2014 - Alicante - Spain

PAGE 2014: Drug/Disease modeling - Other topics
Mats Magnusson

A Population PK/PD Analysis of Nebivolol and Valsartan Combination Therapy

Mats O. Magnusson (1), E. Niclas Jonsson (1), Chun Lin Chen (2), Timothy Carrothers (2), Parviz Ghahramani (2)

(1) Pharmetheus, Uppsala, Sweden (2) Forest Research Institute, Jersey City, NJ, USA

Objectives: To develop population PK/PD models that describe the effects of placebo, nebivolol and valsartan as single drugs and as a fixed-dose combination (FDC) on both sitting cuff-measured blood pressure (BP) and 24-hour ambulatory measured BP (diastolic and systolic BP), and to use the model to further simulate possible combinations of dosages.

Methods: Population PK/PD models were developed in NONMEM V7.2.0 [1] based on data from the Phase 3 study (NAC- MD-01), utilizing estimates of exposure for each individual derived from a population PK model. The exposure-response population included 761 patients for the sitting cuff-measured BP and 746 patients for ambulatory measured BP. The BP lowering drug effects of nebivolol and valsartan were described by inhibitory Emax models with a parameter (alpha) to account for the interaction between the compounds. The diurnal rhythm in the ambulatory BP data was described by the sum of cosine functions [2].

Results: For the sitting cuff-measured BP model, total effect was comprised of the drug effect and placebo effect. There was no detectable placebo response in the ambulatory BP data. The maximum drug effects were 13 mmHg for sitting cuff-measured BP, and 14 and 19 mmHg for diastolic and systolic ambulatory measured BP, respectively. The exposure-response relationships were statistically significant in the valsartan and nebivolol monotherapy arms as well as in all FDC arms, suggesting increasing efficacy with increasing PK exposures. The alpha parameter was estimated to be positive in all sitting cuff-measured and ambulatory measured BP models, and this effect was further characterized as partially additive.

The PK/PD model was used to simulate possible dosages in mono-, and combination therapy, including that of dosing strengths that were not studied in the NAC-MD-01 study. The model-predicted change from baseline in diastolic and systolic BPs demonstrated that all studied FDC doses (5/80 mg, 5/160 mg, 10/160 mg, 10/320 mg, 20/320 mg, and 40/320 mg) had greater BP reductions compared to their monotherapy components in a partially additive manner. An exposure response relationship was also evident between the FDC dose groups.

Conclusions: Modeling and simulation of nebivolol and valsartan combination therapy demonstrate a partially additive effect on diastolic and systolic sitting cuff-measured BP as well as on diastolic and systolic 24-hour ambulatory measured BP among all FDC doses studied.

[1] Beal SL, Sheiner LB, Boeckmann AJ & Bauer RJ (Eds.) NONMEM Users Guides. 1989-2011. Icon Development Solutions, Ellicott City, Maryland, USA.
[2] Hempel G, Karlsson MO, de Alwis DP, Toublanc N, McNay J and Schaefer HG, 1998, Population pharmacokinetic-pharmacodynamic modeling of moxonidine using 24-hour ambulatory blood pressure measurements. Clin Pharmacol Ther vol. 64: 622–635.

Reference: PAGE 23 (2014) Abstr 3162 [www.page-meeting.org/?abstract=3162]
Poster: Drug/Disease modeling - Other topics
Click to open PDF poster/presentation (click to open)