RCMI International Symposium on Health Disparities
November 12-16, 2000
San Juan, Puerto Rico

Invited Speakers

HOW DOES ETHNICITY AFFECT
THE PREDISPOSITION AND PROGNOSIS
OF SYSTEMIC LUPUS ERYTHEMATOSUS (SLE)
AND SYSTEMIC SCLEROSIS (SSc)?



John D. Reveille, M.D.
The University of Texas-Houston Health Science Center
Houston, TX

Epidemiologic studies have suggested that both SLE and SSc occur significantly more frequently in African-Americans than in Caucasians. Such data in Hispanics, however, are lacking. Moreover, studies of prognosis in SLE and SSc have suggested that non-Caucasians fare worse in these diseases. Whether this be due to sociodemographic, behavioral, genetic or other factors associated with ethnicity is not clear. We are currently engaged in two studies of outcome in Hispanics (Mexican or Central Americans), African-Americans and Caucasians, designated LUMINA (for LUpus in MInorities, NAture versus nurture) and GENISOS (for Genetics vs. ENvironment In Scleroderma Outcome Study). Both studies enrolled patients of defined ethnicity (four grandparents from the same ethnic group) of disease duration of less than five years, and meeting American College of Rheumatology criteria for the respective diseases. Patients are followed at regular intervals (up to twice a year). LUMINA began enrollment in April, 1994 and GENISOS in February, 1998.

In both studies Caucasians had significantly higher income, education level and a higher likelihood of having private insurance. Caucasians were also more likely to smoke and consume alcoholic beverages. Non-Caucasians were more likely to exhibit abnormal illness behaviors (as indicated by the Illness Behavior Questionnaire of Pilowsky and Spence) as well as to have poorer social support (as measures by the Interpersonal Support Evaluation List (ISEL).

In the LUMINA study we have enrolled 285 patients, including 85 Hispanics, 90 Caucasians and 110 African Americans from Alabama and Texas. Data from disease onset and from study entry demonstrate that non-Caucasians have a higher frequency of renal and cardiovascular involvement and greater disease activity (as measured by the SLAM) even after correcting for sociodemographic, behavioral and genetic (HLA-class II, CR1 size polymorphisms, C4 null alleles). Differences were also seen for specific autoantibody frequencies (anti-dsDNA and U1-RNP) and in the distribution of HLA class II alleles. Followup studies have suggested a greater mortality and accumulation of new SLE-related damage (as measured by the SLICC) in non-Caucasians, especially in Hispanics. Of particular note was the finding on multivariable analysis of ethnic-specific factors independently predicting mortality and damage accrual, suggesting that findings in one ethnic group may not be generalizable to another.

In the GENISOS study 170 patients have been enrolled, including 54 Hispanics, 28 African-Americans, 80 Caucasians and eight from other ethnic groups from three medical centers in Texas. Baseline studies show a tendency, albeit nonsignificant, toward greater skin and pulmonary involvement in non-Caucasians, as well as striking differences in the frequencies or specific autoantibodies (anti-centromere, anti-U1-RNP, anti-fibrillarin) between the three major ethnic groups. On the other hand, similar HLA class II allele associations with SSc as well with autoantibody subsets thereof were seen in the three ethnic groups. However, different ethnic associations were seen for non-HLA-genes (i.e. fibrillin, osteonectin). As followup data on this cohort is still rather limited, the impact of these factors on outcome remains to be determined.

These data suggest that non-Caucasians tend to have more aggressive disease and a worse prognosis, even after correcting for the less favorable sociodemographic profiles seen in these ethnic groups. Moreover, at least in SLE, the finding of ethnic-specific factors affecting outcome suggests a need for focusing different interventions in different ethnic groups in order to improve the prognosis of these rheumatic diseases.