To the Editor:
B cell depletion therapy using rituximab (RTX) has emerged as a promising therapy in systemic sclerosis (SSc). Our group reported a clinical benefit on skin thickening when RTX (2 × 1000 mg) in combination with methylprednisolone (100 mg) was administered at months 0 and 61. The rationale behind the use of B cell depletion in SSc treatment is based on the growing evidence that B cells play a role in the pathogenesis of the disease2,3. In addition, SSc-associated antinuclear antibody (ANA) are present in the majority of patients, but are generally accepted to be rather static markers, primarily important in diagnosis rather than in followup4. Moreover, conflicting results have been published5 on the change in disease-specific ANA titers in relation to clinical response in other connective diseases (e.g., systemic lupus erythematosus and rheumatoid arthritis) after RTX. To our knowledge, no studies in SSc are available documenting the longterm effect of RTX on these autoantibody levels.
In this letter we report the 2-year serologic followup data (months 0, 3, 6, 12, 15, 18, and 24) on the same 8 patients with SSc with diffuse skin involvement (dcSSc) who were included in our initial RTX studies1,6. The idea was to evaluate the relationship between SSc-ANA level changes and clinical response. To document this latter relationship, we performed specific SSc-ANA analysis [anticentromere B, antitopoisomerase I, and anti-RNA polymerase III immunoglobulin G (IgG)] on all serial serum samples (months 0, 3, 6, 12, 15, 18, and 24) of each patient using fluorescent enzyme immunoassay on the Immunocap 250 system (EliA, Thermofisher Scientific)7. On the same serum samples, Epstein-Barr virus (EBV) IgG and total IgG levels were also determined using ELISA (Enzygnost, Siemens Healthcare Diagnostics Products GmbH) and nephelometry on the BNII analyzer (N Antisera to human immunoglobulins IgG, Siemens Healthcare Diagnostics Products GmbH), respectively. For the specific ANA analysis, all timepoint samples of 1 patient were performed in duplicate in the same run to avoid inter-run variability5.
Seven patients had SSc-ANA [3 patients had antitopoisomerase I, 3 patients had anti-RNA polymerase III, and 1 patient (number 6) showed double reactivity with anticentromere B and antitopoisomerase I]. Mean levels of topoisomerase I antibodies and RNA-polymerase III antibodies were 717 (SD 1179) and 134 (SD 107) EliA units/ml, respectively. Of particular interest for our report, we previously documented for the same patient group a clinically and statistically significant change in skin score [from a mean modified Rodnan skin score (mRSS) of 24.8, SD 3.4 at baseline to 13.6, SD 5.6 at Month 24]1. In our current study, we observed a disproportional effect of RTX on the different types of (auto)antibodies, with the SSc-ANA levels clearly more affected than the levels of the protective antibodies (see Figure 1 for the evolution of the antibody titers). There was a statistically significant overall decrease of the SSc-ANA levels over the 2-year followup period (mixed model analysis p = 0.001; see Appendix 1 for details on the statistical analysis), with a mean decrease of 50% (95% CI 12–72) at Month 24 compared with baseline. A significant decrease compared with baseline was also shown for each timepoint in between starting from Month 12 onward (Table 1). Comparable results were found when the SSc-ANA levels were corrected for total IgG [ratio; overall decrease, mixed model analysis p = 0.008; mean percentage decrease at Month 24 compared with baseline 48% (95% CI 9–72), mixed model analysis p = 0.003]. In contrast, no statistically significant evolution of total IgG and EBV IgG could be shown at most timepoints compared to baseline (Table 1).
Because both variables showed a significant decrease after a 2-year period following RTX therapy (see above), the association between the SSc-ANA levels and the mRSS over time was also evaluated. Appendix 2 shows the evolution of the mRSS, the SSc-ANA titers, and the total IgG titers for the individual patients over time as expressed by the percentage decrease of baseline values. The percentage decrease in mRSS and SSc-ANA levels had similar kinetics in the individual patients, and the extent of mean percentage decrease in mRSS (45% at Month 24 vs baseline) and autoantibody level (50% at Month 24 vs baseline) were comparable. Moreover, there was a significant association between the percentage decrease of the mRSS and the percentage decrease of the SSc-ANA level at a given timepoint (mixed model analysis p = 0.007). Contrarily, no significant association between the total IgG and the mRSS could be shown (mixed model analysis p = 0.063).
We observed a selective decline in SSc-ANA levels in dcSSc, suggesting that the B cell clones responsible for the production of these antibodies are particularly more sensitive for B cell depletion therapy. Moreover, the effect of RTX on the SSc-associated humoral immunity was associated with skin improvement. Further, large and longterm prospective studies will clarify the potential clinical added value of monitoring of the SSc-ANA levels after RTX treatment.
APPENDIX 1.
Statistical analysis
Mixed model analysis with random intercept for patient was used to evaluate the serologic changes over time and the association between percentage decrease in systemic sclerosis (SSc)-associated antinuclear antibody (ANA) levels and clinical activity (as expressed by the modified Rodnan skin score). When appropriate, heterogeneous variances over time were allowed. In case of double SSc-associated ANA reactivity, only 1 antibody per patient was included in the analysis (randomly selected). Mixed model analysis on the serologic changes over time was applied on lntransformed values. All analyses were performed using SPSS 21 (SPSS Inc.). A statistical significance level of 0.05 was used. In case of multiple testing, the Bonferroni correction was applied.
Footnotes
Supported by a research grant from the Fund for Scientific Research, Flanders, to CB and VS. ClinicalTrials.gov number NCT00379431.
REFERENCES
- 1.
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.