Juvenile localized scleroderma (jLS) is the most common form of childhood scleroderma. Because of its chronicity and association with extracutaneous involvement, children are at risk for major morbidity including hemiatrophy, arthritis, and seizures. A sensitive clinical disease activity assessment tool is essential to optimize treatment of inflammation and thereby reduce the risk of fibrosis and associated damage. Several clinical tools have been developed; all include skin thickening and at least one other feature. No study has been done to evaluate the relative importance of scored features. Our Localized scleroderma Clinical and Ultrasound Study group (LOCUS), based in the Childhood Arthritis and Rheumatology Research Alliance (CARRA), conducted a longitudinal study to evaluate the specificity and relative importance of lesion features for disease activity.

Methods:

We conducted a multicenter prospective observational cohort study of jLS patients with either active or inactive disease (minimum ratio 2:1). Using a standardized evaluation form, we scored 1 specified study lesion per subject for defined features, at 3 visits over 6 months. Physicians scored global assessments for this lesion’s activity and damage (PGA-A, PGA-D, respectively). Wilcoxon test, 2-sample t-test, correlation and regression analyses were used to evaluate assessed variables and their correlation with PGAs.

Results:

Of the 103 subjects enrolled, 66 were classified as active, 24 as inactive, and 13 were excluded from analysis because of incomplete data. Active and inactive patients had similar age of onset, gender, and lesion subtypes. The most common subtype was linear scleroderma (62% in active, 58% in inactive). Erythema, violaceous color, tactile warmth, new lesions, enlargement of lesion size, and abnormal skin texture were found to occur predominantly in active lesions. While skin thickening correlated with lesion PGA-A within the active group, neither presence nor level of skin thickening (at both lesion edge and center) differentiated active from inactive lesions. No single variable was found to track tightly with PGA-A in all lesions, indicating that multiple lesion features need to be evaluated when assessing disease activity. Multiple regression analysis assigned highest weights to new or enlarged lesion and erythema.

Conclusion:

We identified several variables strongly associated with disease activity, and their relative contribution to physician scoring. Use of skin thickening as an activity variable is limited by its lack of specificity. These results aid development of a sensitive activity tool for comparative effectiveness treatment studies.