Category: Clinical Pharmacology
Purpose: Interleukin 23 (IL-23) has been implicated in a variety of immunological conditions and suppressing it has shown clinical efficacy in several auto-immune diseases, among which inflammatory bowel disease and psoriasis. Several anti-IL-23 antibodies have been approved in psoriasis – ustekinumab (Stelara®, targeting the p40 subunit of IL-23, common with IL-12), tildrakizumab (Ilumya™, targeting the p19 subunit of IL-23), guselkumab (Tremfya®, anti- p19), and risankizumab (Skyrizi™, anti-p19). A few others have also been in clinical trials – briakinumab (anti-p40) and mirikizumab (anti-p19). The purpose of this study is to relate suppression of IL-23 with efficacy data through a mathematical model.
Methods: Data regarding the levels and turnover of IL-23 in psoriasis was gathered from literature. A site-of-action mathematical model was constructed to account for antibody distribution and binding of the target in the skin. Population pharmacokinetics two-compartment model parameters, binding affinity, and clinical efficacy data (PASI75) were available for ustekinumab, briakinumab, tildrakizumab, guselkumab, and risankizumab. Efficacy data was placebo subtracted. Based on dosing regimens, average IL-23 suppression in the skin was projected at the time points at which efficacy data was gathered. A sigmoid curve (with domain 0 to 1) was fit through a plot of the IL-23 suppression on the x-axis and efficacy on the y-axis. The equation of the curve is E=Emax*((1+E_x^h )*(TC/100)^h)/((TC/100)^h+E_x^h ), where E is PASI75 in %, Emax is the maximum efficacy achieved at 100% suppression of IL-23, Ex is an inflexion-related parameter, TC is the projected IL-23 target coverage in the skin, in %, and h is a Hill-like power coefficient.
Results: Clinical data combined psoriasis and plaque psoriasis data and comprised three studies for ustekinumab (a Ph2 and Ph3 Phoenix 1 & 2), two studies for briakinumab (Ph2 and Ph3), three studies for guselkumab (Ph1, Ph2, and Ph3 VOYAGE 1 & 2), three studies for risanskizumab (Ph2 and Ph3 ULTIMMA 1 & 2, ustekinumab data from these studies was excluded), and three studies for tildrakizumab (Ph2b and Ph3 reSURFACE 1 & 2). Few data points related efficacy to low level of IL-23 suppression – most dosing regimens are projected to achieve >70% IL-23 suppression and >50% PASI75. This is reflected in the wide 95% confidence interval near the origin. Data was not binned by time points – results range from efficacy and projected IL-23 suppression between 6 weeks and 52 weeks. The curve fits the data well, implying strong relationship between IL-23 suppression and PASI75.
Conclusion: The gathered efficacy data and the fit of the curve suggest that suppression of IL-23 is correlated to efficacy in psoriasis, which is also evident from the list of approved drugs. The stronger the suppression, the higher the efficacy, and, given that risankizumab’s projected IL-23 suppression is well above 95%, there is little room for improvement in PASI75 score by suppressing IL-23 further than what has been achieved already.