Patients with chronic inflammatory diseases experience higher exposure of small-molecule drugs than healthy volunteers: A population PK model-focused analysis of recently approved drugs
Johannes Tillil1,2, Sathej Gopalakrishnan3, Jennifer Dong4, Karthik Venkatakrishnan4, Wilhelm Huisinga1, Lena Klopp-Schulze3
1Institute of Mathematics, University of Potsdam, 2PharMetrX Graduate Research Training Program: Pharmacometrics & Computational Disease Modelling, Freie Universität Berlin and University of Potsdam, 3Quantitative Pharmacology, Merck Healthcare KGaA, 4EMD Serono Research & Development Institute, Inc., an affiliate of Merck KGaA
Background and Objectives: Elevated levels of proinflammatory cytokines in patients with chronic inflammatory diseases can suppress the expression and activity of cytochrome P450 (CYP) enzymes, potentially leading to reduced metabolic clearance of small-molecule drugs [1, 2]. Since many small-molecule drugs in clinical use are metabolized by CYP enzymes, this can pose a challenge in drug development, particularly during dose translation from healthy volunteers (HVs) to patients. Several clinical studies have observed a decrease in apparent clearance (CL/F) of small-molecule drugs in patients with chronic inflammatory diseases compared to HVs. However, to our knowledge no systematic literature search of the extent of this phenomenon in drug development has been conducted. This study addresses this gap by evaluating reported differences in CL/F of recently approved extravascularly administered small-molecule drugs between patients with chronic inflammatory diseases and HVs over a 15-year period (2010–2024). Methods: We conducted a systematic literature search by screening the Drugs@FDA database for new drug applications (NDAs) of “Type 1: New Molecular Entity” (NME) between 01/2010 and 12/2024 for small-molecule drugs approved for the chronic inflammatory diseases rheumatoid arthritis (RA), psoriasis (PsO), atopic dermatitis (AD) psoriatic arthritis (PsA), multiple sclerosis (MS), ulcerative colitis (UC), biliary cholangitis (BC), and vasculitis. The NDAs were searched for author-reported differences in PK and for population pharmacokinetic (PopPK) models that included disease status (patient vs. HVs) as a binary covariate on CL/F. Results: A total of 19 drugs indicated for any of the pre-specified chronic inflammatory diseases were included in the analysis. For 14 drugs, authors discussed the PK difference between patients and HVs. No PK difference was reported for 8 drugs, a decrease in CL/F from HVs to patients was reported for 4 drugs, and increased exposure in patients compared to HVs was reported for 2 drugs. For 3 drugs, no PopPK model was available. Thus, 16 PopPK models were analyzed. A total of 8 drugs were included in the comparison: 7 with PopPK models where disease status was included as a binary covariate on CL/F and additionally tofacitinib, indicated for RA, where no HV data was included in the PopPK modeling dataset but where the authors reported a pronounced difference in CL/F and where a direct patient (PopPK estimate based only on patient data) to HV (NCA estimate based only on HV data) comparison of CL/F was still possible. All 8 drugs included in the comparison showed linear, time-invariant PK and demonstrated a decrease in CL/F in patients of chronic inflammatory diseases compared to HVs. The mean decrease was 28.2%, ranging from 5.6% (ozanimod in UC) to 47.3% (tofacitinib in RA). 5 drugs demonstrated a reduction of more than 20%, equivalent to an increase in exposure (AUCss) of more than 25%. All 5 of these drugs with a considerable reduction in CL/F are metabolized with major contribution from CYP3A4, CYP2C19, or CYP2C9. Drugs for which no difference in PK was reported or that show a lower reduction in CL/F are generally either not extensively metabolized (e.g. baricitinib in RA where no difference was reported and which is metabolized by CYP3A4 but 84% of the dose is excreted unchanged) or are metabolized with major contribution by other CYP and non-CYP enzymes (e.g. ozanimod in UC where a low decrease of 5.6% was reported and which is metabolized by aldehyde dehydrogenase (ALDH) in addition to CYP3A4). Of note, none of the analyzed PopPK models included any inflammation-related biomarkers as covariates. Conclusion: This review provides evidence that chronic inflammation has the potential to significantly reduce the apparent clearance of small-molecule drugs, potentially due to reduced activity of metabolic enzymes (particularly certain CYP enzymes), leading to higher drug exposures in patients compared to HVs. Furthermore, a recent review of the clinical pharmacokinetics of antibody-based therapeutic proteins (TPs) in chronic inflammatory diseases found that in 9 out of 30 TPs, patients with chronic inflammatory diseases appeared to have mostly higher drug clearance than HVs, resulting in lower drug exposure [3]. These findings highlight the need to consider chronic inflammation as potential influential factor on PK and to utilize PK modeling and simulations in early-stage clinical study design and dose optimization to ensure safe and robust dose translation from HVs to patients. Further research is needed to better understand the underlying mechanisms and clinical implications of these differences.
[1] Jover, R., Bort, R., Gómez-Lechón, M.J. and Castell, J.V., 2002. Down-regulation of human CYP3A4 by the inflammatory signal interleukin 6: molecular mechanism and transcription factors involved. The FASEB Journal, 16(13), pp.1-29. [2] Aitken, A.E., Richardson, T.A. and Morgan, E.T., 2006. Regulation of drug-metabolizing enzymes and transporters in inflammation. Annu. Rev. Pharmacol. Toxicol., 46(1), pp.123-149. [3] Tian, X., Yu, Y., Neeli, H. and Jappar, D., 2024. Impact of Chronic Inflammatory Diseases on Clinical Pharmacokinetics of Antibody-Based Therapeutic Proteins. Clinical Pharmacology & Therapeutics.