Abstract
Objective. To assess the prevalence of disease-specific autoantibodies in patients with limited cutaneous systemic sclerosis (lcSSc).
Methods. Sera from 180 patients with lcSSc were analyzed for antinuclear antibody (ANA). Clinical characteristics were compared in the presence or absence of specific autoantibodies.
Results. SSc-specific antibodies were detected in 135 patients (75%). Associations were found between anticentromere antibody and age at lcSSc diagnosis, telangiectasia, reduced creatinine clearance, and selective reduction in DLCO, and between antitopoisomerase-I and pulmonary fibrosis.
Conclusion. The majority of patients with lcSSc belong to distinctive serologic subsets, potentially with prognostic significance.
Systemic sclerosis (SSc) is characterized by disease-specific antinuclear antibodies (ANA) that help identify disease subsets1. We studied whether previously reported associations with hallmark ANA held true in a more tightly defined cohort of patients with limited cutaneous SSc (lcSSc).
MATERIALS AND METHODS
A total of 180 patients fulfilling the criteria for lcSSc2 were recruited from 20 UK centers for the QUINS trial (ISRCTN 57984704)3. The study was approved by the Multicentre Research Ethics Committee (MREC Wales 00/09/19); all patients provided written informed consent.
The patients were recruited between March 2001 and February 2004 and followed prospectively for up to 3 years. Clinical and serological data were obtained as in Gliddon, et al3.
Serum samples taken at the baseline QUINS Trial visit were sent to a central laboratory for analysis of ANA and antinucleolar antibodies (ANoA). They were initially screened using indirect immunofluorescence (IIF) on Hep-2 cells using a serum dilution of 1/40 and polyvalent FITC-conjugated secondary antibody. IIF patterns were interpreted by 2 independent, experienced observers. Positive samples were then further investigated for their autoantibody specificity. Ouchterlony immunodiffusion was performed using a commercial antigen source (Bio-Diagnostics, Upton-upon-Severn, UK) and rabbit thymus extract (Pel-Freez; Rogers, AR, USA). Commercial ELISA (The Binding Site, Birmingham, UK) were used for confirmatory purposes for anti-U1-RNP, anti-Sm, anti-Ro, anti-La, and antitopoisomerase I (anti-topo I). Protein or RNA immunoprecipitation of K562 cells was used to detect anti-RNA polymerase antibodies and other ANoA such as anti-PM-Scl, anti-U3-RNP, and anti-Th/To, as described in Betteridge, et al4.
Clinical characteristics were compared in the presence or absence of the 4 most common ANA. Fisher’s exact test was used for categorical variables, t test or Mann-Whitney U test for continuous variables, and multiple regression to correct creatinine clearance for age.
RESULTS
All 180 patients [mean age 55 yrs (range 21–82), 154 women] were white except for 3 Asians, 2 African Caribbeans, and 2 people of mixed race. The median duration of Raynaud’s phenomenon (RP) was 10.7 years (range < 1 to 65 yrs) and median disease duration from fulfilling SSc criteria was 4 years (range < 1 to 46 yrs). The median number of organ systems affected by lcSSc was 3 (range 1 to 7). The median skin score was 5 (range 0 to 23).
ANA were detected by IIF in 96% of subjects. SSc-specific antibodies were identified in 75%, with the most frequent being anticentromere antibody (ACA; Table 1). Mostly, these antibodies were mutually exclusive (1 patient with ACA had anti-topo I and 1 patient with ACA had anti-PM-Scl). A further 10% had anti-U1-RNP or anti-Ro and/or anti-La, antibody specificities recognized as markers of connective tissue diseases but with less specificity for SSc. In 12%, the autoantibody specificity could not be identified.
Clinical associations are summarized in Tables 2 and 3. Patients who were ACA-positive were significantly older at diagnosis, while anti-U1-RNP-positive patients were significantly younger. Age of onset and frequency and severity of RP were not significantly different in any of the serological groups (data not shown). Skin score was higher in anti-topo I-positive patients and less in ACA-positive patients. A direct comparison (Mann-Whitney U test) of skin score in ACA and anti-topo I patients showed a significant difference (p = 0.017). ACA positivity was associated with telangiectasia (OR 2.4, 95% CI 1.3–4.3) but the median number of organ systems affected, calcinosis, and levels of hemoglobin, creatinine, and proteinuria showed no association with ANA subtype. However, creatinine clearance was significantly reduced in patients with ACA. The proportion of ACA-positive patients with an isolated reduction of DLCO, assessed by ratio of forced vital capacity (FVC)/DLCO of > 1.4 (OR 1.9, 95% CI 1.05–3.6) was significantly higher than in non-ACA patients. The percentage of ACA-positive patients with pulmonary artery pressure > 30 mm Hg by transthoracic echocardiography was higher (but not significantly so) than that of ACA-negative patients. Of the 13 patients who had high estimated pulmonary artery pressure, 8 had ACA, 4 had unidentifiable ANA, and 1 had anti-U3-RNP antibody. Anti-topo I was associated with lung fibrosis (OR 6.0, 95% CI 1.9–18.7).
DISCUSSION
In this predominantly white population of patients with lcSSc, drawn from multiple centers across England and Wales, ACA was the predominant autoantibody specificity, occurring in a proportion of patients similar to previous reports5. Therefore, serologically, this was a heterogeneous population in which mostly there was mutual exclusiveness of SSc-specific autoantibodies.
Autoantibodies considered more commonly associated with diffuse cutaneous SSc, such as anti-topo I, anti-RNA polymerase antibodies, and anti-U3-RNP, were found in 14%. However, the relationship between these autoantibodies and the extent and rapidity of skin involvement is not absolute, and at least 1 other study has reported that over 30% of patients positive for anti-topo I had lcSSc6.
Similar to other reports7, patients who were ACA-positive were older at lcSSc diagnosis and, by inference, demonstrated a longer interval between the onset of RP and the development of other SSc features. There was a significant association with telangiectasia, as described by others1,8, and reduced creatinine clearance. This result is similar to studies using 51CrEDTA clearance as a measure of renal function9. Patients who were ACA-positive also demonstrated selective reduction of DLCO, indicated by an FVC/DLCO ratio > 1.410. This ratio may predict isolated pulmonary hypertension, especially if it exceeds 1.811. In our patients who were ACA-positive, 9% had an estimated pulmonary artery pressure > 30 mm Hg that was not significantly greater than in the other patients with lcSSc. However, in terms of studying predictors for disease progression, the length of followup was not very long.
Although all patients in our study satisfied criteria for lcSSc2, the skin score in patients with anti-topo I was significantly higher than in those with ACA. Anti-topo I-positive patients had a significantly higher risk of developing pulmonary fibrosis. This supports an observation7 that the association between anti-topo I and pulmonary involvement occurs irrespective of the extent of skin involvement and consequently, disease classification. The presence of anti-To/Th has also been reported to identify patients with lcSSc at risk of pulmonary fibrosis12 and with a worse prognosis13. However, clinical correlations were not found in this study and only 1 of 14 patients with anti-To/Th RNP had pulmonary fibrosis.
Our study shows that most patients with lcSSc fall into distinct serologic subsets. Since these autoantibodies are a very early feature, persisting throughout the course of the disease14, ANA identification is important for predicting SSc in patients presenting with, for example, RP. Additionally, individual antibodies have been associated with a particular constellation of clinical features and potentially have prognostic significance.
Acknowledgment
We acknowledge the staff at QUINS Trial participating hospitals.
APPENDIX
List of study collaborators
The QUINS Trial Study Group, in alphabetical order: T. Barnes, University Hospital Aintree; C. Black, Royal Free Hospital, London; A. Borg, Nevill Hall Hospital, Gwent; J. Camilleri, University Hospital, Cardiff; K. Chakravarty, Harold Wood Hospital, Romford; C. Denton, Royal Free Hospital, London; P. Emery, Chapel Allerton Hospital, Leeds; B. Griffiths, Freeman Hospital, Newcastle; A. Herrick, Hope Hospital, Salford; N. Hopkinson, Christchurch Hospital, Christchurch; P. Hickling, Derriford Hospital, Plymouth; P. Lanyon, Queen’s Medical Centre, Nottingham; C. Laversuch, Taunton and Somerset Hospital, Taunton; T. Lawson, Princess of Wales Hospital, Bridgend; R. Mallya, Halton General Hospital, Runcorn; R. Moots, University Hospital, Aintree; M. Nisar, Queen’s Hospital, Burton-upon-Trent; C. Rhys-Dillon, Royal Glamorgan Hospital, Llantrisant; T. Sheeran, Cannock Chase Hospital, Cannock.
Footnotes
-
Funded by Arthritis Research UK.
- Accepted for publication November 26, 2010.