Evaluating Risk Versus Benefit of Antiarrhythmic Therapy

The benefits of antiarrhythmic drug therapy in the acute setting are clearly evident when a patient presents with a highly symptomatic, sustained, arrhythmia that drug therapy terminates rapidly. In this setting, symptomatic benefits are obvious, and the risk is minimized because patients are monitored and exposed to the drug only for a brief period. By contrast, with chronic therapy, the balance between benefit and risk is more difficult to evaluate. Although chronic therapy in a patient with heart disease can be beneficial initially, efficacy can be lost over time because of a changing electrophysiological milieu. Indeed, in virtually all clinical trials to date, chronic antiarrhythmic therapy has been incompletely effective. Furthermore, the consequences of arrhythmia recurrence are clinically important and can vary from recurrent hospitalization to SCD, and the risk of serious adverse effects appears to increase over time. Therefore, antiarrhythmic drugs retain a primary role in the acute termination of arrhythmias and in selected chronic arrhythmia settings, such as AF, in which arrhythmia recurrence might not be catastrophic. By contrast, device-based and ablative therapies are more desirable from a risk-to-benefit point of view in such settings as ventricular tachycardia (VT)/VF, where antiarrhythmic drugs have mainly assumed a secondary role. However, as ablative therapies for AF continue to evolve, it is possible that drugs might assume a secondary role here as well. Nevertheless, the development of new drug therapies for AF that are highly effective and, more importantly, safe during chronic therapy is still highly desirable because ablative therapies are likely limited to a small section of patients with the arrhythmia.

Historical Perspective on Current Antiarrhythmic Drugs and Their Limitations

Drugs that are currently marketed as antiarrhythmics were initially developed in whole-animal models by screening for their effects on normal cardiac electrophysiology. However, decades elapsed before the initial use of older antiarrhythmic drugs in humans and more importantly the definition of their molecular mechanisms of action. Antiarrhythmic drugs are only partly effective and have many adverse effects, the most important of which is the potential to generate new life-threatening arrhythmias (proarrhythmia). The increasing appreciation of ventricular arrhythmias as a marker of underlying heart disease and, therefore, a potential drug target, led to the development of multiple new antiarrhythmic drugs in the late 1970s and early 1980s. An early animal model that was used to screen for antiarrhythmic activity was the “Harris dog,” which demonstrates extremely high-frequency ventricular ectopic activity after two-stage coronary artery ligation. Ventricular arrhythmias are especially sensitive to sodium channel blockers; therefore, the first wave of new antiarrhythmic drugs were derived from existing structures and demonstrated activity in this and other models (i.e., primarily sodium channel blockers).

The Cardiac Arrhythmia Suppression Trial (CAST) represents a landmark trial result, not only for antiarrhythmic drug development but for drug development in general.⁴ The CAST tested the prevalent hypothesis that because ventricular ectopy following myocardial infarction (MI) is a risk factor for SCD, suppression of ventricular ectopy would reduce the incidence of SCD. Mortality among patients randomized to ventricular ectopic suppression with encainide or flecainide in the CAST was approximately threefold higher than that of patients randomized to receive placebo. The study also emphasized the importance of a randomized, placebo-controlled trial with a primary “hard” endpoint, such as death, as opposed to a surrogate end point, such as ectopic beat suppression, to determine whether a drug was beneficial. The drugs tested in CAST were potent sodium channel–blocking agents.

For basic and clinical scientists interested in arrhythmia mechanisms, the CAST result provided a strong impetus to further work that defined the way in which loss of sodium channel function was arrhythmogenic. It is thought that blocking sodium channels increases the risk of SCD by slowing conduction or increasing the heterogeneity of repolarization, both of which can be arrhythmogenic, especially in diseased hearts.5,6 Multiple lines of evidence support the hypothesis that loss of sodium current (I_Na) slows cardiac conduction and may be arrhythmogenic; such loss can be structural (e.g., mutations in the cardiac sodium channel gene SCN5A leading to loss of peak I_Na),^7–9 pharmacologic,^7,10 or reflect activation of second messenger systems. For example, protein kinase A (PKA) activation has been reported to increase peak I_Na, at least in part because of altered trafficking,¹¹ whereas protein kinase C activation decreases it.¹²

Another consequence of CAST was that development of sodium channel–blocking drugs came to a rapid halt and QT-prolonging agents then assumed the limelight. Just as procainamide and lidocaine provided the structural starting point of a range of sodium channel–blocking molecules, two compounds with prominent action potential (AP)-prolonging properties formed the framework here: N-acetyl procainamide (NAPA) and sotalol. Both compounds also did not block the sodium channel and gave rise to a generation of QT-prolonging drugs, such as d-sotalol, dofetilide, almokalant, and sematilide. Although the screening assay to develop these agents usually was APD in a guinea pig papillary muscle, and as with newer sodium channel blockers, these compounds were not synthesized to interact with a specific predefined molecular target. Nevertheless, subsequent studies identified block of one specific potassium current, termed I_Kr (the rapid component of the cardiac delayed rectifier potassium current), as the major mechanism underlying QT prolongation by antiarrhythmic drugs. I_Kr is encoded by KCNH2 (also known as HERG). However, blocking I_Kr carries the risk of marked QT interval prolongation and a distinctive form of polymorphic ventricular tachycardia, termed torsades de pointes. Indeed, potassium-channel blocking antiarrhythmic drugs have been tested in CAST-like trials and did not prevent more deaths than did placebo.13–15 Moreover, the same mechanism underlies the development of torsades de pointes in response to noncardiovascular drugs, such as certain antihistamines, antibiotics, and antipsychotics.¹⁶

Classification of Antiarrhythmic Drugs

The earliest schemes classified drugs into classes based on shared efficacies and toxicities. Some antiarrhythmic drugs share important electrophysiological properties, notably block of sodium channels or of the potassium current I_Kr, and these can provide the basis for predicting shared or class actions and toxicities. Block or enhancement of other ionic currents, such as I_Ks, the transient outward current (I_TO), or acetylcholine-activated current (I_K-Ach), can also contribute to clinical drug actions in some cases and are not usually considered in broad classification schemes. An alternative approach is to classify drugs based on key electrophysiological mechanisms involved in arrhythmogenesis, thereby allowing specific drugs to be chosen to target these mechanisms. This approach is adopted by the “Sicilian gambit” investigators and is nicely exemplified by the definition of new molecular mechanisms in congenital arrhythmia syndromes, such as congenital long QT syndrome (LQTS), and the way in which these then lead naturally to mechanism-based therapies (discussed later). As understanding of the molecular and cellular basis of arrhythmias evolves, a more rational choice of key molecules whose targeting is likely to be safe and effective in the therapy of a particular arrhythmia should be possible.

The term antiarrhythmic drugs has traditionally included drugs targeting ion channels in cardiac myocytes (sodium channel blockers, calcium channel blockers, and QT prolonging drugs, generally potassium channel blockers), β-adrenergic receptor blockers, and a series of drugs with diverse mechanisms used primarily for the therapy of arrhythmias, such as digoxin, amiodarone, magnesium, and adenosine. However, recent studies have demonstrated that other widely used cardiovascular therapies, such as angiotensin-converting enzyme and hydroxy-methyl-glutaryl (HMG) Co-A reductase inhibitors, can also have important antiarrhythmic effects. These effects can include reduction of SCD, an arrhythmic event that represents the final common pathway for many potential disease pathways,¹⁷ and prevention of AF. These studies provide novel potential therapies and implicate new signaling pathways in the pathogenesis of arrhythmias and therefore as potential targets for the development of effective antiarrhythmic interventions.

Pharmacokinetic Principles

One important mechanism underlying variability in response to antiarrhythmic drugs is variability in pharmacokinetics, the result of the processes of drug absorption, distribution, metabolism, and elimination. Variability in pharmacokinetics contributes substantially to variability in efficacy or toxicity when there is a narrow therapeutic window (the margins between doses and plasma concentrations that are associated with efficacy and those associated with toxicity) and when the drug is metabolized or excreted by a single pathway; these criteria apply to many antiarrhythmic drugs. Sotalol and dofetilide are examples of drugs with narrow therapeutic windows and for which the risk of toxicity, namely torsades de pointes, increases with increasing doses. In addition, both drugs are largely excreted by a single pathway: renal excretion. As a result, drug doses need to be adjusted in renal failure to avoid increased risk of torsades de pointes. By contrast, the toxicity of flecainide is also concentration-dependent, but the drug is eliminated by hepatic metabolism and renal excretion. As a result, impaired metabolism or renal dysfunction alone do not generally alter response to the drug, although rare cases of flecainide toxicity owing to inhibited drug metabolism in patients with renal dysfunction have been reported. Table 111-1 identifies drugs that have narrow therapeutic windows and are eliminated by a single pathway, in addition to circumstances under which the meeting of these two clinical conditions can result in serious and often unexpected drug toxicity.

Cytochrome P450s and Other Drug Elimination Molecules

Drug metabolism, elimination, and disposition are accomplished by specific gene products, most commonly drug metabolizing enzymes (primarily members of the cytochrome P450 superfamily, or CYPs) and drug transport molecules. DNA variants that alter the activity of these proteins are increasingly well recognized; however, some exert only subtle effects on protein function, and in other cases an individual may totally lack enzymatic activity. Polymorphisms that modulate drug metabolizing enzymes or transport molecules are especially important if the affected pathway is critical for elimination of a drug with a narrow therapeutic margin. Furthermore, concomitant drug therapy can modulate the activity of the drug metabolizing and transport molecules. In most cases, such drug interactions result in inhibition of the elimination pathway. Occasionally, however, concomitant drug therapy can induce expression of drug metabolism and thus accelerate elimination. Under this circumstance, an increase in the drug dose may be required to maintain a therapeutic effect.

The major drug-metabolizing enzymes for antiarrhythmic drugs are CYP2D6, CYP2C9, and CYP3A4/5. Approximately, 5% to 10% of whites and African Americans are homozygous for loss of function alleles in CYP2D6; these individuals totally lack enzymatic activity and are designated poor metabolizers (PMs). CYP2D6 PMs have markedly decreased propafenone clearance and accumulate the parent drug-to-plasma concentrations high enough to produce clinically significant β-blockade. As a result, asthma can be a risk in these subjects. Similarly, CYP2D6 PMs also have higher concentrations of timolol and metoprolol. Propafenone and quinidine are CYP2D6 inhibitors and can therefore alter the effects of these CYP2D6 substrates. Amiodarone is a potent CYP2C9 inhibitor, and doses of warfarin must therefore be adjusted downward with amiodarone therapy.

CYP3A4/5 are two closely related enzymes that are the most abundant cytochromes in the liver (and in other sites, such as enterocytes) and are responsible for the metabolism of the majority of currently used antiarrhythmic drugs, including quinidine, disopyramide, propafenone, and dofetilide. Individuals totally lacking CYP3A activity have not been described, but there is substantial interindividual variability in the activity. However, CYP3A activity can be nearly totally inhibited by concomitant drug therapy, notably with certain azole antifungals (ketoconazole), macrolide antibiotics (erythromycin), HIV protease inhibitors (ritonavir), amiodarone, diltiazem, verapamil, and large doses of grapefruit juice. CYP3A activity can also be induced by rifampin, phenytoin, and Saint John’s wort; reduction in plasma concentrations and loss of drug effects can occur under these conditions. Inhibition of CYP3A-mediated elimination by these drug interactions was the major cause of terfenadine accumulation in plasma, leading to cases of torsades de pointes that eventually prompted the drug’s withdrawal from the market.

N-acetyltransferase (NAT) activity is responsible for the elimination of procainamide. There are two NAT genes: NAT1 and NAT2. NAT1 is expressed in all individuals, but loss of functional alleles has been reported in NAT2. As a result, patients can be divided into rapid and slow acetylators. Slow acetylators have a higher incidence of procainamide-induced lupus syndrome during chronic therapy.¹⁸

In addition to metabolizing enzymes, transport across biological membranes is another crucial determinant of drug disposition that has received increasing attention over the last few years. The multi–substrate efflux carrier P-glycoprotein is the most well-studied to date in terms of cardiovascular drugs. P-glycoprotein is the product of the multi-drug resistance 1 (MDR1; ABCB1) gene, and is a member of the ABCB subfamily of ABC (ATP-binding cassette) transmembrane proteins, which have been implicated with drug resistance in cancer. P-glycoprotein is expressed in drug-resistant cancer cells and in many organs such as the gut and kidney, where it has an important role in distribution and elimination. It acts as an efflux pump and in the gut, preventing the entry of toxic compounds, whereas in the liver and kidney it serves to remove xenobiotics from the circulation. Although P-glycoprotein substrates are diverse, there is considerable overlap between the substrates that are transported by P-glycoprotein and those metabolized by CYP3A4/5. One important role for P-glycoprotein in cardiovascular medicine is that it transports cardiac glycosides. A synonymous single-nucleotide polymorphism in exon 26 (C3435T) in the MDR1 gene has been associated with expression of the transporter and variable digoxin concentration.¹⁹ Furthermore, correlation of the MDR1 genotype and digoxin uptake in vivo has been described.²⁰ Many drugs inhibit P-glycoprotein activity, and their use with digoxin can lead to increased digoxin plasma concentrations and toxicity; amiodarone, quinidine, verapamil, itraconazole, cyclosporine, and erythromycin are examples.²¹

Pharmacodynamic Principles

Experimental studies before the cloning era indicated that ion channels, the targets of antiarrhythmic drugs, have specific drug binding sites. Furthermore, drug binding to these “receptor” sites on the channels to modify their function was modulated by the state of the channel (open, closed, or inactivated). These observations led to the formulation of the modulated receptor hypothesis to analyze drug-channel interactions. Recent studies have demonstrated the existence of specific drug-binding sites on ion channel proteins, and the interaction of drugs with these sites can be modulated by channel protein conformation (or “state”). In some cases, ion channels are blocked by direct binding in the pore region (a common mechanism for most I_Kr blockers), whereas in others, drug binding to other regions of the protein alters its function, which is an allosteric effect.

Most sodium channel–blocking drugs inhibit open or inactivated states of the channel; therefore, during each cardiac cycle, they associate with and block channels and then dissociate during diastole, ultimately reaching a steady state level of block. If the heart rate is increased, there is less time for dissociation, so that channel block is enhanced. Furthermore, the rate of dissociation from the channel varies among drugs. For drugs like lidocaine with “fast-off” kinetics, very little block occurs even at fast rates. In contrast, for “slow-off” drugs like flecainide, block accumulates even at physiologic rates. As sodium channel block slows intraventricular conduction, it prolongs QRS duration on the surface electrocardiogram (ECG). This result explains why QRS widening is apparent at normal heart rates during flecainide (but not lidocaine) therapy, and it widens further if the heart rate is increased. In addition, conduction slowing promotes reentry; therefore, slowing conduction by the drug at faster rates might explain cases of flecainide-induced VT during exercise.

Mechanism-Based Approach to Antiarrhythmic Drugs

One universal principle of pharmacologic therapy is that the best treatment is that targeted specifically to disease mechanisms. As the understanding of the molecular and cellular basis of arrhythmias has evolved, so too has the list of arrhythmias for which specific mechanisms have been defined, and therefore specific antiarrhythmic drug therapies may be indicated (Table 111-2). However, it is the recent advances in the understanding of molecular mechanisms in specific, rare, inherited monogenic arrhythmia syndromes that has provided insight into common arrhythmias, which might also prompt consideration of specific mechanism-based therapies.

The inherited arrhythmia syndrome catecholaminergic polymorphic ventricular tachycardia (CPVT) provides a good contemporary example of how a mechanism-based approach for drug therapy can be used. This syndrome is characterized by VT induced by adrenergic stress in the absence of structural heart disease and a high incidence of SCD. Mutations in two causative genes—RYR2, encoding the cardiac ryanodine receptor (RyR2) Ca²⁺ release channel, and CASQ2, encoding cardiac calsequestrin—have been associated with CPVT.²² Mutations in these two genes destabilize the RyR2 Ca²⁺ release channel complex in sarcoplasmic reticulum and result in spontaneous Ca²⁺ release through RyR2 channels leading to delayed afterdepolarizations (DADs), triggered activity, and bidirectional/polymorphic VT.²³ Although ICDs are recommended for the prevention of SCD in patients with CPVT,²⁴ painful shocks can trigger further adrenergic stress and arrhythmias; deaths have occurred despite appropriate ICD shocks. Treatment with β-adrenergic blockers has been shown to reduce arrhythmia burden and prevent SCD, but it is not completely effective.²⁵ However, because Ca²⁺ leakage through ryanodine channel is a common mechanism of CPVT, ryanodine channel block can have a therapeutic effect. This effect led Watanabe et al.²⁶ to assess whether direct inhibition of RyR2 channels by flecainide could be used to treat CPVT. They discovered that flecainide directly inhibits RyR2 channels and suppresses DADs and triggered activity in Casq2 null cardiomyocytes.^26,27 Sodium channel block by flecainide can also be effective in preventing CPVT. Flecainide prevents spontaneous VT and inducible VT by exercise or isoproterenol in vivo. Based on these experimental findings, the effects of flecainide in two patients carrying a CPVT-linked mutation in either RYR2 or CASQ2 were examined, and it was found that flecainide suppressed exercise-induced arrhythmias and recurrences of VT/VF.²⁸

Unfortunately, for most common arrhythmias, such as VT associated with myocardial disease or AF, arrhythmia mechanisms have been difficult to define; therefore, the choice of drug therapy continues to remain largely empiric. In such settings, drugs shown by clinical experience or controlled trials to be effective often target multiple mechanisms, including modifying the arrhythmia-prone substrate or inhibiting known or putative triggers of arrhythmias. Because specific mechanisms for many arrhythmias have not been defined, the expectation that all patients with AF or VT will respond to therapy in a similar fashion makes the assumption that underlying mechanisms are homogeneous across patients. However, as genetic, molecular, and cellular studies continue to demonstrate, this assumption is largely unfounded. Therefore, the incomplete efficacy of drugs in treating these arrhythmias, which appear to represent a spectrum of arrhythmia mechanisms, is not unexpected. Large clinical trials provide the best evidence for choosing among drug therapies and dosages in settings like AF and VT. The use of evidence-based principles should be complemented by consideration of patient-specific characteristics that might make one drug more or less desirable than another.

Proarrhythmia Syndromes

Proarrhythmia represents an extreme example of the phenomenon that drug effects vary widely among individuals. The recognition that drugs designed to suppress arrhythmias can, in some patients, actually increase arrhythmias or provoke new ones is probably the most important factor governing the selection and use of antiarrhythmic drugs. Insight into mechanisms leading to proarrhythmia has had important implications for understanding arrhythmogenesis, rational use of antiarrhythmic therapies, selection of patients for specific therapies, and drug development. Furthermore, because proarrhythmia often seems to develop in the absence of clear risk factors, a role for genetics in predisposing to this adverse drug reaction is increasingly being appreciated.²⁹ Proarrhythmia is defined as the generation of new or worsened arrhythmias with drug therapy, but it might not be readily apparent that a drug is, in fact, responsible for an arrhythmia exacerbation. This is a particular problem in patients with advanced heart failure and in whom the spontaneous development of frequent, serious arrhythmias is common even in the absence of drugs. The following sections review three well-recognized proarrhythmia syndromes: digoxin toxicity, drug-induced torsades de pointes, and toxicity related to sodium channel block. These syndromes serve to illustrate gene-drug interactions potentially mediating proarrhythmia risk (Table 111-3).

Drug-Induced Torsades de Pointes

The examples of high-risk pharmacokinetics described in this chapter have in common drugs whose elimination depends on a single pathway: propafenone, sotalol, dofetilide. In each case, risk is conferred by absence of redundancy of drug-elimination pathways for that specific drug. The example of drug-induced torsades de pointes is similar; cardiac repolarization is ordinarily a highly redundant process and can be accomplished by I_Ks, I_Kr, and probably other mechanisms. Subjects who exhibit marked QT prolongation with I_Kr-blocking drugs must therefore represent a subset in whom repolarization is highly I_Kr dependent and thus have lost redundancy in repolarization mechanisms (Figures 111-1, 111-2). This phenomenon has been referred to as “reduced repolarization reserve”^16,34 (Figure 111-3) and suggests that multiple often redundant mechanisms maintain normal repolarization, so that minor alterations in function might not be obvious at baseline. For example, a minor reduction in repolarizing current (e.g., owing to a genetic lesion resulting in a small reduction in I_Kr) might be without consequence because other mechanisms help to maintain a near-normal QT interval; however, such a reduced reserve can become obvious when further stressors to repolarization are superimposed, such as drug challenge, bradycardia, or hypokalemia.

Figure 111-1 Continuous recording from a patient who recently started receiving sotalol. During atrial fibrillation, there is irregularity of the ventricular response creating frequent short-long-short cycles, but there is minimal change in QT intervals (top panel). After DC-cardioversion, the QT interval has increased dramatically to 0.64 s (middle panel), and an episode of torsades de pointes is triggered (bottom panel). (Reprinted with permission from Darbar D, Kimbrough J, Jawaid A, et al: Persistent atrial fibrillation is associated with reduced risk of Torsades de Pointes in patients with drug-induced long QT syndrome. J Am Coll Cardiol 51:836–842, 2008.)

Figure 111-2 Mechanisms of proarrhythmia associated with HERG blockade. Blockade of the HERG channel produces prolongation of the QT interval (blue) and generates an early afterdepolarization (red) in the cardiac action potential. These changes, which are heterogeneous across the ventricular wall, create a substrate for torsades de pointes. In this example, torsades de pointes degenerates into ventricular fibrillation. (Adapted with permission from Roden DM, Viswanathan PC: J Clin Invest 115(8):2025–2032, 2005.)

Figure 111-3 The concept of reduced repolarization reserve. Patient 1 is prescribed a QT-prolonging drug and has minimal or no prolongation of the QT interval. In contrast, patient 2 was given the same QT-prolonging drug at the same dosage and developed marked prolongation of the QT interval; this has been termed reduced repolarization reserve and has been discussed further in the text. Repolarization is a physiologically redundant process such that I_Kr block will not result in markedly prolonged repolarization—that is, the system displays some “reserve.” However, otherwise subclinical lesions in other components of the repolarization system, such as reduction of I_Ks owing to genetic factors, hypokalemia can become apparent as marked QT prolongation when I_Kr is reduced. (Reprinted with permission from Kannankeril et al., 2010.)

Other risk factors for torsades de pointes appear pharmacodynamic in nature: female sex, advanced heart disease or left ventricular hypertrophy, bradyarrhythmias, hypokalemia, and recent conversion from AF.¹⁶ Each of these can be interpreted in the context of reduced repolarization reserve, and underlying molecular mechanisms have been proposed in some cases (Table 111-4). Thus, hypokalemia can reduce I_Kr

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Genetic Testing
2. Supraventricular Arrhythmias in a Realistic 3D Model of the Human Atria
3. Implantable Cardioverter Defibrillator: Clinical Aspects
4. Epigenetics in Cardiac Rhythm Diseases

Stay updated, free articles. Join our Telegram channel

Join