ESK-001, an allosteric TYK2 inhibitor, inhibits TYK2-and psoriasis-relevant biomarkers in skin using spatial transcriptomics

Summary: Abstract Body: The pathogenesis of plaque psoriasis is driven by dysregulated signaling of the IL23/IL17 axis. ESK-001 is an oral, highly selective small molecule allosteric inhibitor of TYK2. In the ESK-001 STRIDE trial, a phase two placebo-controlled 12-week trial in 228 moderate-to-severe plaque psoriasis patients, the primary and secondary endpoints were met at the highest doses with clear dose dependent improvement in efficacy. STRIDE undertook bulk and spatial transcriptomic analysis in patients to characterize TYK2- and psoriasis-relevant biomarker inhibition in response to ESK-001. Skin punch biopsies from 40mg BID, 40mg QD, and placebo dose arms were collected at baseline and week 12 and split for bulk and spatial transcriptomics. Spatial transcriptomics was performed using the 10X Genomics Visium HD platform and analysis with Space Ranger. For bulk RNAseq, reads were aligned to the genome and transcriptome using STAR, and transcripts were quantified with Salmon. Spatial transcriptomics show high expression of TYK2- and psoriasis-relevant biomarkers in the epidermis of baseline lesional samples. Expression of biomarkers KRT16 and TYK2-relevant cytokines are decreased in baseline non-lesional samples compared to lesional. After 12 weeks of ESK-001 administration, lesional samples show significant decrease in expression of KRT16 and other psoriasis-relevant biomarkers in the epidermis in 40mg BID and 40mg QD dose groups compared to placebo. These results align with bulk RNAseq analysis of paired punch biopsies where the same biomarkers returned to baseline non-lesional expression post ESK-001 administration. Disease-relevant bulk and spatial transcriptomic analysis in skin tissue shows decreased expression of key psoriasis biomarkers in non-lesional skin compared to lesional. Analysis after ESK-001 administration further indicates a return of key psoriasis biomarkers to non-lesional baseline levels. Nicole Narayan¹, Joshua D. Hoffman¹, Claire L. Langrish¹, Sibel Ucpinar¹, Pedro Corpuz Jr¹, Nicholas E. Vlahakis¹, Mera K. Tilley¹ 1. Alumis, South San Francisco, CA, United States. Bioinformatics, Computational Biology, and Imaging