Pemetrexed exposure predicts toxicity in advanced non-small-cell lung cancer: A prospective cohort study.

ABSTRACT

BACKGROUND: We explored whether total exposure to pemetrexed predicts effectiveness and toxicity in advanced non-small-cell lung cancer (NSCLC). Furthermore, we investigated alternative dosing schedules. METHODS: In this prospective cohort study, patients with advanced NSCLC receiving first- or second-line pemetrexed(/platinum) were enrolled. Plasma sampling was performed weekly (cyclePK) and within 24 h (24hPK) after pemetrexed administration. With population pharmacokinetic/pharmacodynamic modelling, total exposure to pemetrexed during cycle 1 (area under the curve during chemotherapy cycle 1 [AUC1]) was estimated and related to progression-free survival (PFS)/overall survival (OS). We compared mean AUC1 (mg·h/L) in patients with and without severe chemotherapy-related adverse events (AEs) during total treatment. Second, different dosing schedules were simulated to minimise the estimated variability (coefficient of variation [CV]) of AUC. RESULTS: For 106 of 165 patients, concentrations of pemetrexed were quantified (24hPK, n = 15; cyclePK, n = 106). After adjusting for prognostic factors, sex, disease stage and World Health Organisation performance score, AUC1 did not predict PFS/OS in treatment-naive patients (n = 95) (OS, hazard ratio [HR] = 1.05, 95% confidence interval [CI] 1.00-1.11; PFS, HR = 1.03, 95% CI 0.98-1.08). Patients with severe chemotherapy-related AEs (n = 55) had significantly higher AUC1 values than patients without them (n = 51) (226 ± 53 vs 190 ± 31, p < 0.001). Compared with body surface area-based dosing (CV 22.5%), simulation of estimated glomerular filtration rate (eGFR)-based dosing (CV 18.5%) and fixed dose of 900 mg with 25% dose reduction, if the eGFR<60 mL/min (CV 19.1%), resulted in less interindividual variability of AUC. CONCLUSIONS: Higher exposure to pemetrexed does not increase PFS/OS but is significantly associated with increased occurrence of severe toxicity. Our findings suggest that fixed dosing reduces interpatient pharmacokinetic variability and thereby might prevent toxicity, while preserving effectiveness.