Proceedings of the 99th Meeting of the Food and Drug Administration Cardiovascular and Renal Drugs Advisory Committee
May 29th and 30th, 2003
The committee reviewed two new drug applications (NDAs) that previously had received “approvable” actions. Initial submissions had shown that these two compounds might have a prolonging effect on the QT interval. Therefore, sponsors were asked to undertake additional studies that incorporated an evaluation of QTc (QT interval corrected) prolongation at plasma concentrations relevant for monotherapy at the proposed labeled doses, as well as at concentrations expected to be achieved as a result of pharmacokinetic interaction with drugs that plausibly might be administered concomitantly.
The first of the drugs, alfuzosin, is an α1-adrenergic receptor antagonist known to improve urinary flow rate in patients with symptoms of benign prostatic hypertrophy. The sponsor presented data from a crossover study of 48 healthy white men between the ages of 18 and 50 years given single doses of alfuzosin targeting therapeutic (10 mg) and supratherapeutic (40 mg) blood levels. Maximum changes in QT interval and QTc interval were measured and compared with changes in individuals who took placebo or moxifloxacin (400 mg), a drug that is known to prolong QT interval. Approximately 10 mg alfuzosin caused a modest dose-related increase in QTc from baseline that was less than the increase caused by moxifloxacin relative to placebo.
The second drug, vardenafil, is a potent phosphodiesterase 5 inhibitor. It was studied in 59 healthy adult men between the ages of 45 and 60 years, using sequential 12-lead ECGs at specified intervals. Studies were conducted with doses targeting therapeutic (10 mg) and supratherapeutic (80 mg) blood levels. As in the alfuzosin trial, both placebo and moxifloxacin (400 mg) were administered as positive controls. A small increase in individually corrected QTc was seen, with a shallow dose-response curve for effects between 10 and 80 mg, and little difference compared with effects of moxifloxacin.
With regard to the postmarketing experience for the two drugs, alfuzosin is currently approved in Europe as well as in Canada and Australia. During this clinical use, no cases of torsade de pointes have been attributed to this agent. Vardenafil has been marketed in Europe since March 2003, and no adverse clinical cardiovascular side effects and, in particular, no torsade de pointes have been reported.
A presentation from Pfizer on a similar phosphodiesterase 5 inhibitor, sildenafil, indicated that no case of QTc prolongation or torsade de pointes has been identified during 5 years of widespread use.
Extensive discussion focused on optimal assessment of the QT interval, which is inversely related to heart rate. The two most commonly used methods to adjust QT for rate are Bazett’s correction (QTcB=QT/RR1/2) and Fridericia’s correction (QTcF=QT/RR1/3). The aim of these methods is to correct the QT interval to what it would be at a heart rate of 60 bpm; if such a correction were perfect, the slope of the line relating heart rate to QTc would be zero. Of these two methodologies, Fridericia’s correction appears preferable because it is less likely to overcorrect for high heart rates. Additional methodologies were described and evaluated, including correction for variations seen in specific study populations and correction for variation seen in individual patients. All of these used standard 12-lead ECG for measurement of QT interval. A novel method based on 24-hour ambulatory ECG recordings was used in the alfuzosin trial (the “Holter Bin Method”), which involves recording QT intervals at varying heart rates achieved spontaneously by study subjects over 24 hours, enabling direct determination of QT at a heart rate of 60 bpm, as well as at other heart rates, with and without intervention. The latter method does not require the use of correction factors.
The committee concluded that the studies using single doses of alfuzosin and vardenafil provided relevant and adequate information so that likelihood of QT prolongation in clinical use could be assessed. The QT information, taken together with observed lack of relevant arrhythmic events and with reported cardiovascular event rates consistent with those epidemiologically expected in the target populations treated with the drugs, caused the committee to conclude that no clinically important arrhythmogenic effects are likely in clinical practice—such that, in terms of cardiovascular effects, the drugs are acceptably safe for their intended uses. The committee noted, however, that no direct information was available to define the effects of the drugs in females and in target populations with comorbid cardiac conditions, cardioactive medications, and/or specific metabolic disturbances that potentially could interact with the new agents to prolong the QT interval and produce arrhythmias. Hence, the relation of benefit to risk in individual patients will need to be judged on the basis of inferences about the putative effects of drug–disease and drug–drug pharmacodynamic interactions.
Notwithstanding these reservations, the committee recognized that the sponsors of both NDAs had conducted studies that were significantly more thorough than those previously performed and had satisfactorily evaluated each drug’s potential for QTc prolongation.