Dong Woo Chae (1,2), Mijeong Son (1,2), Yukyung Kim (1,2), Hankil Son (1,2), Nick Holford (3), Kyungsoo Park (1,2)
(1) Department of Pharmacology, Yonsei University College of Medicine, Seoul, Korea (2) Brain Korea 21 Plus Project for Medical Science, Yonsei University, Seoul, Korea (3) Department of Pharmacology and Clinical Pharmacology, University of Auckland, New Zealand
Objectives: Telmisartan, a potent angiotensin II type-1 (AT1) receptor blocker (ARB), is indicated for the treatment of essential hypertension. Its effect is classified into rapid response due to vasodilation and slow response due to volume loss steming from natriuresis [1]. PK-PD models of telmisartan have been published in rats [2,3], but no such model reported in human. This study aimed to develop a mechnitic model of telmsartan drug effect in human using non-invasive markers, which can be readily applicable in real clinical situations.
Methods: Data was acquired from a previous PK study where telmisartan 80 mg was given once daily for 6 days. Systolic and diastolic blood pressure and heart rate (HR) were measured before dosing for Day 1 to 5 and serially after the last dose. PK was modeled with weight incorporated using allometry. With posthoc estimates of PK parameters, a mechanistic PD model was built considering followings: (i) MAP formulated as MAP = C·TPR·PP·HR, based on MAP = CO·TPR, CO = SV·HR, and C = SV/PP, with CO, TPR, SV and C being cardiac output, total peripheral resistance, stroke volume and compliance, respectively, where C = SV/PP is a simplified version of Windkessel model and C assumed to be constant over the study period; (ii) negative feedback of MAP on turnover rates of PP, HR and TPR; (iii) a common circadian rhythm with MAP, HR and CO in parallel fluctuations and TPR in the mirror image of the others; (iv) drug plasma concentration linked through a first-order process to Cb, the effect site concentration binding to AT1 receptor; (v) drug effect, a function of Cb, assumed to inhibit turnover rates of TPR (rapid effect) and PP (slow effect), with no influence on HR. The prediction of MAP was computed using the estimates of C·TPR, PP and HR, where C·TPR was estimated from ‘derived’ data. The model was validated using a visual predictive check (VPC) and analyses were made using NONMEM 7.3.
Results: Two compartment model with 1st order absorption and lag time was chosen for PK. Feedback of MAP was negligible in PP and HR, and an inhibitory Imax model successfully described the time courses of C·TPR, PP and HR (and MAP accordingly). VPC showed 95% of observations lied within 95% prediction interval.
Conclusions: This work demonstrated the feasibility of using non-invasive cardiac indices in deriving a mechanistic model of telmisartan effect in human. The developed model can be applied to antihypertensive drugs other than ARBs.
References:
[1] Wolfgang Wienen et al., A Review on Telmisartan: A Novel, Long-Acting Angiotensin II-Receptor Antagonist. Cardiovascular Drug Reviews, Vol. 18, No. 2, pp 127-154, 2000
[2] Kun HAO et al., Pharmacokinetic-pharmacodynamic modeling of telmisartan using an indirect response model in spontaneously hypertensive rats, Acta Pharmacol Sin 2007 May; 28(5): 738-743
[3] N Snelder et al., PKPD modeling of the interrelationship between mean arterial BP, cardiac output and total peripheral resistance in conscious rats, British Journal of Pharmacology (2013) 169 1510-1524
Reference: PAGE 24 () Abstr 3640 [www.page-meeting.org/?abstract=3640]
Poster: Drug/Disease modeling - Other topics