Discussion
This study demonstrated that benralizumab resulted in significant clinical improvement in subjective and objective measures in 78% (7/9) of CSU patients unresponsive to H1-antihistamines who completed the study, which was sustained for two months after the last treatment dose. Furthermore, an overall difference of UAS7 greater than the MDC value of 14.7 between baseline and end of study post-benralizumab treatment implies that this finding is also clinically meaningful (Table 1) . These results are consistent with the intent-to-treat primary endpoint analysis recently reported24. Comparison of benralizumab responders (n=7) to non-responders (n=2) revealed that the average baseline UAS7 among treatment responders was less than non-responders, although this difference was not statistically significant (27 vs. 37.5, p=0.5). This suggests that clinical improvement among responders was independent of baseline disease severity. These observations support the use of benralizumab for treatment of SGAH unresponsive CSU and provide evidence for a pathogenic role for infiltrating eosinophils. Although significant differences were observed between responders and non-responders at baseline for relative blood eosinophil (2.2% vs. 7.9%, p=0.001) and basophil counts (0.3% vs, 1.5%, p=0.01), these differences were minimal and no longer apparent post-treatment (eos%: 0.0 vs. 0.0, p=0.99; basophil%: 0.4 vs. 0.2, p=0.88). Thus, non-responsiveness to benralizumab as measured by changes in UAS7 are likely due to other mechanistic factors unrelated to the presence of eosinophils which are eliminated by blocking IL-5R. Further investigation is required to determine the relationship between baseline peripheral blood eosinophil levels and benralizumab’s response in CSU.
Not surprisingly, as seen in previous asthma studies, the effect of benralizumab on eos% was observed immediately after the first dose and was sustained for 8-weeks after the last dose.34 The more prolonged effect of benralizumab in the complete responders after discontinuation, compared to what has been reported after discontinuation of omalizumab in controlled CSU patients, may be explained by modification of biologic pathways important for promoting persistent cutaneous inflammation associated with CSU. Additional objective evidence to support this conclusion is the observed effect of benralizumab in decreasing the numbers of lymphocytes observed in non-lesional skin of complete responders compared to baseline lesional biopsies. Eosinophils were also significantly reduced in the lesional skin of non-responder and partial responder CSU subjects (n=4) 4-weeks after the last dose of benralizumab compared to baseline lesional skin biopsies, as would be expected given benralizumab’s primary mechanism of action (Figure 3 D-E ).
A decrease in UAS7 should reflect an improved CSU-related QoL since both outcomes measure wheals and pruritus, the two main clinical features of CSU. As expected, the CU-QoL and UAS7 post-treatment values were significantly correlated (r2=0.9, p<0.0001), which support the internal consistency of our data. Besides the pruritus and wheal component, the most significant improvement in CU-QoL were reported for urticarial interference with physical activities, sleep and spare time (S-Table 6) .
Improvements in UAS7 and CU-QoL for this pilot study are comparable to the efficacy of omalizumab in CSU, as determined in a previous pilot study (S-Table 7 ), with nearly the same percentage of complete responders to benralizumab as omalizumab.35 It is important to note, that when the larger Phase 3 omalizumab study (ASTERIA I trial) for the treatment of CSU was conducted based on the results of the earlier feasibility study, the number of patients with 50% reduction in mean UAS7 from baseline revealed a 35.8% complete response and a 51.9% partial response.36 This demonstrates that the omalizumab pilot study was a good indicator of expected outcomes in the larger multicenter clinical trial.
Given benralizumab’s mechanism of action as an IL-5R antagonist, its effect in the treatment of CSU is not outwardly apparent. However, DEGs in blood before and after benralizumab has provided greater insight into some biologic pathways that may explain its efficacy in the treatment of CSU (S-Table 4 and 5 ). Urticarial wheals are clinical manifestations of altered vasoactive and cellular infiltration kinetics resulting from inappropriate activation of dermal mast cells in up to 50% of cases by IgG autoantibodies directly targeting the high affinity IgE mast cell receptor ( FcER1-alpha subunit) or IgE affixed to these receptors.37 The role of possible autoimmune mechanisms in CSU, although not directly investigated in this study, could be explained by our blood transcriptomic data which demonstrated that benralizumab downregulated Indoleamine-2,3-Dioxygenase (IDO) pathway expression. IDO is an interferon-γ-inducible enzyme that catalyzes tryptophan degradation and drives Th2-mediated inflammation.38 Elevated tryptophan degradation has been shown to correlate with disease activity previously reported in other autoimmune rheumatological conditions.39,40Thus, benralizumab-mediated inhibition of IDO activity may improve CSU by restoring the basal tryptophan metabolic state from a hypercatabolic state leading to immune homeostasis41 consistent with what other studies have reported.42
A minimal persistent inflammation in CSU is maintained by infiltrating inflammatory cells that release proinflammatory mediators, cytokines/chemokines and adhesion molecules capable of recruiting and activating other effector cell types.14 The sustained clinical improvement by benralizumab in 5/9 CSU subjects until at least 4-weeks after the last dose, versus baseline, is reflected by a significant reduction of peripheral blood inflammatory cell counts, which was also seen in partial- and non-responders (n=4) (Figure 3 ). This is supported by benralizumab’s downregulation of DEGs in blood transcriptomes that control cytokine/chemokine signaling pathways (S-Table 4 ) and SIGLEC-8 gene expression. Since SIGLEC-8 is expressed on human eosinophils and mast cells, this finding further supports a role for IL5-receptor-mediated signaling of inflammatory responses involving mast cells and eosinophils.43
The study helps to better understand potential alternative medications in patients not responding to high dose SGAH. The number of complete responders was remarkably high. The blood DEGs that were significantly associated with response to benralizumab were IDO1, PTGDR2, ALOX15, IL1RL1, and SLC29A1 all of which are regulated upstream by IL4. In this context, it is worth mentioning that previous studies have reported increased serum concentrations of IL4 during the appearance of urticarial lesions in CSU patients. 44 Among those DEGs, the expression of prostaglandin D2 receptor (PTGDR2) was most evidently related to the response status (i.e., complete, partial or no-response to benralizumab). Complete responders had under-expression of PTGDR2, indicated by a negative z-score (z-score= -0.65); but, partial or no-responders had persistent over-expression of PTGDR2, indicated by positive z-scores, even after treatment with benralizumab. The expression levels of this gene were correlated with UAS7, and also with absolute and relative eosinophil or basophil scores. Previous studies have reported increased prostaglandin D2 in venous blood draining urticarial wheals.16 In this context, over-expression of PTGDR2 might perpetuate or exaggerate the inflammatory condition that may be unresponsive or less responsive to monotherapy with an IL-5 antagonist alone. Fevipiprant (NVP-QAW039, Novartis), a potent and selective PTGDR2 antagonist which has previously been assessed for safety and tolerability in human studies45 could be studied in partial or non-responsive benralizumab treated patients to confirm the relevance of this genetic biomarker.