Alejandro Pérez-Pitarch (1), Beatriz Guglieri-López (1), Virginia Merino (1), Amparo Nacher (1), Rafael Ferriols-Lisart (2), Matilde Merino-Sanjuán (1).
(1) Department of Pharmacy and Pharmaceutical technology, Faculty of Pharmacy, University of Valencia. Valencia. Spain. (2) Pharmacy Department. University Clinical Hospital of Valencia. Valencia. Spain.
Objectives: To develop a model describing the time-to-progression distribution of a sample of non-small-cell lung cancer (NSCLC) patients treated with erlotinib. To identify covariates as progression-free survival predictors (including drug exposure).
Methods: Time-to-progression data and relevant covariates were available from 26 NSCLC patients treated with erlotinib at the University Clinical Hospital of Valencia. Erlotinib plasma concentration data were simulated for the 26 NSCLC patients using a previously published pharmacokinetic (PK) model [1]. Time-to-progression was characterized with a time-to-event (TTE) model. TTE modelling and PK simulation was performed in NONMEM v 7.1.0 [2]. Exponential and Weibull TTE distributions were tested. Covariates, such as age, gender, NSCLC subtype, presence and location of metastases and tumour mutations, were evaluated as progression-free survival predictors and inclusion was performed on basis of objective function value (OFV) decrease and graphical improvement of visual predictive check diagnostic graphics.
Results: The Weibull distribution model fitted the observed data significantly better than the exponential distribution model (ΔOFV=-6.13).
h0(t) = λ α ( λ t ) α – 1
h0(t): hazard at time t; λ: scale parameter; α: shape parameter.
Among the tested covariates, the following were included in the model in a stepwise manner: 1) epidermal growth factor receptor (EGFR) mutation on the shape parameter (α) (ΔOFV=-27.33); 2) erlotinib minimum plasmatic concentration on hazard (h) for patients with mutated EGFR (ΔOFV=-4.36); 3) presence of central nervous system metastases on the scale parameter (λ) (ΔOFV=-5.08). The model that best described the relationship between erlotinib minimum plasma concentration (Cmin) and progression hazard was:
ha(t) = h0·(1-(Cmin/(Cmin + CE50))
ha(t): hazard depending on drug exposure; CE50: minimum plasma concentration to achieve 50% of the maximum effect.
Conclusions: The described model supports therapeutic drug monitoring of erlotinib based on the evidenced relationship between drug trough concentrations and progression hazard. It is concluded that TTE modelling of disease progression has the potential to improve the efficacy of NSCLC treatment with erlotinib.
References:
[1] Lu JF, Eppler SM, Wolf J, Hamilton M, Rakhit A, Bruno R and Lum BL. Clinical pharmacokinetics of erlotinib in patients with solid tumors and exposure-safety relationship in patients with non-small cell lung cancer. Clin Pharmacol Ther 2006; 80: 136-45.
[2] Beal SL SL, Boeckmann AJ & Bauer RJ (Eds.) NONMEM Users Guides. Icon Development Solutions: Ellicott City, Maryland, USA, 1989-2011.
Reference: PAGE 24 () Abstr 3365 [www.page-meeting.org/?abstract=3365]
Poster: Drug/Disease modeling - Oncology