R.J. Boosman 1, T.P.C. Dorlo 1, N. de Rouw 2,3, M.M. van den Heuvel 2, A.C. Dingemans 4,5, L.E.L. Hendriks 4, B. Biesma 3, J.G.J.V. Aerts 5, R.H.J. Mathijssen 5, J.A. Burgers 1, A.D.R. Huitema 1,6, R. ter Heine 2
1 The Netherlands Cancer Institute, Amsterdam, the Netherlands 2 Radboud University Medical Center, Nijmegen, the Netherlands 3 Jeroen Bosch Hospital, ‘s-Hertogenbosch, the Netherlands 4 Maastricht University Medical Center, Maastricht, the Netherlands 5 Erasmus Medical Center, Rotterdam, the Netherlands 6 University Medical Center Utrecht, Utrecht, the Netherlands
Introduction: Pemetrexed is a widely used cytostatic agent for the treatment of non-small cell lung cancer (NSCLC) and mesothelioma. Dosing of pemetrexed is currently based on body surface area (500 mg/m2 every 3 weeks) although renal function is the main determinant of systemic exposure. Pemetrexed is currently contraindicated in patients with a creatinine clearance <45mL/min and, therefore, a large proportion of patients (about 30%) is withheld effective treatment. Previously, the linear relationship between pemetrexed plasma concentration and inhibition of proliferating neutrophils has been demonstrated, indicating that administration of a dose adjusted to renal function, might be a safe dosing strategy in patient with renal dysfunction [2]. We recently included three patients in a study evaluating a novel renal function based dosing regimen for pemetrexed in patients with renal dysfunction (renal functions of 35, 36 and 8 mL/min). However, we observed a grade 1, 3 and 4 neutropenia, respectively, in these patients, indicating that the exposure-toxicity relationship might not be linear. To allow safe dosing of pemetrexed in this patient group, it is pivotal to unravel the exposure-toxicity relationship of pemetrexed.
Objective: To investigate the pemetrexed exposure-neutropenia relationship.
Methods: Phase I pharmacokinetic (PK) data of pemetrexed and pharmacodynamic (PD) data on neutrophil counts after pemetrexed exposure were provided by Eli Lilly and supplemented with data from our renal impairment study. A previously developed population PK model with renal function as covariate on clearance described the plasma concentrations. A chemotherapy-induced myelosuppression model was used to model the neutrophil data [1]. The drug effect on the proliferation of the neutrophils was modelled as previously suggested with a linear relation (slope) between drug concentration and proliferation rate [2]. Nonlinear mixed effects modelling was performed using NONMEM (v. 7.3). A posterior predictive check of the model was done using the neutrophil counts as reported in the phase I study of pemetrexed, where pemetrexed was given at a dose of 0.2-5.2 mg/m2 for 5 consecutive days in a Q3W cycle in patients with adequate renal function [3].
Results: The final dataset included 555 pemetrexed plasma concentrations and 1514 neutrophil counts from 109 patients with estimated glomerular filtrations ranging from 8 to 155 mL/min. Typical value for baseline neutrophil count (BANC) and mean transit time (MTT) including relative standard error (RSE) were 5.32*109/L (4%) and 103h (4%). For the dose stimulus (DS; a linear proportionality constant relating the pemetrexed concentration to its cytotoxic effect on the proliferation) and feedback parameter (FP) typical values (RSE) were 0.221 L/mg (9%) and 0.166 (7%). Vitamin B11 and B12 supplementation led to a decrease of 4.3% and 28.9% in MTT and DS, respectively. The corresponding between-patient variabilities were: BANC 37.7%, MTT 19.0%, DS 54.9% and FP 27.0%. The previously demonstrated slope model [2] fitted the PD data for patients with adequate renal function well, indicating that there is sufficient internal validity for this model regarding this patient group. The hematological toxicity observed in patients with impaired renal function was, however, not adequately captured. According to the model, when the dose of pemetrexed is adjusted to target the typical area under the concentration time curve (AUC), the probability of neutropenic events to occur is low. However, as stated above, all three patients in our clinical study developed some grade of neutropenia. In addition, the slope model did not predict the development of neutropenia (≥ grade 1) as described in the phase I study (83%, n=6) with a dose of 4.0 mg/m2 for 5 consecutive days, that was established as the maximum tolerated dose. In simulations, the model predicted a low probability (0%, n=50) for development of neutropenia for this pemetrexed dose regimen.
Conclusions: There is weak external validity for the previously suggested exposure-neutropenia relationship for pemetrexed in patients with a prolonged (5 days), yet low, exposure to pemetrexed, e.g. in patients with impaired renal function or patients with a low daily dose. We are now developing a more mechanistic model to enable the prediction of pemetrexed toxicity in patients with impaired renal function to develop a safe dosing regimen in this vulnerable patient group.
References:
1. Friberg LE, Henningsson A, Maas H, Nguyen L, Karlsson MO. Model of chemotherapy-induced myelosuppression with parameter consistency across drugs. J Clin Oncol. 2002; 20(24):4713-21
2. Latz JE, Schneck KL, Nakagawa K, Miller MA, Takimoto CH. Population pharmacokinetic/pharmacodynamic analyses of pemetrexed and neutropenia: effect of vitamin supplementation and differences between Japanese and Western patients. Clin Cancer Res. 2009; 15(1): 346-54
3. McDonald AC, Vasey PA, Adams L, Walling J, Woodworth JR, Abrahams T et al. A phase I and pharmacokinetic study of LY231514, the multitargeted antifolate. Clin Cancer Res. 1998; 4(3):605-10
Reference: PAGE () Abstr 9585 [www.page-meeting.org/?abstract=9585]
Poster: Oral: Drug/Disease Modelling