SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Objective

Systemic lupus erythematosus (SLE) disease manifestations are highly variable among patients, and the prevalence of individual clinical features differs significantly by ancestry. Serum tumor necrosis factor α (TNFα) levels are elevated in some SLE patients and may play a role in disease pathogenesis. The aim of this study was to look for associations between serum TNFα levels, clinical manifestations of SLE, autoantibodies, and serum interferon-α (IFNα) levels in a large multiancestral SLE cohort.

Methods

We studied serum TNFα levels in 653 SLE patients (214 African Americans, 298 European Americans, and 141 Hispanic Americans). TNFα was measured using an enzyme-linked immunosorbent assay, and IFNα was measured with a functional reporter cell assay. Stratified and multivariate analyses were used to detect associations in each ancestral background separately, with meta-analysis when appropriate.

Results

Serum TNFα levels were significantly higher in SLE patients than in non–autoimmune disease controls (P < 5.0 × 10−3 for each ancestral background). High serum TNFα levels were positively correlated with high serum IFNα levels when tested in the same sample across all ancestral backgrounds (odds ratio range 1.76–1.86, P = 4.8 × 10−3 by Fisher's combined probability test). While serum TNFα levels alone did not differ significantly among SLE patients of different ancestral backgrounds, the proportion of patients with concurrently high levels of TNFα and IFNα was highest in African Americans and lowest in European Americans (P = 5.0 × 10−3). Serum TNFα levels were not associated with autoantibodies, clinical criteria for the diagnosis of SLE, or age at the time of sampling.

Conclusion

Serum TNFα levels are high in many SLE patients, and we observed a positive correlation between serum TNFα and IFNα levels. These data support a role for TNFα in the pathogenesis of SLE across all ancestral backgrounds and suggest important cytokine subgroups within the disease.

Systemic lupus erythematosus (SLE) is characterized by a wide variety of clinical manifestations, including inflammation of the dermal, renal, hematologic, and musculoskeletal organ systems. Differences in the prevalence of particular clinical and serologic manifestations of the disease according to ancestral background have long been appreciated (1). Some of the clinical differences observed in patients of different ancestral backgrounds likely represent differences in biologic pathways related to disease pathogenesis, although little is currently known about the molecular mediators of these differences.

Previous studies have documented elevated serum tumor necrosis factor α (TNFα) levels in some patients with SLE, and these levels have been correlated with clinical disease activity and anti–double-stranded DNA (anti-dsDNA) antibodies (2, 3). TNFα is also overexpressed in renal tissue in lupus nephritis (4). Although TNFα is present at sites of inflammation, the role of TNFα in human SLE pathogenesis is a subject of controversy. The role of TNFα in murine models of SLE is a subject of similar debate. In some models, TNFα improved disease features, while in others, TNFα blockade was beneficial (4).

Interferon-α (IFNα) and TNFα appear to cross-regulate each other in vitro (5). TNFα inhibits peripheral dendritic cell generation and secretion of IFNα by these cells (5). In healthy peripheral blood mononuclear cells, culture with etanercept led to an increase in IFNα and IFNα-inducible genes, and IFNα inhibited secretion of TNFα (5, 6). Many lines of evidence support the idea that IFNα is a primary pathogenic factor in SLE, including the development of SLE in patients given recombinant IFNα to treat viral infections and malignancy, and familial aggregation of high serum IFNα levels in SLE (7, 8). Thus, there is some reasonable concern that SLE-like features that have arisen during anti-TNFα therapy may relate to increased IFNα levels (9), and that this increase in IFNα could catalyze a change in the clinical syndrome from rheumatoid arthritis to SLE. Clinical trials in SLE patients suggest that short-term TNFα blockade may have benefits in lupus nephritis, as well as transient benefits in lupus arthritis (4), but some significant side effects were reported in a small group of patients who received long-term anti-TNFα therapy (10).

In the present study, we examined relationships between serum TNFα levels and simultaneously obtained IFNα levels, serologic parameters, and clinical parameters in SLE. Given the interrelated nature of many of the clinical and serologic characteristics in SLE and the potential for a relationship between TNFα and IFNα, we used multivariate regression models to account for between-variable relationships.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Patients, samples, and data.

Serum samples were obtained from 653 SLE patients (214 African American, 298 European American, and 141 Hispanic American patients) from the Lupus Family Registry and Repository at the Oklahoma Medical Research Foundation. All patients met the American College of Rheumatology (ACR) criteria for the diagnosis of SLE (11), and information regarding the presence or absence of these criteria as well as of SLE-associated autoantibodies (antinuclear, anti-Ro, anti-La, anti-Sm, anti-RNP, and anti-dsDNA antibodies) was available for all patients. The 62 controls included in the study were unrelated individuals who were screened by medical record review for the absence of autoimmune disease. The controls were of similar age (mean ± SD 45.6 ± 12.9 years), sex (90.3% female), and ancestral background (39% African American, 44% European American, 15% Hispanic American) as the SLE patients. All subjects provided informed consent, and the study was approved by the institutional review board at the respective institutions.

Measurement of serum IFNα activity.

We have developed a sensitive and reproducible bioassay for detecting serum IFNα activity (8), since enzyme-linked immunosorbent assay (ELISA) methods for the detection of IFNα in human serum have been limited by low sensitivity and specificity. In this bioassay, reporter cells (WISH cells; ATCC catalog no. CCL125) are used to measure the ability of sera to cause IFN-induced gene transcription. In this study, the reporter cells were cultured with patient sera for 6 hours and then lysed, and 3 canonical IFNα-induced transcripts (IFN-induced protein with tetratricopeptide repeats 1, MX-1, and RNA-dependent protein kinase) were measured using reverse transcriptase–polymerase chain reaction. Relative expression data of the 3 transcripts were then normalized to healthy donor sera (n = 141), run in the same assay, and presented as an IFNα activity score (for further details, see ref.8).

Measurement of TNFα levels.

Serum TNFα was measured in all of our samples using a commercial ELISA kit (Pierce). According to the manufacturer, addition of TNFα receptor types I and II (40 mg/ml) does not interfere with this assay. Patient samples were considered positive if TNFα levels were >2 SD above the mean level of our nonautoimmune control population.

Statistical analysis.

The SLE cohort was stratified by self-reported ancestral background (African American, European American, and Hispanic American), and each ancestral background was analyzed separately. Quantitative TNFα data were compared using a Mann-Whitney nonparametric U test. Categorical analyses were performed using chi-square test statistics to compare proportions among groups. In addition to stratification by ancestral background, in some analyses, patients were further stratified categorically into high-level versus low-level TNFα and IFNα subgroups (levels were considered high if they were >2 SD above the mean level of healthy donors). This resulted in 4 groups representing all 4 possible combinations of the 2 categorical variables, as follows: high IFNα/high TNFα, low IFNα/high TNFα, high IFNα/low TNFα, and low IFNα/low TNFα.

Multivariate modeling was performed using logistic regression. The model included European versus non-European ancestry, age at recruitment, each of the ACR criteria, each autoantibody, and high IFNα versus low INFα levels as predictor variables, and high TNFα versus low TNFα levels as the outcome variable. The ACR criterion for antinuclear antibodies was not included in the regression because it was almost uniformly positive in SLE patients, and the criterion for immunologic disease was not included because the individual components anti-dsDNA and anti-Sm were already part of the model. Results from the multivariate analysis that were significant at P < 0.0029 would still be significant after Bonferroni correction for multiple comparisons to account for the number of variables tested.

RESULTS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

High serum TNFα levels in SLE patients versus non–autoimmune disease controls.

When quantitative serum TNFα levels were examined, levels were significantly higher in SLE patients than in non–autoimmune disease controls across all ancestral backgrounds (Figure 1A). There were no significant differences between SLE patients of different ancestral backgrounds (P > 0.09 for each comparison). The non–autoimmune disease controls were well matched by age, sex, and ancestral background to the SLE patients, and TNFα levels were not significantly different between controls of different ancestral backgrounds.

thumbnail image

Figure 1. Serum tumor necrosis factor α (TNFα) levels in systemic lupus erythematosus (SLE) patients and healthy controls. A, Quantitative TNFα levels in SLE patients stratified by ancestral background. Data are shown as box plots. Each box represents the interquartile range. Lines inside the boxes represent the median, and error bars represent the 10th and 90th percentiles. P values were calculated by Mann-Whitney U test. B, Proportion of SLE patients with high levels versus low levels of TNFα and interferon-α (IFNα) across the different ancestral backgrounds. P values were calculated by chi-square test. ∗ = P < 0.05; ∗∗ = P < 0.005; ∗∗∗ = P < 0.0005. AA = African American; EA = European American; HA = Hispanic American.

Download figure to PowerPoint

Observation of different combinations of high versus low levels of TNFα and IFNα in patients with different ancestral backgrounds.

As shown in Figure 1B, we divided each ancestral background into 4 groups, stratifying by high levels versus low levels of both TNFα and IFNα. The proportion of patients with concurrently high TNFα and IFNα levels was highest among African Americans and lowest among European Americans (P < 0.005). Conversely, the proportion of patients with concurrently low TNFα and IFNα levels was highest among European Americans and lowest among African Americans (P < 0.0005).

Association of high TNFα levels with high IFNα levels across all ancestral backgrounds, but not with clinical or serologic features.

When comparing clinical characteristics of the patients across the 3 backgrounds, we confirmed many of the previously demonstrated differences between ancestral backgrounds (Table 1). Similar to the results of the quantitative analysis, the categorical analysis revealed no statistically significant difference in the proportion of patients with high TNFα levels across the 3 ancestral backgrounds. We determined the prevalence of the various clinical characteristics and autoantibodies when stratified by high versus low TNFα levels in each background (data available upon request from the corresponding author). Among African Americans, patients with low TNFα levels were more likely to have photosensitivity than those with high TNFα levels. The only consistent trend across ancestral backgrounds was that patients with high TNFα levels were more likely to have high IFNα levels. The odds ratio (OR) for high TNFα levels as a predictor of high IFNα levels was strikingly similar in each ancestral background (OR range 1.76–1.86, P = 4.8 × 10−3 by Fisher's combined probability test).

Table 1. Prevalence of various clinical characteristics, autoantibodies, TNFα, and IFNα in SLE patients stratified by ancestral background*
CharacteristicEuropean AmericansHispanic AmericansAfrican Americansχ2P
  • *

    Differences in proportions across the 3 ancestral backgrounds were determined by chi-square test (χ2[2] = 0.00), and differences in age were assessed by one-way analysis of variance. Values are the percentage, except for age, which is shown as the mean ± SD years at the time of sampling. TNFα = tumor necrosis factor α; IFNα = interferon-α; SLE = systemic lupus erythematosus; ANA = antinuclear antibody; anti-dsDNA = anti–double-stranded DNA.

Female sex86.688.792.13.770.15
Age44.0 ± 14.639.3 ± 13.841.8 ± 13.00.0039
Photosensitivity50.746.131.818.599.2 × 10−5
Discoid rash9.4014.926.727.461.1 × 10−6
Malar rash59.449.637.424.165.7 × 10−6
Oral ulcers33.930.524.35.480.065
Serositis42.640.433.64.320.12
Hematologic disorders56.758.967.36.080.048
Neurologic disorders12.19.915.42.510.29
Renal disorders35.246.842.15.920.052
Arthritis82.961.774.323.478.0 × 10−6
Immunologic disorders80.980.184.11.20.55
ANA98.796.599.55.470.065
Anti-dsDNA35.629.831.31.830.40
Anti-Sm3.364.2611.214.477.2 × 10−4
Anti-RNP10.417.038.862.023.4 × 10−14
Anti-Ro22.125.526.21.270.53
Anti-La6.049.226.071.760.41
High TNFα29.236.236.43.740.15
High IFNα32.641.851.418.431.0 × 10−4

Multivariate analysis of associations between TNFα, IFNα, autoantibodies, clinical features, and demographic features in SLE patients.

We performed a multivariate analysis to detect independent associations with high serum TNFα levels in SLE patients, including all of the available variables for each patient, such as ancestry, age, clinical features, autoantibodies present versus absent, and high versus low levels of serum IFNα. As shown in Table 2, only high IFNα levels were significantly associated with high TNFα levels after statistical correction for multiple comparisons (P < 0.0029). To explore clinical associations further, we examined the prevalence of clinical manifestations in patient subgroups representing each of the 4 possible combinations of high versus low TNFα and IFNα shown in Figure 1B, but no significant relationships were observed that were related to the TNFα levels (data not shown).

Table 2. Results of multivariate regression to detect independent associations with high tumor necrosis factor α (TNFα) levels across all ancestral backgrounds*
Characteristicβ coefficientStandard errorP
  • *

    Each of the patient characteristics was used as a predictor variable in the logistic model, with high versus low TNFα levels used as the outcome variable. Anti-dsDNA = anti–double-stranded DNA; IFNα = interferon-α.

Non-European ancestry−0.2700.1910.15
Age at recruitment−0.0030.0070.71
Photosensitivity−0.2710.1860.14
Discoid rash−0.3700.2480.14
Malar rash0.3680.1860.048
Oral ulcers0.0900.1970.65
Serositis−0.1670.1840.36
Hematologic disorders−0.0890.1830.63
Neurologic disorders0.3880.2540.13
Renal disorders0.0130.1860.94
Arthritis−0.1820.2100.39
Anti-dsDNA−0.4810.1960.014
Anti-Sm−0.3710.4120.37
Anti-RNP0.0570.2470.82
Anti-Ro0.2090.2340.37
Anti-La0.3460.3820.37
IFNα0.6110.1890.0012

Complement CH50 association with cytokine parameters.

Given that immune complexes may be related to TNFα and IFNα levels, we looked for a correlation between complement CH50 levels and TNFα and IFNα. Data on CH50 were available for most patients, and in European Americans, lower CH50 levels were associated with increased TNFα and IFNα levels (data available upon request from the corresponding author). In African Americans and Hispanic Americans, CH50 levels were inversely correlated with IFNα levels, but not with TNFα levels. The lack of relationship with TNFα in patients with these backgrounds could be due to smaller sample size, although no strong trends were observed, and this may represent a biologic difference.

DISCUSSION

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

Our study confirms that elevated serum TNFα levels are a frequent finding in human SLE and are slightly less prevalent than high IFNα levels in SLE, using the same criteria (12). Interestingly, we found a positive correlation between TNFα and IFNα levels but no correlation between the TNFα level and any clinical or serologic features of SLE. This suggests that TNFα may cooperate with IFNα or other cytokine or environmental factors not measured in this study to affect clinical features of the disease. Some previous studies have documented a relationship between high TNFα levels and anti-dsDNA antibodies (4), but we did not replicate this finding in our study. It is possible that the previously described association between high TNFα levels and anti-dsDNA antibodies may have been secondary to high IFNα levels, as high IFNα levels are strongly associated with both of these characteristics. In the current study, disease activity data for the time of sampling were not available. The correlation between CH50 and TNFα in patients with European American ancestry could indicate a relationship between TNFα and disease activity, which could warrant further large-scale longitudinal studies.

While in vitro studies suggest cross-regulation between TNFα and IFNα (5), not all studies of autoimmune disease populations fit this model. For example, up-regulation of both type I IFN and TNFα has been shown in synovial biopsy samples from rheumatoid arthritis patients (13). In juvenile dermatomyositis, the promoter polymorphism in the TNFα gene that has been linked to higher TNFα expression was associated with increased serum IFNα levels (14). Thus, it is likely that cross-regulation of TNFα and IFNα in humans in vivo will be complex. In our cross-sectional study of SLE sera, we did not find evidence of cross-regulation; we instead observed that levels of the 2 cytokines were positively correlated, even when controlling for clinical and demographic variables. Mechanistically, this could relate to immune complexes triggering Toll-like receptors, which could result in production of inflammatory cytokines such as TNFα and IFNα (15). Correlations observed between CH50, a proxy for immune complexes, and the 2 cytokines may relate to this phenomenon. We have not tested IFNα levels in SLE patients receiving anti-TNFα drugs, so we do not know if IFNα levels would increase. This could be of interest, since some of the hesitation to initiate anti-TNFα treatment in SLE stems from the idea that blocking TNFα may result in a compensatory increase in IFNα, which could worsen SLE disease activity.

It is not clear whether high TNFα levels predispose patients to the development of SLE, or if levels rise after disease is established. One previous study identified a difference in simultaneous TNFα and IFNα levels related to the PTPN22 genotype (16), although the variant classically associated with SLE risk was associated with lower TNFα levels. Additionally, the promoter polymorphism in the TNFα gene that has been linked to higher TNFα expression has been associated with SLE susceptibility, although the TNFα locus is in the HLA region, which is characterized by multiple association signals that are difficult to resolve due to high linkage disequilibrium in the region. Given the frequent elevation of TNFα levels in SLE sera, it seems likely that TNFα plays a role in the disease process. Further understanding of this role will increase our knowledge of disease pathogenesis and may allow for the definition of a subset of patients who could benefit from TNFα blockade, which has proven to be safe and effective in multiple other inflammatory diseases.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES

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. Niewold 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. Weckerle, Harley, Niewold.

Acquisition of data. Weckerle, Mangale, Franek, Kelly, Kumabe, James, Moser, Harley, Niewold.

Analysis and interpretation of data. Weckerle, James, Niewold.

REFERENCES

  1. Top of page
  2. Abstract
  3. PATIENTS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. AUTHOR CONTRIBUTIONS
  7. REFERENCES