Laurent Claret

Model-Based Prediction of Phase III Overall Survival in Colorectal Cancer on the Basis of Phase II Tumor Dynamics

PURPOSE We developed a drug-disease simulation model to predict antitumor response and overall survival in phase III studies from longitudinal tumor size data in phase II trials. METHODS We developed a longitudinal exposure-response tumor-growth inhibition (TGI) model of drug effect (and resistance) using phase II data of capecitabine (n = 34) and historical phase III data of fluorouracil (FU; n = 252) in colorectal cancer (CRC); and we developed a parametric survival model that related change in tumor size and patient characteristics to survival time using historical phase III data (n = 245). The models were validated in simulation of antitumor response and survival in an independent phase III study (n = 1,000 replicates) of capecitabine versus FU in CRC. RESULTS The TGI model provided a good fit of longitudinal tumor size data. A lognormal distribution best described the survival time, and baseline tumor size and change in tumor size from baseline at week 7 were predictors (P < .00001). Predicted change of tumor size and survival time distributions in the phase III study for both capecitabine and FU were consistent with observed values, for example, 431 days (90% prediction interval, 362 to 514 days) versus 401 days observed for survival in the capecitabine arm. A modest survival improvement of 39 days (90% prediction interval, -21 to 110 days) versus 35 days observed was predicted for capecitabine. CONCLUSION The modeling framework successfully predicted survival in a phase III trial on the basis of capecitabine phase II data in CRC. It is a useful tool to support end-of-phase II decisions and design of phase III studies.

Evaluation of Tumor-Size Response Metrics to Predict Overall Survival in Western and Chinese Patients With First-Line Metastatic Colorectal Cancer

PURPOSE To assess new metrics of tumor-size response to predict overall survival (OS) in colorectal cancer (CRC) in Western and Chinese patients. PATIENTS AND METHODS Various metrics of tumor-size response were estimated using longitudinal tumor size models and data from two phase III studies that compared bevacizumab plus chemotherapy versus chemotherapy as first-line therapy in Western (n = 923) and Chinese (n = 203) patients with CRC. Baseline prognostic factors and tumor-size metrics estimates were assessed in multivariate models to predict OS. Predictive performances of the models were assessed by simulating multiple replicas of the phase III studies. RESULTS Time to tumor growth (TTG) was the best metric to predict OS. TTG fully captured bevacizumab effect. Chinese ethnicity had no impact on OS or on the TTG-OS relationships. The model correctly predicted OS distributions in each arm as well as bevacizumab hazard ratio (model prediction, 0.75 v 0.68 observed in Western patients; 95% prediction interval, 0.62 to 0.91). CONCLUSION TTG captured therapeutic benefit with bevacizumab in first-line CRC patients. Chinese ethnicity had no impact. Longitudinal tumor size data coupled with model-based approaches may offer a powerful alternative in the design and analysis of early clinical studies.

Tumor growth modeling from clinical trials reveals synergistic anticancer effect of the capecitabine and docetaxel combination in metastatic breast cancer

PurposeMost of the cancer chemotherapy treatments employ drugs in combination. For combination treatments, it is relevant to assess interaction between two or more anticancer agents used in clinics. Based on clinical data and using modeling techniques, the work analyzes the pharmacodynamic interaction between capecitabine and docetaxel used in combination in metastatic breast cancer.MethodsWe developed mathematical models to describe tumor growth inhibition profile under treatment based on Phase II and Phase III clinical data of capecitabine and docetaxel in metastatic breast cancer. Model parameters were estimated by population approach with NONMEM® on single-agent and combination data. Simulations were performed using MATLAB.ResultsCapecitabine and docetaxel combination in metastatic breast cancer results in a synergistic effect as compared with the simple additive effects of single-agent treatments. Docetaxel is more efficient than capecitabine at the start of treatment but develops resistance faster. Modeling revealed no resistance of capecitabine for the combination data.ConclusionsModeling could be a powerful tool to design the most advantageous combination regimen for capecitabine and docetaxel in metastatic breast cancer in order to increase the time before regrowth and decrease the tumor size at regrowth.

A Modeling and Simulation Framework to Support Early Clinical Drug Development Decisions in Oncology

• J Clin Pharmacol 2011;51:6-8 T success rate of new molecules in oncology is the lowest in any therapeutic area. The high failure rate and cost associated with oncology drug development may reflect poor predictability of in vivo preclinical tumor xenograft models and lack of quantitative approaches to guide both preclinical and clinical development. Antitumor activity in early clinical studies is typically evaluated using objective response rate (ORR) or progressionfree survival (PFS). However, these estimates in typical small, noncomparative phase 1 or phase 2 trials are generally imprecise and uninformative to efficiently support go–no go decisions and design of phase 3 clinical trials. The phase 3 trial failure rate is particularly high in oncology, and there clearly is a need for more quantitative approaches to improve the success rate of oncology drugs consistent with the FDA recent initiatives. To address these issues, we are developing a drugdisease modeling framework that has been successfully applied to predict expected clinical response and survival in cancer patients in a number of clinical settings. This modeling framework focuses on efficacy, and the core of this framework is constituted by an exposure-driven tumor growth inhibition (TGI) model that uses the full longitudinal tumor size data as opposed to categorizing data, as in the calculation of ORR. Change in tumor size from baseline is used as a biomarker of drug effect to predict survival in a drug-independent survival model. It is therefore an informative and predictive patient-level continuous end point that can be assessed in early phase 1 or 2 clinical studies. The proposed modeling framework (drug-specific TGI model coupled with drugindependent survival model) can enhance learning from early clinical studies compared with the traditional approach of estimating ORR and PFS.

Fractal Volume of Drug Distribution: It Scales Proportionally to Body Mass

AbstractPurpose. To develop the physiologically sound concept of fractal volume of drug distribution, vf, and evaluate its utility and applicability in interspecies pharmacokinetic scaling. Methods. Estimates for vf of various drugs in different species were obtained from the relationship:

Modeling and simulations relating overall survival to tumor growth inhibition in renal cell carcinoma patients

vf = (v - V_{pl} )\frac{{V_{ap} - V_{pl} }}{{V_{ap} }} + V_{pl}

Prediction of overall survival or progression free survival by disease control rate at week 8 is independent of ethnicity: Western versus Chinese patients with first-line non-small cell lung cancer treated with chemotherapy with or without bevacizumab

where v is the total volume of the species (equivalent to its total mass assuming a uniform density 1g/mL), Vpl is the plasma volume of the species and Vap is the conventional volume of drug distribution. This equation was also used to calculate the fractal analogs of various volume terms of drug distribution (the volume of central compartment, Vc, the steady state volume of distribution, Vss, and the volume of distribution following pseudodistribution equilibrium, Vz). The calculated fractal volumes of drug distribution were correlated with body mass of different mammalian species and allometric exponents and coefficients were determined. Results. The calculated values of vf for selected drugs in humans provided meaningful and physiologically sound estimates for the distribution of drugs in the human body. For all fractal volume terms utilized, the allometric exponents were found to be either one or close to unity. The estimates of the allometric coefficients were found to be in the interval (0,1). These decimal values correspond to a fixed fraction of the fractal volume term relative to body mass in each one of the species. Conclusions. Fractal volumes of drug distribution scale proportionally to mass. This confirms the theoretically expected relationship between volume and mass in mammalian species.

Nonparametric density estimation applied to population pharmacokinetics

The pharmacokinetics of cyclosporin (CsA) are unusual because of several heterogeneous features which include the presence of more than one conformer, considerable accumulation in erythrocytes and lipoproteins, extensive plasma protein binding, distribution into deep tissues, biliary secretion and hepatic clearance involving a large number of metabolites. In this study, a stochastic compartmental model was developed to describe the heterogeneous elimination kinetics of CsA. This new approach relies on a probabilistic transfer model with a gamma distributed probability intensity coefficient for drug elimination. For comparative purposes both the stochastic model and compartmental deterministic models were fitted to real post infusion data from patients receiving CsA as a 2-hr intravenous infusion. The criteria for selecting the best model showed that the stochastic model, although simpler than the compartmental deterministic models, is more flexible and gives a better fit to the kinetic data of CsA than the compartmental deterministic models. The stochastic model with a random rate intensity coefficient adequately describes the heterogeneous pharmacokinetics of CsA.