Juliane Staudinger1, Marco Orleni2, Sara Gagno2, Bianca Posocco2, Erika Cecchin2, Georg Hempel1
1Institute of Pharmaceutical and Medical Chemistry, Department of Clinical Pharmacy, University Münster, 2Experimental and Clinical Pharmacology Unit, CRO Aviano, National Cancer Institute
Objective Olaparib is an inhibitor of poly ADP ribose polymerase (PARPi) used in treatment of ovarian cancer with BRCA1/2 mutation, HER2-negative breast cancer, pancreatic cancer, and metastatic prostate cancer. The standard dose of 300 mg BID is reduced for patients with moderate renal impairment [1] or after the occurrence of a severe adverse event (AE). Other individual factors like phenotypes of the main metabolizing enzymes are not considered when dosing decisions are made. This one dose fits all approach bears the risk of toxic levels accompanied by AEs leading to the question whether the need for a dose reduction can be anticipated based on routine measurements as part of therapeutic drug monitoring (TDM). For further dose individualization phenotypes of the main metabolizing enzymes from the group of cytochrome P450 (CYP) enzymes [2] and genotypes of efflux transporters whose activity is responsible for olaparib-resistance in cancer cells (ATP binding cassette (ABC) transporters ABCB1 [3] and ABCG2 [4]) are subject to this population pharmacokinetic (popPK) analysis. Methods Pharmacokinetic data was obtained at regular checkups of patients treated with olaparib. Overall, 32 patients were included receiving 300 to 600 mg of olaparib per day. In addition to geno- and phenotyping of CYP3A4/5 and ABCB1/ABCG2, demographic data such as height and weight were documented for each patient, and written consent was obtained before inclusion. For the popPK analysis a published model by Zhou et. al. [5] was rebuilt and evaluated. The software packages NONMEM version 7.4 (ICON Development Solutions, Hanover, MD, USA) were used for modeling and simulation. R version 4.4.2 (R Foundation) was used for data preparation, analysis, and statistical summaries. Perl-speaks-NONMEM (PsN) and Pirana (Certara) were used to implement the NONMEM runs. The first-order conditional estimation (FOCE) method in NONMEM was employed for all model runs. Results The two compartment-model includes a sequential zero- and first-order absorption and a first-order elimination. All patients were in steady-state so the clearance was reduced by 15% and all administered doses were above 100 mg so the associated factor was added. To ensure the best fit of the model, the clearance and proportional error were re-estimated. The re-estimation of the clearance (Cl: 5.94 L/h) showed a significant improvement in the performance compared to the model with the reported parameters by Zhou et al. [5] (Cl: 4.22 L/h) by Chi square test. All other used parameters were fixed. The new parameter for the clearance was confirmed by a bootstrap analysis (N=1000) with the following result: median of 6.13, 2.5% of 4.77 and 97.5% of 8.09. Especially the clearance is of interest as the clearance mainly determines the overall exposure and the trough concentrations. Furthermore, the metabolizing enzymes and efflux transporters of interest might have an influence on the clearance. The popPK model was evaluated through goodness-of-fit plots. None of covariates under investigation were found to be significant in the analyzed patient group. No definitive statement could be made on the influence of the metabolizing enzymes as the available data only included four and two out of 32 patients in one of the groups, respectively. The group size limits the possibilities of the popPK analysis. In the conducted analysis, the SNPs of the efflux transporters ABCB1 and ABCG2 did not show a significant influence on the model and the individual predictions. Olaparib shows an exposure-toxicity relationship with a safety threshold of 2.5 mg/L at trough level (t = 12 h) [6]. The popPK model was used for the according prediction with the result that five out of the 32 patients exceed the safety threshold which is accompanied by a higher risk for AEs. Conclusion This popPK model serves as a crucial addition to the TDM as it allows more flexible sampling time points. We propose further investigation of the effect of the metabolizing enzymes as CYP3A4/5 are crucial for drug metabolizing in general and an expansion of the search for covariates as the random effects are rather high.
[1] Rolfo C, Vos-Geelen J de, Isambert N, et al. Pharmacokinetics and Safety of Olaparib in Patients with Advanced Solid Tumours and Renal Impairment. Clin Pharmacokinet 2019; 58(9): 1165–74 [https://doi.org/10.1007/s40262-019-00754-4][PMID: 30877569] [2] Dirix L, Swaisland H, Verheul HMW, et al. Effect of Itraconazole and Rifampin on the Pharmacokinetics of Olaparib in Patients With Advanced Solid Tumors: Results of Two Phase I Open-label Studies. Clin Ther 2016; 38(10): 2286–99 [https://doi.org/10.1016/j.clinthera.2016.08.010][PMID: 27745744] [3] Vaidyanathan A, Sawers L, Gannon A-L, et al. ABCB1 (MDR1) induction defines a common resistance mechanism in paclitaxel- and olaparib-resistant ovarian cancer cells. Br J Cancer 2016; 115(4): 431–41 [https://doi.org/10.1038/bjc.2016.203][PMID: 27415012] [4] Song Y-K, Park JE, Oh Y, et al. Suppression of Canine ATP Binding Cassette ABCB1 in Madin-Darby Canine Kidney Type II Cells Unmasks Human ABCG2-Mediated Efflux of Olaparib. J Pharmacol Exp Ther 2019; 368(1): 79–87 [https://doi.org/10.1124/jpet.118.250225][PMID: 30396915] [5] Zhou D, Li J, Bui K, et al. Bridging Olaparib Capsule and Tablet Formulations Using Population Pharmacokinetic Meta-analysis in Oncology Patients. Clin Pharmacokinet 2019; 58(5): 615–25 [https://doi.org/10.1007/s40262-018-0714-x][PMID: 30357650] [6] van der Kleij MBA, Guchelaar NAD, Mathijssen RHJ, et al. Therapeutic Drug Monitoring of Kinase Inhibitors in Oncology. Clin Pharmacokinet 2023; 62(10): 1333–64 [https://doi.org/10.1007/s40262-023-01293-9][PMID: 37584840]
Reference: PAGE 33 (2025) Abstr 11686 [www.page-meeting.org/?abstract=11686]
Poster: Drug/Disease Modelling - Oncology