed according to the manufacturer's instructions.

Statistical analysis.

All values are presented as the median (range). For dichotomous data, we used the paired McNemar's test. For continuous data, we used the nonparametric paired Wilcoxon's signed rank test. For comparison of proportions, we applied a chi-square test. P values less than or equal to 0.05 were considered significant.

RESULTS

Induction of ANAs and anti-dsDNA antibodies by infliximab.

We investigated whether the induction of ANAs and anti-dsDNA autoantibodies that was observed after 6 months of infliximab treatment in patients with SpA (8) was further increased at 1 and 2 years. As shown in Figure 1A, the year 1 data confirmed the significant ANA induction, as reflected by the high number of patients with newly induced ANAs (21 of 34 patients [61.8%]) and those with a significant increase in ANA intensity (21 of 34 patients [61.8%]) compared with baseline. Accordingly, there was a significant induction of anti-dsDNA antibodies during the first year of infliximab treatment (24 of 34 patients [70.6%]). These data indicate that there is no additional increase of ANA and anti-dsDNA antibody positivity after 6 months (8), which was confirmed by the absence of further induction of ANA or anti-dsDNA antibodies between year 1 and year 2 (Figure 1). In addition, there was a trend toward a decrease in anti-dsDNA antibody positivity in the infliximab-treated SpA cohort between year 1 (73.5%) and year 2 (55.9%; P = 0.07).

Figure 1.

Effect of tumor necrosis factor α blockade on antinuclear antibodies (ANAs), anti–double-stranded DNA (anti-dsDNA) antibodies, anti–extractable nuclear antigen (anti-ENA) antibodies, and antihistone antibodies in patients with spondylarthropathy (SpA) and patients with rheumatoid arthritis (RA). A, Percentage of ANA-positive patients (intensity ≥2+) in the 3 cohorts at baseline, year 1, and year 2. B, Percentage of anti-dsDNA antibody–positive patients. C, Percentage of patients with a positive test result for anti-ENA, antihistone, or antinucleosome antibodies. DF, Results of isotyping of the anti-dsDNA antibodies at different time points in the 3 cohorts. Anti-dsDNA antibodies of the IgG subclass were nearly absent (only 1 RA patient tested positive, at year 2) (D). The autoantibody profiles in the etanercept-treated SpA patients were not analyzed at year 2. Anti-ENA, antihistone, and antinucleosome antibodies were not tested in the infliximab-treated RA patients at year 2. ∗ = P ≤ 0.05 versus baseline.

Confirming these findings in an independent disease background, we found that among 59 infliximab-treated patients with RA, 24 (40.7%) had newly positive ANAs and 24 (40.7%) had a significant increase in ANA intensity between baseline and year 1. Anti-dsDNA antibodies were induced in 29 of 59 patients (49.2%) (Figure 1). As in patients with SpA, these values closely resemble the data at 6 months (8), which were further confirmed by the stable profiles between years 1 and 2 (Figure 1). Of interest, the induction of anti-dsDNA antibodies tended to be more pronounced in SpA than in RA at year 1 (P = 0.039).

Induction of ANAs and anti-dsDNA antibodies by infliximab and etanercept, and comparison with infliximab.

In order to assess potential differences in autoantibody induction between different TNFα blockers, we next analyzed 20 patients with SpA who were treated with etanercept and who fulfilled the same inclusion criteria as those fulfilled by the previously reported infliximab-treated SpA cohort. There was no significant ANA induction after 1 year of etanercept treatment: only 3 patients (15%) developed newly positive ANAs, and only 1 patient (5%) had a significant increase in ANA intensity. These values were significantly lower than those observed in the infliximab group (61.8% of infliximab-treated patients had newly positive ANAs [P < 0.001], and 61.8% had a significant increase in ANA intensity [P < 0.001]). Similarly, the induction of anti-dsDNA antibodies during 1 year of anti-TNFα treatment was significantly lower in the etanercept-treated SpA patients (3 of 20 patients [15%]) than in the infliximab-treated SpA patients (24 of 34 patients [70.6%]; P < 0.001).

Induction of IgM and IgA anti-dsDNA antibodies by TNFα blockade.

In order to confirm that the anti-dsDNA antibodies induced by infliximab treatment in RA as well as in SpA were exclusively of the IgM and/or IgA isotypes (8), and that isotype switching to IgG did not occur at later time points, we isotyped all induced anti-dsDNA antibodies. As shown in Figure 1, infliximab-induced anti-dsDNA antibodies in SpA as well as RA were predominantly of the IgM isotype at both 1 and 2 years. Among patients with SpA who were anti-dsDNA antibody positive, a combination of IgM and IgA isotypes was found in 44% at year 1 and in 11% at year 2. Among anti-dsDNA antibody–positive patients with RA, this combination of isotypes was found in 10% at year 1 and in 25% at year 2. Anti-dsDNA antibodies of the IgA class alone were not detected. Of major importance, no IgG anti-dsDNA antibodies were detected in either of the infliximab-treated cohorts over 2 years, with the exception of a single RA patient who was characterized by IgM anti-dsDNA antibodies at year 1 and a combination of IgM, IgA, and IgG anti-dsDNA antibodies at year 2. Interestingly, the anti-dsDNA antibodies observed in 3 etanercept-treated patients were also of the IgM isotype, and neither IgA nor IgG anti-dsDNA antibodies were detected. Because we detected high frequencies of IgM and virtually no IgG anti-dsDNA antibodies in the TNFα blockade–treated patients, we analyzed sera from 15 patients with SLE as an additional control. Nine of these SLE patients demonstrated anti-dsDNA reactivity, which was of the IgG isotype in all 9 patients, with associated IgM and/or IgA anti-dsDNA antibodies in 4 patients. Taken together, these data indicate that the pronounced induction of anti-dsDNA antibodies in SpA and RA is not associated with the occurrence of lupus-specific IgG anti-dsDNA antibodies.

Disappearance of infliximab-induced ANAs and anti-dsDNA antibodies after treatment withdrawal.

Because short-term humoral immunity was suggested by the fact that anti-dsDNA antibodies were mainly of the IgM isotype, and that anti-dsDNA antibody levels fluctuated over time in individual patients and tended to decrease in patients with SpA after 1 year of treatment, we tested the persistence of induction over time in an additional 7 patients who were treated with infliximab for at least 5 months and in whom treatment had to be interrupted. As shown in Table 2, 4 patients had newly positive ANAs and 5 patients had newly induced anti-dsDNA antibodies at the time when infliximab therapy was stopped. Isotyping revealed that the anti-dsDNA antibodies were predominantly of the IgM class. One patient also had IgG and IgA anti-dsDNA antibodies. A new analysis that was performed several months after infliximab treatment was stopped revealed no isotype switching to IgG anti-dsDNA antibodies in the other patients. In contrast, the ANA intensities decreased and the anti-dsDNA antibodies disappeared in all 5 patients.

Table 2. Persistence of infliximab-induced autoantibodies over time in 7 patients in whom treatment with infliximab was interrupted*
PatientNo. of months receiving infliximabReason for stopping infliximabNo. of months after stopping infliximabANA intensity scoreNo. of patients with anti-dsDNA
BaselineUpon withdrawal∼1–3 years after withdrawalBaselineUpon withdrawal∼1–3 years after withdrawal
  • *

    Patients were tested at baseline, at the time infliximab was stopped, and ∼1–3 years after withdrawal from treatment. ANA = antinuclear antibody; anti-dsDNA = anti–double-stranded DNA. ANAs were considered positive if the ANA intensity score was ≥2+ on a scale of 0 to 5+. RA = rheumatoid arthritis; SpA = spondylarthropathy.

RA         
 112.5Pneumonia241+01+000
 214Withdrawal of consent332+2+1+01 (IgM)0
 312.5Fever, esophagitis, cystitis10.504+1+01 (IgM)0
 45Agranulocytosis3103+1+01 (IgG/IgM/IgA)0
 512.5Herpes zoster33001+000
SpA         
 17.5Withdrawal of consent34.503+2+01 (IgM)0
 222.5Pneumonia/abscess1603+2+01 (IgM)0

TNFα blockade and anti-ENA, antihistone, and antinucleosome antibodies.

In consideration of the induction of ANAs and anti-dsDNA antibodies, we investigated other lupus-related autoantibodies including anti-ENA, antihistone, and antinucleosome antibodies (22, 23). As shown in Figure 1, the prevalence of anti-ENA and antihistone antibodies did not increase during TNFα therapy, but these levels fluctuated somewhat in individual patients. Similarly, antinucleosome antibody positivity developed in only 2 infliximab-treated patients (1 with SpA and 1 with RA) and in no etanercept-treated patients. In contrast, the control group of 15 patients with SLE who were not treated with TNFα blocker was characterized by multiple anti-ENA, antihistone, and antinucleosome antibody reactivities (15 anti-SSA, 5 anti-SSB, 6 anti-Sm, 3 anti-RNP, 1 anti–ribosomal P, 3 antihistone, 2 antinucleosome), confirming the different autoantibody profiles in patients treated with TNFα blockade compared with the profiles in untreated patients with SLE.

Association of TNFα blockade–induced ANAs and anti-dsDNA antibodies with lupus-like characteristics.

Because the described biologic characteristics of the induced autoantibody profiles bring into question their clinical relevance in SLE, we systematically assessed other possible nonserologic lupus-like characteristics in all patients during longer-term followup. As shown in Table 3, no lupus-like features were observed in the etanercept-treated patients with SpA. Among the 34 patients with SpA who were treated with infliximab, mild leukopenia (>3,200 cells/μl) developed in 4 patients, and mild lymphopenia (>1,200 cells/μl) developed in 4 patients. However, these hematologic abnormalities were transient, occurred both in patients with (n = 6) and those without (n = 2) induction of anti-dsDNA antibodies, and were not associated with other lupus-like features. Lymphopenia was observed in nearly half of the infliximab-treated RA patients (n = 24) and was more pronounced in RA than in SpA. TNFα blockade was associated with mild leukopenia (>2,800 cells/μl) in 4 patients and with mild proteinuria in 4 patients. Again, however, these biologic abnormalities were not associated with clinical symptoms, were transient in most patients, and lymphopenia occurred both in patients with (n = 16) and those without (n = 8) induction of anti-dsDNA antibodies. None of the patients had to discontinue infliximab treatment due to development of lupus-like symptoms. The only patient with anti-dsDNA IgG antibodies at year 2 also had lymphopenia but no other associated clinical or biologic abnormalities. Because 3 RA patients and 5 SpA patients withdrew from the original infliximab studies before year 1 and were thus not included in the present serologic study, we also assessed the clinical followup files of these patients to exclude a possible selection bias; none of these patients experienced development of a lupus-like syndrome.

Table 3. Nonserologic lupus-like characteristics that occurred in the SpA and RA patients during TNFα blockade*
CharacteristicSpA (n = 34)RA (n = 59)SpA (n = 20)
  • *

    Values are the number of patients. SpA = spondylarthropathy; RA = rheumatoid arthritis; TNFα = tumor necrosis factor α.

  • Anemia was defined as hemolytic anemia with reticulocytosis; leukopenia was defined as a white blood cell count of <4,000/μl on at least 2 occasions; lymphopenia was defined as a lymphocyte count of <1,500/μl on at least 2 occasions; thrombocytopenia was defined as a thrombocyte count of <100,000/μl; proteinuria was defined as a protein concentration of >0.2 gm/liter.

TNFα blockerInfliximabInfliximabEtanercept
Followup period2 years2 years1 year
Malar rash000
Discoid rash000
Photosensitivity010
Oral ulcers000
Serositis (pleuritis/pericarditis)000
Neurologic disorder (seizures/psychosis)000
Hematologic disorder   
 Anemia000
 Leukopenia440
 Lymphopenia4240
 Thrombocytopenia000
Renal disorder (proteinuria)040

TNFα blockade and aCL concentrations.

In order to investigate whether the induction of IgM but not IgG responses might also apply to other (auto)antibody systems, we analyzed another lupus-related reactivity, induction of aCL. In the infliximab-treated patients with SpA, the IgG aCL concentrations did not change significantly (5.46 units/ml at baseline, 5.21 units/ml at year 1, and 4.91 units/ml at year 2), whereas the IgM aCL concentrations increased significantly (1.69 units/ml at baseline, 2.14 units/ml at year 1, and 2.69 units/ml at year 2) (for baseline versus year 1, P = 0.043; for year 1 versus year 2, P = 0.007) (Figure 2A). We confirmed these results in an independent experiment using commercially available ELISA kits (Orgentec) (data not shown).

Figure 2.

Effect of tumor necrosis factor α blockade on IgG and IgM anticardiolipin antibodies in A, infliximab-treated patients with spondylarthropathy (SpA), B, infliximab-treated patients with rheumatoid arthritis, and C, etanercept-treated patients with SpA. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines inside the boxes represent the medians. Lines outside the boxes represent the 10th and the 90th percentiles. Outliers are not displayed. ns = not significant. ∗ = baseline versus year 1; † = year 1 versus year 2.

Similarly, in the infliximab-treated patients with RA, the IgG aCL concentrations did not change (5.50 units/ml at baseline versus 5.40 units/ml at year 1), whereas the IgM aCL concentrations increased significantly (1.62 units/ml at baseline versus 2.59 units/ml at year 1; P < 0.001) (Figure 2B). In the etanercept-treated patients with SpA, the IgG aCL concentrations decreased between baseline and year 1 (4.53 units/ml and 3.47 units/ml, respectively; P = 0.028), with a similar trend for IgM aCL concentrations (1.69 units/ml versus 1.42 units/ml; P = 0.056) (Figure 2C). In terms of the aCL system, these data confirm that infliximab but not etanercept increases selectively the IgM autoantibody response without an isotype switch to IgG. However, the data also bring into question the clinical relevance, because the levels remained clearly below the diagnostic cutoff value associated with a phospholipid syndrome.

DISCUSSION

Although the fact that ANAs are induced during infliximab treatment is well known (1, 7–12), the clinical and biologic correlates in autoimmune arthritis are not yet fully understood. The present study confirms and extends results from our previous short-term study (8) by indicating that the infliximab-induced autoantibody profiles are not critically dependent on the disease background and remain stable over a 2-year treatment period in both SpA and RA. Of interest, the induction of anti-dsDNA antibodies tended to be more pronounced in patients with SpA than in patients with RA. This effect could be related to the use of concomitant methotrexate therapy in patients with RA, because methotrexate can lead to a decrease in circulating autoantibodies in cutaneous lupus erythematosus (24). However, this influence of methotrexate could not be confirmed during infliximab treatment (7, 9, 11, 12). Alternatively, the observed differences could be explained by the higher dosage of infliximab in patients with SpA, although the influence of the total dose of infliximab on new autoantibody formation remains controversial (7, 9). In the present study, 15 patients with RA received an extra (100-mg) vial of infliximab at each infusion, beginning at week 30, but this had no significant influence on autoantibody induction (data not shown).

The first major new finding of the present study is that etanercept did not lead to similar induction of ANAs and anti-dsDNA antibodies. Whereas the occasional appearance of ANAs and anti-dsDNA antibodies in a few etanercept-treated SpA patients is consistent with previous data in etanercept-treated patients with RA (4, 14, 25), this first prospective head-to-head comparison of both TNFα blockers indicates clearly that the induction of both ANAs and anti-dsDNA antibodies is far more pronounced during infliximab therapy than during etanercept treatment. Several explanations for this difference could be envisaged. First, the difference might be related to a general, nonspecific B cell activation by infliximab (26). However, neither infliximab-treated nor etanercept-treated patients with SpA showed significant changes in serum IgG, IgA, or IgM levels (data not shown). Second, the induction of ANAs has been suggested to be linked to alterations of apoptosis and the release of nuclear antigens (7, 27). Although infliximab but not etanercept induces apoptosis of monocytes and T lymphocytes in Crohn's disease (28), both drugs induced apoptosis in RA synovium (29). Third, because the clearance of nuclear debris is also of crucial importance for ANA induction (30–32), down-regulation of the C-reactive protein (CRP) level could potentiate autoimmunity by reducing this clearance (7). However, the CRP level was profoundly decreased in both the infliximab-treated and etanercept-treated groups and was comparable in patients with and those without autoantibody induction (data not shown).

Besides the difference in autoantibody induction by infliximab and etanercept, a second biologic phenomenon of major importance is that the induction of autoantibodies was largely restricted to anti-dsDNA antibodies of the IgM or associated IgM–IgA subtype. Based on published data in human SLE and animal models of lupus, as well as our own SLE control group, one would expect the induction of antibodies of different isotypes, including IgG, against a variety of nuclear antigens released during apoptosis (22, 23, 33). Further studies have been undertaken to investigate whether the intriguing absence of other antinuclear reactivities and autoantibodies of the IgG subtype could be related to the T cell–dependent or independent nature of the nuclear antigen (34–39). This could also be related to modulation by TNFα blockade of specific B cell populations, especially the so-called IgM memory B cells, which are prone to differentiation in plasma cells producing predominantly high-affinity IgM and IgA against T cell–independent antigens (40). Independently of the exact mechanisms underlying the restriction of the induced responses to IgM and IgM/IgA anti-dsDNA antibodies, this profile suggests short-term, nonpathogenic humoral autoimmunity rather than a genuine SLE-associated signature (7, 22, 23, 41–43). This is confirmed by 2 clinically important observations, as follows.

First, the level of induced ANAs decreased after therapy was withdrawn, with complete disappearance of the anti-dsDNA antibodies (including IgG anti-dsDNA antibodies in 1 patient). Second, we observed no lupus-like syndromes over a 2-year followup period, even in the rare case involving induction of IgG anti-dsDNA antibodies. The observed leukopenia and lymphopenia were mostly mild and transient, occurred specifically in the RA patients who received concomitant methotrexate therapy, and were not associated with the induced autoantibodies. Although more than 40 cases of TNFα blockade–related lupus-like syndromes have been reported, the present data provide evidence that ANAs and anti-dsDNA antibodies as well as mild hematologic abnormalities can hardly be considered as lupus criteria in this context, and confirm that clinically relevant lupus erythematosus induced by TNFα blockade is rare (4, 20, 44, 45).

A last important observation is that the differential effect on IgM and IgG antibody induction may be a more generalized effect of infliximab. Indeed, infliximab treatment induced an increase in IgM aCL concentrations in both SpA and RA but failed to increase IgG aCL, even during longer-term followup. In contrast, etanercept treatment decreased the aCL concentrations, indicating that infliximab and etanercept have not only quantitatively but also qualitatively different effects on the humoral autoimmune responses. Although the effect of TNFα blockade on aCL is controversial (10–12, 46), the restricted increase of IgM aCL concentrations during therapy with infliximab but not etanercept parallels our findings for anti-dsDNA antibodies. Although these data have no direct clinical implication, because the aCL concentrations remained below the diagnostic cutoff, they indicate that the modulation of the humoral immune response by TNFα blockade is not restricted to the ANA/anti-dsDNA antibody system.

Taken together, the results of this study indicate that the prominent ANA and anti-dsDNA autoantibody response is not a pure class effect of TNFα blockers, is independent of the disease background, and is not associated with clinically relevant lupus-like symptoms. Furthermore, the restricted induction of anti-dsDNA antibodies of the IgM isotype and the similar findings for aCL suggest a broader biologic effect of TNFα blockade on humoral immunity.

Acknowledgements

We thank Annette Heirwegh, Jeanine Discart, and Virgie Baert for excellent technical assistance. Lydie Meheus (Innogenetics, Zwijnaarde, Belgium) kindly provided the INNO-LIA ANA update.

Ancillary