Impact of rilzabrutinib on gene-expression of inflammatory factors in patients with warm autoimmune hemolytic anemia who participated in the Phase 2B lumina 2 study

Abstract Introduction: Warm autoimmune hemolytic anemia (wAIHA) is a rare form of anemia caused by the destruction of red blood cells (RBC) primarily due to IgG (warm antibody) binding to RBC antigens. Mechanisms of wAIHA pathophysiology include extravascular hemolysis (antibody-dependent RBC destruction in the spleen, complement-dependent RBC destruction in the liver), intravascular hemolysis (complement cascade activation), and variable bone marrow compensation. Overactive cellular immune effectors, free hemoglobin (Hb), microparticle (vesiculation) and dysregulation of cytokines support the crucial role of inflammation in wAIHA. Rilzabrutinib is an oral, reversible, covalent Bruton tyrosine kinase inhibitor with multi-immune modulatory effects that demonstrated a 64% Hb response with a favorable safety profile in part A of the open-label, multicenter phase 2b LUMINA 2 study in adults with primary wAIHA relapsed/refractory to or dependent on corticosteroids (NCT05002777). Here, we evaluate the impact of rilzabrutinib on inflammatory and complement biomarkers at gene expression level in patients (pts) with wAIHA from the LUMINA2 part A study. Methods: In part A of LUMINA 2, pts received oral rilzabrutinib 400 mg twice daily for 24 weeks. In this post-hoc analysis, gene expression was assessed in pts with wAIHA (n=9) and compared with that of healthy volunteers (HVs, n=9) and with paired post-treatment whole blood samples. Blood samples were taken at baseline and 24 weeks after first dose of rilzabrutinib. The mRNA expression of 15 genes related to inflammation, neutrophil function, complement activation, and cell adhesion was analyzed using data from bulk RNAseq transcriptomics. Association of gene expression with clinical response (overall Hb response [complete or partial] by week 24 of treatment) was analyzed. Linear models were used to compare gene expression levels post-rilzabrutinib at week 24 and HVs to baseline. False discovery rate adjusted P-values were used to control multiple testing. Results: At data cutoff of November 21, 2024, samples for transcriptomic analyses were available for 9 pts with wAIHA (5 durable responders and 4 nonresponders) at baseline and at week 24, and for 9 HVs. For pts with wAIHA and HVs, the median (range) age was 67 years (33-80) and 44 years (22-70), respectively. For pts with wAIHA, the median (range) Hb level at baseline was 8 g/dL (5.8-9.7), time since diagnosis was 10 years (0.1-47.2), and the number of prior wAIHA medications was 3 (1-5). Of 15 selected genes, there was significantly higher mRNA expression of complement C5 (C5), selenoprotein P (SELENOP), interleukin 18 (IL-18), myeloperoxidase (MPO), and elastase neutrophil expressed (ELANE) at baseline in pts with wAIHA compared to HVs (adjusted P-value <0.05). Conversely, mRNA expression of tumor necrosis factor (TNF) and intracellular adhesion molecule 1 (ICAM1) at baseline were significantly lower in pts with wAIHA compared to HVs (adjusted P-value <0.05). Following 24 weeks of rilzabrutinib treatment in pts with wAIHA, overall C5 mRNA level was significantly reduced from baseline (log fold change -2.62, adjusted P-value: 0.011). Seven out of 9 pts had decreased SELENOP mRNA levels (adjusted P-value 0.083) after rilzabrutinib treatment. Of the 9 treated pts, all 5 responders exhibited reduced mRNA levels of both C5 and SELENOP; of the 4 non-responders, 3 had decreased levels of C5 mRNA and 2 had decreased levels of SELENOP mRNA. Conclusion: In this study, elevated mRNA levels of inflammatory biomarkers were observed in pts with wAIHA compared with HVs suggesting an association between wAIHA and the activation of inflammatory cellular processes. C5 mRNA levels in pts with wAIHA were significantly diminished after 24 weeks of rilzabrutinib treatment, indicating a reduction in complement activation; the reductions were more pronounced in responders. Although not significant after multiple adjustments, there was a trend of SELENOP mRNA level reduction at week 24 of rilzabrutinib treatment with a more relevant decrease in responders, suggesting regulation of oxidative stress and inflammation by rilzabrutinib. Rilzabrutinib, through its multi-immune modulatory effects, may address the underlying inflammatory pathophysiology in wAIHA. This study is limited by small sample sizes, and further analysis is required to better understand the impact of rilzabrutinib on immune-related pathways in pts with wAIHA.