Abstract
BACKGROUND: Thrombocytopenia (TP) is a hematological manifestation of systemic lupus erythematosus (SLE) and is associated with unfavorable prognostic outcomes. This study aimed to develop a risk prediction model for new-onset TP in SLE patients. METHODS: Based on the multicenter prospective Chinese SLE Treatment and Research Group (CSTAR) registry, newly diagnosed SLE patients without TP at registration were enrolled. The least absolute shrinkage and selection operator (LASSO) method was used for variable selection. The final model was developed using multivariate Cox regression and displayed as a nomogram. Internal validation was achieved using enhanced Bootstrap resampling. RESULTS: During follow-up, thrombocytopenia developed in 80 (3.52%) of 2270 lupus patients. The final risk prediction model incorporated six predictors: baseline SDI score ≥ 1 (HR 2.207, 95% CI 1.350-3.609, p = 0.002), hemolytic anemia (HR 1.953, 95% CI 1.017-3.750, p = 0.044), low complement level (HR 2.351, 95% CI 1.004-5.505, p = 0.049), positive anti-β2GPI antibody (HR 1.805, 95% CI 1.084-3.004, p = 0.024), positive Coombs test (HR 1.878, 95% CI 1.123-3.141, p = 0.017), and positive anti-histone antibody (HR 1.595, 95% CI 1.017-2.587, p = 0.059). The model's performance was indicated by C-index values for risk prediction at one, two, and three years, which were 0.741 (0.660-0.823), 0.730 (0.655-0.805), and 0.710 (0.643-0.777), respectively; and Brier scores of 0.018 (0.012-0.024), 0.025 (0.017-0.032), and 0.037 (0.027-0.046), respectively. Calibration curves were drawn and situated near the diagonal line. CONCLUSIONS: This study developed the first risk prediction model for TP onset in lupus patients. Patients with baseline organ damage, hemolytic anemia, low complement, positive anti-histone antibody, positive anti-β2GPI antibody, or positive Coombs test were identified as being at high risk for thrombocytopenia and require further clinical attention.