Systemic lupus erythematosus (SLE) is a clinically diverse, complex autoimmune disease which may present with coincident onset of many criteria or slow, gradual symptom accrual. Early intervention has been postulated to delay or prevent the development of more serious sequelae. One option for treatment in this setting is hydroxychloroquine. Using 130 US military personnel who later met ACR SLE criteria, a retrospective study of onset, development and progression of SLE with and without pre-classification hydroxychloroquine (n = 26) use was performed. Patients treated with hydroxychloroquine prior to diagnosis had a longer (Wilcoxon signed rank test, P = 0.018) time between the onset of the first clinical symptom and SLE classification (median: 1.08 versus 0.29 years). Patients treated with prednisone before diagnosis also more slowly satisfied the classification criteria (Wilcoxon signed rank test, P = 0.011). The difference in median times between patients who received NSAIDs before diagnosis, as opposed to those who did not, was not different (P = 0.19). Patients treated with hydroxychloroquine also had a lower rate of autoantibody accumulation and a decreased number of autoantibody specificities at and after diagnosis. These findings are consistent with early hydroxychloroquine use being associated with delayed SLE onset. A prospective, blinded trial testing the capacity of hydroxychloroquine to delay or prevent SLE in high risk populations is warranted.

Many aspects of the immune maturation are uncharted. For ordinary human autoimmune systems there are no complete descriptions of the progression from an initial antigenic epitope to a maximally complex immune response. In this study we have exploited a large serial collection of human sera to investigate the development of the anti-Sm autoimmune response in systemic lupus erythematosus (SLE). The results suggest a similar, if not virtually identical, stepwise progression in the early humoral immune maturation of anti-Sm. The amino acid sequence PPPGMRPP comprises the first epitope in the anti-Sm B/B'response and its close relative, PPPGMRGP, the second. Epitopes are subsequently enlarged by the incorporation of neighbouring amino acids. The third and fourth epitopes are also recognised by an antibody in a nearly identical sequence in different lupus patients. A column absorption with PPPGMRPP demonstrates that the epitope spreading among the first four early epitopes appears to occur by the sequential generation of cross-reactive antibodies. Unexpectedly, epitope spreading in this system occurs in a predictable fashion by involving essentially the same sequence of antigenic structures from person to person. In addition, these data support the lupus anti-Sm antibodies originating against a single antigenic structure and, hence, strongly support a unifying mechanism in the generation of these autoantibodies.