Molecular characterisation of hidradenitis suppurativa clinical clusters through serum proteomic profiling: Insights from the SUNSHINE and SUNRISE phase 3 trials
Johann E. Gudjonsson,1* Aashil A. Batavia,2 Anna Passera,3 David Demanse,3 Ben Roediger,2 Frank Kolbinger,2 Christian Loesche,3 Magdalena B. Wozniak,4 Valeria De Luca,2 Andre Nogueira da Costa,2 Enrico Ferrero2
Objective 1
To investigate the differences in the serum proteome of clinically defined clusters of patients with hidradenitis suppurativa previously identified from the SUNSHINE and SUNRISE phase 3 trials
Objective 2
To investigate the clinical relevance of molecular differences between clusters of patients with hidradenitis suppurativa
Introduction
Hidradenitis suppurativa (HS) is a debilitating, inflammatory skin disease with heterogenous presentation. Previous analyses from the SUNSHINE and SUNRISE trials of secukinumab in patients with moderate to severe HS described three HS patient clusters (termed Cluster 1, Cluster 2, and Cluster 3) with distinct demographic and disease characteristics,1 and associated treatment response to secukinumab.2 Here, a post hoc analysis of serum proteomic data was performed to identify molecular characteristics of these clusters.
Methods
Hierarchical clustering on principal components of 40 baseline demographic and clinical characteristics was previously applied to 1084 patients enrolled in the SUNSHINE and SUNRISE trials to identify patient clusters.1 Baseline serum samples were obtained from 1036 patients and relative expression of 6597 proteins (7596 aptamers) was assessed using SomaScan v4.1 to investigate the underlying molecular characteristics of each cluster. Linear models were applied to determine differentially expressed aptamers between clusters.
Results
In Cluster 3, which was characterised by increased lesion counts and worse patient-reported outcomes, inflammatory markers and insulin growth factors (e.g., SAA1, SAA2, IGFBP2 and IGFL1) were overexpressed versus Clusters 1 and 2 (adjusted p≤0.05; LFC>0.4). Gene set enrichment and over representation (i.e., significantly differentially expressed) analyses further supported the involvement of proinflammatory pathways in Cluster 3. Few differences were identified between Clusters 1 and 2 regarding their serum proteome (MCEE, CKMT1A, ELAC1, PIGR, PZP; adjusted p≤0.05; absolute LFC>0.4).
Conclusion
These data provide insights into the serum proteomic differences accompanying the clinical heterogeneity of HS and may inform future molecular endophenotyping of patients with HS.
Saverio Minervini