The pharmacological potency of various AT 1 antagonists assessed by Schild regression technique in man

Transcription

1 Paper Keywords: angiotensin II antagonists, dose ratio, Schild regression technique, candesartan, irbesartan, losartan, telmisartan, valsartan * Centre for Cardiovascular Pharmacology, ZeKaPha GmbH, Mainz, Germany Takeda Pharma GmbH, Aachen, Germany Correspondence to: Professor Gustav G Belz, Centre for Cardiovascular Pharmacology, ZeKaPha GmbH, Mathildenstr. 8, D Mainz, Germany Tel: Fax: Accepted for publication 12th December 2000 JRAAS 2000;1: The pharmacological potency of various AT 1 antagonists assessed by Schild regression technique in man Gustav G Belz,* Kerstin Breithaupt-Grögler,* Raunhild Butzer,* Winfried Fuchs, Christian Hausdorf,* Christian Mang* Abstract Rationale A quantitative technique was used to compare the pharmacological potency in healthy volunteers of angiotensin II receptor antagonists (AIIA): candesartan cilexetil, losartan, irbesartan, valsartan, and telmisartan. Methods In a randomised, double-blind, parallel-group (4x12 subjects) study, single oral doses of candesartan cilexetil 4, 8 and 16 mg, losartan potassium 25, 50 and 100 mg, valsartan 40, 80 and 160 mg, and irbesartan 75, 150 and 300 mg were administered on three consecutive days. Telmisartan 20, 40 and 80 mg was similarly evaluated in 12 volunteers in an open amendment. Angiotensin II (Ang II) antagonistic effects were determined in vivo from rightward shifts in Ang II dose-response curves for diastolic blood pressure (BP) and dose ratios were calculated. Apparent K i -doses, i.e. doses (in mg) required to induce a two-fold shift in Ang II dose-response curves (equivalent to approx. 50% blockade of receptors) were determined, using Schild regression analysis. Results All treatments dose-dependently attenuated increases in diastolic BP induced by infusion of exogenous Ang II. Candesartan cilexetil appeared to have a more pronounced increase in effect following cumulative dosing. At 24 hours, apparent K i -doses were: candesartan cilexetil 6 mg, irbesartan 123 mg, valsartan 93.5 mg, and telmisartan 54 mg. It was not possible to determine an apparent K i -dose for losartan at 24 hours. Conclusion Consistent with results from experimental pharmacology, candesartan cilexetil displayed the highest pharmacological potency (i.e. antagonistic activity per mg substance) of the AIIAs tested. Apparent K i -doses at 24 hours were within the dose range recommended for clinical use in patients with hypertension. Introduction Blockade of the effects of Ang II in the reninangiotensin-aldosterone system (RAAS) remains one of the major targets for therapeutic intervention in cardiovascular disease. The RAAS is implicated in the pathophysiology of hypertension as well as in end organ damage such as left ventricular hypertrophy (LVH). 1 converting enzyme inhibitors (ACE-I) and, more recently, Ang II AT 1 -receptor antagonists (AIIAs) have been extensively studied. 2 Drugs affecting the RAAS, such as ACE-I and possibly AIIAs, may be considered as first line therapy for the treatment of hypertension. Unlike ACE-I, which are associated with dry cough and, in rare cases, angioneurotic oedema,aiias are reported to be very well-tolerated in clinical use, with a safety profile similar to that of placebo. 3 9 Although there are many similarities between the prototype AIIA, losartan potassium, and newer members of this class, important differences exist between individual drugs, not least in their pharmacological and pharmacokinetic properties. 10 Moreover, clinical experience with the AIIAs has highlighted a number of distinguishing features, for example dose-related blood pressure (BP) reduction with candesartan cilexetil 7,11,12 and irbesartan, 9 but not with losartan 5 or valsartan. 8 The degree of BP reduction achieved with AIIAs may also depend on the choice of agent, with evidence of superior efficacy for newer drugs Several in vivo studies have investigated the Ang II antagonistic properties of AIIAs, directly comparing their effects on parameters such as diastolic BP and receptor occupancy in the presence of increasing doses of exogenous Ang II, 16 and variability of pharmacokinetic-pharmacodynamic response. 17 The study by Belz et al. 16 in healthy volunteers demonstrated greater potency and more prolonged antagonism of the BP-lowering effects of Ang II with irbesartan vs. valsartan or losartan after single and repeated dosing. These findings were supported by radioreceptor assay data. Candesartan cilexetil has also been shown to have more prolonged effects than losartan in clinical use, 7,18 compatible with its tight binding and slow dissociation from AT 1 -receptors. 19 In a recent study in healthy volunteers, 20 the Ang II antagonistic effects and radioreceptor kinetics of candesartan were evaluated using Schild regression.this classical procedure compares agonist dose-response curves before and after dosing of an antagonist. 21 Typically, dose-response curves are shifted to the right in the presence of the antagonist and calculation of dose ratios allows this effect to be quan- 336

2 tified. 22,23 The study by Malerczyk et al. 20 confirmed the high potency of candesartan at AT 1 -receptors and supported clinical observations of prolonged activity (sustained 24-hour BP reduction). The Schild technique has been used successfully in several in vivo settings in man to characterise the effect and duration of the antagonistic properties of drugs The present study in healthy volunteers was conducted to compare the Ang II antagonistic effects of various AIIAs: candesartan cilexetil, losartan, valsartan, irbesartan and telmisartan. The BP response to infusion of exogenous Ang II was determined for incremental oral doses of each compound. Methods The study was carried out in accordance with the Declaration of Helsinki and reviewed and approved by an independent ethics committee.all volunteers gave their written informed consent. Design and study procedures A total of 61 healthy male volunteers participated in this clinical trial,with 60 individuals included in the efficacy analyses.volunteers had a mean (±SD) age 26.5±2.7 years, weight 77.5±8.5 kg, and height 182.6±6.3 cm. The study had a randomised, double-blind parallel-group design. Single, escalating, oral doses of candesartan cilexetil 4, 8, and 16 mg, irbesartan 75, 150, and 300 mg, losartan 25, 50, and 100 mg, or valsartan 40, 80, and 160 mg, were administered on three consecutive days in a fourway schedule (12 subjects per group).in an amendment to the original study protocol, telmisartan 20, 40, and 80 mg was administered in open fashion in 12 subjects, using the same investigational procedures as in the main double-blind study. Dosages were chosen according to information provided by European package inserts, encompassing the range from lowest to highest recommended doses for the treatment of hypertension. Evaluations The pre-study screening included informed consent, medical history, physical examination, ECG and laboratory analysis, as well as checks for inclusion and exclusion criteria. On study days, volunteers arrived at the study centre in the morning, having fasted the previous night, and remained fasting until completion of the 4-hour measurements. They remained supine until completion of the 4- hour measurements. Between 4 hours and 8 hours after dosing of the AIIAs, subjects were ambulatory within the study centre, after which they were dismissed until the next morning following formal assessment of their physical status. Subjects received cumulative oral doses of the allocated AIIAs on three consecutive study days.ang II was infused in stepwise increasing doses over the range µg/minute to obtain dose effect curves of diatolic BP. Each dose step was applied for 3 minutes. These dose effect curves were established before the first oral dosing and at four, eight and 24 hours after the first, second and third doses. Time points relating to the first dose of an AIIA on the first day were 0, 4, 8, 24, 28, 32, 48, 52, 56 and 72 hours. Ang II dosing was not further increased when systolic or diastolic BP had increased by 25 mmhg or when diastolic BP was >110 mmhg, and/or heart rate fell below 40 beats/minute or the dose of 20 µg/minute Ang II had been reached. 20,28 Physical examination, ECG and laboratory analysis were repeated at post-study follow-up. Angiotensin II dose-response curves Infusion of Ang II is known to induce a dosedependent increase in diastolic and systolic BP. Individual dose-response curves may be derived by plotting the time course of BP vs. dose of Ang II. In previous studies, pressor responses to Ang II were altered by AIIAs, such that a higher dose of Ang II was required to induce the same increase in BP, thus dose-response curves were shifted to the right. 20,28 As in earlier studies, a curve-fitting procedure was performed to objectively quantify these shifts in dose-response curves. All dose-response curves of one subject (time points 0 72 hours) were fitted simultaneously, sharing the same slope and thus minimising bad fitting due to lack of data points in solitary dose-response curves. Fitted dose-response curves before and after dosing of an Ang II receptor antagonist are illustrated by the example shown in Figure 1. Following blockade of Ang II receptors, a rightward shift in Ang II dose-response curves occurred, which declined with time after dosing and approached pre-dosing baseline again. The fitting procedure based on the sigmoid E max model according to Hill. 29 E = E min + ((E max - E min )*D m / (ED 50 m + D m )) (where E is the effect (BP) at the corresponding Ang II dose D, E max the theoretical maximum of effect, E min the pre-infusion blood pressure, ED 50 the dose of Ang II at which the effect is 50%, and m the slope parameter.) An iterative process of optimisation was performed on the parameters D, m, E min and ED 50 in order to minimise the residual sum of squared differences between the observed and the predicted values of the dependent variable. Dose ratios The dose ratios (DR) provide a quantitative measure of the dose-dependent rightward shifts of the Ang II dose-response curves at the respective time points after drug administration in relation to the baseline curve. The DR provides the factor by which the antagonist (AIIA) shifts the agonist (Ang II) response curve to the right.the DR is the multiple of the Ang II dose necessary to induce the same increase in BP as that achieved prior to administration of the antagonist. For mathematical reasons, values for dose ratio are normally quoted as DR minus 1 (DR-1); the dose ratio is 1 (and thus 337

3 Figure 1 Example (Subject No. 40) of Ang II dosediastolic blood pressure response curves before (open circles) and at several time points after administration of candesartan cilexetil (closed symbols). Rightward shifts in dose-response curves, indicated by arrows, are quantified as dose ratios (DR). Rightward shifts mean after dosing of an AIIA a multi-fold Ang II dose is necessary to obtain a similar BP increase than at 0 h. Figure 2 Time course of median and mean±sd (insert) dose ratios (quantification of pharmacodynamic effect given as DR-1) following consecutive administration of oral doses of mg candesartan ( ), mg irbesartan ( ), mg losartan (ο), mg telmisartan ( ), and mg valsartan ( ). D1 represents the administration of the lowest, D2 of the middle and D3 of the highest doses, respectively.time axis starts after administration of D1. the DR-1 is 0) for a compound with no antagonist activity.a value of DR-1 of 1 represents a two-fold rightward shift of the dose-response curve, i.e. 50% blockade of receptors in vivo. Dose ratios were calculated by dividing the ratios of ED 50 values after dosing by those prior to dosing. In this context, ED 50 values were wholly theoretical, serving only as the basis for determining dose ratios. 20 Apparent K i -doses Apparent K i -doses are doses (in mg) necessary to shift the Ang II dose-response curve two-fold to the right, or double the dose of Ang II required to obtain the same increase in diastolic BP as achieved prior to administration of the antagonist. The degree of in vivo blockade that is clinically relevant remains to be fully elucidated. Apparent K i -doses were derived from Schild plots (regression of DR-1 onto dose on a double-log scale). A Schild regression with a slope not different from unity indicates a competitive mechanism between substrate (or agonist) and inhibitor (antagonist). 27 Apparent K i -doses were calculated for y=dr 1=1 by extrapolation from the regression equations log(y)=a+b*log(x). Regression equations and apparent K i -doses were determined per treatment and time point after dosing using individual and median data. Statistical methods Statistical analysis was entirely descriptive and exploratory in nature. All tests of statistical hypotheses were two-sided and carried out at the 5% significance level. The time courses of mean and median diastolic dose ratios (given as DR-1) were compared in order to determine dose-dependency of effects and maximum ceiling effects.the data of the main study were analysed for statistical significance of treatment, dose, time and interaction effects using a repeated measures analysis of variance (GLM procedure). The diastolic DRs of telmisartan were compared with those of candesartan cilexetil using an unpaired t-test. Results Efficacy All five AIIAs studied dose-dependently inhibited the Ang II-induced increase in diastolic BP, thereby shifting Ang II dose-response curves to the right and resulting in an increasing dose ratio (DR). Figure 2 depicts mean (±SD) and median time courses of DR-1.The increase in pharmacodynamic effect with cumulative dosing appeared to be somewhat more pronounced with candesartan cilexetil, thus demonstrating a steeper and more distinct dose-response relationship. Analysis of variance revealed significant dose and time effects. All median DR-1 with candesartan cilexetil from 28 hours onwards (4 hours after administration of 8 mg) were higher than those obtained with the other compounds tested. The highest median and mean effects occurred at 4 hours and 8 hours after administration of candesartan cilexetil 16 mg.the median maximum DR-1 with candesartan cilexetil was 22, compared with a value of 18 for irbesartan, 19 for losartan and 16 for valsartan. The median maximum DR-1 following administration of telmisartan was eight; this was reached 4 hours after administration of the 80 mg 338

4 Figure 3 Schild regression plots (median DR-1 vs. dose on a double-log scale) for 4 hours (A), 8 hours (B) and 24 hours (C) after dosing of candesartan cilexetil ( ), irbesartan ( ), losartan (ο), telmisartan ( ) and valsartan ( ). A B C dose. There was considerable interindividual variability in results obtained in this study. Analysis of variance failed to identify a statistically significant treatment effect overall, with no statistically significant differences between candesartan cilexetil, losartan, irbesartan and valsartan. However, using the t-test, telmisartan was found to differ significantly from candesartan cilexetil. Figure 3 presents the median DR-1 data in the format of Schild regressions in relation to the applied oral doses. Candesartan cilexetil produced antagonistic effects at the lowest doses, i.e. had the greatest pharmacological potency. After 24 hours, the effects of all compounds were markedly reduced, although most of the higher doses still exerted an antagonism with a DR-1 of >1. The apparent 24-hour K i -doses at DR-1=1 (dose required to be administered at 0 hours to induce a two-fold rightward shift in the Ang II doseresponse curves at 24 hours post-dose) were 6 mg for candesartan cilexetil, 123 mg for irbesartan, 54 mg for telmisartan and 93 mg for valsartan (data derived taking into account all individual data).no K i -value could be calculated for losartan at 24 hours post-dose (all values of median DR-1 <1 on Schild regression plot). Tolerability Of the 61 volunteers participating in the study, 23 (38%) experienced an adverse event following administration of irbesartan (n=6), valsartan (n=5), telmisartan (n=4), losartan (n=3), or candesartan cilexetil (n=2). One subject withdrew from the study due to an adverse event related to an intercurrent infection (randomised to losartan 25 mg).headache was the most commonly reported adverse event. Discussion Using a classical technique from in vitro pharmacology, 21 this study quantitatively compared and characterised the pharmacological potency of various antagonists at angiotensin AT 1 -receptors.the underlying hypothesis assumes that an agonist exerts a response proportionate to receptor occupancy, irrespective of the presence of an antagonist. 22,25 Thus, by titrating a given agonist effect (increase in diastolic BP with Ang II in this context) in the absence of the antagonist as well as in the presence of different doses, values of DR-1 may be derived which reflect the antagonist's actual activity. Over the range of doses studied, all investigational compounds displayed competitive antagonism at AT 1 -receptors,bearing in mind that in vivo studies in man are unlikely to reveal unsurmountable antagonism, as this would require titration to maximum effect. Such a procedure would involve far higher increases in BP than are clinically reasonable or ethically sound. By definition, the maximum effect of the agonist dose-response curve remains unchanged following administration of a competitive antagonist, whereas the maximum effect is reduced in the case of an unsurmountable antagonist. 22,23 The K i -doses derived in this clinical study are referred to as apparent K i to clearly differentiate these results from those of experiments with defined in vitro systems. 25 All five AIIAs dose-dependently reduced the diastolic BP response to exogenously infused Ang II. The strongest antagonistic effects occurred 4 hours (irbesartan, telmisartan) or 8 hours (candesartan cilexetil, losartan, valsartan) after administration (Figure 3). At 24 hours, apparent K i -doses ranged from 6 mg for candesartan cilexetil to

5 Abbreviations Ang II Angiotensin II AT 1 Angiotensin II receptor subtype T 1 BP Blood pressure D Dose of agonist (angii) (Hill equation) DR Dose ratio DR-1 Dose ratio minus 1 ECG Electrocardiogram E Effect (Hill equation) E max Maximum of effect (Hill equation) E min Pre-infusion baseline (Hill equation) ED 50 Dosage of the agonist (angii) at which the effect is 50% (Hill equation) GLM General linear model (analysis of variance) HR Heart rate K i Dissociation constant m Slope parameter of Hill equation SD Standard deviation mg for irbesartan, with no apparent K i available for losartan at 24 hours.thus, for candesartan cilexetil, irbesartan, telmisartan and valsartan, apparent K i - doses at 24 hours post-dose were within the recommended oral maintenance dose for the treatment of hypertension. The apparent K i -dose and the drug s half-life characteristics (or clearance, respectively) determine the dosage and dosage interval required for long-term daily administration. 25,30 The technique of agonist-antagonist titration has been shown to be useful, not only for the assessment of antagonist potency but also in order to derive the pharmacological elimination characteristics. 24,28 In contrast to single dose agonist challenges, this technique is independent of the antagonist dose given. Therefore, it allows a true quantification of the kinetics of pharmacodynamic effects without bias due to antagonist dose. 30 In the present study, the number of time points was too small to allow exact pharmacokinetic calculations. Taking these limitations into account, we can estimate that, from 8 hours to 24 hours after dosing, the effects of candesartan cilexetil, telmisartan, valsartan and irbesartan decline with a half-life in the range of 8 10 hours. For losartan, this value is less than 5 hours. Thus, from a kinetic point of view, losartan appears to be more appropriate for twice daily administration in order to obtain reliable 24-hour blockade of AT 1 -receptors in the arterial system without resorting to excessive doses. Cumulative administration of increasing doses of AIIAs in this study introduced potential bias from carryover effects on study days two and three, although these effects appeared to be negligible. Another potential source of bias was the use of a parallel-group rather than crossover design. However, a crossover design was not thought to be feasible, as it would have required at least eight weeks of study participation per subject, thereby introducing a new bias, that of variability over time. Data for telmisartan were obtained in an open extension to the original protocol, since the drug was not initially available on the market. Nevertheless, results seen with this recently introduced compound suggest a classic dose-proportionate antagonistic effect. It should be considered that the Schild regression technique requires steady-state conditions and a competitive type of antagonism. Previously, we have demonstrated steady-state conditions for an infusion model similar to the one used in this study. 31 Overall, the Schild regression plots produced in this study (Figure 3) did not differ significantly from unity, so we may conclude that, at least over the dose range examined,there were no major deviations from a competitive type of antagonism. Conclusion This study has clearly demonstrated that, compared with other compounds in the same class, candesartan cilexetil demonstrated the highest pharmacological potency, i.e. direct Ang II antagonistic activity per mg in vivo in man.this finding is in agreement with data on candesartan cilexetil s high receptor affinity in in vitro models. Taken together, these results may in part explain the marked efficacy of candesartan cilexetil in lowering BP in patients with hypertension. Acknowledgements The study was supported by a research grant from Takeda Pharma GmbH, Germany.The results were excerpted in part from a medical thesis done by C. Hausdorf at the Johannes Gutenberg-University Mainz, Germany. References 1. Stroth U, Unger T. The renin-angiotensin system and its receptors. J Cardiovasc Pharmacol 1999;33(suppl 1):S21 S28; discussion S41 S Messerli FH,Weber MA, Brunner HR.Angiotensin II receptor inhibition. A new therapeutic principle. Arch Intern Med 1996;156(17): Kirch W. Angiotensin II Rezeptorantagonisten im Vergleich. Arzneimitteltherapie 1998;16: Goldberg AI, Dunlay MC, Sweet CS. Safety and tolerability of losartan potassium, an angiotensin II receptor antagonist, compared with hydrochlorothiazide, atenolol, felodipine ER, and angiotensin-converting enzyme inhibitors for the treatment of systemic hypertension. Am J Cardiol 1995;75(12): Goa KL, Wagstaff AJ. Losartan potassium: a review of its pharmacology, clinical efficacy and tolerability in the management of hypertension [published erratum appears in Drugs 1996;52(4):540]. Drugs 1996;51(5): Belcher G, Hubner R, George M, Elmfeldt D, Lunde H. Candesartan cilexetil: safety and tolerability in healthy volunteers and patients with hypertension. J Hum Hypertens 1997;11(suppl 2):S85 S McClellan KJ, Goa KL. Candesartan cilexetil.a review of its use in essential hypertension. Drugs 1998;56(5): Markham A, Goa KL. Valsartan. A review of its pharmacology and therapeutic use in essential hypertension. Drugs 1997;54(2): Gillis JC, Markham A. Irbesartan. A review of its pharmacodynamic and pharmacokinetic properties and therapeutic use in the management of hypertension. Drugs 1997;54(6):

6 10. Csajka C, Buclin T, Brunner HR, Biollaz J. Pharmacokineticpharmacodynamic profile of angiotensin II receptor antagonists. Clin Pharmacokinet 1997;32(1): Elmfeldt D, George M, Hubner R, Olofsson B. Candesartan cilexetil, a new generation angiotensin II antagonist, provides dose dependent antihypertensive effect. J Hum Hypertens 1997;11(suppl 2):S49 S Sever PS. Key features of candesartan cilexetil and a comparison with other angiotensin II receptor antagonists. J Hum Hypertens 1999;13(suppl 1):S3 S10; discussion S33 S Andersson OK, Neldam S.The antihypertensive effect and tolerability of candesartan cilexetil, a new generation angiotensin II antagonist, in comparison with losartan. Blood Press 1998;7(1): Hedner T, Oparil S, Rasmussen K et al. A comparison of the angiotensin II antagonists valsartan and losartan in the treatment of essential hypertension. Am J Hypertens 1999;12(4 Pt 1): Kassler-Taub K, Littlejohn T, Elliott W, Ruddy T, Adler E. Comparative efficacy of two angiotensin II receptor antagonists,irbesartan and losartan in mild-to-moderate hypertension. Irbesartan/Losartan Study Investigators [published erratum appears in Am J Hypertens 1998;11(6 Pt 1):736] [see comments]. Am J Hypertens 1998;11(4 Pt 1): Belz GG, Butzer R, Kober S, Mang C, Mutschler E. Time course and extent of angiotensin II antagonism after irbesartan, losartan, and valsartan in humans assessed by angiotensin II dose response and radioligand receptor assay. Clin Pharmacol Ther 1999;66(4): Azizi M, Chatellier G, Guyene TT, Menard J. Pharmacokinetic-pharmacodynamic interactions of candesartan cilexetil and losartan. J Hypertens 1999;17(4): Lacourciere Y, Asmar R. A comparison of the efficacy and duration of action of candesartan cilexetil and losartan as assessed by clinic and ambulatory blood pressure after a missed dose, in truly hypertensive patients: a placebo- controlled, forced titration study. Candesartan/Losartan study investigators. Am J Hypertens 1999;12(12 Pt 1 2): Morimoto S,Ogihara T.TCV-116:a new angiotensin II type- 1 receptor antagonist. Cardiovasc Drug Rev 1994;12: Malerczyk C, Fuchs B, Belz GG et al. Angiotensin II antagonism and plasma radioreceptor-kinetics of candesartan in man. Br J Clin Pharmacol 1998;45(6): Arunlakshana O, Schild HO. Some quantitative uses of drug antagonists [classical article]. Br J Pharmacol 1997;120(4 suppl):s151 S161; discussion S48 S Gaddum JH.Theories of drug antagonism. Pharmacol Rev 1957;9: Van den Brink FG. General theory of drug-receptor interactions. Drug-receptor interaction models. Calculation of parameters. In: van Rossum JM, editor. Kinetics of drug action. Handbook of Experimental Pharmacology, vol. 47. Berlin- Heidelberg, New York: Springer, 1977: Wellstein A, Essig J, Belz GG. Inhibition of angiotensin-i response by cilazapril and its time course in normal volunteers. Clin Pharmacol Ther 1987;41(6): Wellstein A, Essig J, Belz GG. A method for estimating the potency of angiotensin-converting enzyme inhibitors in man. Br J Clin Pharmacol 1987;24(3): Belz GG, Essig J, Wellstein A. Hemodynamic responses to angiotensin I in normal volunteers and the antagonism by the angiotensin-converting enzyme inhibitor cilazapril. J Cardiovasc Pharmacol 1987;9(2): Wellstein A, Belz GG, Palm D. Beta adrenoceptor subtype binding activity in plasma and beta blockade by propranolol and beta-1 selective bisoprolol in humans. Evaluation with Schild-plots. J Pharmacol Exp Ther 1988;246(1): Breithaupt-Groegler K, Malerczyk C, Belz GG et al. Pharmacodynamic and pharmacokinetic properties of an angiotensin II receptor antagonist - characterization by use of Schild regression technique in man. Int J Clin Pharmacol Ther 1997;35(10): Hill AV. The possible effects of the aggregation of the molecules of haemoglobin on its dissociation curves. J Physiol 1910;40:iv vii. 30. Wellstein A, Palm D, Pitschner HF, Belz GG. Receptor binding of propranolol is the missing link between plasma concentration kinetics and the effect-time course in man. Eur J Clin Pharmacol 1985;29(2): Essig J, Belz GG,Wellstein A.The assessment of ACE activity in man following angiotensin I challenges: a comparison of cilazapril, captopril and enalapril. Br J Clin Pharmacol 1989;27(suppl 2):S217 S