Modulation of inflammatory gene transcripts in psoriasis vulgaris: Differences between ustekinumab and etanercept

Since the advent of biologic therapy for psoriasis in 2003, continued development of cytokine antagonists has yielded progressively better clinical outcomes, as assessed using the Psoriasis Area and Severity Index (PASI) or Patient's Global Assessment. In addition, more targeted therapies seemingly afford better patient adherence. For example, the anti–IL-12/23 mAb ustekinumab has demonstrated much better “drug survival” than the TNF antagonists etanercept and adalimumab.1No D.J. Inkeles M.S. Amin M. Wu J.J. Drug survival of biologic treatments in psoriasis: a systematic review.J Dermatolog Treat. 2018; 29: 460-466Crossref PubMed Scopus (48) Google Scholar Previous molecular profiling showed that etanercept strongly suppressed genes associated with the IL-23/type 17 T-cell pathway and psoriasis disease transcriptome (PSTR), but patients with resolved lesions retained expression of certain inflammatory genes (“molecular scar”) of psoriasis that could potentially reignite pathogenic skin inflammation on drug discontinuation.2Suárez-Fariñas M. Fuentes-Duculan J. Lowes M.A. Krueger J.G. Resolved psoriasis lesions retain expression of a subset of disease-related genes.J Invest Dermatol. 2011; 131: 391-400Abstract Full Text Full Text PDF PubMed Scopus (144) Google Scholar As psoriasis treatments continue to advance, it becomes even more important to identify differences between therapeutic antagonists in their ability to control pathological gene activation, particularly as treatment goals evolve toward complete disease normalization. For therapies developed after TNF antagonists, including ustekinumab, molecular profiling studies of drug effects (using clinically relevant doses and time points) are lacking. In this study, we determined the PSTR from whole-genome profiling of skin biopsies on Affymetrix U133 2.0Plus in 89 patients with moderate-to-severe psoriasis (ustekinumab 45 mg [N = 19] or 90 mg [N = 33] or etanercept 50 mg twice weekly [N = 37]) who participated in a phase 3, randomized, blinded trial designed to compare the efficacy of etanercept and ustekinumab (Active Comparator (CNTO 1275/Enbrel) Psoriasis Trial [ACCEPT]; ClinicalTrials.gov identifier, NCT004545843Griffiths C.E.M. Strober B.E. van de Kerkhof P. Ho V. Fidelus-Gort R. Yeilding N. et al.Comparison of ustekinumab and etanercept for moderate-to-severe psoriasis.N Engl J Med. 2010; 362: 118-128Crossref PubMed Scopus (716) Google Scholar; microarray raw data-Accession Number GSE106992). Pre- and posttreatment biopsies were taken from the same psoriasis plaque within a patient but varied as to trunk versus extremity locations across patients. Each drug's effect on mRNAs expressed in active lesions was assessed after 1 or 12 weeks of treatment in 45 patients who achieved greater than or equal to 75% improvement in the Psoriasis Area and Severity Index (PASI75 response) with ustekinumab 90 mg (n = 23) or etanercept (n = 22) treatment and compared drug-induced changes to the baseline PSTR (for methodology, see this article's Online Repository at www.jacionline.org). This analysis was exploratory in the overarching phase 3 trial and not based on a prespecified analysis plan. The baseline PSTR comprised 5051 mRNAs with increased or decreased expression in skin lesions compared with nonlesional skin (Fig 1, A; see Table E1 in this article's Online Repository at www.jacionline.org). Ustekinumab modulated 5020 mRNAs after 12 weeks, whereas etanercept modulated 4567 mRNAs. Both drugs modulated a common set of 3054 mRNAs in the PSTR, but each also uniquely modulated additional genes within and outside of the baseline PSTR (703 and 375 transcripts were unique to ustekinumab and etanercept, respectively; see Table E2 in this article's Online Repository at www.jacionline.org). Using RT-PCR, IL-23 subunits, IL-17A (low abundance mRNAs not well quantified on microarrays), and IL-17– modulated genes were found to be strongly suppressed at week 12 among PASI75 responders for both drugs (see Fig E1 and Table E3 in this article's Online Repository at www.jacionline.org). Higher abundance genes demonstrated good correlation between differential expression measured via microarrays versus RT-PCR (see Fig E2 in this article's Online Repository at www.jacionline.org). For genes upregulated in the PSTR, overall improvements of 91% and 97% were observed for etanercept and ustekinumab, respectively (Fig 1, C; see Table E4 in this article's Online Repository at www.jacionline.org). Because both drugs induced more than 90% improvement in this study-specific disease transcriptome, one could view both agents as highly effective with relatively small mechanistic differences. Next, improvement in consensus PSTR and specific cytokine-response pathways (see Table E5 in this article's Online Repository at www.jacionline.org) was determined for both drugs (Fig 1, B; see Table E6 in this article's Online Repository at www.jacionline.org). Ustekinumab improved overall disease-associated genes and genes modulated by IL-22, IFNs, TNF, IL-1, and IL-17 to a greater extent than did etanercept (P < .05-.001 for listed pathways), establishing stronger suppression of disease-associated inflammatory gene transcripts by IL-12/23 p40 blockade versus TNF blockade. By its mechanism, stronger suppression of genes induced by IFN-γ would be predicted for ustekinumab (due to IL-12 antagonism), but stronger modulation of TNF-induced genes by ustekinumab (P < .001) is surprising. The residual disease expression profile or “molecular scar” for a given treatment is defined as PSTR genes modulated by less than 75% toward baseline nonlesional levels. As shown in Fig 1, D, and Tables E7 and E8 (in this article's Online Repository at www.jacionline.org), 18% of transcriptome genes continued to be expressed in ustekinumab-treated skin lesions versus 23% for etanercept, but only 11% of the residual genes were common to both agents (12% and 7% were unique to etanercept and ustekinumab, respectively). Particularly notable is the likely function of residual gene transcripts differing between ustekinumab and etanercept, with high expression of proinflammatory transcripts only in the etanercept molecular scar (Fig 2). Note that the most prominent residual pathway identified in etanercept-treated lesions is “Role of IL-17A in Psoriasis” (Fig 2, A). We conclude that ustekinumab reduces expression of inflammation-related gene transcripts to a far greater degree, yielding less residual expression of inflammatory gene products, than etanercept. Furthermore, we hypothesize that more durable control of psoriasis over years of treatment, that is, “drug survival,”4Costanzo A. Malara G. Pelucchi C. Fatiga F. Barbera G. Franchi A. et al.Effectiveness end points in real-world studies on biological therapies in psoriasis: systematic review with focus on drug survival.Dermatology. 2018; 234: 1-12Google Scholar may be related to lower expression of cytokines, chemokines, and inflammatory mediators in treated lesions and thus a more stable environment for preventing reinitiation of psoriasis from likely pathogenic TH17 T-cell clones retained in treated skin lesions.5Matos T.R. O'Malley J.T. Lowry E.L. Hamm D. Kirsch I.R. Robins H.S. et al.Clinically resolved psoriatic lesions contain psoriasis-specific IL-17-producing αβ T cell clones.J Clin Invest. 2017; 127: 4031-4041Crossref PubMed Scopus (151) Google Scholar An alternative explanation is that IL-17 production from T cells is more consistently reduced by direct ustekinumab IL-23 blockade, while etanercept reduces IL-23 indirectly by modulating IL-23–producing dendritic cells in psoriasis lesions, which could be subject to costimulatory inputs or Toll-like receptor sensors capable of reinitiating IL-23 production over treatment periods of months-to-years. Overall, study results establish that patients with psoriasis with similar degrees of clinical improvement (PASI75 response) can exhibit different degrees of suppression of the molecular disease profile (PSTR) and significantly different qualitative suppression of inflammatory gene transcripts (molecular scar) based on the drug used. The maximal attainable improvement in psoriasis associated with each mode of cytokine antagonism remains unknown. In future studies, it will be important to understand how effectively the new generation of selective IL-23 or IL-17 antagonists, which generally have superior clinical performance to TNF and IL-12/IL-23 antagonism,6Sbidian E. Chaimani A. Garcia-Doval I. Do G. Hua C. Mazaud C. et al.Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis.Cochrane Database Syst Rev. 2017; : CD011535PubMed Google Scholar modulates psoriasis as defined by cellular and molecular pathways that collectively change tissues and clinical signs/symptoms. Understanding these molecular mechanisms may be particularly important to appreciate the basis for long-term disease clearance observed with next-generation IL-23 and IL-17 antagonists in some studies.6Sbidian E. Chaimani A. Garcia-Doval I. Do G. Hua C. Mazaud C. et al.Systemic pharmacological treatments for chronic plaque psoriasis: a network meta-analysis.Cochrane Database Syst Rev. 2017; : CD011535PubMed Google Scholar The data set reported herein will provide an important reference for comparison with next-generation cytokine antagonists used to treat psoriasis. Download .docx (.02 MB) Help with docx files Online Repository text Download .pdf (.45 MB) Help with pdf files Fig E1 Download .pdf (.36 MB) Help with pdf files Fig E2 Download .pdf (.27 MB) Help with pdf files Fig E3 Download .pdf (.2 MB) Help with pdf files Fig E4 Download .docx (.19 MB) Help with docx files Table E1 Download .docx (14.63 MB) Help with docx files Table E2 Download .docx (.02 MB) Help with docx files Table E3 Download .docx (.01 MB) Help with docx files Table E4 Download .docx (2.59 MB) Help with docx files Table E5 Download .docx (.11 MB) Help with docx files Table E6 Download .docx (.38 MB) Help with docx files Table E7 Download .docx (.3 MB) Help with docx files Table E8

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.