S. Chabaud(1,2), P. Girardd(1,2), P. Nony(1,2), JP. Boissel(1,2) and THERMOS group(3)
1 Service de Pharmacologie Clinique, EA643, Lyon , France; 2 THErapeutic MOdelling and Simulation group: SPC Lyon, France; IRIS, Servier, Courbevoie, France; Department of applied Mathematics and Informatics, Ghent, Belgium; School of Pharmacy and Pharmaceutical Sciences, University of Manchester, G.B.
Introduction: Monte Carlo simulation is used for determining the dose and number of subjects needed for a phase III clinical trial of a new bradycardic agent. Surrogate effects and clinical endpoints are used in the model.
Objectives: Ivabradine (S16257) is a new bradycardic agent with potential effect on stable angina pectoris. To derive the best compromise between efficacy, safety and drug regimen, we developed a full therapeutic model linking pharmacokinetics (PK), pharmacodynamics (PD) and clinical effect (CE) models. This model allows to compute the reduction in number of occurrences of chest pain in treated patients and the number of patients to include in a phase III clinical trial, and to perform Monte-Carlo simulations.
Methods : The therapeutic model connects the onset of chest pain at time t, to doses administrated up to this time. The binary clinical outcome is simulated using a physiologic model where the cardiac reserve is derived from heart rate (HR), the surrogate outcome, and compared to a theoretical threshold of angina randomly allocated to each subject [1]. Safety as well is defined as HR dependent, not falling below a randomly chosen threshold. A PK-PD model, previously developed using phase I data, is used for simulating the decrease in HR consecutive to 5 oral ivabradine doses (2.5, 5, 10, 20, and 40 mg) given once (OID) or twice daily (BID) compared to no treatment at all [2,3]. New individual PK-PD parameters are sampled from normal distributions. The model is used to alter real 24h ambulatory HR profiles sampled from the OCTAVE2 epidemiological database [4]. One hundred randomised parallel clinical trials (100 patients with ivabradine and 100 patients untreated) are simulated. The result of each trial is expressed as the percentage of patients presenting at least one clinical event during the 24h period in the 2 groups.
Results: Up to 10 mg OID, only 25% of the simulated trials show a significant effect of ivabradine. Efficacy is shown in 48% and 55% of the simulated trials for the 10 mg BID and 20 mg OID doses respectively. With 40 mg daily dose or higher, efficacy is shown in more than 80% of the simulated trials. For safety, 4 % of patients had at least one adverse event in the untreated group. This percentage increases respectively from 5 to 13% in the treated groups from the lowest to highest doses. 10%, 27% and 52% of the simulated trials show a significant increase in adverse effects with 20, 40 and 80 mg daily doses, respectively. Based on this sole simulation, the number of subjects to include in the future trial, in order to show a 16% decrease in episodes of chest pain under the assumption of 68% basal risk, would be 239 subjects by group with 10 mg BID or 196 with 20 mg OID for a 18% decrease of the event rate. This number has the same magnitude as number of patients that are usually included in this type of trials.
Conclusion: That work illustrates the potential interest of introducing in clinical trial simulations a PK-PD model as well as a physiopathologic model describing the relationship between a surrogate criteria and the clinical outcome using a mechanistic approach.
References:
1. F Kappel, RO Peer. A mathematical model for fundamental regulation processes in the cardiovascular system. J Math Biol, 31: 611-31, 1993.
2. I Ragueneau, C Laveille, R Jochemsen, G Resplandy, C Funck-Brentano, P Jaillon. Pharmacokinetic-pharmacodynamic modeling of the effects of ivabradine, a direct sinus node inhibitor, on heart rate in healthy volunteers. Clin.Pharmacol.Ther. 64:192-203, 1998.
3. S Chabaud, P Girard, P Nony , C. Laveille, JP Boissel. PK-PD model of ivabradine, a bradycardic agent, an its metabolite in healthy volunteers. PAGE 1998, Wuppertal Germany, June 19-20 1998.
4. F. Gueyffier, F. Boutitie, C. Cornu, G. Julien, P. Poncelet, A. Sebaoun et al. Présentation d’Octave II. Une étude épidémiologique en cours en France sur la valeur pronostique ajoutée par la mesure ambulatoire de la pression artérielle. Arch Mal Coeur Vaiss. 89:1381-1388, 1996.
Supported by grant BIOMED PL 962640 from E.C..
Reference: PAGE 8 () Abstr 74 [www.page-meeting.org/?abstract=74]
Poster: oral presentation