Biomarkers Predict Thrombosis in Lupus Save

A composite risk score that included three biomarkers predicted thrombotic events among patients with systemic lupus erythematosus (SLE), a cross-sectional study found.

The presence of C4d on platelets, low C3, and lupus anticoagulant yielded a risk score for thrombosis of 1.93 compared with a score of 0.81 when none of these markers were detected (P<0.01), according to Michelle A. Petri, MD, of Johns Hopkins University School of Medicine in Baltimore, and colleagues.

In addition, each of the three markers conferred an odds ratio of 5.2 (95% CI 2.5-10.7, P<0.01) for the occurrence of any thrombosis, the researchers reported online in Lupus Science & Medicine.

"SLE is an immune complex disease linked to classical complement pathway activation, hyperconsumption of C3 and C4 proteins, and production of C4d split fragments covalently bound to a variety of hematopoietic cells, and the pathogenesis of the disease involves interactions between the complement system, coagulation pathways, and platelets," they explained.

Accordingly, patients with SLE have an elevated risk for thrombotic events for reasons including abnormalities of complement, especially low C3, the presence of lupus anticoagulant, and the use of prednisone. In addition, one of the most specific complement markers has been identified as deposition of C4d fragments on platelets, rather than on B lymphocytes and erythrocytes; this has been associated not only with thrombosis but also mortality.

To explore whether combining these markers could predict thrombosis risk, Petri and colleagues enrolled 149 patients with SLE from the Hopkins Lupus Center whose mean age was 48. A total of 16 of the patients had a history of a thrombotic event within the previous five years, with ten patients having had a venous thrombosis and eight patients having arterial events (two patients had both). Composite risk scores were calculated for the cumulative presence of the three biomarker abnormalities.

African-American patients more often had thrombosis (50% vs 32%), and those taking prednisone had three times the risk (OR 3, 95% CI 1.1-9, P=0.029). Other demographic variables and disease activity were not associated with thrombosis.

Individually, low C3 and the presence of C4d on platelets were associated with thrombosis. For venous thrombosis, the odds ratios for low C3 and platelet-bound C4d were 10.5 (95% CI 2.8-39.5) and 19.2 (95% CI 4.2-84.7, P<0.05 for both), while for arterial thrombosis, the odds ratios were 5.4 (95% CI 1.28-23.17) and 4 (95% CI 1-15.8, P<0.05 for both).

Moreover, all patients with venous thrombosis were positive for lupus anticoagulant, but that factor was not associated with arterial thrombosis. Because it can be difficult to interpret tests for lupus anticoagulant among patients undergoing anticoagulation, the researchers also assessed antiphosphatidyl serine/prothrombin (anti-PS/PT) complex antibody as an alternative. On that test, a positive association was seen for arterial but not venous thrombosis. "While it is tempting to suggest that lupus anticoagulant and anti-PS/PT may have different selectivity for venous and arterial thrombosis, larger studies will be required," Petri and colleagues cautioned.

On a multivariate analysis, independent associations for any type of thrombosis were seen for the three primary biomarkers:

• Platelet-bound C4d, OR 4.2 (95% CI 1.1-16, P=0.04)
• Low C3, OR 6.20 (95% CI 1.3-29.6, P=0.02)
• Lupus anticoagulant, OR 7 (95% CI 1.3-38.5, P=0.02)

The composite risk score calculated from those three factors, which was significantly higher for any thrombosis, also was higher for venous thrombosis (2.30 vs 0.83, P<0.001) and for arterial thrombosis (1.62 vs 0.90, P=0.05).

"Both the composite risk score and the individual markers may be of value in clinical practice to identify patients likely to benefit from interventions, including hydroxychloroquine therapy, that can significantly reduce the risk of thrombosis in SLE," the researchers concluded.

Limitations of the study included its cross-sectional design and the small number of thrombotic events.

Three of the co-authors are employees of Exagen.