Autoantibody and Clinical Profiles in Patients With Discoid Lupus and Borderline Systemic Lupus FREE

Discoid lupus erythematosus (DLE) is 1 of 11 American College of Rheumatology (ACR) systemic lupus erythematosus (SLE) diagnostic criteria,¹ of which 4 are required to fulfill the diagnosis of SLE. By meeting mainly skin-related ACR criteria, patients with DLE can be easily diagnosed as having SLE without having other organ involvement. These patients are defined as having borderline DLE/SLE. To determine whether SLE diagnosis is appropriate in these patients, we compared their autoantibody and clinical profiles with those of DLE patients who do not have SLE (DLE only) and those who do have SLE (DLE/SLE).

A pilot cross-sectional study was conducted comparing age- and sex-matched borderline DLE/SLE, DLE-only, and DLE/SLE patients selected from the Cutaneous Lupus Erythematosus Registry at The University of Texas Southwestern (UTSW) Medical Center in Dallas. For all patients, the diagnosis of DLE, which represented the discoid subtype of chronic cutaneous lupus, was confirmed by clinicopathologic correlation. Borderline DLE/SLE patients had either 3 or 4 skin-related ACR SLE diagnostic criteria (including DLE, self-reported malar eruption, oral ulcers, and photosensitivity) and positive findings of antinuclear antibody (ANA) to fulfill the SLE diagnosis. The DLE-only patients had fewer than 4 ACR SLE diagnostic criteria. The DLE/SLE patients met at least 4 ACR SLE diagnostic criteria with at least 1 non–skin-related, non–ANA-related criterion. Patients with absence of DLE, history of drug-related DLE, or DLE diagnosis before age 18 years were excluded. The study was approved by the UTSW institutional review board.

Autoantibodies were quantified using (1) an enzyme-linked immunosorbent assay (ELISA) kit manufactured by INOVA Diagnostics for ANA; (2) an ELISA kit manufactured by ORENTEC Diagnostika for double-stranded DNA (dsDNA) and single-stranded DNA (ssDNA) antibodies; and (3) a QUANTA Plex fluorescent immunoassay system manufactured by INOVA Diagnostics for ribonucleoprotein (RNP), Smith (Sm), Scl-70, 60-kDa SS-A, 52-kDa SS-A, and SS-B antibodies. Treatments, disease activity,^2- 3 lesion distribution, medical and family histories, and laboratory values were also collected. Unsupervised cluster analyses were performed using Cluster software, version 2.1, and Treeview software, version 1.6 (http://rana.lbl.gov/EisenSoftware.htm). Continuous and categorical predictive variables among groups were compared using Kruskal-Wallis and Fisher exact tests, respectively.

Table 1 summarizes the patient characteristics. All borderline DLE/SLE patients had 3 skin-related ACR SLE diagnostic criteria with positive ANA findings except 1 patient, who had 4 skin-related ACR SLE diagnostic criteria with positive ANA findings.

Unsupervised clustering analysis of 9 autoantibody levels revealed 2 distinct clusters of patients (Figure 1). One cluster was composed mostly of borderline DLE/SLE patients (6 of 8) and DLE-only patients (11 of 12). Another cluster was dominated by DLE/SLE patients (8 of 12). Borderline DLE/SLE and DLE-only patients had similarly low levels of Sm (P < .001), RNP (P = .01), and ssDNA autoantibodies (not shown) (P = .08) compared with DLE/SLE patients (Figure 2A and B). Compared with the other 2 groups, borderline DLE/SLE patients had moderate dsDNA autoantibody (P = .04) and ANA levels (P = .01) (Figure 2C and D). No significant differences in 60-kDa SS-A, SS-B, Scl-70, and 52-kDa SS-A autoantibodies were seen (data not shown).

Localized disease (or lesions above the neck) was more common in borderline DLE/SLE and DLE-only patients than in DLE/SLE patients (P = .03) (Figure 3A). Compared with DLE/SLE patients, borderline DLE/SLE and DLE-only patients had lower numbers of treatment types (P = .05) (Figure 3B). The DLE/SLE patients were most likely to be undergoing oral immunosuppressive therapy (P = .01) (Figure 3C). Disease activity, personal and family history of autoimmunity, and serum urea nitrogen and creatinine levels were not significantly different among the groups (data not shown).

Our pilot study demonstrated that borderline DLE/SLE and DLE-only patients have similar autoantibody, treatment, and lesion distribution profiles, calling into question their SLE diagnosis. Requiring a noncutaneous, non-ANA diagnostic criterion would reduce overdiagnosis of SLE.

Levels of Sm and RNP autoantibodies, which were low in borderline DLE/SLE and DLE-only patients compared with DLE/SLE patients, may be important in distinguishing these groups in future studies. Moderate ANA and dsDNA values in borderline DLE/SLE patients imply that they are somewhat distinct fromDLE-only patients and could be at higher risk for systemic spread of their disease. Comparable treatment history and lesion distribution in borderline DLE/SLE and DLE-only patients suggest similar disease courses. Since generalized DLE is associated with systemic disease,⁴ the low incidence of this clinical finding in borderline DLE/SLE patients favors limited disease.

Limitations of this study include cross-sectional design and small sample size, which would be addressed with a larger prospective study. Self-reporting of malar eruption, oral ulcers, and photosensitivity affects patient classification.

Overall, borderline DLE/SLE patients express autoantibody and clinical features more comparable with DLE-only patients than with DLE/SLE patients, implying that SLE diagnosis may be misleading in borderline DLE/SLE patients. To improve diagnostic methods of SLE, additional studies are needed to define which cutaneous features are most related to SLE.