Eduard Schmulenson (1), Markus Joerger (2), Laurent Arthur Decosterd (3), Carlo R. Largiadèr (4), Daniela Bärtschi (5), Roger von Moos (6), Sara Bastian (6)§, Ulrich Jaehde (1)§
(1) Institute of Pharmacy, Department of Clinical Pharmacy, University of Bonn, Bonn, Germany (2) Department of Medical Oncology and Hematology, Cantonal Hospital St. Gallen, St. Gallen, Switzerland (3) Laboratory of Clinical Pharmacology, Lausanne University Hospital and University of Lausanne, Lausanne, Switzerland (4) Department of Clinical Chemistry, Inselspital, Bern University Hospital and University of Bern, Bern, Switzerland (5) SAKK Coordinating Center, Bern, Switzerland (6) Cantonal Hospital Graubünden, Chur, Switzerland §shared senior authorship
Objectives: Despite anticancer therapy, patients with locally advanced rectal cancer frequently suffer from relapses. A recent study in which sorafenib, a tyrosine kinase inhibitor (TKI) was added to the standard therapy consisting of capecitabine, a prodrug of the antimetabolite 5-fluourouracil, showed promising results [1]. Regorafenib, a multitarget TKI approved for the use in rectal cancer, has a broad mechanism of action and was investigated in combination with capecitabine in the phase Ib trial “Neoadjuvant treatment with Regorafenib and Capecitabine combined with radiotherapy in locally advanced rectal cancer (SAKK 41/16 RECAP)” (NCT02910843). This trial investigated if similar results could be achieved as with sorafenib. Until then, the combination of capecitabine and regorafenib has never been investigated before. Population pharmacokinetic (PopPK) models of both regorafenib and capecitabine and its metabolites were implemented in this study to analyze possible drug-drug interactions (DDI).
Methods: Pharmacokinetic data from 12 patients enrolled in the RECAP trial were used for the development of the PopPK models using NONMEM®. Patients received 825 mg/m² capecitabine twice daily from day 1 to 38. Regorafenib was administered in three dose levels (40, 80 and 120 mg) once daily from day 1 to 14 and day 22 to 35. Plasma samples were collected on day 1 (0.5, 1, 2, 3, 4 and 6 h after dosing), followed by random sampling on day 2, 4, 8, 15, 22, 29 and 36. Plasma concentrations of regorafenib and its metabolites M-2 and M-5 as well as plasma concentrations of capecitabine and its metabolites 5’-deoxy-5-fluorocytidine (DFCR) and 5’-deoxy-5-fluorouridine (DFUR) were quantified using validated assays by liquid chromatography coupled to tandem mass spectrometry [2, 3]. PopPK parent-metabolite models were built in three sequential steps. After establishing the parent drug model, its parameters were fixed and the subsequent metabolites were included stepwise. Eventually, all parameters were estimated simultaneously. These models were then used to generate drug exposure parameters (unbound and total concentrations over time, cumulative area under the curve [AUC] over time). Finally, these parameters were tested as covariates within the respective other PopPK model in order to investigate possible DDI. Model evaluation was performed using goodness-of-fit plots and visual predictive checks.
Results: Regorafenib, M-2 and M-5 pharmacokinetics were adequately described by linear two-compartment models. A transit compartment model was used for the description of regorafenib absorption and the formation of M-2 from regorafenib was best described by presystemic metabolism. Since the fractions metabolized were unknown, it was assumed that M-2 and M-5 volumes of distribution were the same as that of regorafenib. Likewise, capecitabine, DFCR and DFUR pharmacokinetics were well described by linear one-compartment models. Parallel first- and zero-order absorption processes were chosen for the capecitabine absorption model. Generally, model parameters were precisely estimated (relative standard errors <30%, except for the shared volume of distribution of regorafenib, M-2 and M-5). Adding regorafenib M-2 cumulative AUC over time as linear covariate significantly improved the capecitabine model fit and reduced capecitabine clearance estimates.
Conclusions: The developed PopPK models of regorafenib, capecitabine and their respective metabolites described the observed data well and were applied to investigate the influence on the pharmacokinetics of the respective other drug. The influence by regorafenib M-2 cumulative AUC over time on capecitabine clearance suggest a decreased formation of the active metabolite 5-fluorouracil by the addition of regorafenib, which deserves further investigation.
References:
[1] Von Moos R, Koeberle D, Schacher S, et al. Neoadjuvant radiotherapy combined with capecitabine and sorafenib in patients with advanced KRAS-mutated rectal cancer: A phase I/II trial (SAKK 41/08). Eur J Cancer 2018; 89: 82-89.
[2] Cardoso E, Mercier T, Wagner AD, et al. Quantification of the next-generation oral anti-tumor drugs dabrafenib, trametinib, vemurafenib, cobimetinib, pazopanib, regorafenib and two metabolites in human plasma by liquid chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci 2018; 1083: 124-136.
[3] Vainchtein LD, Rosing H, Schellens JHM, et al. A new, validated HPLC-MS/MS method for the simultaneous determination of the anti-cancer agent capecitabine and its metabolites: 5′-deoxy-5-fluorocytidine, 5′-deoxy-5-fluorouridine, 5-fluorouracil and 5-fluorodihydrouracil, in human plasma. Biomed Chromatogr 2010; 24: 374-386.
Reference: PAGE 29 (2021) Abstr 9738 [www.page-meeting.org/?abstract=9738]
Poster: Drug/Disease Modelling - Oncology