Modelling change in tumour size, survival and new lesions appearance in patients with ovarian cancer treated with carboplatin monotherapy or in combination with gemcitabine
Chiara Zecchin (1), Ivelina Gueorguieva (1), Nathan H. Enas (2), Lena E. Friberg (3)
(1) Lilly Research Laboratories, Sunninghill Road, Windlesham, Surrey, UK, (2) Lilly Research Laboratories, Indianapolis, Indiana, USA , (3) Department of Pharmaceutical Biosciences, Uppsala University, Sweden
Objectives: Change in tumour size (CTS) is a marker of cytotoxic drug effects and there is growing interest in using this metric as primary endpoint [1], allowing earlier evaluation of treatment outcome compared to conventional metrics such as overall survival (OS). The objective of this study is to develop a model to quantify CTS during therapy and to investigate the predictive value of CTS, lesions location on trial enrolment and time of new lesion appearance on OS in metastatic ovarian cancer (MOC).
Methods: Data from a Phase III randomized study, comparing the efficacy of gemcitabine plus carboplatin versus carboplatin monotherapy in patients with recurrent MOC, was available for analysis [2]. The database included 336 patients, (173 followed up until death, 163 censored). A modelling approach was applied to characterise the CTS time course, evaluating several exposure measures to describe drug effects. Parametric time-to-event (TTE) models were investigated to predict appearance of metastasis, OS and dropout probability as functions of CTS and other covariates.
Results: The CTS model [3,4,5] successfully described the data. Resistance to treatment was however not statistically significant and the two drugs promoted tumour shrinkage with independent additive effects. Drug exposure was incorporated as the per-cycle AUC predicted from the doses and literature PK models [6,7]. Metastasis appearance, OS and dropout probabilities were described using parametric TTE models, with a Weibull hazard increasing with time. Two time-varying covariates, describing tumour evolution during treatment, were included in the OS model: the predicted relative CTS up to week 12 (and thereafter rCTS at week12), and the appearance of new lesions. Other included (time-constant) covariates were tumour size and ECOG status at baseline. The rCTS at the end of the first treatment cycle was a significant predictor in the metastasis appearance model.
Conclusions: Metrics from the developed CTS model, quantifying the effect of carboplatin monotherapy and when combined with gemcitabine, could successfully predict metastasis appearance and OS probability in MOC. In addition to appearance of new lesions, predicted rCTS(t) up to week 12 was a significant predictor of OS probability and better than rCTS at fixed time points such as week 6 or 8. Predicted rCTS after first treatment cycle was the best predictor for appearance of new metastasis.
This work is supported by the DDMoRe project.
References:
[1] Jaki T et al. “Designing exploratory cancer trials using change in tumour size as primary endpoint." Stat Med 32(15):2544-54, 2013.
[2] Pfisterer J et al. “Combination therapy with gemcitabine and carboplatin in recurrent ovarian cancer” Int J Gynecol Cancer (2005) 15(S1):36-41
[3] Claret L et al. “Model-based prediction of phase III overall survival in colorectal cancer on the basis of phase II tumor dynamics.” J Clin Oncol (2009) 27(25):4103-8
[4] Claret L et al. “Population pharmacokinetic/pharmacodynamic modeling for the time course of tumor shrinkage by motesanib in thyroid cancer patients.” Cancer Chemother Pharmacol (2010) 66(6):1151-8
[5] Houk BE et al. “Relationship between exposure to sunitinib and efficacy and tolerability endpoints in patients with cancer: results of a pharmacokinetic/pharmacodynamic meta-analysis.” Cancer Chemother Pharmacol (2010) 66(2):357-71
[6] Lindauer A et al., “Population pharmacokinetics of high-dose carboplatin in children and adults.” Ther Drug Monit, (2010) 32(2): 159-68
[7] Zhang L et al., “Model-based drug development: the road to quantitative pharmacology.” J Pharmacokinet Pharmacodyn (2006) 33(3):369-93
