Background Belimumab is used as add-on therapy in patients with refractory systemic lupus erythematosus (SLE). While shared features guide patient eligibility, whether clinical and laboratory variables may identify distinct candidate subgroups remain unclear. Aim To identify clinical clusters among SLE patients initiating belimumab and assess their relationship with clinical outcomes. Methods Data were derived from the Italian multicentric Lupus Italian REgistry (LIRE), including patients starting belimumab added to standard treatment. Factor Analysis of Mixed Data (FAMD) with 150 baseline variables was followed by Hierarchical Clustering on Principal Components using Ward's criterion and Euclidean distance to define clusters. Disease activity was assessed by SLEDAI-2 K, damage by SLICC Damage Index (SDI), and clinical response as cSLEDAI-2K = 0 at 6 and 12 months (T6, T12). Results Among 125 patients, clustering identified four main groups. Cluster 1 ( n = 55) exhibited lower SDI; Cluster 2 ( n = 28) included patients with renal involvement ( p 0.001); Cluster 3 ( n = 14) displayed neuropsychiatric/cardiovascular activity, highest baseline SDI ( p = 0.006), enriched neurological/cardiovascular damage ( p 0.001), and lowest hydroxychloroquine use (78.6% vs. 95.2%, p = 0.05); Cluster 4 ( n = 28) showed predominant hematological abnormalities. Clinical responses were comparable across clusters, though Cluster 2 showed numerically higher cSLEDAI-2K = 0 rates at T12 (70.8% vs. 52.3%, p = 0.082). Final SDI was highest in Cluster 3 ( p 0.01), while damage accrued faster in Cluster 1 0 (0–2) vs. 1 (1, 2), p 0.001. Conclusion Unsupervised clustering revealed clinical heterogeneity among belimumab-treated SLE patients. Renal/serologically active profiles showed a numerically higher response rate, whereas pre-existing damage and low hydroxychloroquine use associated with higher damage burden despite biologic therapy.
Gatto et al. (Thu,) studied this question.