The present and future of antiarrhythmic drugs

Classification of AADs and general principles

The saga of antiarrhythmic drugs (AADs) has just exceeded a century. Antiarrhythmic pharmacotherapy developed gradually and was marked by empiricism for the first fifty years. Then, the Singh–Vaughan–Williams (SVW) classification organized available knowledge in a convenient, yet fundamentally flawed format, because it prioritized electrophysiologic drug properties over the underlying arrhythmogenic mechanisms, which were incompletely understood and widely contested. ¹ Yet paradoxically, the same classification that initially catalyzed AAD development, later constrained it. ² Such a reductionist approach yielded limited success and resulted in the discontinuation of many drugs during phase II or III trials (e.g., the Cardiac Arrhythmia Suppression Trial—CAST).^3,4

In 1991, the ESC Task Force on Arrhythmias made a strategic opening move to counter the SVW paradigm by issuing an operational framework called the Sicilian Gambit. ⁵ It was meant as a functional algorithm that allowed clinicians to choose the appropriate AAD based on the most pharmacologically accessible determinants of each arrhythmia, also called vulnerable parameters.^1,5 Despite its conceptual merit, the Sicilian Gambit gained limited traction, likely due to entrenched clinical inertia and greater complexity. Moreover, recent updates by Lei and colleagues ⁶ and the EHRA practical compendium ⁷ extended the SVW framework to accommodate more recently developed or repurposed molecules, thus perpetuating an obsolete concept. The classification retains a broad appeal due to familiarity and simplicity, but falls short in terms of practical clinical applicability.

The stalled pharmacological progress and overall inertia become even more evident as the global burden of arrhythmias continues to escalate. From 1990 to 2021, the absolute number of atrial fibrillation (AF) and atrial flutter (AFL) incident cases more than doubled and mortality nearly tripled, and this trend is expected to continue throughout the following decade. ⁸ Population growth and demographic aging contribute to this epidemic, rather than disease acceleration. ⁸ Despite its superiority over AADs in reducing symptom burden and arrhythmia recurrence, catheter ablation reaches only a relatively small fraction of eligible AF or AFL patients, even in high-income countries. ⁹ Rhythm control strategies are clinically beneficial and remain largely dependent on AADs even in the ablation era, despite the fact that current agents carry significant proarrhythmic risks. ¹⁰ A recent Cochrane network meta-analysis reconfirms the efficacy of Na⁺ channel blockers for AF conversion, which is, however, inferior to that of electrical cardioversion and has a less favorable safety profile in patients with structural heart disease. ¹¹

Notably, ventricular tachyarrhythmias (VT) account for more than half of sudden cardiac deaths, responsible for a significant share of global cardiovascular mortality. ¹² Clinical outcomes vary substantially across regions, reflecting disparities in access, infrastructure, and resources. ¹³ Besides beta-blockers and amiodarone, no currently available AAD has demonstrated a reliable mortality reduction in ventricular arrhythmia management. ¹⁴ Due to a narrow therapeutic window, even modest fluctuations in drug concentration (e.g., generic formulations) can increase proarrhythmic risk or reduce treatment efficacy. AADs continue to be a valuable therapy in conjunction with ICD implantation and/or ablation therapy of VT (hybrid therapy).

There is a persistent unmet need for safer and more targeted pharmacological agents. Additional challenges include patient personalized therapy, including selection and treatment initiation conditions, but equally critical, recognizing the limits of AADs and knowing when to switch to nonpharmacologic strategies or consider interventional electrophysiology as a first-line therapy.^15,16

The present of “classical” AADs

Arrhythmogenic mechanisms, vulnerable parameter identification, and drug characteristics collectively determine the efficacy of the antiarrhythmic pharmacotherapy, but the underlying substrate is the main determinant of safety. The concept of substrate refers to structural, functional, and genetic components. Each can independently or synergically influence proarrhythmic risk. Clinicians need to simultaneously integrate these elements, especially because in the post-CAST era, safety is considered even more important than efficacy. Given the limited available pharmacological armamentarium, the role of each AAD is constantly reassessed, tailoring efficacy and safety profiles according to arrhythmia and substrate.

Inactivated state Nav1.5 inhibitors with rapid dissociation have an increased efficiency in tachyarrhythmias with short cycle lengths, such as ventricular arrhythmias, particularly in partially depolarized myocardium. ¹⁷ The pharmacological profile of lidocaine supports its lasting role for terminating ventricular arrhythmias generated on ischemic substrate. ¹⁸ During cardiopulmonary resuscitation in patients with pulseless VT or VF refractory to electrical shock, guidelines recommend antiarrhythmic treatment with either amiodarone or lidocaine, and a compelling case can be made for either agent.^19–21 Although unpopular, mexiletine remains the only drug from its subclass that is available for chronic outpatient use. It becomes a practical option for patients who experience recurrent ventricular tachyarrhythmias despite ablation or antiarrhythmic treatment with other agents. ⁷ Off-label, it can shorten the QT interval in congenital or acquired long QT syndromes, particularly when bradycardia complicates the use of beta-blockers. ²²

Some agents that also have Na⁺ channel blocking properties were initially developed for non-arrhythmic indications. Ranolazine has an attractively balanced multichannel activity, as it targets I_Na,L and I_Kr, but does not negatively impact heart rate or hemodynamics.^23,24 It offers significant protection against AF incidence in patients with chronic coronary syndrome, and even enhances their activity performance.^23,24 Also, ranolazine is currently investigated alongside other drugs for their RyR2 blocking properties. ²⁵

I_f blockade was developed to lower the sinus rate without a negative impact on inotropy. Besides the inhibition of HCN channel, at therapeutic concentrations, ivabradine also inhibits the Kv11.1 channel responsible for I_Kr, which raises some concern when combined with drugs that prolong the QT interval. ²⁶ Despite inferior efficacy compared to digoxin, ivabradine showed modest potential for rate control in patients with AF, a role that might be supported by future evidence. ²⁷ Additional research is also needed to establish the therapeutic value of ivabradine in postural orthostatic tachycardia syndrome. ²⁸

Digoxin has been an intuitive choice for rate control in AF and AFL, especially given the overlap between atrial arrhythmias and heart failure with reduced ejection fraction, as it can also improve myocardial contractility. Its safety profile remains a subject of ongoing debate, and a recent target emulation trial reported higher all-cause mortality and cardiovascular mortality in patients with AF and HF treated with digoxin compared to beta-blockers. ²⁹ However, a recent RCT positively repositioned digitoxin on the safety scale. ³⁰

Selective K⁺ channel blockade has been associated with significant pro-arrhythmic risks, which led to the discontinuation of drugs such as almokalant, sematilide, or d-sotalol after the SWORD trial.^31,32 For dofetilide, despite a highly selective I_Kr blockade, the torsadogenic risk can be mitigated through close monitoring and individualized dose adjustment. ⁷ For the moment, its role remains limited to atrial arrhythmias.^33,34 Less channel selective, ibutilide is successfully used for the pharmacological cardioversion of AF and AFL, but requires continuous monitoring of the QT interval for several hours after the procedure. ⁷

A concomitant multi-channel blockade favorably balances efficacy and safety, even though numerous past attempts to develop such drugs failed due to bioavailability issues or non-cardiovascular adverse effects. ³² However, there are several agents that concomitantly block multiple ion channels. Vernakalant is used mainly in the emergency departments for its unique rapid action and efficacy in converting recent-onset AF to sinus rhythm.^35,36 Despite occasional concerns that it might cause heart failure or malignant ventricular arrhythmias, available evidence supports a reliable safety profile compared to other AADs (except hypotensive patients).^11,37 Although still preferred in some settings, sotalol should be used mainly for ischemic patients with normal ejection fraction. Sotalol is proarrhythmic even in sinus rhythm and in the presence of a normal cardiac substrate. Therefore, the continuous QT monitoring after initiating sotalol therapy is mandatory. ³⁸

Amiodarone is often perceived as versatile and safe, yet it causes thyroid, hepatic, pulmonary, neurological, ocular, and dermatological toxicity, among others. ³⁹ Despite its use for several decades, the mechanisms of serious side-effects such as pulmonary fibrosis have only recently been elucidated. ³⁹ Toxicity induced by amiodarone is cumulative and carries significant prognostic implications. Thus, it is unsuitable as a long-term strategy. Some indications for amiodarone need to be reconsidered. Such is Chagas cardiomyopathy, highly prevalent in countries such as Brazil, which carries a significant risk for sudden cardiac death predominantly due to ventricular fibrillation. Although amiodarone effectively reduces ventricular arrhythmia burden in these patients, it does not offer sufficient protection against sudden cardiac death, and ICD implantation should be considered as a complementary therapy. ⁴⁰

Anticoagulants and AADs are routinely prescribed together in AF. Yet this drug combination has raised concern for excess bleeding, mostly regarding the concomitant use of amiodarone and DOACs.^41,42 An increased risk for major bleeding emerged in a meta-analysis based mainly on retrospective studies, but was not confirmed in a more recent one, which also included post hoc analyses from the pivotal DOAC trials ARISTOTLE, ROCKET AF, and ENGAGE AF TIMI 48.^43,44 This concern extends to other AADs as well, although combinations with DOACs appear safer than with vitamin K antagonists.^45,46

Until novel agents emerge, there is an increased interest in combinations of AADs, but these can exponentially increase proarrhythmic risk. ⁴⁷ In silico models have a high potential to contribute toward the development of safe and synergic AAD combinations, but translation to clinical practice requires rigorous validation. ⁴⁸ Early clinical data from a small cohort of patients with AF suggest favorable safety outcomes over a mean follow-up of one year, with combinations such as flecainide-amiodarone or flecainide-dronedarone. ⁴⁹ Another rational approach involves pairing a Na⁺ channel blocker with a K⁺ channel blocker in order to counterbalance their respective adverse effects. ⁵⁰ Flecainide-ibutilide could be effective, without more significant QT interval prolongation and less torsadogenic risk.^51,52 However, routine prescription of such combinations is strongly discouraged pending robust safety validations.

Historically, proarrhythmic risk was often identified only after prolonged clinical exposure, at the cost of patient safety, which led to reputational and financial liabilities. Current testing frameworks, such as the Comprehensive in vitro Proarrhythmia Assay (CiPA), ⁵³ make proarrhythmic risk assessment more reliable. ⁵⁴ Nowadays, computational tools integrate safety screening directly into the drug research process. ⁵⁴ These developments will hopefully serve as catalysts for further progress.

Resurrection of some classical AADs

In 1997, amiodarone and sotalol were considered the best antiarrhythmic drugs. ⁵⁵ Unfortunately, three decades later, amiodarone is still regarded by many physicians as a panacea for arrhythmias, possibly due to risk aversion and defensive prescribing. Such perspectives should be left behind, allowing evidence to restore confidence in other AADs. This is the case with some Na⁺ channel blockers and dronedarone. The CAST population no longer reflects the contemporary arrhythmia patient. As clinical profiles evolve, a periodic reassessment of historical evidence is not only justified, but necessary. ²

Flecainide maintains an important role in the pharmacologic cardioversion of AF.^7,37 There is an increasing body of evidence to also support its safety in treating AF even in patients with nonobstructive coronary disease ⁵⁶ or other profiles of structural heart disease, such as arrhythmogenic cardiomyopathy and cardiomyopathy induced by premature ventricular contractions (PVCs).^57,58 Flecainide addition to beta-blocker therapy in catecholaminergic polymorphic ventricular tachycardia significantly reduces arrhythmic events. ⁵⁹ Additionally, its pediatric application for PVC treatment is supported by demonstrated efficacy compared to metoprolol. ⁶⁰ The most recent study with orally inhaled flecainide, with the aim to reduce the onset of action, reflects broader efforts to reformulate legacy molecules, but was stopped in evolution due to less promising results than anticipated. ⁶¹ Some drug reformulations may offer clinical advantages.

Oral propafenone is just as effective as intravenous amiodarone in converting AF to sinus rhythm, but has a faster onset of action and a more practical administration route.^11,62 For rhythm control in AF, propafenone is comparable to amiodarone, including for patients with heart failure, although until further research, there are insufficient data on risk stratification according to ejection fraction. ⁶³ Similar results were also seen in patients with supraventricular arrhythmias and septic shock.^64,65

The declining preference for amiodarone has been accompanied by a repositioning of dronedarone in recent years. The unfavorable design of the PALLAS trial contributed to a disproportionate rejection of dronedarone, despite the consistent benefits demonstrated by the previous ATHENA trial. ² Supported by recent evidence, confidence in this drug is steadily recovering, regaining its former momentum. ² New data support its safe use in non-permanent AF for an early rhythm control strategy. ⁶⁶ In AF, evidence seems to support dronedarone over sotalol, with a lower proarrhythmic effect in terms of both ventricular tachyarrhythmias and severe bradyarrhythmias. ⁶⁷ The drug is also very effective for rate strategy in AF.

Quinidine and hydroquinidine are also among the resurrected legacy agents. However, their current indications are limited to Brugada syndrome and potentially short QT syndrome, mainly due to proarrhythmic risk, low tolerability, and availability. ⁷

New directions for development

Novel molecules emerge through the pipeline at various development stages, though only a fraction of initial candidates advance beyond early phases.^68–70 The scarcity of novel AADs is driven by poor financial incentives, industry inertia, and lack of prioritization.^2,71 Even when available, new drugs face substantial delays before reaching routine clinical practice. ⁷¹

Some of the new drugs are already on the market. Etripamil is a short-acting L-type Ca²⁺ channel blocker, recently approved by the US FDA for terminating episodes of paroxysmal supraventricular tachycardia via intranasal insufflation.^72–74 It enables patients to conveniently and promptly benefit from self-administered treatment in the outpatient setting. Its approval facilitates other ongoing investigations of etripamil for acute rate control in AF with rapid ventricular response. ⁷⁵ Landiolol is an ultra-short-acting and ultra-specific β1-blocker (superior to esmolol) that has also been recently approved for rapidly reducing ventricular rate in supraventricular arrhythmias such as AF or AFL. ⁷⁶ Since 2019, it has been a pharmacological alternative for treating VT or VF in patients who do not respond to other AADs, such as amiodarone, with a good safety profile, including for HF patients. ⁷⁷

The modern management of AF focuses on triggers and re-entry. ⁷⁸ Triggers may be responsible for inducing arrhythmia, but perpetuation also relies heavily on the substrate, an umbrella term that integrates form and function.^79,80 The success of antiarrhythmic therapy is also critically dependent on timing. Interventions might become less impactful once substrate remodeling evolves.

Triggered activity in AF is often driven by Ca²⁺ overload in the sarcoplasmic reticulum, which causes abnormal Ca²⁺ leak through RyR2 channels. ⁸¹ This in turn activates the sodium-calcium exchanger (NCX) and generates delayed afterdepolarizations that can initiate ectopic beats. CaMKII further amplifies this process through RyR2 hyperphosphorylation. Consequently, RyR2 inhibition or stabilization and CaMKII inhibition represent mechanistically rational strategies to suppress triggered activity in AF. ⁸¹ Ongoing studies investigate whether some repurposed or novel agents could perform well as antiarrhythmics using these targets. ²⁵ Currently, some available drugs, such as flecainide, propafenone, ranolazine, carvedilol, or dantrolene, are evaluated for their ability to also inhibit the RyR2 channel. ⁸² CaMKII inhibition is still at a very early stage because of uncertain effects on the nervous system and reproduction. ⁸³

Multiple K⁺ channel subtypes contribute to atrial repolarization imbalances that sustain re-entry. Small-conductance Ca²⁺-activated K⁺ channels, also known as KCa₂ or SK channels, are upregulated in AF and prolong the action potential duration and effective refractory period. ⁸⁴ Because they contribute to AF stabilization, SK channels remain among the leading targets that are actively pursued for AAD development. ⁷⁰ Preliminary data with intravenous AP30663 show encouraging results in restoring sinus rhythm in patients with recent-onset AF. ⁸⁵ The END-AF-1 phase 2 trial investigating the oral compound AP31969 is ongoing with active patient enrollment (NCT07267949).

Sulcardine sulfate is a new hERG channel inhibitor that could offer superior efficacy and torsade protection due to a balanced multiple ion channel blockade that involves the I_Kr, I_Na, and I_Ca,L channels. ⁶⁸ Its favorable pharmacokinetic profile, which allows rapid onset and elimination, makes it suitable for cardioversion. Results with the intravenous drug HBI-3000 are expected from the ongoing trial NCT04680026, while its oral equivalent HBI-3020 is still evaluated in the pre-clinical phase.

Supported by pathological pathways and atrial specificity, the respiratory stimulant doxapram is evaluated and could be repurposed for AF cardioversion due to its ability to inhibit the two-pore-domain K⁺ channel TASK-1 (K2P3.1). These channels are predominant in the atrial myocardium, where they shorten the action potential and might facilitate AF. ⁸⁶ Good initial results on porcine models offered support for the DOCTOS trial, which will provide reliable evidence. ⁸⁶

Budiodarone is a non-iodinated amiodarone analog developed to retain the properties of multiple channel inhibition and avoid long-term tissue accumulation through a rapid esterase-mediated metabolism. Early randomized data in paroxysmal AF demonstrated significant reductions in AF burden and were followed by defined endpoints for a phase 3 study that integrates continuous rhythm monitoring using wearables.^87,88

Substrate targeting is a very prolific area of research. Emerging evidence regarding risk factors and mechanistic mediators will provide novel therapeutic targets and significantly change the future management of AF and other arrhythmias.^10,89 Atrial cardiomyopathy, characterized by electrical, functional, structural, and prothrombotic remodeling, is a concept that bridges substrate and atrial arrhythmias. Its definition, markers, risk factors, and their demonstrated associations with HF, atrial arrhythmias, stroke, and vascular dementia are already well-documented.^90–92

Substrate is also particularly important in ventricular arrhythmias, and is critical not only for the AAD selection, but it can also guide the strategy for catheter ablation. ⁹³ Whether ventricular arrhythmias emerge secondary to ischemia, structural heart disease, congenital syndromes, or extracardiac causes will determine the adequate AAD selection.^94,95

Antifibrotic therapy is the least developed area in drug research, despite its potential to directly influence arrhythmogenesis. THRV-1268, an oral inhibitor of serum- and glucocorticoid-regulated kinase 1 (SGK1), targets upstream pathways for pro-fibrotic and pro-inflammatory signaling involved in atrial remodeling. ⁹⁶ In preclinical AF models, SGK1 inhibition reduced atrial fibrosis and AF inducibility, and a first phase 1 study of THRV-1268 in humans has been completed. ⁹⁶ Further research is needed to define the extent of its antifibrotic and antiarrhythmic properties, metabolic off-target effects, and long-term safety. Only time will tell which molecular targets yield the safest and most effective results. ⁸³

Shifting the focus toward a timely detection of structural heart disease could lead to arrhythmia prevention using upstream therapy, which exerts secondary antiarrhythmic properties by modulating the underlying substrate. For this reason, RAAS inhibitors, mineralocorticoid receptor inhibitors, SGLT2 inhibitors, or statins are now part of the updated classification of AADs, but their exact role and contribution are still evaluated. ⁷ Other drugs that are currently studied for their secondary role as down- or upstream agents in antiarrhythmic therapy are canakinumab, colchicine, rilonacept, anakinra, etanercept, metformin, botulinum toxin type A, while others are still under development.^25,68

The future of antiarrhythmic therapy will rely on the accurate endotyping of patients. Artificial intelligence and in silico myocardial reconstructions, culminating in digital twins, will play an important role in a more individualized antiarrhythmic pharmacotherapy, based on reliable patient information and constant feedback on safety using wearables, which are convenient and increasingly accurate. ⁹⁷

Conclusion

AADs represent one of the greatest challenges—and, at the same time, one of the most persistent frustrations—in the evolution of drug molecules. Arrhythmias are not only extremely prevalent; they also constitute a considerable health, social, and financial burden due to their chronic, symptomatic, or potentially malignant nature. Nevertheless, the development of new AADs has lagged significantly behind other therapeutic fields. Despite the warnings raised by the Cardiac Arrhythmia Suppression Trial (CAST) and by the “Sicilian Gambit” approach, a substantial gap persists between practical clinical needs and the availability of current effective AADs.

The rapid development of interventional electrophysiology has, to some extent, acted as a brake on the advancement of pharmacological therapies for arrhythmias, even though ablation techniques and implantable devices cannot fully address the demands of clinical practice, neither in terms of patient volume nor in terms of the complexity of arrhythmic substrates. Therefore, pharmacological therapy remains the first-line approach in the management of arrhythmias.

However, as Milton Packer noted, “Discouraged by payers and plagued by clinical inertia and public neglect, physicians who treat cardiovascular diseases often wait years before prescribing a new agent.” ⁷¹

Current progress and near-future perspectives in antiarrhythmic pharmacotherapy follow three main directions. First, the repositioning of classical AADs according to specific arrhythmic endotypes, in line with contemporary advances in the mechanistic understanding of arrhythmias. Second, the reevaluation of established antiarrhythmic agents, particularly with respect to safety, such as flecainide and dronedarone. Finally, novel agents such as vernakalant have entered clinical practice, while the search for new ones continues.

The rapid expansion of knowledge regarding novel pathophysiological mechanisms, paralleled by advances in basic research and early-phase (Phase I and II) clinical studies, has fostered the development of highly specific molecular targets, an evolution further accelerated by artificial intelligence-driven drug discovery. These developments are expected to bridge the gap left by the stagnation of the late twentieth and early twenty-first centuries. However, the pharmaceutical industry will not invest in innovation without assurance of return on investment, and such assurance must ultimately be supported by increased demand arising from everyday clinical practice.