Antinuclear antibodies following infliximab treatment in patients with rheumatoid arthritis or spondylarthropathy

Abstract

Objective

To investigate the effect of infliximab treatment on antinuclear antibodies (ANAs), anti–double-stranded DNA (anti-dsDNA), antinucleosome, antihistone, and anti–extractable nuclear antigen (anti-ENA) antibodies in rheumatoid arthritis (RA) and spondylarthropathy (SpA) patients.

Methods

Sera from 62 RA and 35 SpA patients treated with infliximab were tested at baseline and week 30 (RA group) or week 34 (SpA group). ANAs were tested by indirect immunofluorescence (IIF) on HEp-2 cells. Anti-dsDNA antibodies were detected by IIF on Crithidia luciliae and by enzyme-linked immunosorbent assay (ELISA) and were further isotyped with γ, μ, and α chain–specific conjugates at various time points. Antinucleosome antibodies were tested by ELISA. Antihistone and anti-ENA antibodies were detected by line immunoassay.

Results

Initially, 32 of 62 RA patients and 6 of 35 SpA patients tested positive for ANAs. After infliximab treatment, these numbers shifted to 51 of 62 (P < 0.001) and 31 of 35 (P < 0.001), respectively. At baseline, none of the RA or SpA patients had anti-dsDNA antibodies. After infliximab treatment, 7 RA patients (P = 0.016) and 6 SpA patients (P = 0.031) became positive for anti-dsDNA antibodies. All 7 anti-dsDNA–positive RA patients had IgM and IgA anti-dsDNA antibodies. Three of the 6 anti-dsDNA–positive SpA patients had IgM and IgA anti-dsDNA antibodies, and 2 had IgM anti-dsDNA antibodies alone. In both diseases, the IgM anti-dsDNA antibodies appeared before the IgA anti-dsDNA antibodies. During the observation period, no IgG anti-dsDNA antibodies or lupus symptoms were observed. The development of antinucleosome, antihistone, or anti-ENA antibodies following infliximab treatment was observed in some patients, but the numbers were not statistically significant.

Conclusion

Infliximab treatment may induce ANAs, and especially IgM and IgA anti-dsDNA antibodies, in RA and SpA patients. However, no anti-dsDNA IgG antibodies or lupus symptoms were observed during the period of observation in this study, and the development of antinucleosome, antihistone, or anti-ENA antibodies was not statistically significant. These observations do not exclude potential induction of clinically significant lupus in the long term, and further followup is therefore mandatory.

The chronic inflammatory joint diseases rheumatoid arthritis (RA) and spondylarthropathy (SpA) are the two most frequent types of autoimmune arthritis, with a prevalence rate of 1–2% (1, 2). The classic treatment schemes for RA consist of a combination of disease-modifying antirheumatic drugs (DMARDs) and nonsteroidal antiinflammatory drugs (NSAIDs). SpA patients are mainly treated with NSAIDs, eventually in combination with sulfasalazine for peripheral arthritis or methotrexate (MTX) for psoriatic arthritis. This approach, however, often leads to only partial control of the inflammation and structural damage (3–5). Although the pathogenesis of RA and SpA remains incompletely understood, much insight into the cellular and molecular mechanisms involved has been gained in the last decade. Based on these insights, new therapies have been developed. One of these new strategies is the use of biologic agents directed against tumor necrosis factor α (TNFα).

Infliximab (Remicade; Centocor, Malvern, PA) is a chimeric anti-TNFα monoclonal IgG1 antibody that neutralizes the soluble cytokine and blocks the membrane-bound cytokine. Infliximab is very effective in reducing the chronic symptoms and indicators of inflammation in patients whose RA has failed to respond to conventional DMARDs (6–11), and there is evidence that infliximab in combination with MTX halts radiologic progression in RA (10). Recently, TNFα blockade has also opened new perspectives for the treatment of SpA: significant clinical and laboratory improvements in patients with active SpA treated with infliximab have been reported (12–16). Also, in patients with treatment-resistant Crohn's disease (CD), infliximab has shown to be effective in reducing the symptoms (17, 18) and healing the enterocutaneous fistulas (19).

Headache, nausea, upper respiratory tract infections, and infusion-related reactions are the most commonly observed adverse events with infliximab (20). However, serious adverse events, such as opportunistic infections (Mycobacterium tuberculosis, fungal and bacterial sepsis) have also been reported (20–22).

Clinical trials reveal that infliximab induces autoantibodies, such as antinuclear antibodies (ANAs) and anti–double-stranded DNA (anti-dsDNA). According to the safety data on infliximab, 63.8% of RA patients and 49.1% of CD patients newly develop positivity for ANA during infliximab treatment, and 13% of infliximab-treated RA patients and 21.5% of infliximab-treated CD patients develop positivity for anti-dsDNA antibodies (23). However, systematic studies concerning the incidence and isotype of induced autoantibodies, and especially anti-dsDNA antibodies, during infliximab treatment are limited with regard to RA and nonexistent for SpA. The aim of this study was to describe the effect of infliximab treatment on ANAs, anti-dsDNA antibodies, and other antinuclear reactivities (antinucleosome antibodies, antihistone antibodies, and anti–extractable nucler antigen [anti-ENA] antibodies), with special reference to the isotyping and the timing of induction of the anti-dsDNA antibodies in the context of infliximab treatment.

PATIENTS AND METHODS

Patients and medication regimen.

Sixty-two patients (24 men and 38 women; mean age at baseline 53.9 years [range 32–76]) with refractory RA, treated with infliximab at one center in the context of an expanded access program, were included in the study (24). All patients fulfilled the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) classification criteria for RA (25). They received 3 mg/kg infliximab intravenously at weeks 0, 2, 6, and every 8 weeks thereafter, in combination with MTX.

Thirty-five patients (23 men and 12 women; mean age at baseline 46.6 years [range 29–66]) with active SpA were treated with infliximab in a placebo-controlled trial in which the placebo group received active treatment with infliximab after 12 weeks (15). All patients fulfilled the European Spondylarthropathy Study Group classification criteria for SpA (26). The treatment schedule was a loading-dose regimen (intravenous infusions at weeks 0, 2, and 6) of 5 mg/kg infliximab, followed by 10 mg/kg infliximab every 14 weeks (15). Originally, 40 SpA patients were included in the placebo-controlled trial with infliximab. In the active treatment group (n = 20), infliximab treatment was stopped in 3 patients before week 34, due to M tuberculosis infection (n = 1), a diagnostic procedure–related adverse event (n = 1), and withdrawal from the study for personal reasons (n = 1). In the placebo group (n = 20), no dropouts were reported during the placebo treatment period, but during the infliximab treatment, 2 patients were withdrawn from the study before week 34, due to M tuberculosis infection (n = 1) and a protocol violation by the patient (n = 1). Since both groups received the same regimen, we pooled the remaining SpA patients treated with infliximab (n = 35).

In the RA patients, serum samples were obtained at baseline prior to infusion and at the time of each infusion up to week 30 (weeks 0, 2, 6, 14, 22, and 30). In the SpA patients, serum samples were obtained at baseline prior to infusion, at the time of each infusion up to week 34 (weeks 0, 2, 6, 20, and 34), and at week 12. Serum was obtained by clotting through centrifugation and was aliquoted at −20°C until further analysis. Serum samples were collected after informed consent from the patients and approval by the local ethics committee were obtained.

Detection of ANAs.

Serum diluted 1:40 in phosphate buffered saline (PBS) was overlaid onto fixed HEp-2 cells (MeDiCa, Carlsbad, CA). Slides were washed twice for 5 minutes each with PBS, overlaid with fluorescence-labeled conjugate (anti-human IgG heavy and light chain specific; MeDiCa), and incubated for an additional 30 minutes. After washing twice, a coverslip was placed over the slide, and the slides were read using a fluorescence microscope at 40× power. The fluorescence intensity was scored semiquantitatively from 1+ to 5+ relative to the intensity of a negative and a positive (4+) control (27). The fluorescence pattern was also reported.

Detection of anti-dsDNA antibodies.

For the initial screening of anti-dsDNA antibodies, serum samples obtained at baseline and at week 30 or 34 were tested by indirect immunofluorescence (IIF) on Crithidia luciliae and by anti-dsDNA enzyme-linked immunosorbent assay (ELISA) (see below), using an anti-human IgG heavy and light chain–specific conjugate in both assays. A patient was classified as having anti-dsDNA antibodies only when a positive result was confirmed in both anti-dsDNA assays. Serum samples from patients with anti-dsDNA antibodies according to this definition were further isotyped with γ, μ, and α chain–specific conjugates and were tested at various time points in order to determine the moment anti-dsDNA antibodies appeared.

IIF on C luciliae.

C luciliae–coated slides (MeDiCa) were used for the detection of anti-dsDNA antibodies. Serum samples diluted 1:20 in PBS were incubated for 30 minutes at room temperature. As in the ANA experiments, slides were washed twice for 5 minutes each with PBS, overlaid with fluorescence-labeled conjugate (anti-human IgG heavy and light chain specific; MeDiCa), and incubated for an additional 30 minutes. After washing twice, a coverslip was placed over the slide, and the slides were read using a fluorescence microscope at 40× power.

IIF on C luciliae was also used for isotyping of the anti-dsDNA antibodies. For the detection of anti-dsDNA antibodies of the IgG, IgM, or IgA class, we used specific conjugates against human IgG heavy (γ) chain (P.A.R.I.S, Compiègne, France), human IgM heavy (μ) chain (Dako, Glostrup, Denmark), and human IgA heavy (α) chain (Dako).

Anti-dsDNA ELISA.

A commercially available kit (Varelisa dsDNA antibodies; Pharmacia Diagnostics, Freiburg, Germany) was used for the detection of anti-dsDNA antibodies. The ELISA was performed according to the manufacturer's instructions. Briefly, the wells were washed once with 300 μl wash buffer immediately prior to use. Subsequently, 100 μl of standards, controls, and patient samples (1:100) were dispensed into the appropriate wells. After 30 minutes of incubation, the fluid was aspirated from the wells and the wells were washed 3 times with wash buffer. In a second step, 100 μl of conjugate (anti-human IgG heavy and light chain specific, coupled with horseradish peroxidase) was added to the wells. After incubation for 30 minutes, the wells were washed 3 times with wash buffer. Thereafter, 100 μl of enzyme substrate (3,3′,5,5′-tetramethylbenzidine) was dispensed into the wells. After incubation for 10 minutes in the dark, 50 μl of stop solution (0.5M H₂SO₄) was dispensed into all wells. The absorbance (optical density [OD]) was read at 450 nm. According to the manufacturer's instructions, a result of ≥35 IU/ml was considered positive.

Detection of antinucleosome antibodies.

A commercially available kit (Medizym antinucleo; Medipan Diagnostica, Selchow, Germany) was used for the detection of antinucleosome antibodies. The ELISA was performed according to the manufacturer's instructions. Briefly, 100 μl of standards, controls, and patient samples (1:50) was dispensed into the appropriate wells. After 30 minutes of incubation at room temperature, the fluid was aspirated from the wells and the wells were washed 3 times with wash buffer. In a second step, 100 μl of conjugate (anti-human IgG heavy chain specific, coupled with horseradish peroxidase) was added to each well. After incubation for 30 minutes at room temperature, the wells were washed 3 times with wash buffer. Thereafter, 100 μl of enzyme substrate (3,3′,5,5′-tetramethylbenzidine in citrate buffer containing hydrogen peroxide) was dispensed into the wells. After incubation in the dark for 10 minutes at room temperature, 100 μl of stop solution (0.3M H₂SO₄) was added to all wells. OD was read at 450 nm. According to the manufacturer's instructions, a result of >25 units/ml was considered positive.

Detection of anti-ENA and antihistone antibodies by line immunoassay.

Sera were analyzed for anti-ENA and antihistone antibodies by line immunoassay (LIA) (INNO-LIA ANA update K1090; Innogenetics, Zwijnaarde, Belgium). This multiparameter assay utilizes 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.

The test was performed according to the manufacturer's instructions. Briefly, nylon strips were incubated with serum at a 1:200 dilution for 1 hour at room temperature. After washing 3 times with wash buffer, 2 ml of conjugate (anti-human IgG heavy and light chain specific, labeled with alkaline phosphatase) was added to all troughs and was allowed to bind to the antigen–antibody complex if one was present. The strips were incubated for 30 minutes at room temperature. Thereafter, they were washed twice with wash buffer and once with substrate buffer before addition of 2 ml 5-bromo-4-chloro-3-indolyl phosphatase/nitroblue tetrazolium to each trough. The enzyme substrate and chromogen produced a dark brown color in proportion to the amount of specific autoantibody in the test sample. The strips were incubated for 30 minutes at room temperature. After aspiration of the liquid, 2 ml of stop solution (0.1M H₂SO₄) was added to each trough to stop the color development. After 10–30 minutes of incubation at room temperature, the fluid was aspirated and the strips were removed from the test troughs. The results were interpreted when the strips had dried completely.

Monitoring for lupus-like symptoms.

History-taking at every visit included questioning for any adverse event and, more specifically, active screening for fever or other symptoms suggestive of infection or rash. At the time of each infusion, erythrocyte sedimentation rate, C-reactive protein level, peripheral blood count, liver and kidney function, and levels of autoantibodies (ANA, anti-dsDNA, anti-ENA, antihistone) were tested. Urine was analyzed for proteinuria at the beginning and the end of the observation period. In order to evaluate lupus-like symptoms, we reviewed the medical charts of the RA and SpA patients who were positive for ANAs and anti-dsDNA antibodies following infliximab treatment. The following components of the ACR classification criteria for systemic lupus erythematosus (SLE) could then be positively confirmed or ruled out (28, 29): malar rash, discoid rash, photosensitivity, renal disorder, and hematologic disorder. The other SLE criteria (oral ulcers, serositis, and neurologic disorder) were assumed to be absent if symptoms were not reported by the patient.

Statistical analysis.

Statistical analysis was performed using SPSS software (SPSS, Chicago, IL). For dichotomous data, we used McNemar's test, which is suitable for paired analysis. When statistical analysis could not be performed because of empty rows in the crosstab, the value 1 was added to one cell. For comparison of proportions, we used a binomial test. P values less than or equal to 0.05 were considered significant.

RESULTS

Effect of infliximab treatment on ANAs in RA and SpA patients.

At baseline, 32 of 62 RA patients (51.6%) and 6 of 35 SpA patients (17.1%) were positive for ANAs, most frequently with a homogeneous pattern (78.1% and 66.7%, respectively). After infliximab treatment, 51 of 62 RA patients (82.3%) had a positive result for ANAs at week 30, and 31 of 35 SpA patients (88.6%) tested positive for ANA at week 34 (both P < 0.001 versus baseline) (Table 1). The fluorescence pattern after infliximab treatment was most frequently homogeneous (92.2% and 93.5% in RA patients and SpA patients, respectively). Induction of ANAs was observed in both diseases, but was more pronounced in SpA than in RA: newly positive results for ANA were found in 20 RA patients (32.3%) and 25 SpA patients (71.4%) (P < 0.001, RA versus SpA). Moreover, an increase in ANA intensity of 2 or more points (on a 0–5+ scale) following infliximab therapy was observed in 28 RA patients (45.2%) compared with 26 SpA patients (74.3%) (P < 0.001, RA versus SpA).

Table 1. Autoantibody profile in RA and SpA patients after infliximab treatment*

Antibody, detection method	RA (n = 62)		SpA (n = 35)
	Week 0	Week 30	Week 0	Week 34
* Values are the number (%). RA = rheumatoid arthritis; SpA = spondylarthropathy; ANA = antinuclear antibody; IIF = indirect immunofluorescence; anti-dsDNA = anti–double-stranded DNA; ELISA = enzyme-linked immunosorbent assay; LIA = line immunoassay. † P ≤ 0.001 versus baseline. ‡ P ≤ 0.01 versus baseline. § P ≤ 0.05 versus baseline.
ANA, IIF on HEp-2 cells	32 (51.6)	51 (82.3)†	6 (17.1)	31 (88.6)†
Anti-dsDNA
IIF on Crithidia luciliae	0 (0)	27 (43.5)†	0 (0)	20 (57.1)†
ELISA	2 (3.2)	10 (16.1)‡	6 (17.1)	6 (17.1)
Both IIF on Crithidia luciliae and ELISA	0 (0)	7 (11.3)§	0 (0)	6 (17.1)§
Antinucleosome, ELISA	1 (1.6)	5 (8.1)	3 (8.6)	4 (11.4)
Antihistone, LIA	3 (4.8)	6 (9.7)	2 (5.7)	5 (14.3)
Anti-SmD, LIA	1 (1.6)	1 (1.6)	0 (0)	0 (0)
Anti-RNP, LIA	0 (0)	0 (0)	0 (0)	0 (0)
Anti-SSA, LIA	0 (0)	0 (0)	0 (0)	0 (0)
Anti-SSB, LIA	1 (1.6)	2 (3.2)	1 (2.9)	1 (2.9)
Anticentromere, IIF/anti-CENP, LIA	1 (1.6)	1 (1.6)	0 (0)	0 (0)

Effect of infliximab treatment on anti-dsDNA antibodies in RA and SpA patients.

Initial screening by IIF on C luciliae and by ELISA revealed that at baseline, none of the 62 RA patients were positive by both assays for anti-dsDNA antibodies. At week 30, 7 RA patients (11.3%) had developed anti-dsDNA antibodies demonstrated in both assays (P = 0.016 versus baseline) (Table 1). Similarly, initial screening with IIF on C luciliae and ELISA showed no patient in the SpA population to have positive results on both assays for anti-dsDNA antibodies at baseline. At week 34, 6 SpA patients (17.1%) were positive for anti-dsDNA antibodies (P = 0.031 versus baseline) (Table 1).

The induced anti-dsDNA antibodies were further isotyped using γ, μ, and α chain–specific conjugates (Table 2). None of the 7 RA patients had anti-dsDNA antibodies of the IgG class after 30 weeks of infliximab treatment, but all 7 were positive for anti-dsDNA antibodies of the IgM and IgA class. At week 34, none of the 6 SpA patients had anti-dsDNA antibodies of the IgG class, but 3 had anti-dsDNA antibodies of both the IgM and IgA class, and 2 had anti-dsDNA antibodies of the IgM class alone.

Table 2. Isotyping of anti-dsDNA antibodies in RA and SpA patients after infliximab treatment*

Patient	Anti-dsDNA antibodies at week 30 (RA patients) or week 34 (SpA patients)
	IgG	IgM	IgA
* See Table 1 for definitions.
RA
1	−	+	+
2	−	+	+
3	−	+	+
4	−	+	+
5	−	+	+
6	−	+	+
7	−	+	+
Total positive	0/7	7/7	7/7
SpA
1	−	−	−
2	−	+	+
3	−	+	−
4	−	+	+
5	−	+	−
6	−	+	+
Total positive	0/6	5/6	3/6

To study the timing of the induction of anti-dsDNA antibodies in relation to infliximab treatment, we analyzed, at various time points during infliximab treatment, serum samples from the 7 RA and 6 SpA patients who had anti-dsDNA antibodies at week 30 or 34 (Tables 3 and 4). None of the 7 RA patients had anti-dsDNA antibodies of the IgM class at baseline. One RA patient was positive for anti-dsDNA antibodies of the IgA class at baseline (by IIF on C luciliae with IgA-specific conjugate), and that patient remained positive up to week 30. Anti-dsDNA antibodies of the IgM class appeared in the majority of RA patients (5 of 7) at week 6, whereas anti-dsDNA antibodies of the IgA class did not appear until week 30 in the 6 patients who were negative for these antibodies at baseline (Table 3). One SpA patient was positive for anti-dsDNA antibodies of the IgM class at baseline, and, in another, anti-dsDNA antibodies of the IgA class were detected at baseline (by IIF on C luciliae with IgA-specific conjugate). Both patients remained positive up to week 34. Induction of anti-dsDNA antibodies occurred over a wider span of time in SpA than in RA. However, similar to the findings in RA, anti-dsDNA antibodies of the IgM class appeared before anti-dsDNA antibodies of the IgA class (Table 4).

Table 3. Detection and isotyping of anti-dsDNA antibodies in RA patients at various time points*

Patient	Anti-dsDNA antibody isotype
	IgG		IgM						IgA
	Week 0	Week 30	Week 0	Week 2	Week 6	Week 14	Week 22	Week 30	Week 0	Week 2	Week 6	Week 14	Week 22	Week 30
* See Table 1 for definitions.
1	−	−	−	−	+	+	+	+	+	+	+	+	+	+
2	−	−	−	−	−	+	+	+	−	−	−	−	−	+
3	−	−	−	−	+	+	+	+	−	−	−	−	−	+
4	−	−	−	−	+	+	+	+	−	−	−	−	−	+
5	−	−	−	−	−	+	+	+	−	−	−	−	−	+
6	−	−	−	−	+	+	+	+	−	−	−	−	−	+
7	−	−	−	−	+	+	+	+	−	−	−	−	−	+
Total	0/7	0/7	0/7	0/7	5/7	7/7	7/7	7/7	1/7	1/7	1/7	1/7	1/7	7/7

Table 4. Detection and isotyping of anti-dsDNA antibodies in SpA patients at various time points*

Patient	Anti-dsDNA antibody isotype
	IgG		IgM						IgA
	Week 0	Week 30	Week 0	Week 2	Week 6	Week 12	Week 20	Week 34	Week 0	Week 2	Week 6	Week 12	Week 20	Week 34
* See Table 1 for definitions.
1	−	−	−	−	+	+	+	−	−	−	−	−	−	−
2	−	−	+	+	+	+	+	+	−	−	−	−	+	+
3	−	−	−	−	+	+	+	+	−	−	−	−	−	−
4	−	−	−	−	−	+	+	+	−	−	−	−	−	+
5	−	−	−	−	−	−	+	+	−	−	−	−	−	−
6	−	−	−	+	+	+	+	+	+	+	+	+	+	+
Total	0/6	0/6	1/6	2/6	4/6	5/6	6/6	5/6	1/6	1/6	1/6	1/6	2/6	3/6

Effect of infliximab treatment on other antinuclear reactivities in RA and SpA patients.

Antinucleosome antibodies detected by ELISA.

One of the 62 RA patients (1.6%) had antinucleosome antibodies at baseline. At week 30, 5 RA patients (8.1%) tested positive for these antibodies (Table 1); the 1 RA patient who had a positive result at baseline remained positive, and 4 other RA patients became positive (P = 0.125 versus baseline). Two of the 7 RA patients with anti-dsDNA antibodies at week 30 were also positive for antinucleosome antibodies at week 30. Among SpA patients, 3 of 35 (8.6%) had a positive result for antinucleosome antibodies at baseline. After 34 weeks of treatment with infliximab, 4 patients (11.4%) were positive (Table 1), including 2 of the 3 who were positive at baseline; 1 SpA patient with a positive result at baseline became negative, and 2 patients exhibited new development of antinucleosome antibodies (P = 1.000 versus baseline). Four of the 6 SpA patients with anti-dsDNA antibodies at week 34 were also positive for antinucleosome antibodies at week 34.

Antihistone antibodies and anti-ENA antibodies detected by LIA.

At baseline, 4 of 62 RA patients (6.5%) had at least 1 fine reactivity on LIA (2 with reactivity to histones, 1 with reactivity to CENP-B and histones, 1 with reactivity to SSB and SmD). At week 30, 8 RA patients (12.9%) had at least 1 fine reactivity on LIA (P = 0.125 versus baseline). The 4 RA patients with a reactivity at baseline remained positive. Four other RA patients developed new reactivities after infliximab treatment (3 patients with reactivity to histones, 1 patient with reactivity to SSB) (Table 1). Three of 35 SpA patients (8.6%) were positive for at least 1 reactivity on LIA at baseline (2 patients with reactivity to histones and 1 with reactivity to SSB). At week 34, 6 of 35 SpA patients (17.1%) had at least 1 fine reactivity on LIA (P = 0.250 versus baseline). The 3 SpA patients with a reactivity at baseline remained positive, and 3 other SpA patients developed new reactivities to histones after treatment (Table 1). In both the RA group and the SpA group, no newly developed reactivities to SmD, RNP, SSA, or CENP-B could be detected.

Detection of lupus-like characteristics in the RA and SpA patients with anti-dsDNA antibodies after infliximab treatment.

Autoantibodies to dsDNA are highly specific for the diagnosis of SLE (30–33) and constitute one of the revised ACR classification criteria for this disease (28, 29). We identified 7 RA and 6 SpA patients who tested positive for ANA and anti-dsDNA antibodies following infliximab treatment. In 1 RA patient (patient 2), anti-SmD antibodies were detected at baseline and that patient remained positive during infliximab treatment. In order to determine if any of these 7 RA and 6 SpA patients developed other, nonserologic lupus-like symptoms during infliximab treatment, we reviewed the medical charts for the ACR SLE classification criteria, as described in Patients and Methods. Since arthritis involving 2 or more peripheral joints was present in all 7 RA and 6 SpA patients as part of the underlying disease, this criterion could not be evaluated. Findings with regard to the remaining criteria indicating newly induced SLE are summarized in Table 5. The 7 RA and 6 SpA patients with anti-dsDNA antibodies exhibited no clinical symptoms suggestive of SLE during the observational period, nor did proteinuria or cellular casts appear. Among the RA patients, 1 developed a mild leukopenia during infliximab treatment and another had an episode of lymphopenia. In SpA patients, no hematologic abnormalities were found.

Table 5. Components of the ACR SLE criteria developing after infliximab treatment in the RA and SpA patients who had ANAs and anti-dsDNA antibodies after treatment*

Criterion	No. of patients
	RA (n = 7)	SpA (n = 6)
* ACR = American College of Rheumatology; SLE = systemic lupus erythematosus (see Table 1 for other definitions). † Lymphopenia (defined as lymphocyte count <1,500/mm^3) in patient 1 (lymphocyte count [/mm^3] 1,850 at baseline, 1,100 at week 2, 3,060 at week 6, 3,180 at week 14, 850 at week 22, 1,290 at week 30); leukopenia (defined as white blood cell count <4,000/mm^3) in patient 4 (white blood cell count [/mm^3] 4,110 at baseline, 4,290 at week 2, 4,050 at week 6, 5,210 at week 14, 3,950 at week 22, 3,780 at week 30).
Malar rash	0	0
Discoid rash	0	0
Photosensitivity	0	0
Oral ulcers	0	0
Serositis	0	0
Renal disorder	0	0
Neurologic disorder	0	0
Hematologic disorder	2†	0

DISCUSSION

Treatment with biologic agents directed against TNFα, such as infliximab and etanercept (Enbrel; Wyeth-Ayerst, Philadelphia, PA), has significant clinical benefits in inflammatory diseases. Infliximab is registered by the Food and Drug Administration (FDA) and by the European Medicines Evaluation Agency (EMEA) as a therapy for treatment-resistant RA (6–11), treatment-resistant moderate-to-severe CD, and CD with fistulas (17–19). More recently, the efficacy of infliximab therapy has been proven in diseases that are within the SpA category (12–16). Etanercept is approved by the FDA and the EMEA for use in treatment-resistant RA, severe active, and progressive RA (34–38), and treatment-resistant polyarticular juvenile chronic arthritis (39, 40), and by the FDA for treatment-resistant psoriatic arthritis (41). A beneficial effect of etanercept treatment in ankylosing spondylitis has also been demonstrated (42). The induction of ANAs (in 63.8% of RA patients and 49.1% of CD patients) and anti-dsDNA antibodies (in 13% of RA patients and 21.5% of CD patients) after infliximab treatment has been reported (23). The development of ANAs and anti-dsDNA antibodies after etanercept therapy in RA patients has also been described (in 11% and 15% of patients, respectively) (43). However, despite the development of ANAs and anti-dsDNA antibodies in a fairly high proportion of patients during anti-TNFα therapy, the incidence of possible lupus-like syndromes is low.

Some case reports of lupus-like symptoms, with or without autoantibodies, occurring in patients receiving infliximab or etanercept treatment have been published (18, 23, 43–49), but it is difficult to determine whether these lupus-like symptoms are related to the anti-TNFα treatment, to the underlying disease, to other drugs being taken concurrently, or whether they may have occurred coincidentally during anti-TNFα treatment. The identification of drug-induced SLE among patients with RA is especially difficult, owing to the overlaps between the two diseases (25, 28).

Charles et al described the induction of anti-dsDNA antibodies in 7% of RA patients treated with infliximab (with or without MTX therapy) and noted that the isotype of these antibodies differed from IgG (44). The current report extends the phenomenon of infliximab-induced ANAs to a new disease background, i.e., SpA. We confirm the high prevalence of ANAs and anti-dsDNA antibodies in the latter disease following infliximab treatment. Moreover, the increase in ANA titers and the induction of anti-dsDNA antibodies are more pronounced in SpA than in RA. However, ANAs might be more expected in untreated RA compared with SpA (50). Absence of associated MTX therapy in SpA may contribute to the observed difference. Of interest in this context is the observation, by Boehm et al, that MTX therapy could lead to a decrease in circulating autoantibodies in patients with cutaneous lupus (51).

In our study, 11.3% of the RA patients and 17.1% of the SpA patients had anti-dsDNA antibodies at the end of the observational period. Sera that converted to anti-dsDNA positivity were further analyzed at different time points and isotyped. Isotyping revealed that none of the anti-dsDNA antibodies had the IgG isotype (in RA, 7 of 7 IgM and 7 of 7 IgA; in SpA, 5 of 6 IgM and 3 of 6 IgA). In both diseases, IgM anti-dsDNA antibodies appeared mostly in the first 6 weeks, prior to the appearance of anti-dsDNA antibodies of the IgA isotype. The diagnostic and prognostic value of these anti-dsDNA antibodies remains controversial since lupus-associated anti-dsDNA antibodies are classically of the IgG isotype (30–33), while IgM or IgA anti-dsDNA antibodies may occur in other diseases as well (mixed connective tissue disease, primary Sjögren's syndrome, scleroderma, RA, ankylosing spondylitis, chronic active hepatitis, primary biliary cirrhosis) (30, 33, 52). The non-IgG isotype of the anti-dsDNA antibodies detected in the present study, and the fact that we could not identify clinical or biologic features suggestive of lupus (other than anti-dsDNA induction), suggest that clinically significant lupus induction did not occur. However, longer-term followup of these patients will be needed to counter that concern completely.

Apart from the study of anti-dsDNA antibodies, other ANA reactivities have not previously been systematically studied in relation to infliximab therapy. In this study, we additionally tested for antibodies to nucleosomes, histones, and ENA. Antinucleosome antibodies of the IgG class are considered to be a more sensitive marker for SLE than anti-dsDNA antibodies, occurring at an earlier stage than anti-dsDNA and antihistone antibodies (53). Although there was an increase in the number of patients with antinucleosome antibodies following infliximab treatment compared with baseline, this increase was not statistically significant. Antihistone antibodies are particularly common in SLE and drug-induced lupus (with a sensitivity of 50–70% and >95%, respectively [32]), but may also occur in other diseases (RA, juvenile chronic arthritis, autoimmune chronic hepatitis, chronic infection) or even in healthy controls (32, 54). In both the RA and the SpA groups, some patients exhibited new development of antihistone antibodies. However, this increase was not statistically significant in either disease group. Anti-ENA antibodies detected by LIA were rarely observed in RA or SpA. Only 1 RA patient developed a new anti-ENA reactivity (anti-SSB antibodies).

In conclusion, we confirm the induction of ANAs, and especially non-IgG anti-dsDNA antibodies, after infliximab treatment in patients with RA, and we present the first published report of the same phenomenon in SpA patients receiving this new treatment. The present data document that this particular serologic side effect is unrelated to the genetic background or the underlying disease process. The absence of IgG anti-dsDNA antibodies, as well as the absence of other lupus-specific ANAs or clinical features in patients who developed anti-dsDNA antibodies, are rather reassuring in terms of the safety of the treatment. However, longer-term observations of anti-dsDNA–positive patients remain mandatory. The mechanism of anti-dsDNA antibody induction with anti-TNFα therapy is still under study.

Acknowledgements

The authors thank Annette Heirwegh for excellent technical assistance.