Large-scale Analyses of Disease Biomarkers and Apremilast Pharmacodynamic Effects

Abstract

Finding biomarkers that provide shared link between disease severity, drug-induced pharmacodynamic effects and response status in human trials can provide number of values for patient benefits: elucidating current therapeutic mechanism-of-action, and, back-translating to fast-track development of next-generation therapeutics. Both opportunities are predicated on proactive generation of human molecular profiles that capture longitudinal trajectories before and after pharmacological intervention. Here, we present the largest plasma proteomic biomarker dataset available to-date and the corresponding analyses from placebo-controlled Phase III clinical trials of the phosphodiesterase type 4 inhibitor apremilast in psoriasis (PSOR), psoriatic arthritis (PsA), and ankylosing spondylitis (AS) from 526 subjects overall. Using approximately 150 plasma analytes tracked across three time points, we identified IL-17A and KLK-7 as biomarkers for disease severity and apremilast pharmacodynamic effect in psoriasis patients. Combined decline rate of KLK-7, PEDF, MDC and ANGPTL4 by Week 16 represented biomarkers for the responder subgroup, shedding insights into therapeutic mechanisms. In ankylosing spondylitis patients, IL-6 and LRG-1 were identified as biomarkers with concordance to disease severity. Apremilast-induced LRG-1 increase was consistent with the overall lack of efficacy in ankylosing spondylitis. Taken together, these findings expanded the mechanistic knowledge base of apremilast and provided translational foundations to accelerate future efforts including compound differentiation, combination, and repurposing.

Figure 1

IL-17A and KLK-7 as biomarkers of psoriasis. (A) Lasso coefficients of analytes to PASI score. Only proteins that were significant in at least one time point are shown. Stars indicate significance (p-value < 0.05). (B) Mean analytes concentration of IL-17A and KLK-7 in pg/mL for each subgroup across time points; (C) Comparison of the change in aggregated IL-17A and KLK-7 expression score in the first 4 weeks in subgroups: placebo (PBO), apremilast responders (AR) and apremilast non-responders (ANR) (***denotes p < 0.01). Aggregated score was significantly lowered by apremilast.

Figure 2

Differential expression (DE) analysis between placebo and treatment arms at specified time points and meta-analysis across all diseases. Significant DE proteins are labeled. The last row “week 4 & week 16“ denotes to mixed-effect modeling on pooled data from the two time points. The last column “Meta-analysis“ denotes to differential expression modeling on pooled data across all three diseases.

Figure 3

CoGAPS patterns across comparisons (A,C,F) and the associated analytes (B,D,G). The color of analytes and patterns is matched in each study. PSOR- psoriasis cohort; PBO - placebo treatment arm; APR - apremilast treatment arm; ANR - apremilast non-responders; AR - apremilast responders; Meta - meta-analysis on pooled data from all three diseases: ankylosing spondylitis, psoriasis and psoriatic arthritis. (G) Geometric mean declines of PEDF, KLK-7, MDC, and ANGPTL4 (weeks 4 and 16) between apremilast non-responders and responders (**denotes p < 0.05).

Figure 4

Systematic literature review of KLK-7 and phenotypes. Columns of panel A and B are aligned and labeled in panel B. (A) The highest HPO parent class(es) for each phenotype that co-occurred with KLK-7 in Medline abstracts. The association frequency of KLK-7 with the abnormality of the integument is the second highest behind neoplasm; (B) Specificity score for each unique gene-phenotype pairing extracted from Medline abstracts based on sentence level co-occurrence of KLK-7 and phenotype terms. Size of a circle indicates the square root of the number of publications covering a gene-phenotype pairing. Top 3 conditions with the highest specificity all of the skin-related diseases.

Figure 5

Survey of KLK-7 mRNA expression patterns in human psoriasis studies. For clarity, only statistically significant contrasts are shown (FDR < 0.01). The fold change corresponds to the change from condition 1 to condition 2.