OBJECTIVE To determine whether African-American race is independently associated with lung disease in scleroderma. METHODS Retrospective review. METHODS University medical center in Baltimore. METHODS One hundred one patients with diffuse cutaneous scleroderma with available serum samples. METHODS Patients underwent lung function testing as part of their routine clinical care. Percent predicted values adjusted for race were calculated for FVC, single-breath carbon monoxide diffusing capacity (Dco), and FEV1. Serum samples were assayed for the presence of antibodies to topoisomerase I and RNA polymerase II. RESULTS Scleroderma patients of African-American race had lower percent predicted values than white patients for FVC (p<0.002), Dco (p<0.0001), and FEV1 (p<0.0001). Antibodies to topoisomerase I but not antibodies to RNA polymerase II were also associated with lung function. African-American scleroderma patients were distinct from white patients in having younger age of onset and higher prevalence of antibodies to topoisomerase I. In multivariate analyses accounting for sex, age, smoking history, years of scleroderma symptoms, and RNA polymerase II antibody status, African-American race and topoisomerase I antibody status independently predicted lower lung function. CONCLUSIONS African-American race and antibodies to topoisomerase I are independent risk factors for scleroderma lung disease.