Trough to peak ratio: current status and applicability

Journal of Human Hypertension (1998) 12, 55 59 1998 Stockton Press. All rights reserved 0950-9240/98 $12.00 REVIEW ARTICLE Trough to peak ratio: current status and applicability Department of Medicine and Therapeutics, Western Infirmary, Glasgow G11 6NT, UK Keywords: trough to peak ratio; antihypertensive drugs Introduction In concept, trough to peak (T:P) ratio is a simple and straightforward parameter which succinctly describes the consistency and duration of a drug s antihypertensive effectiveness across its recommended dosage interval. Additionally, it provides information about the suitability of the drug for its recommended dosage interval for which there is a clear current preference for once-daily dosing. Thus a T:P ratio approaching 100% informs the clinician that drug X in a dose of Y mg is well suited to its dosage interval and that the blood pressure (BP) reduction persisting at trough (which, by definition, is immediately prior to the administration of the next dose) is closely similar in magnitude to the BP reduction measured at peak (ie, at the time of maximal or peak pharmacological activity). By inference, and in terms of clinical reality, the trough BP measurement corresponds to the lowest level of pharmacological activity and, usually, the residual plasma drug concentration. It is implicit that the magnitude of the trough BP reduction is clinically useful and intellectual arguments about a drug with a T:P ratio of 100% which is achieved with a trough BP reduction of only 1 2 mm Hg are completely spurious. Unfortunately, this straightforward and useful parameter has been misconstrued, misunderstood and misrepresented in much of the published literature to such an extent that its meaning and usefulness have been significantly compromised. T:P ratio the FDA perspective It is not difficult to appreciate that a drug with a relatively short duration of action might deliberately be administered in a very high dose with the sole and specific intention of demonstrating that significant BP-lowering activity persists until the end of the dosage interval. In this way, the drug is entitled Correspondence:, Department of Medicine and Therapeutics, Western Infirmary, Glasgow G11 6NT, UK Revised and accepted 10 August 1997 to be licensed as an effective antihypertensive agent (for once daily administration). Thus, an acceptable trough BP reduction has been generated via an unnecessarily large peak BP reduction. It is fundamental to the concept of T:P ratio that an unnecessarily large peak BP reduction is not appropriate when it constitutes a specific stratagem to avoid having to administer lower doses via a multiple daily dosing regimen. The essence of T:P ratio, therefore, is to identify the treatment regimen (dose and dosing frequency) which is the most effective (and safest) for routine use in an unselected hypertensive population. The trough BP reduction itself remains the critical index of antihypertensive efficacy for regulatory purposes. Figure 1 illustrates the BP reduction profiles for two different drug treatment regimens and it can be seen that the once-daily treatment produces a measurable effect at trough (sufficient for licensing as a once-a-day drug) and that this has been achieved in association with an obvious and pronounced peak BP reduction. In contrast, the twicedaily treatment regimen not only increases the magnitude of the trough BP reduction but also avoids an excessive peak reduction in BP. In summary, and with specific respect to the overall level of BP control, the twice daily regimen is superior and this is directly reflected in the high T:P ratio of greater than 70%. These BP reduction profiles are derived from treatment with enalapril 20 mg once daily (T:P ratio of 48%) and enalapril 10 mg b.d. (T:P ratio of 74%). 1 Methodological issues principles and problems a) Different studies, different results: Specific issues of methodology are discussed elsewhere in this issue but it is important to note that the FDA did not define the most appropriate methodology although they did make two specific and important recommendations: that the antihypertensive effect should be assessed under steady state dosing conditions and that it should be adjusted for placebo effects. In conventional clinical trials practice, therefore, a parallel group design comparing the achieved BPs under placebo and active treatment conditions

Trough to peak ratio: current status 56 Figure 1 A comparison of systolic BP reductions throughout 24 h during treatment with two different antihypertensive drug treatment regimens. would seem to be the simplest and most obvious approach. Alternatively, and in some ways preferably, a randomised crossover comparison of the achieved BPs following periods of placebo and active treatment. Because the FDA failed to detail the methodology a variety of approaches have been employed and many of these contain serious misrepresentations: for example, there are no grounds for calculating T:P ratios for the placebo effect itself, relative to the pre-treatment or baseline BP; there are no grounds for calculating T:P ratio via the offset of antihypertensive effect following temporary treatment withdrawal. These are examples of the different and often flawed approaches which have led to disparate published results and created confusion. The confusion created by the use of a variety of methodologies, allied to some fundamental misconceptions about the clinical pharmacology of antihypertensive drugs, makes it virtually impossible to compare T:P ratios for different competitor drugs on the basis of different studies, by different researchers, by different methodologies, etc. However, where the differing methodologies are sound and the studies well conducted, it may sometimes be possible to compare different study results and some degree of concordance would be expected 2 (Table 1). However, even in this illustrative metaanalysis (with its acknowledged limitations) there are discrepancies which, in most cases, reflect dosedependent differences in T:P ratio and/or different methodologies. b) Definition of the peak blood pressure reduction: The peak BP reduction constitutes the Table 1 T:P ratios with ACE inhibitor drugs Drug T:P ratio (%) Benazepril 10 40 Captopril 0 40 Enalapril 40 or 50 or 70 or 80 Lisinopril 40 or 50 or 60 or 70 or 80 Perindopril 30 Quinapril 30 or 40 Ramipril 40 or 50 Trandolapril 50 or 100 maximum BP reduction attributable to the antihypertensive drug but it is not necessarily the lowest achieved BP (which is often attained during deep sleep). This is illustrated in Figure 2 with systolic BP profiles following placebo and following an antihypertensive drug (upper panel). This also highlights the importance of placebo rather than baseline correction for calculating the BP reduction directly attributable to the drug itself. In this illustration, although the lowest BP relative to baseline, occurs at approximately 6 h post-dose, this is not the peak antihypertensive effect which occurs at about 2 h post-dose, relative to placebo (lower panel). The maximum or peak BP reduction reflects the maximum pharmacological effect of the drug (which in turn is often related to the maximum plasma drug concentration) and across a range of, say, 2 3 h it will occur at about the same time in the great majority of patients. However, it will not occur at exactly the same time in every patient: this is an argument in favour of T:P studies in individual

Trough to peak ratio: current status 57 Figure 3 Comparison of the drug concentration-time profiles for two different formulations of nifedipine. Figure 2 The systolic BP responses to placebo and an antihypertensive drug (upper panel) and the BP corrected for the baseline (pre-treatment) BP and for the corresponding placebo profile (lower panel). patients, rather than groups. Furthermore, it will often be difficult to identify the exact time of the maximum BP reduction and several different strategies have been advocated to permit the accurate definition of the peak BP response: multiple BP measurements in the controlled and reproducible environment of a clinical research unit, 3 for example, or the use of ambulatory BP measurement with a smoothing analytical approach and averaging of 2 3 h of BP measurements to eliminate rogue values and avoid undue reliance on single time points. 4 Not only are there difficulties in clearly identifying the peak BP response but there are also reproducibility problems with ambulatory BP measurement for repeat measurements in the same individual. Thus, although average values may be sufficiently reproducible to suggest that the ambulatory technique has no placebo effect, the consensus view now indicates that there is a significant placebo or acclimatisation effect, which is most obvious during the first few hours of measurement. 5 9 Whatever the technique of BP measurement, therefore, there is a clear requirement for an appropriate parallel group or crossover design to define the BP levels during a placebo treatment and to take account of circadian changes. clearly there will be different T:P ratios for the tablet and for the GITS formulation. The published results indicating differing T:P ratios, of respectively about 50% 10 and approximately 100%, 11 are entirely consistent with these differing pharmacokinetic profiles. b) Dose and concentration dependent effects: The impact of dose on the T:P ratio is fundamentally dependent on the underlying drug concentrationeffect relationship. In simplistic terms, the T:P ratio will not vary with dose when there is a linear concentration-effect relationship (and when there is a well-designed drug formulation with a consistent drug delivery system). This applies to calcium antagonist drugs in general: thus, for nifedipine delivered via the tablet formulation the T:P ratio falls in the range 48 50% for the twice-daily administration of 10 mg, 20 mg and 30 mg doses 10 (Figure 4). Similar results are obtained with the GITS formulation with reported T:P ratios of about 100% with both the 30 mg and 60 mg doses. 11 In contrast, drugs which have more complex concentration-effect characteristics such as those involving E max relationships, for example ACE inhibitors, have dose-dependent T:P ratios. This has The consistency of the T:P ratio a) Formulation dependent results: The plasma drug concentration-time profiles for two formulations of nifedipine are illustrated in Figure 3 and Figure 4 Effect of dose on the T:P ratio for nifedipine (tablet formulation).

58 Trough to peak ratio: current status been very well illustrated (Figure 5) in a study with lisinopril where the low dosages had low T:P ratios well below the minimum target of 50% and the higher dosages were just over 50%. 12 Quite clearly, therefore, low doses of lisinopril should probably be administered via a twice-daily dosing regimen whereas, at least arguably, the higher doses might be administered once daily. Trough:peak ratio implications for treatment The validity of T:P ratio as an index of duration of action, and of the drug s suitability for its recommended dosage interval, is largely but crucially dependent upon a satisfactory methodology. 13 From a practical point of view, however, if the T:P ratio is appropriately characterised then a ratio consistently in excess of 50% (ie, consistent between different individuals and consistent in the published literature) is indicative of a drug with a BP response whose magnitude will be relatively consistent throughout the nominated dosage interval. Thus, approximately the same level of BP control will be consistently maintained throughout the full 24-h period (assuming once-daily dosing), including an overnight BP reduction if the patient is a nondipper and including persisting antihypertensive efficacy in the waking and early working day. With particular reference to concerns about excessive reductions in nocturnal BP, when the levels might already be inherently low, there again appear to be misconceptions. The absolute magnitude of the BP reduction in response to an antihypertensive agent is, in general, determined not only by the prevailing drug concentrations but also by the level of the untreated BP. 14,15 This reflects a mathematical/statistical phenomenon whereby the higher the (untreated) pressure the greater the fall (in response to treatment). 16 Thus, if nocturnal BP is normal or low then there will be a relatively small further reduction in response to antihypertensive treatment. The vexed question of the ideal T:P ratio, however, is unanswerable. The basic message is that antihypertensive drugs with values less than 50% require to be administered more frequently: drugs with ratios in the range 66 100% are well suited to their recommended dosing frequencies, presumably once daily. For agents with ratios in the range 50 Figure 5 Effect of dose on the T:P ratio for lisinopril. 66% there clearly would be differences of opinion concerning their suitability for single or multiple dosing regimens and other factors, such as the range of T:P values, might influence the decision about once- or twice-daily dosing. Some of the practical issues are discussed in detail elsewhere. 17,18 Overall, although there is no definitive proof that an antihypertensive profile characterised by a high T:P ratio is preferable, it is interesting to note that thiazide diuretics which are arguably the most successful agents in the published clinical trials of antihypertensive drugs have, in general, high T:P ratios. Conclusions The T:P ratio is not a fixed and immutable single value which can be used to label each antihypertensive drug. It is probably best expressed as a mean value with some indication of inter-individual variability in order to provide some information about the consistency of the drug, and the appropriateness of its dosing regimen, within a hypertensive population. Furthermore, where there are dose-dependent differences it will be necessary to provide a series of values for each of the dosages recommended in routine clinical practice. Thus, T:P ratio is one of several parameters to be taken into account alongside, say, half-life and the concentration effect relationship, when dosage regimens are being determined. Ideally, the T:P ratio should be determined in early clinical development via two or three appropriately designed and well conducted studies. Thus, T:P ratio is a useful parameter and an important indicator of dose and dose frequency when it is quantified correctly and applied appropriately. It is not, and was never intended to be, the yardstick by which an antihypertensive drug is judged. Misconceptions and misrepresentations, allied to the application of sub-optimal methodology (in fact, sometimes wholly inappropriate methodology) has sullied an otherwise useful and clinically relevant measurement which contributes to our overall assessment of an antihypertensive drug. References 1 Meredith PA, Donnelly R, Elliott HL, Reid JL. Prediction of response to enalapril. J Hypertens 1990; 8: 1085 1090. 2 Zannad F. Trandolapril: how does it differ from other angiotensin converting enzyme inhibitors? Drugs 1993; 46 (Suppl 2): 172 182. 3 Elliott HL, Meredith PA. Calculation of trough-to-peak ratio in the Research Unit setting. Am J Hypertens 1996; 9: 71S 75S. 4 Omboni A et al. Calculation of trough-to-peak ratio of antihypertensive treatment from ambulatory blood pressure; methodological aspects. J Hypertens 1995; 13: 1105 1112. 5 Mutti E et al. Effect of placebo on 24-h non-invasive ambulatory blood pressure. J Hypertens 1991; 9: 361 364. 6 Mancia G et al. Limited reproducibility of hourly blood pressure values obtained by ambulatory blood pressure monitoring: implications for studies on antihypertensive drugs. J Hypertens 1992; 10: 1531 1535.

7 Gerin W, Rosofsky M, Pieper C, Pickering TG. A test of reproducibility of blood pressure and heart rate variability using a controlled ambulatory procedure. J Hypertens 1993; 11: 1127 1131. 8 Prasad N, MacFadyen RJ, Ogston SA, MacDonald TM. Elevated blood pressure during the first two hours of ambulatory blood pressure monitoring: a study comparing consecutive twenty-four-hour monitoring periods. J Hypertens 1995; 13: 291 296. 9 Staessen JA et al (on behalf of the Syst-Eur Investigators). Ambulatory pressure decreases on longterm placebo treatment in older patients with isolated systolic hypertension. J Hypertens 1994; 12: 1035 1039. 10 Meredith PA, Donnelly R, Elliott HL. Prediction and optimisation of the antihypertensive response to nifedipine. Blood Press 1994; 3: 303 308. 11 Zanchetti A et al. Antihypertensive effects of nifedipine GITS on clinical and ambulatory blood pressures in essential hypertensives. J High Blood Press 1994; 3: 45 56. 12 Menard J, Bellet M, Brunner HR. Clinical development Trough to peak ratio: current status of antihypertensive drugs: can we perform better? In: Larach JH, Brenner BM (eds). Hypertension: Pathophysiology, Diagnosis and Management, 1st edn. Raven Press Limited: New York, 1990, pp 2331 2350. 13 Elliott HL, Meredith PA. Methodological considerations in calculation of the T:P ratio. J Hypertens 1994; 12 (Suppl 8): S3 S7. 14 Meredith PA, Donnelly R, Elliott HL. Prediction and optimisation of the antihypertensive response to nifedipine. Blood Press 1994; 3: 303 308. 15 Meredith PA et al. Age and the antihypertensive effect of verapamil: an integrated pharmacokineticpharmacodynamic approach. J Hypertens 1987; 5 (Suppl 5): S125 S221. 16 Sumner DJ et al. Blood pressure and correlations. Lancet 1985; 1: 1110 1111. 17 Elliott HL, Meredith PA. Analysis of T:P ratio and the assessment of antihypertensive drug action. J Hum Hypertens 1995; 9: 423 427. 18 Elliott HL, Meredith PA. T:P ratio: clinically useful or practically irrelevant. J Hypertens 1995; 13: 279 283. 59