Mourad HAMIMED (1,2), Mélanie DONNETTE (1,3), Laure FARNAULT (4), Christel PISSIER (3), L’Houcine OUAFIK (5), Caroline SOLAS (3,6), Joseph CICCOLINI (1,2,3), Raphaelle FANCIULLINO (1,2,7)
(1) SMARTc Unit, Cancer Research Center of Marseille, Inserm U1068 – CNRS UMR 7258 – Aix-Marseille University U105, Marseille, France, (2) Inria – Inserm COMPO team, Centre Inria Sophia Antipolis – Méditerranée, Inserm U1068 – CNRS UMR 7258 – Aix-Marseille University U105, Marseille, France, (3) Clinical Pharmacokinetics and Toxicology Laboratory, La Timone University Hospital of Marseille, APHM, Marseille, France , (4) Hematology-Oncology Unit, La conception University Hospital of Marseille, Marseille, France , (5) Transfert Oncology Laboratory, Nord University Hospital of Marseille, Marseille, France , (6) Unité des Virus Émergents (UVE), Aix-Marseille Univ-IRD 190-Inserm 1207, Marseille, France, (7) Pharmacy, La Conception University Hospital of Marseille, Marseille, France
Objectives: Cytarabine-based chemotherapy represents a cornerstone in the current leukemia’s therapeutic arsenal, yet the outcomes remain unsatisfactory due to toxicity and variable clinical response. Growing evidence suggests that cytidine deaminase (CDA), an enzyme that catalyzes the detoxification of cytarabine (Ara-C) into a 1-β-D-arabinofuranosyluracil (Ara-U), may partially explain the large between subject variability (BSV) reported for Ara-C pharmacokinetics [1,2]. The effect of CDA activity on Ara-C exposure has never been quantified. This work aimed to develop a parent-metabolite joint population pharmacokinetic model and investigate the factors, including CDA activity and genetic polymorphisms, that are impacting exposure to Ara-C.
Methods: A prospective population pharmacokinetic study of forty-seven patients with acute myeloid leukemia was undertaken. Patients received the standard combination (7+3) with a 7-day continuous infusion of Ara-C at the dosage of 200 mg/m² per day on days 1 to 7 and daunorubicin/idarubicin on days 1 to 3. The quantification of Ara-C and Ara-U plasma levels was performed using a validated ultra-performance liquid chromatography with tandem mass spectrometric detection method (UPLC-MS/MS)[3]. Population PK models were developed in Monolix (version 2021R1). Parameters were estimated by the SAEM algorithm. Seventeen SNPs on PK-related genes were genotyped. The effect of potential factors on PK parameters was investigated using a stepwise covariate model procedure (SCM), while their impact on exposure (AUC and Cmax) was evaluated using univariate predictions. A forest plot was constructed using the coveffectsplot R package version 0.0.9.1[4].
Results: Thirty-seven patients (382 samples) were available and used in the population PK analyses. The median (range) age and CDA activity were 62 (22-78) years and 2.1 UA/mg (0.6 – 17.4), respectively. For Ara-C, a 2-compartment model best described the data, whereas for Ara-U, a 1-compartment model best fitted the observations. The population estimate (and BSV, CV%) for clearance and volume of distribution of the central compartment (VC) for Ara-C were 159 L/h (72.5%) and 12.4 L (50%), respectively, and 10 L/h (42.8%) and 87.3 L (83.9%), respectively, for Ara-U. Model diagnosis showed good descriptive and predictive performances. The covariates that were associated with the largest impact on AUC and Cmax were CDA activity and percentage of blasts for Ara-C and creatinine clearance for Ara-U. There was no impact of anthracycline co-administration nor sex on Ara-C pharmacokinetics.
Conclusions: Population PK models characterized the disposition for both molecules. CDA activity was a significant covariate for Ara-C. Patients with above-normal CDA values may receive sub-optimal treatment doses due to elevated Ara-C clearance. Linking the significant relationships between CDA and kidney function and Ara-C/ Ara-U clearances to clinical end points would support dose individualization.
References:
[1] Donnette M, Ciccolini J, Pissier C, Costello R, Duffaud F, Salas S, et al. High incidence of CDA deficiency in patients with hematological malignancies: perspectives and therapeutic implications. Annals of Oncology. 2021;32:684–6.
[2] Pharmacokinetics and pharmacogenetics of liposomal cytarabine in AML patients treated with CPX-351. Journal of Controlled Release. Elsevier; 2021;338:244–52.
[3] Donnette M, Solas C, Giocanti M, Venton G, Farnault L, Berda-Haddad Y, et al. Simultaneous determination of cytosine arabinoside and its metabolite uracil arabinoside in human plasma by LC-MS/MS: Application to pharmacokinetics-pharmacogenetics pilot study in AML patients. Journal of Chromatography B. 2019;1126–1127:121770.
[4] Mouksassi S, Attali D. coveffectsplot: Produce Forest Plots to Visualize Covariate Effects [Internet]. 2020 [cited 2021 Aug 6]. Available from: https://CRAN.R-project.org/package=coveffectsplot
Reference: PAGE 30 (2022) Abstr 10230 [www.page-meeting.org/?abstract=10230]
Poster: Clinical Applications