Differential RNA Transcriptome in Ustekinumab Responders vs Non-Responders

February 2023 | Volume 22 | Issue 2 | 219 | Copyright © February 2023


Published online January 24, 2023

Amy G. Johnson a*, Dustin P. DeMeo a,c*, Brian Richardson b, Jacklyn B. Golden b, Mark J. Cameron b, Andrew B. Young c, Thomas S. McCormick c, Kevin D. Cooper MDa

aDepartment of Dermatology, University Hospitals Cleveland Medical Center, Cleveland, OH
bDepartment of Population and Quantitative Health Sciences, Case Western Reserve University, Cleveland, OH
cDepartment of Dermatology, Case Western Reserve University, Cleveland, OH

Abstract
Despite numerous available psoriasis treatments, no "one size fits all" regimen provides complete disease control without side effects, logistical obstacles, and/or expense. Despite increasingly efficacious drugs, only 20-25% of patients treated with biologic therapies achieve completely clear skin (PASI 100) and even fewer achieve this if they have experienced failures of multiple biologics.

CASE

Despite numerous available psoriasis treatments, no "one size fits all"' regimen provides complete disease control without side effects, logistical obstacles, and/or expense. Despite increasingly efficacious drugs, only 20-25% of patients treated with biologic therapies achieve completely clear skin (PASI 100) and even fewer achieve this if they have experienced failures of multiple biologics.1,2

The balance of cytokines and chemokines produced in psoriatic lesions- exemplified by Interleukin-12 (IL-12) and IL-23- is implicated in disease phenotypic heterogeneity and lesion severity.3,4 In humans, IL-12 is a heterodimer of the IL-12p40 subunit with the IL-12p35 subunit, which signals through the IL-12 receptor (IL-12R) leading to Janus kinase 2 (JAK2) and tyrosine kinase 2 (TYK2) activation and phosphorylation of signal transducer and activator of transcription (STAT) family molecules.5 IL-23 is composed of the IL-23p19 subunit and the common IL-12p40 subunit, which signals through IL-23R also activating JAK and STAT signals.6 Combined, the fascinating interplay between IL-12 and IL-23 accounts for activation of TNFα, IFNγ and IL-17 through activation of myeloid and T helper (Th) specific pathways, now common targets for numerous biologic agents.

However, psoriasis patients often experience a counterintuitive failure to respond to p40 blockade of IL-12 and IL-23 via ustekinumab, yet successfully respond to p19 blockade of the IL-23 pathway.7 This observation is not limited to psoriasis as Crohn’s Disease (CD) patients exhibit similar discordant outcomes. Although in CD some headway has been made to predict patient response to IL-12 and IL-23 blockade by ustekinumab, validated biomarkers are still lacking, and a clear gap in understanding the mechanism(s) responsible for such heterogeneous responses persists.8

Our lab has previously reported distinct psoriasis patient endotypes associated with unique clinical phenotypes and responses to treatment based on whole blood transcriptome analyses of psoriasis and psoriatic arthritis (PsA) patients.9 This analysis identified a general psoriasis signature featuring upregulated interferons that trended toward an increase with age. Other psoriatic endotypes were identified using proinflammatory signals that clustered patients by increasing PASI score, gender and ethnicity.10 We hypothesize that discovery of differentially expressed gene (DEG) signatures will inform more individualized care and that an integrated 'omic and clinical approach may identify classifiers of psoriasis and PsA patients that may be used to individualize their treatment.

In the current report, we utilized a similar whole blood transcriptome analysis to investigate the intersection of psoriasis endotypes and treatment response. Based on a cohort of psoriasis patients at our institution who failed to improve on ustekinumab, we performed RNA transcriptome analysis of six self-reported Responders (R) and four self-reported Non-Responders (NR) to examine DEGs between the groups that may contribute to treatment response. Patients were of similar age yet mostly male with only one non-white patient. Ustekinumab responders exhibit elevated cardiovascular risk scores, although differences with non-responders did not reach statistical significance (Figure 1). Gene set variation analysis (GSVA) using the Molecular Signatures Database (MSigDB) hallmark pathways to identified genes and associated annotations.

Heat map visualization of the top 50 DEGs by P value suggests strong separation between ustekinumab R and NR patients by hierarchical clustering (Figure 2). Each identified gene has significant power to discern groups, suggesting that they may be good sentinels or biologically relevant to the pharmacological mechanism of action associated with ustekinumab. Furthermore, intragroup similarity for NR is also high, such that an inadequate response may be especially uniform across patients. There are roughly equal numbers of upregulated (shades of red) versus downregulated genes (shades of purple), with some variability