Background/Purpose: Recent studies on large cohorts of patients with systemic sclerosis (SSc) have shown that lowering the upper threshold of the modified Rodnan skin score (mRSS) as a study inclusion criterion leads to cohort enrichment with patients with progressive skin disease. Limitations of these studies were lack of racial diversity and low proportion of patients with anti-RNA-Polymerase III (Pol3) antibodies. As the Texas-based GENISOS cohort is an ethnically diverse cohort, including a large proportion of Pol3-positive patients, this study aimed to assess the effect of different mRSS cut-off values at baseline on progression of skin fibrosis after one year of follow-up.

Methods: We extracted data from GENISOS for patients fulfilling the 1980 ACR criteria for SSc and the Le Roy criteria for diffuse cutaneous SSc, who had a minimum mRSS of 7 at inclusion and a follow-up visit with documented mRSS at 12±2 months. Progressors were defined as having an increase in mRSS >5 points and ≥25% from the baseline to 2nd visit, while regressors were defined as having a decrease in mRSS of >5 points and ≥25%. To identify the optimal cut-off of baseline mRSS that yields the highest sensitivity for progressive skin fibrosis, we developed ROC curves and logistic regression models with “progression” as outcome variable and a binary variable of baseline mRSS cut-off point as predictor.

Results: 152 patients (age and disease duration [median, Q1-Q3, years] 49.5, 40.2-57.3 and 2.2, 1.1-3.3 respectively, 22.4% males) matched the inclusion criteria. The proportion of patients of African American ethnicity was 32/152 and 50/152 patients were Pol3-positive.

After one year of follow-up, 17 patients (11.2%) classified as progressors and 51 (33.6%) as regressors. Progressors were more frequently positive for anti-topoisomerase antibodies (37.5% vs. 15.3%, p=0.028), negative for anti-Pol3 antibodies (93.8% vs. 62.3 %, p=0.012), had a shorter disease duration (median, Q1-Q3: 1.3, 0.5-2.2 vs. 2.4, 1.1-3.5 years, p=0.005) and lower mRSS (median, Q1-Q3: 21, 11-25 vs. 24, 16-31, p=0.012) than the remainder of the patients.

Sixteen of 17 progressors, but only 33 of 51 regressors had a baseline mRSS≤27. The mRSS cut-off of ≤27 had the highest probability of progression (odds ratio 9.1, 95% confidence interval 1.2-70.9, p=0.035, area under the curve 0.652). Using this cut-off as an inclusion criterion in a clinical trial (vs. no cut-off) would have included 94% of all progressors, but only 65% of all regressors and 67% of all patients. The figure displays absolute numbers of progressors and regressors at 1 year for each mRSS cut-off.

Conclusion: This analysis reconfirmed, in a population rich in patients of African American origin and with high prevalence of Pol3 antibodies, that setting a lower upper threshold of mRSS at study inclusion increases the proportion of progressors and reduces the absolute number of regressors.