Francis W. Ojara (1,2), Andrea Henrich (2), Niklas Hartung (3), Wilhelm Huisinga (3), Markus Joerger (4), Charlotte Kloft (1)
(1) Dept. Clinical Pharmacy & Biochemistry, Institute of Pharmacy, Freie Universitaet Berlin, Germany, (2) Graduate Research Training Program PharMetrX, Germany, (3) Institute of Mathematics, University of Potsdam, Germany, (4) Medical Oncology and Clinical Pharmacology, Dept. of Internal Medicine, Cantonal Hospital St. Gallen, Switzerland
Objectives: Peripheral neuropathy (PN), a dose-limiting paclitaxel-related toxicity, affects >20% of patients on paclitaxel (PTX) therapy and negatively impacts quality of life. The risk of PN was shown to increase with higher PTX dose and exposure (time of plasma concentration above 0.05 μM), by comparing odds of PN for patients at different dose or exposure levels [1,2]. Using parametric time-to-event analysis, accounting for time of occurrence of PN on the risk of PN, we established the relationship between PTX dose and the risk of PN, grades 2 or 3 (PN2+), based on data from the CEPAC-TDM study [3,4]. In this analysis we extend the established time-to-event analysis framework to evaluate the impact of paclitaxel exposure on the risk of PN2+ to support dose adaptation and hence reduce the occurrence of PTX-associated PN.
Methods: Patients (n=365) from the CEPAC-TDM study, who received
3-weekly PTX dosing for ≤ 6 cycles either with standard 200 mg/m2 PTX or a PK/PD-guided dosing approach were included [4]. Carboplatin or cisplatin were co-administered. PTX PK data was available in the PK/PD-guided dosing arm (n=183). PN symptoms, severity, start and end dates were recorded and classified using the common toxicity criteria (version 4.0) [5]. The risk of 1st-occurrence of PN2+ during treatment was described using a cycle-varying hazard model, including impact of covariates, PTX dose, age, weight, sex, and smoking status [3]. In this analysis, PTX exposure (instead of dose), age, sex, and smoking status were jointly evaluated in a full covariate model (FCM). Since the BSA-guided dosing arm had no PTX PK, PTX exposure metrics (AUC(0-∞,cycle), TC>0.01/ 0.05/ 0.1 μM) was derived using the multiple imputation (MI) approach [6]. 50 sets of PTX exposure metrics were simulated using a PTX PK model [7] and the CEPAC-TDM dataset. For each set, FCM parameters were evaluated. FCM parameters were averaged across imputations [6]. Assessment for the PTX exposure metric that provided the best improvement in model fit (ΔOFV) was first based on FCM evaluation with single imputated PTX PK. The PTX dose-exposure-PN2+ relationship was subsequently explored by simulating incidence of PN2+ for three clinically relevant dosing schedules: 200 mg/m2 and 175 mg/m2 (both 3-weekly, 6 cycles, i.e. q3w) and 80 mg/m2 (weekly for 3 weeks and a week off, 6 cycles i.e. qw). The endpoint of interest was incidence of PN2+ with the different dosing schedules. Randomness of the TTE model and parameter uncertainty were the levels of variability included. Dataset formatting was performed in R (3.4.3) and TTE analysis in NONMEM (7.3.0).
Results: PTX exposure data from MI in the PK/PD-guided dosing arm were closely aligned with estimated PTX exposure based on the PTX concentrations observed in the PK/PD-guided dosing arm: the estimated and typical exposure from single imputation largely fell within the interquartile range of MI PTX exposure, showing reliability of the imputation procedure. Amongst the investigated PTX exposure metrics AUC(0-∞,cycle) was the most predictive of PN2+. A 17.1% increase in risk of PN2+ with change in PTX AUC[0-∞, cycle] from 7.34 (P0.025) to 24.6 (P0.975) µM.h was predicted. The 80 mg/m2 (qw) was associated with higher risk of PN2+ compared to 200 mg/m2 (q3w). A 38% increase in risk of PN2+ with 80 mg/m2 (qw) over 200 mg/m2 was predicted: hazard ratio [95% CI] of 1.38 (1.05, 1.78). The risk of PN2+ increased with increase in dose for the 3-weekly dosing i.e. an 11% increase in risk of PN2+ with 200 mg/m2 over 175 mg/m2 was predicted: hazard ratio [95% CI] of 1.11 [1.03, 1.24].
Conclusions: We established a quantitative relationship between PTX exposure and the risk of 1st-occurrence of PN2+, accounting for effects of age, sex and smoking status. Using the multiple imputation approach enabled exposure-PN2+ evaluation for the entire dataset, including patients with no PK data. We found that PTX AUC(0-∞,cycle) described better the occurrence of PN2+ compared to TC>0.05 μM, reported in literature [2]. Weekly PTX dosing i.e. 80 mg/m2 (qw) was associated with a significantly higher risk of PN2+ compared to 3-weekly PTX dosing i.e. 200 mg/m2 (q3w), whereas within the 3-weekly dosing schedule the risk of PN2+ increased with increase in PTX dose. The developed model enables quantification and comparison of the individual risks of PN2+ for commonly used PTX dosing regimens apriori for decision making.
References
[1] A. Argyriou et al. Crit. Rev. Oncol. Hematol. 66: 218 (2008)
[2] S. Zhang et al. Cancer Tansl. Med. 2: 131-136 (2016)
[3] F.W. Ojara et al. PAGE 27: 8610 [www.page-meeting.org/default.asp?abstract=8610], (2018)
[4] M. Joerger et al. Ann. Oncol. 27: 1895 (2016)
[5] National Institute of Health, National Cancer Institute, Common Terminology Criteria for Adverse Events (CTCAE) (2009)
[6] A.M. Johansson et al. AAPS J. 15: 1035-1042 (2013)
[7] A. Henrich et al. J Pharmacol Exp Ther. 362: 347-358 (2017)
Reference: PAGE 28 (2019) Abstr 9054 [www.page-meeting.org/?abstract=9054]
Poster: Drug/Disease Modelling - Oncology